Proceedings of 2002 General Practice and Primary Health Care Research Conference:

Proceedings of 2002 General Practice and Primary Health Care Research Conference:

Proceedings of

2002 General Practice and Primary Health Care

Research Conference:

Research — making a difference to health and health care

29 to 31 May 2002

Prepared by the Primary Health Care Research and Information Service

September 2002

© Primary Health Care Research and Information Service 2002

This work is copyright apart from any use as permitted under the Copyright Act 1968. No part may be reproduced by any process without prior written permission.

ISBN 0 9578862 7 6

The views expressed in these proceedings are those of the authors and not necessarily those of the Primary Health Care Research and Information Service or the Commonwealth

Department of Health and Ageing.

Published by the Primary Health Care Research and Information Service, Department of

General Practice, Flinders University, Adelaide

Printed by Flinders Press

September 2002

Contents

X Introduction

Libby Kalucy

Professor Louis Pilotto

X Acknowledgments

X Conference Advisory Committee membership

X Peer review of conference abstracts

1

SECTION 1 PRE-CONFERENCE WORKSHOPS....................................................................7

X New methods to discover best practices in health care

X Bridging the gap: communication between researchers and policy makers

X Issues in measuring health inequalities, including social capital perspectives

X Engaging the grassroots: making research networks a reality

X Evaluation report for the pre-conference workshops

9

11

12

13

14

3

4

5

SECTION 2 CONFERENCE PROCEEDINGS .......................................................................19

Opening session.............................................................................................................21

X Opening address

Senator the Hon. Kay Patterson

21

Keynote papers ..............................................................................................................24

X Health services research in primary care: influencing health policy, service development and professional practice

Professor David Wilkin

X Using research to inform and change primary care

Professor James Dunbar

X Clinical practice improvement: a scientific methodology to discover best medical practice

Dr Susan Horn

X More than just political correctness: making consumer participation in primary health care research a productive reality

Hilda Bastian

24

28

35

45

Plenary paper abstracts .................................................................................................50

50

X Designing research to maximise impact

Professor Paul Glasziou

X How can research turn clinical general practice questions into clinical practice?

Professor Chris Del Mar

X Researching new models of care: the impact and economics of the Glasgow Heart

Failure Nurse Liaison Service

Assoc Professor Simon Stewart

51

57

2002 General Practice and Primary Health Care Research Conference — Proceedings

iii

X Divisions and researchers: developing viable and sustainable partnerships for quality systems of care in general practice

Dr Peter Del Fante

X Collaborations for relevance, depth and insight: meeting the challenge of different language and perspectives between research disciplines

Dr Rosemary Aldrich

65

70

X Challenges of community-based interventions to improve nutritional status of

Indigenous people in remote communities

Professor Kerin O’Dea

72

X Too risky not to rethink: research and policy

Professor Mark Harris

81

Summary session...........................................................................................................86

X Moving PHC research forward — summary

Libby Kalucy

86

Concurrent paper abstracts............................................................................................88

Concurrent poster abstracts.........................................................................................128

Breakfast sessions .......................................................................................................195

195

X Joining the IMET (n=1 trials) program: a new program of research needs collaboration

X Are health inequalities widening or narrowing across the SES gradient? Current issues in health inequalities and implications for a national research agenda 196

Conference evaluation report.......................................................................................198

INDEX OF PRESENTERS................................................................................................207

iv

2002 General Practice and Primary Health Care Research Conference — Proceedings

Introduction

Libby Kalucy

Co-Director & Senior Research Fellow, PHC RIS

Professor Louis Pilotto

Co-Director, PHC RIS

Professor and Head, Department of General Practice, Flinders University

The Minister for Health and Ageing, Senator the Hon. Kay Patterson opened the GP & PHC

Research Conference 2002 by remarking on the importance of this conference in bringing together people from health professions, research, policy makers and consumer organisations, to exchange views, present their projects, and extend their network of contacts. The presence of these diverse groups was particularly relevant as the theme of this year’s conference, ‘Research: Making a difference to health and health care’ reflected world-wide interest in bridging research, practice and policy, in order to make a difference to health care and to the health of the public.

Conference presenters were invited to address four key areas: identifying research questions relevant to consumers, practitioners, managers and policy makers; selecting rigorous methodologies appropriate to purpose and setting; developing viable partnerships between researchers and other groups; and building the capacity of the research workforce to do research and others to use research findings. These areas were addressed in various combinations. Some particularly interesting papers illustrated the way indigenous research has matured through meaningful partnerships between researchers and communities, selection of appropriate methodologies, and selection of relevant issues.

We were fortunate in attracting to this conference two distinguished international speakers, David

Wilkin from UK and Susan Horn from USA, who presented stimulating pre-conference workshops, memorable keynote addresses, and were available throughout the conference for informal discussions with delegates. David Wilkin spoke on partnerships between policy and research, and the need for multiple research disciplines to address enduring research questions. Susan Horn stimulated the large audience at her full day workshop, and a different audience during the conference with her accounts of the results of large-scale observational research. The common point between the two speakers was a realistic approach — developing rigorous evidence to link outcomes, processes and patient/group characteristics.

These, and the other plenary speakers who live in Australia and have international reputations, generated lively discussion about selection of rigorous qualitative and quantitative methodologies, for different purposes and settings. James Dunbar presented his experiences in conducting research to change practice with practising GPs in Australia and UK. Hilda Bastian used some wonderful cartoons to illustrate aspects of consumer participation in research, especially the contribution of consumers as new partners in research to help researchers start to look at old things in a new way. The impact of rigorous economic evaluation of outreach cardiac nurses in

Scotland was well illustrated by Simon Stewart’s excellent presentation, which also provided a refreshing nursing perspective. Prof Kerin O’Dea (who appeared by videolink from Darwin to the relief of all those involved in organising this) gave a vividly illustrated account of the barriers to doing research and to changing health and health care in an indigenous community in the

Northern Territory.

One of the enduring features of the GP & PHC Research Conference is the opportunity it offers for experienced and inexperienced researchers to proffer abstracts about their projects, completed and in progress. The opportunity to present a paper or poster encourages people to actively participate in the conference, which contributes to delegates feeling part of the research community. It also justifies conference attendance, as some organisations will sponsor attendance only if people are active presenters at conferences.

2002 General Practice and Primary Health Care Research Conference — Proceedings

1

In 2002 PHC RIS received a record number of 155 varied and interesting abstracts, from authors from across the General Practice and Primary Health Care spectrum — university departments; state and national policy and research centres; professional, practice, and consumer organisations.

Although the conference aimed at a broader audience than General Practice, 43% of abstracts came from university departments of general practice, and 16% from GP organisations such as

Divisions and RACGP.

In contrast to past years when most abstracts related to GPEP funded research, this year’s abstracts were funded by a wide range of sources. The most frequently identified source was the

Commonwealth Department of Health and Ageing (17%), followed by PHC RED funding through university departments of general practice and rural health (10%), national sources such as

NHMRC and ARC (10%), State governments (8%), and Divisions of General Practice (6%). Only four authors identified GPEP funding as the main source. 20% did not state their funding sources, and 24% were funded through their own organisation or a wide variety of other sources.

Most abstracts submitted to the GP & PHC Research Conference were consistent with four of the priority areas identified in 2001 for PHC research — models of PHC delivery, translating evidence into practice, quality of care, and integration. Relatively few abstracts related to health promotion, health inequalities or health economics research. Research into workforce and GP wellbeing, an important area of research that did not fit easily into the PHC RED priority framework, was represented at this conference by 20 abstracts.

The conference organising committee applied a peer review process to these abstracts for the first time, rejecting the few that did not meet conference criteria. The remainder were included either as 15’ papers in concurrent sessions, or as posters for 5’ presentations. Both poster and paper sessions were grouped into themes as far as possible. This year we planned the program to maximise the number of papers, as this is the preferred mode of delivery.

Pre-conference workshops were included in the program for the first time in 2002. Susan Horn’s full day workshop was very successful in terms of attendance and impact, resulting in much discussion and an energetic start to the conference. Three other half-day workshops were less well attended, but all provided a good opportunity to address complex issues with adequate time and with active participation.

A highlight of the conference for those lucky enough to be there was the dinner entertainment,

Coco’s Lunch, The women singing were as pleased with their rapt audience (unusual in the middle of a dinner) as we were with their moving songs — a refreshing interlude from the rigours of talking to delegates and listening to conference presentations no matter how interesting.

During the conference, various local speakers expressed some serious concerns about current issues in general practice, primary health care and research. However, David Wilkin as an observer from the UK summed up the conference by saying

I’ve had two fascinating and really very interesting days, hearing and seeing a lot of excellent primary care research covering such a wide range of subject materials, methods, disciplines that it seems to me primary care research in Australia is in many respects in an extremely healthy state.

PHC RIS would like to thank everyone who contributed to this year’s conference — the Department of Health and Ageing for much more than funding, the Conference Advisory Committee for their suggestions, those people who chaired sessions so well, and the research community of researchers, practitioners, policy makers and consumers for their papers, posters, ideas and debate about general practice and primary health care research.

2

2002 General Practice and Primary Health Care Research Conference — Proceedings

Acknowledgments

This Conference is made possible through the support of:

Department of Health and Ageing

General Practice Branch

Research and Quality Section

Divisions Section

GPO Box 9848

Canberra ACT 2601 www.health.gov.au

Conference Secretariat

Conference Logistics

PO Box 201

Deakin West ACT 2600

Ph: 02 6281 6624

Fx: 02 6285 1336

Email: [email protected]

Conference Organiser

Primary Health Care Research and Information Service

Department of General Practice

Flinders University

GPO Box 2100

Adelaide SA 5001

Ph: 08 8204 5399

Fx: 08 8204 4690

Email: [email protected]

www.phcris.org.au

2002 General Practice and Primary Health Care Research Conference — Proceedings

3

Conference Advisory Committee membership

Professor Chris Del Mar / Dr Alison Ward

Australian Association of Academic General Practice (AAAGP)

Dr Steve Clark

Australian Divisions of General Practice (ADGP)

Mr Gawaine Powell Davies

Centre for GP Integration Studies

A/Professor Ann Larson

Combined Universities Department of Rural Health, WA

Ms Margaret MacDonald

Commonwealth Department of Health and Ageing (DoHA)

Mr Robin Wells

Commonwealth Department of Health and Ageing (DoHA)

Mrs Robin Toohey AM

Consumers’ Health Forum

Dr Jan Davies

National Institute of Clinical Studies (NICS)

Dr Rosemary Aldrich

Newcastle Institute of Public Health

Professor Louis Pilotto

Primary Health Care Research & Information Service (PHC RIS)

Ms Libby Kalucy

Primary Health Care Research & Information Service (PHC RIS)

4

2002 General Practice and Primary Health Care Research Conference — Proceedings

Peer review of conference abstracts

For the first time in 2002 all conference abstracts received by the due date under-went peer review. The scientific committee selected to peer review the 155 abstracts received consisted of:

Department of General Practice, Flinders University

Prof Louis Pilotto

Assoc Prof Liz Farmer

Ms Libby Kalucy

Dr Ellen McIntyre

Dr Kath Weston

Department of General Practice, University of Adelaide

Prof Justin Beilby

Dr Nigel Stocks

This group comprised both general practice and primary health care researchers with skills in quantitative and qualitative methodology. They applied a blind review process using an abstract review form developed and adapted by Liz Farmer and Ellen McIntyre. Upon completion of this review process, rejected abstracts were reviewed again by Prof Louis Pilotto, Libby Kalucy and

Dr Ellen McIntyre. Only two abstracts were finally rejected.

2002 General Practice and Primary Health Care Research Conference — Proceedings

5

Section 1

Pre-conference workshops

X

New methods to discover best practices in health care

X

Bridging the gap: communication between researchers and policy makers

X

Issues in measuring health inequalities, including social capital perspectives

X

Engaging the grassroots: making research networks a reality

X

Evaluation report for the pre-conference workshops

New methods to discover best practices in health care

Facilitator and speaker

X

Dr Susan Horn, Institute for Clinical Outcomes Research, Utah, USA

The workshop consisted of the first day of a six-day workshop series prepared by Susan Horn. This is the Senior Leadership day, designed to introduce concepts to senior leaders, rather than go into methodology in great detail.

Participants

The workshop was attended by 87 people, most of whom (65) attended only the pre-conference workshop, not the conference itself. Delegates were from universities, Commonwealth and State governments, hospitals, health services, pharmaceutical companies, and other organisations.

After Libby Kalucy welcomed participants on behalf of PHC RIS, Margaret MacDonald, Director of

Research and Quality Section, GP Branch, Commonwealth Department of Health and Ageing, introduced Dr Susan Horn.

Key ideas:

X

Clinical Practice Improvement (CPI) methods aim to accelerate and enhance current quality improvement efforts.

X

CPI uses a study design which

– controls for patient factors (diseases, severity, psychosocial issues),

– improves and standardises process factors (management strategies, interventions, medication) and

– measures outcomes (clinical, health statues, costs/length of stay/encounters).

X

Multidimensional statistical analysis is used to put together a comprehensive picture.

X

CPI relies on involvement of all health professionals involved in patient care, doctors, nurses, therapists.

X

Studies are clinically driven, research is done by and with clinicians, who define processes, characteristics, analyses, guidelines.

Outcomes research uses large databases (eg claims, insurance) to evaluate specific treatment methods, medical technologies and providers. Challenges: identifies outcomes failures only, and does not relate those failures to process steps under practitioner control. Interventions are not uniform, and patients are not comparable — they have disease labels without adjustment for severity. The major PORT studies in USA (Patient Outcomes Research Team studies) found places with better outcomes but could not determine why or how to improve. PORT studies always found huge variation in practice.

Guidelines: not decidable, due to vague description of patients; not executable due to menu of process steps, not connected to outcomes. When reading guidelines, clinician is not aware for which patient, which steps should be followed for best outcome.

Randomised Controlled Studies (RCTs) and CPI are complementary not competing. CPI can generate RCTs, and also study practice effects of RCT results.

2002 General Practice and Primary Health Care Research Conference — Proceedings

9

Comparison between CPI and RCT

Process

Outcome

Object of study

Evidence

CPI

Measure and feedback to develop dynamic protocol

Dynamic improved based on fact

Effectiveness

Observational studies: said to have less validity, overestimate treatment effects: contradicted by

NJM 2000 342 1887–92

JAMA 2001 286–7, 821–830

RCT

Explicitly specify every element of process of care

Change based on fact

Efficacy

Highest grade of evidence

Relative costs of RCT and CPI

Medical outcomes study and Health insurance experiment: 6000 patients were followed for up to 8 years in 1980s, costing $35 million. SUPPORT study of end of life care followed 9000 patients, 6 months. Dying with dignity sites were randomised. The study found nothing, no benefit in treatment. Through the discovery process, this led to the hospice CPI.

A managed care outcomes project on asthma, hypertension, arthritis, stomach ulcers, and otitis media followed 13000 patients over 1 year, addressing more variables, and cost $1 million. No randomising or screening, just measure actual process of care. This was an investment in future quality improvement.

Case studies

Susan described a number of specific examples of CPI methodology and their results, including

Paediatric bronchiolitis, prevention of pressure ulcers. After lunch, she provided further case studies of the method.

Issues raised by workshop participants

Susan Horn responded to many questions raised by workshop participants. These included:

X

How to engage clinicians who believe that what they are doing is right:

Need the level of data to include any variable that clinicians can suggest. Existence of variation is cause for study. The study does not profile the best or the worst, but learns and examines the steps. Many participants in these studies have been amazed how easy it is to get people involved and collect useful data.

X

Does this lead to change in clinician behaviour?

Because they have been involved in the whole process and have been able to challenge the analyses, CPI has led to change in behaviour.

X

How could this work where there is turnover and rotation of intern and nursing staff, who cannot all be involved in the project

Clinical information system will make it easier to do new rather than old documentation.

X

Validity of observational studies

Susan based validity on

Repeatability in more than one setting. Some examples have been repeated three times, and same significant variables are associated with outcomes.

Change in practice: if results are correct, should get better practice, and do.

Multiple associations, that lead to change in practice leads to greater acceptance of causation rather then mere association.

10

2002 General Practice and Primary Health Care Research Conference — Proceedings

X

CPI is not a study, but an audit, a management tool, very different from an RCT in terms of data compilation and analysis

Thorough documentation is an important feature, but clinical team not administrative team decides what and how to collect the data, how to define, and how to analyse them. Each item of data is collected because a clinician believes it is important. Analysis is not data mining through a priori ‘fishing’ expeditions.

Conclusion

The workshop created a great deal of lively debate among participants, and considerable enthusiasm for the methods used and results obtained, particularly amongst participants from

State health departments. Many expressed their interest in attending a longer workshop, such as the six-day program that Susan Horn conducts, and which may be provided in Canberra later in the year.

Bridging the gap: communication between researchers and policy makers

Facilitator

X

Libby Kalucy, Primary Health Care Research & Information Service (PHC RIS)

Speakers

X

X

X

David Wilkin, National Primary Care Research and Development Centre, University of

Manchester (UK)

Bronwen Harvey, Medical Advisor, GP Branch, Commonwealth Department of Health and

Ageing

Ellen McIntyre, Senior Research Fellow, PHC RIS

Professor David Wilkin from National Primary Care Research and Development Centre, Manchester, discussed how the Centre supports long term, quality research — through a research agenda that is responsive to policy customer needs, but established by researchers (through consultation).

They use a Research Liaison Officer to aid communication between these groups. However, there is tension created by the political nature of research, with the need to publish (researcher) and put results into public domain (policy). Sometimes a choice has to be made between media or refereed journal which can lead to problems related to researcher career progression. David mentioned that policy is influenced by pressures in the wider arena and that researchers need to communicate through a wide range of sources that influence policy including liaison/collaborations with politicians, pressure groups, professional organisations, health care providers, private sector, voluntary sector, and the media.

Dr Bronwen Harvey, Medical Advisor, GP Branch, Commonwealth Department of Health and

Ageing, provided the perspective of a policy maker/research user. She talked about some of the practical aspects of the Commonwealth policy environment, and suggested some strategies to help researchers communicate more effectively with the Commonwealth. Researchers need to understand government and policy making processes and cycles. She suggested they talk to the section heads/medical advisors of Government departments who implement the policies, and manage the programs of the government/Minister before starting a research project. This helps to frame the research question/s and negotiate the area to be researched.

Dr Ellen McIntyre, Senior Research Fellow, PHC RIS, presented results of a recent study of

Australian GP and primary health care researchers (which includes communication with policy makers). With a focus on investigator-driven research, common approaches researchers currently use to inform policy makers of their findings included peer review articles, conference and seminar presentations, research reports, networking, and advisory groups. Factors which helped researchers inform policy makers of their findings included passion and commitment, networking

2002 General Practice and Primary Health Care Research Conference — Proceedings

11

and collaboration, peer support, an understanding of the policy system, targeting the approach and utilising opportunities. Factors that hindered included lack of resources, limited results, poor linkage to policy system, and lack of continuity of policy makers.

Lively discussion throughout the workshop raised many issues including that fact that in considered policy (policy developed over time) the working group thinks about: what they need to know, who they can get information from, who is working in the area, what the research evidence shows, what is in the grey literature/published literature, how can existing networks help, is the minister trying to develop a policy in this area? and would this be something the minister would support?

Issues in measuring health inequalities, including social capital perspectives

Facilitator

X

Health Inequalities Research Collaboration, Primary Health Care Research and Development

Network

Speakers

X

X

Stephen Begg, Epidemiologist, Health Outcomes Section, Department of Human Services,

Victoria

Anna Ziersch, Research Officer, South Australian Community Health Research Unit, South

Australia

The two speakers presented papers on recent research relating to health inequalities/inequities

(summarised below).

The Victorian Burden of Disease Study (Stephen Begg)

While death rates and morbidity rates provide some information about diseases and can provide information on priority health areas, mortality rates are not a good reflection of illness in a population. The Victorian Burden of Disease (BOD) study used health expectancy measures (HALE and DALY), which adjust life expectancy for time spent with a disability. Mortality data are readily available (from ABS), and for non-fatal outcomes, all disabled states that are a consequence of disease (e.g. diabetes may result in several disabled states, such as blindness, foot problems, etc), are looked at individually for their incidence, their average duration and weighting for severity. The scores are driven by underlying epidemiology, which is drawn from the ABS, disease registers, population surveys, epidemiological studies, health facility data, expert opinion, and disease modelling. Results of the study showed that mental disorders are a big burden in Victoria (25% male and 50% female) and that 50% of the burden results in death. Governments and communities can use the results of BOD studies to identify regions where health inequalities may exist. Further details of the study can be found on the Victorian Burden of Disease website: www.dhs.vic.gov.au/phd/lgabod/index.htm.

Measuring health inequities: a social capital perspective (Anna Ziersch)

Anna presented the results of her PhD, which measured social capital and its relationship to health, in two inner western suburbs of Adelaide. The social capital approach connects social relationships to economic and other outcomes; promotes an interdisciplinary approach; uses quantitative and qualitative methodologies; and can consider the individual/community interface. Social capital was measured by looking at the infrastructure (formal and informal social networks, and values), and resources (including help, support and assistance, sense of community, belonging and acceptance, civic participation, and control of life in general), available to individuals within the communities.

The main findings of the study were that within the communities, there were differences in the access to infrastructure and resources; only some elements of social capital were related to health;

12

2002 General Practice and Primary Health Care Research Conference — Proceedings

other individual and area-level factors were important to health; and that social capital is multidimensional and complex.

Discussion session

Following the presentations, we divided into three groups, discussed aspects of health inequalities of personal interest, and reported back to the whole group. Of particular interest were: an example of the use of a local BOD study in an indigenous community to help the community set funding priorities; and the need to have integrated policy to reduce health inequities, which recognises how other factors affect access to health care and individuals’ willingness to seek care (e.g. a whole government approach that integrates health with education, transport, housing, etc).

Engaging the grassroots: making research networks a reality

Facilitator

X

NSW PHC Research Capacity Building Program

X

X

X

X

X

Speakers

Professor David Lyle, Department of Rural Health (Broken Hill), University of Sydney

Frances Boreland, Department of Rural Health (Broken Hill), University of Sydney

Angela Bettess, Lecturer, Department of General Practice, University of Sydney

Caroline Laurence, Research Fellow, Department of General Practice, University of Adelaide

Nabil Sulaiman, Senior Lecturer, Department of General Practice, University of Melbourne

This workshop was for those involved in, or interested in, developing research networks from the practitioner level, i.e. the grassroots. The session provided an opportunity to hear from those who have developed research networks, share experiences, learn about common barriers and problems and identify solutions.

A welcome and introduction was provided by Professor David Lyle. He pointed out that currently there are several barriers to research by general practitioners such as time, workload and support.

Part of the aims of developing research networks is to deal with and assist in overcoming these barriers. Frances Boreland then spoke about the what, who and how of research networks. She discussed the large diversity of PHC research networks and how links to academic departments increase the probability that funding will be provided for research. She also commented on the major challenges for research networks. These include (i) maintaining commitment and enthusiasm over the long-term, (ii) the sustainability of the network over time, (iii) evaluating the impact and outcomes of research networks, and (iv) ensuring broad engagement with other research networks.

Several presentations were then given describing existing research networks in Australia. These included (i) the Evaluation and Research Interest Group (EARIG), presented by Dr Angela Bettess,

(ii) the University Family Practice Network, presented by Caroline Laurence, (iii) RuralNet, presented by Frances Boreland, and (iv) the North Melbourne Research Network, presented by Dr

Nabil Sulaiman.

The group then divided into four smaller groups and were given scenarios relevant to the establishment or maintaining of a research network. These scenarios were discussed and possible solutions generated. Professor David Lyle, Department of Rural Health, Broken Hill then provided a summary of the workshop. He pointed out that for the current activity in terms of establishing and maintaining research networks to be successful, many resources are required as well as an assessment of the value of research networks. On the basis of this evaluation, the sustainability of the infrastructure of research networks needs to be considered. It was suggested there should be a recommendation to the Commonwealth that these initiatives should not be allowed to die and that it may be far too early to evaluate such initiatives. In addition it was pointed out that

2002 General Practice and Primary Health Care Research Conference — Proceedings

13

Divisions are currently not seeing research as a core part of their business and that perhaps one of the next steps is to look and see what they do see as their business and fit research into this.

Evaluation report for the pre-conference workshops

About the workshops

Four workshops were held the day before the 2 day GP & PHC Research Conference. The workshops aimed to complement the conference program through interactive sessions which provided more time to explore issues in-depth.

Workshop attendance and evaluation

There were 150 workshop attendees in total, 81 who also attended the 2-day conference.

Evaluation forms were given to all attendees and 62 evaluation forms were completed (41%). The respondents rated various aspects of the workshops with 5-point Lickert scales, ranging from 1

(Poor) to 5 (Excellent).

Participants mostly heard about the workshops by word of mouth (44%), in a PHC RIS Email Alert

(26%) and on the PHC RIS website (14.5%).

New methods to discover best practices in health care

(full day workshop, 85 participants)

The objectives of this workshop were to:

X explain clinical practice improvement methodology

X discuss severity of illness conceptualisation and the Comprehensive Severity Index

X describe clinical practice improvement studies and discuss their clinical and financial impacts.

Speaker

X

Dr Susan Horn, Institute for Clinical Outcomes Research/Dept of Medical Informatics,

University of Utah, USA.

Ratings

Speaker

Susan Horn

Overall rating

Expectations met

Rating of discussion

Rating of venue

Rating of organisation

Quality of content

Quality of delivery

N

36

37

33

34

33

37

36

Min

2

1

3

3

2

2

4

Max

5

5

5

5

5

5

5

Mean

4.42

4.38

3.73

3.35

4.03

4.43

4.69

Comments

Limit numbers to facilitate interaction and greater discussion by all. (Research/Project Officer)

More opportunity to debate. (Policy Maker)

Shorter sessions please — I can’t concentrate for more than 50 minutes! Let me sit down to eat my lunch. (Health Manager)

Excellent. Would love to hear the government’s response to this. (Health Practitioner)

14

2002 General Practice and Primary Health Care Research Conference — Proceedings

Very open and confident in presentation. Was able to communicate the data in an interesting and enlivened way. Able to see other points of view. Best of all was the big picture focus!!

(Pharmaceutical industry)

Bridging the gap: Communication between researchers and policy makers

(½ day, 24 participants)

This interactive workshop aimed to develop some practical approaches to improve communication between researchers and policy makers to enhance the impact of research.

Speakers

X

Professor David Wilkin, National Primary Care Research and Development Centre, University of

Manchester, UK

X

Dr Bronwen Harvey, Medical Adviser, GP Branch, Commonwealth Department of Health and

Ageing

X

Dr Ellen McIntyre, Primary Health Care Research and Information Service, Flinders University

Speaker

David Wilkin

Bronwen Harvey

Ellen McIntyre

Overall rating of w/shop

Expectations met

Rating of discussion

Rating of venue

Rating of organisation

Quality of content

Quality of delivery

Quality of content

Quality of delivery

Quality of content

Quality of delivery

6

6

6

6

6

N

6

6

6

6

6

6

3

3

3

4

3

Min

3

4

2

4

4

3

5

5

5

5

5

Max

5

5

5

5

5

5

4.33

4.67

4.33

3.67

4.00

Mean

4.00

3.67

4.5

4.17

4.33

4.00

Comments

Pre-reading would be useful. (Academic GP)

Well organised — great facilities. Very worthwhile — Thank you. (Research/Policy Officer)

2002 General Practice and Primary Health Care Research Conference — Proceedings

15

Issues in measuring health inequalities, including social capital perspectives

(½ day, 22 participants)

The aim of the workshop was to provide participants with the information on techniques for researching health inequalities within the data and framework of the Victorian Burden of Disease

Study, and within a social capital framework.

Speakers

X

Stephen Begg, Epidemiologist, Health Outcomes Section, Department of Human Services,

Victoria

X

Anna Ziersch, Research Officer, South Australian Community Health Research Unit

Speaker

Stephen Begg

Anna Ziersch

Overall rating of w/shop

Expectations met

Rating of discussion

Rating of venue

Rating of organisation

Quality of content

Quality of delivery

Quality of content

Quality of delivery

N

7

6

7

6

7

6

7

7

7

Min

4

4

3

2

3

4

3

3

2

Max

5

5

5

5

5

5

4

4

5

Mean

3.57

3.57

3.57

3.71

4.00

4.43

4.5

4.14

4.17

Comments

The two presenters were both excellent, but there was no integration of their work and very little group participation. A less formal structure may have helped (rather than two paper presentations). (Research/Project Officer)

Maybe some options of a series of questions as discussion group ‘seed’/starters — given the short time frame and it took a while for group dynamics to warm up and discussion to get started.

Diverse background of group — different perspectives. (Research/Project Officer)

I was expecting a ‘workshop’ not two lectures on people’s research, followed by an unfocused discussion. (Academic)

The discussion session after the presentations was not well organised — perhaps a more guided discussion would have been useful — it felt like preaching to the converted. (Research/Project

Officer)

I was more interested in ‘where to from here?’ (Research/Project Officer)

I really enjoyed the interaction and the speakers. (Project Manager)

16

2002 General Practice and Primary Health Care Research Conference — Proceedings

Engaging the grassroots: making research networks a reality

(½ day, 19 participants)

This workshop aimed to provide people with the opportunity to hear from those who have developed research networks, share experiences, learn about common barriers and problems and identify solutions.

Speakers

X

Professor David Lyle, Department of Rural Health (Broken Hill), University of Sydney

X

Frances Boreland, Department of Rural Health (Broken Hill), University of Sydney

X

Angela Bettess, Lecturer, Department of General Practice, University of Sydney

X

Caroline Laurence, Research Fellow, Department of General Practice, University of Adelaide

X

Nabil Sulaiman, Senior Lecturer, Department of General Practice, University of Melbourne

Speaker

Frances Boreland

Angela Bettess

Caroline Laurence

Nabil Sulaiman

Overall rating of w/shop

Expectations met

Rating of discussion

Rating of venue

Rating of organisation

Quality of content

Quality of delivery

Quality of content

Quality of delivery

Quality of content

Quality of delivery

Quality of content

Quality of delivery

N

8

8

8

8

9

8

9

9

9

8

11

11

10

Min

3

3

3

3

3

3

3

3

3

3

3

3

3

Max

5

5

5

5

5

5

5

5

5

5

5

5

5

Mean

4.13

4.11

4.13

4.25

4.25

4.13

3.91

4.09

3.8

4.00

4.25

4.11

4.11

Comments

Small group discussion is valuable, we could have done some more. (Academic)

More time for general discussion (probably not possible in this format) (Research/Policy Officer)

All the speakers were good, some better than others, but interesting material and useful to me personally. (Academic)

Future workshops

Suggestions included:

X

DoHA [and] PHC RED people. (Academic GP)

X

Talking through case studies for practical/pragmatic approaches to applied research. (Project

Manager)

X

Get a smoke free environment for the next [workshop]. (GP)

2002 General Practice and Primary Health Care Research Conference — Proceedings

17

Section 2

Conference proceedings

X

Opening session

X

Keynote papers

X

Plenary papers

X

Summary session

X

Breakfast sessions

X

Concurrent abstracts

X

Poster abstracts

X

Conference evaluation report

Opening session

Opening address

Senator the Hon. Kay Patterson

Minister for Health and Ageing

Thank you very much Professor Pilotto. Welcome to Professor David Wilkin, Hilda Bastian,

Professor Susan Horn and Professor James Dunbar. To those of you from overseas, especially to our international speakers, welcome to Australia.

I am delighted to be here today to open the conference on primary health care and research. It’s an area about which I feel very passionate. I think it’s really important that this conference has brought together people from the health professions, from research, some of whom are both health professionals and researchers, the policy makers and consumers. I know that there is an emphasis on the consumers of health services having input into research and I think it’s a very important initiative.

Through my work at Lincoln Institute, School of Health Sciences in Melbourne I became aware of how important it was to get these groups together.

Lincoln Institute, [which for those of you who come from other states, was like Cumberland

College in Sydney], combined all the old colleges of OT, physio, speech, occupational therapy, podiatry, medical records administration, prosthetics, orthoptics, and nursing. There was a School of Behavioural Sciences and a School of Biological Sciences servicing these clinical schools. I joined the teaching staff of the School of Behavioural Sciences not long after the colleges had come together. I remember very clearly that it was still very hospital-like, we had a morning tea lady and we all went over and had scones for morning tea in the morning tearoom. Some of the people who had been teaching in the clinical schools for a long while, thought that the people in Biological and Behavioural Science areas were a little alternate. They were not enamoured of the course in research methodology, nor were they very encouraging to some of the students who ventured across the road to do research methodology. Initially students who came in to do physio, OT or speech therapy were not very interested in learning about chi squared or T tests, or analysis of variance. But I think that one of the things that happened in that period of time was the change in our professions. The people with the clinical experience turned us into health psychologists and health sociologists, and we turned them into researchers.

I believe that psychology has played a very important part in developing research methodology in not only health sciences but in medicine and other areas. I guess it was our defence to try and demonstrate that we were a science. I remember my first psychology textbooks discussed plots of wheat in fields and how they improved when you gave them different fertilisers. I found it difficult to understand how it had any relevance to my being a psychologist. Fortunately, as time went on, one of the contributions we made at Lincoln was that we developed research methodology textbooks that had actually clinical examples which were relevant to students from nursing or in medicine.

I think we’ve come a long way in the last twenty years in terms of research and research teaching for people working in primary health care.

It’s absolutely vital that we inform what we do with research. We must be able to demonstrate that what we are doing achieves outcomes. Some of the work we were doing has been intuitive rather than based on fact. However as we gain more evidence our practice becomes more evidence-based, economies in time and in cost can be delivered to primary health care. So it is vital that what we do is informed and that work that you’re doing informs policy.

2003 General Practice and Primary Health Care Research Conference — Proceedings

21

While you can do all the research work you like, if it’s not informing policy then you know it really isn’t necessarily going to achieve as much as it can. That’s one of the reasons why it’s important that we have this forum. So that the Department that assists me, sees and hears about the research.

The conference also gives you the opportunity for sharing ideas. Sometimes research can be quite a simple project; other research work can have profound effects. Let me share an example of a study that caught my imagination during my years on the backbench.

There was a study by a GP working in the north coast of NSW where he recruited a sample from his practice of people over 75, and he divided them into two groups and tried to match them. As you know this is difficult given a small sample size in the practice, but he did it to the best of his ability. One group joined a walking group that met at the local pharmacy three times a week, and the pharmacist was involved and encouraged them. After they had been doing this for, I think, 12 months, the doctor measured the rate of admissions to hospital, the number of falls, and the medication that they were taking. He saw that there was quite a significant difference between the two groups.

Now, while this was a project that was just done within a practice, it has messages for larger studies. It also has messages for what general practitioners can do in their own practices.

In the time that I’ve been involved in the health sciences, I think we have also seen some interesting changes in the ability of all the health professions to work together in a much more coordinated and cooperative way.

This has been evident in influencing policy. One example is in our mental health initiatives for general practice where we can involve psychologists. Another is in the Home Medicines Review program where a GP and an accredited pharmacist collaborate to provide a comprehensive medication review for a patient in their own home. Once the GP identifies a patient who would benefit from such a review, he or she contacts the patient’s pharmacy of choice, who arranges a visit to the person’s home to go through all the medicines they are taking. By being in the patient’s home the pharmacist can take into account all the patient’s herbal and over the counter preparations, together with medicines the patient may no longer be taking. Now, you know, most probably a doctor’s surgery is not the best place for that to happen when we’re all a little nervous when we visit the doctor. But to have someone sit down quietly in your own home and go through your medication is a great opportunity. The pharmacist then prepares a full report back to the GP so the GP and the patient can agree on a medication management plan. Based on this team approach to their care patients gain a better understanding of their medication regimen and how best to take their medicines.

This is the first time we’ve seen pharmacists receive a Medicare payment for collaboration with

GPs for the provision of Home Medicines Reviews. The pharmacist is paid a fee through provisions under the Third Community Pharmacy Agreement. And already I’ve had anecdotal evidence, including one pharmacist who grabbed me in an airport and told me how successful her first home medicine review had been. This is an example of real integration between the health professions working in cooperation to achieve the same goals of better use of medicines and better health.

It is vital that we have the opportunity of sharing — that we ensure those of you who are doing research set the directions for the next group of health professionals coming behind you — the doctors and nurses, the physios and OTs and the whole gamut of disciplines involved in primary health care. These people will benefit from your shared experiences and also be motivated to do research. I think there’s been a significant change in the past ten to fifteen years in terms of the skill base of health professionals and the willingness to actually do the research so we have greater opportunities for evidence-based practice and policy.

I know some of you are working in more remote areas, and I want to say I’ve seen some very interesting primary health care research being done recently in the Northern Territory. It is very important for us to work together to develop the very best outcomes for all Australians wherever they might be. Each of you will have a passion about what you’re doing which is fantastic. I

22

2002 General Practice and Primary Health Care Research Conference — Proceedings

acknowledge that it is a challenge to make time in hectic schedules and congratulate you on being here. The fact that you are here at the conference demonstrates you are committed to evidencebased medicine improving primary health care.

In closing I would like to wish you a very enjoyable conference. I’m sure you’ll benefit not only from the stimulation of papers and posters but also from the out of session conversations. I wish you all the best, and I look forward to hearing from the Departmental officers the outcome of the conference. I’m sure they’ll have lots of ideas for me to consider for practice and for policy.

I have much pleasure in opening this conference.

2003 General Practice and Primary Health Care Research Conference — Proceedings

23

Keynote papers

Health services research in primary care: influencing health policy, service development and professional practice

Professor David Wilkin

Professor of Health Services Research, National Primary Care Research and Development

Centre, University of Manchester, UK

Abstract

Drawing on examples from UK health services research in primary care, the presentation will highlight the importance of engaging with stakeholders (policy ‘customers’, managers, health professionals and service users), formulating appropriate research questions, incorporating different disciplinary perspectives, selecting appropriate research designs and methods, timely reporting, and effective dissemination for different audiences.

Making a difference to health and health care: evidence-based policy and practice

X

Our goal should be evidence based:

– health policy

– service organisation and delivery

– clinical practice.

X

Evidence should be derived from objective, value free scientific research. The randomised controlled trial represents a ‘gold standard’ for evidence based health care.

X

The job of researchers is to produce scientific evidence.

X

It is the responsibility of policy makers, managers and health professionals to base their decisions on the scientific evidence.

Issues for health services research in primary care

X

Scope and purpose of health services research in primary care.

X

X

Stakeholders in health services research.

Shaping the agenda for HSR in primary care.

X

X

X

X

X

Disciplinary perspectives.

Research questions, designs and methods.

Doing HSR in the real world.

Disseminating research findings.

Contributing to knowledge and understanding.

Scope and purposes of HSR in primary care

X

The funding, content, organisation and delivery of health care, from the individual patient/professional interaction to the organisation of health care systems.

X

Research into service organisation and delivery deals with complex social and political processes in the real world.

24

2002 General Practice and Primary Health Care Research Conference — Proceedings

X

Purpose is to enhance knowledge and understanding to support:

– policy formulation and implementation

– management of health care

– professional practice.

X

Research is only one source of evidence. Decisions will be underpinned by knowledge and understanding derived from a variety of sources.

Stakeholders in HSR in primary care

X

Users of health care.

X

Policy makers: politicians, civil servants, managers and health professions.

X

Commissioners/funders of services.

X

Service providers: organisations, managers, health professions.

X

Researchers.

X

Research funders.

X

Research subjects.

Shaping the agenda for HSR in primary care

X

What gets onto the research agenda and why?

X

X

The role of research funders.

The role of the research community.

X

X

X

Mechanisms for involving other stakeholders in shaping the research agenda.

Investigator driven research versus the customer-contractor model.

Values, politics and power in shaping the research agenda.

Research questions

X

Involving stakeholders in formulating research questions.

X

Locating research questions in theory and current knowledge — disciplinary perspectives.

X

Formulating questions to ensure that they address enduring issues as well as short term requirements.

X

Interaction between research design/methods and research questions.

Research designs and methods

X

Matching the design and methods to the research questions?

X

Selecting appropriate designs and methods. HSR commonly requires mixed methods.

X

Experimental designs rarely feasible when evaluating complex interventions in the real world.

X

Resource constraints and feasibility of the research.

X

Defining and measuring inputs, processes and outcomes.

2003 General Practice and Primary Health Care Research Conference — Proceedings

25

Improving services for orthopaedic patients: Stakeholders, questions, disciplinary perspectives, designs and methods (Fulop et al 2001)

X

Stakeholders: Patients, clinicians, managers, policy makers.

X

Questions:

What is the best way of delivering orthopaedic care?

How can change in existing services be implemented?

How can care be further improved though organisational change?

X

Disciplinary perspectives: medicine, nursing, epidemiology, economics, sociology, history, social policy, psychology.

X

Designs and methods: randomised trial, economic evaluation, case study, participant observation, interview, secondary analysis, operational research, action research, policy analysis, etc.

What is the best way of delivering orthopaedic care?

X

Will telemedicine be as clinically effective in terms of accuracy of diagnosis? (epidemiological research/RCT).

X

Will telemedicine be as efficient for the health service? (economic evaluation).

X

What impact will telemedicine have on the relationship between the GP and specialist?

(historical research).

X

What impact will telemedicine have on the patient’s relationship with the GP and specialist?

(micro-sociology).

How can change to existing services be implemented?

X

What effect would the method of funding orthopaedic services have on implementing telemedicine? (organisational economics).

X

What impact will telemedicine have on the functioning of the orthopaedic service? (operational research).

X

What impact will telemedicine have on the behaviour, attitude and emotions of clinical staff?

(organisational psychology).

X

What happens in practice when telemedicine is implemented? (action research).

How can care be improved further by changes at the level of the organisation?

X

What incentives exist to encourage integration and what is the cost of greater integration?

(organisational economics).

X

What factors facilitate or hinder integration of services (organisational studies).

X

Which policies might facilitate greater integration? (policy analysis).

Doing HSR in the real world

X

Mismatch between the timetables of stakeholders and the research.

X

X

X

Adapting to changes in policy, service organisation, delivery, staffing, etc.

Working with research subjects.

Flexibility, adaptability and responsiveness.

26

2002 General Practice and Primary Health Care Research Conference — Proceedings

Disseminating research findings

X

Commitment to dissemination by all parties.

X

X

Intellectual property, copyright and freedom to publish.

Planning and managing the dissemination process.

X

X

X

X

X

Working with customers and other stakeholders.

Multiple methods of dissemination.

Exploiting opportunities and handling the media.

Spin and censorship.

Funding dissemination.

Knowledge based health policy, service organisation and delivery

X

Decisions should be based on values, knowledge and understanding, supported by evidence.

X

Health services research is but one (important) source of evidence, knowledge and understanding.

X

Good HSR draws on many disciplines and commonly requires a variety of methods. The RCT has an important, but limited, contribution to make.

X

Effective HSR requires dialogue with stakeholders, flexibility and responsiveness.

X

A more mature relationship between research, policy and practice will enhance our capacity to make a difference.

Suggested reading

X

Fulop, N. Allen, P. Clarke, A. and Black, N. (2001) Studying the organisation and delivery of

health services. London: Routledge (ISBN 0-415-2563-8).

X

Mays, N. Wyke, S. Malbon, G. and Goodwin, N. (2001) The purchasing of health care by

primary care organisations. Buckingham: Open University Press. (ISBN 0-335-20900-9).

X

Pawson, R. and Tilley, N. (1997) Realistic Evaluation. London: Sage.

2003 General Practice and Primary Health Care Research Conference — Proceedings

27

Using research to inform and change primary care

Professor James Dunbar

Director, Greater Green Triangle University Department of Rural Health, Deakin

University, Warrnambool and Flinders University, Adelaide

Abstract

This presentation will explore the emerging methodology for quality improvement in health care.

Primary health care research is faced with a conundrum. Classical research methods are difficult to apply to important questions. Consequently traditionally-minded research funding bodies are disinclined to fund primary health care research. Undue emphasis is placed on randomised controlled trials.

To some extent the situation has improved with increasing acceptance of qualitative research.

Nevertheless difficulties remain in researching some of the most important issues that face the health care system. The conundrum will be illustrated using obesity as an example.

During the early nineties the Institute for Health Care Improvement in Boston led by Professor

Donald Berwick came up with the methodology for quality improvement in health care. This methodology, which has a basis in statistical process control, can be applied to important questions about improving the health care system.

Years lost due to disability (YLD) by sex and disease group, Victoria 1996

Other

Cardiovascular

15%

Diabetes

4%

Musculoskeletal

6%

6%

Injuries

10%

Chronic respiratory

16%

10%

8%

25%

Cancer

Mental disorder

Neurological

Males: 143,821 YLD

44% of total DALYs

Other

14%

Diabetes

Musculoskeletal

10%

4%

Injuries

3%

9%

Chronic respiratory

18%

Neurological

Cardiovascular

7%

7%

Cancer

28%

Mental disorder

Females: 147,229 YLD

50% of total DALYs

28

2002 General Practice and Primary Health Care Research Conference — Proceedings

Areas with high and low DALY rates: heart disease and diabetes high low

New challenges for better health by 2010

X

Lifestyle and behaviours are key contributors to health and disease patterns

What is the research question?

What is the research method?

Levels of evidence

X

A — meta-analysis, systematic review or RCTs

X

X

X

B — case control or cohort studies

C — extrapolated from case control cohort studies

D — case reports or expert opinion

EBM — the fallacy

X

General practice is holistic

X

Absence of evidence is not evidence of absence

X

Who pays?

X

Common sense and experts

X

Selective publication and publication bias

X

Lifestyle risk factors and population approaches

2003 General Practice and Primary Health Care Research Conference — Proceedings

29

40

Prevalence

%

Risk

%

60

BP distribution, risk for coronary heart disease or stroke, and number of such morbid events in relation to blood pressure during 13.5 years’ follow-up of 855 men aged 50 at entry.

Morbid events n

Risk

60

BP distribution

30

40 40

20

10

20 20

Morbid events

0 0

Wilhelmsen

100 120 140 160 180 200

0

Systolic BP

Mortality changes in North Karelia in 25 years (35–64, age adjusted, men)

Mortality Rate in 1970 (per 100 000) Change in 25 years (%)

Total

All CVD

Coronary

Cancer

Lung cancer

1556

912

695

293

167

–45

–68

–73

–45

–71

Risk factor changes in North Karelia 1972 and 1992 (age 25–59)

Male

Female

Smoking (%)

S-cholesterol (mmol/l)

Blood pressure (mmHg)

Smoking (%)

S-cholesterol (mmol/l)

Blood pressure (mmHg)

1972

52

6.9

149/92

10

6.8

153/92

X

X

X

X

X

X

X

EBM — the fallacy

General practice is holistic

Absence of evidence is not evidence of absence

Who pays?

Common sense and experts

Selective publication and publication bias

Lifestyle risk factors and population approaches

Economic evidence, patient’s views and implementation

1992

32

5.8

142/85

17

5.6

135/80

30

2002 General Practice and Primary Health Care Research Conference — Proceedings

Identify gaps in Research and

Development

Evidence of

Clinical

Effectiveness

Evidence of

Cost Effectiveness

Effective

Healthcare

Evidence of

Effective Service

Delivery

Patient Priorities

&

Satisfaction

X

X

X

‘Evidence’ in CVD

Prava- or simvastatin

Other choice of drug

Dietary advice

X

X

X

X

X

X

X

X

Publishing quality improvement

Context

Outline of problem

Key measures of improvement

Process for gathering information

Analysis and interpretation

Strategy for change

Effects of change

Next steps

X

X

X

X

X

X

X

Change

Not all change is improvement, but all improvement is change.

Real improvement comes from changing systems not changing within systems.

To make improvements we must be clear about what we are trying to accomplish, how we will know what change has led to improvement and what change we can make that will result in improvement.

The more specific the aim the more likely the improvement.

Concentrate on meeting the needs of patients rather than the needs of organisations.

Measurement is the best for learning rather than for selection, reward or punishment.

Effective leaders challenge the status quo by insisting that the current system cannot remain and by offering clear ideas about superior alternatives.

2003 General Practice and Primary Health Care Research Conference — Proceedings

31

X

X

X

Fundamental questions for improvement

What are we trying to accomplish?

How will we know that a change is an improvement?

What changes can we make that will result in improvement?

What are we trying to accomplish?

How will we know that a change is an improvement?

What changes can we make that will result in an improvement?

AIM

MEASUREMENT

All change does not lead to improvement, but all improvement requires change

CHANGE IDEAS

Model for improvement

O

O

O

What are we trying to accomplish?

How will we know that a change is an improvement?

What changes can we make that will result in an improvement?

Act Plan

Study Do

Incremental improvement

• low investment per project (small projects, but in large numbers)

• grass roots based; empowering (builds morale, customer satisfaction)

• needs reward and recognition system (reinforces improvement vision)

• 100% workforce participation

32

Time

2002 General Practice and Primary Health Care Research Conference — Proceedings

Secondary prevention of coronary vascular disease

An example of improvement

X

X

X

X

CVD Project

Involved 105 GPs in 37 practices.

All data shared unanonymously.

Chosen as a model later applied to diabetes and hypertension.

Became multidisciplinary.

Aim: the original project

Improve secondary prevention of CHD by developing and introducing a local guideline and auditing clinical management before and after introduction.

Lifestyle

X

Smoking habits

19% (n-191) current smokers

– only 3 on nicotine replacement therapy

– on re-audit, rate down 1% (0–31%)

X

X

66% (n=702) received dietary advice at least once.

On reaudit increased to 73%.

X

X

X

X

X

Secondary prevention project: current components

Re-audit of practice activity December

Patient-held record cord

Resources pack

Introduction and evaluation of Heartscore patient-interactive software

Nurse training

Cholesterol and statins

X

X

41 to 62% had cholesterol level below 5.2 mmol/l

46 to 63% have cholesterol level above 5.2 mmol/l and are on statins

Blood pressure (31% (n=302) diagnosed with hypertension

89%

88%

87%

86%

85%

84%

83%

82%

81%

80%

83%

Sysolic BP below 160 mm Hg

88%

Diasolic BP below 90 mm Hg

2003 General Practice and Primary Health Care Research Conference — Proceedings

33

Aspirin

100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%

87%

Yes (n=862)

8%

No (n=78)

5%

Contra-indicated (n=46)

X

X

X

X

X

X

Professions involved in training

Dietician

Diabetic physician, cardiologist, rehab medicine specialist

General practitioners

Health promotion staff

Physiotherapist

Pharmacist.

X

X

X

X

X

X

Nurse training

Behaviour change skills (2 Days)

Smoking cessation

Diet and statin drugs to lower cholesterol

Physical activity and angina management

Diabetes/hypertension

Clinic management

Main outcomes

X

10% reduction in admissions in first year.

X

X

X

X

Peer review

Facilitated inter-practice groups

Collegiate approach

Provided with good information

Protected time

34

2002 General Practice and Primary Health Care Research Conference — Proceedings

Clinical practice improvement: a scientific methodology to discover best medical practice

Dr Susan Horn

Senior Scientist, Institute for Clinical Outcomes Research, Research Professor,

Department of Medical Informatics, University of Utah, Salt Lake City, Utah, USA

Good morning. It is my pleasure to be here today to share a new approach to discovering what treatments are best for specific types of patients. The title I have given this presentation is ‘A scientific methodology to discover best medical practice.’ First, I will give you a brief description of the new methodology. We call it Clinical Practice Improvement or CPI. Then I will share with you several examples of findings from past CPI studies across various areas of health care. Finally, I will point out that if we only had better clinical information systems, what I am talking about would be a lot easier to do. I hope it will stimulate you to get better information systems in your practices.

The idea behind Clinical Practice Improvement is to develop a comprehensive picture of what is going on in the treatment of a particular health care condition in order to analyse the content — what we do — and the timing — when we do it — of individual steps of the health care process with the goal of discovering what is associated with the best medical outcomes. We find that if we get the best medical outcomes — and if there are several different treatments that get the best outcomes, which are the least costly — we can have both better quality and lower cost in health care. So our approach is to focus on achieving cost savings by getting better quality of care.

In CPI studies, we use data from three different dimensions or ‘boxes’ of information. The first is outcomes. We clearly have to figure out the extent of variation in study outcomes. Second is patient factors. In trying to understand what is impacting outcomes, a big part has to do with who you are treating: sicker patients, in particular, are more likely to have poor outcomes, longer length of stay, more visits, more health services utilisation, etc. Also patients may have psychosocial problems making them more challenging to treat. Thus, we try to measure all the factors that could make patients different. Third is treatments or interventions. We try to measure all the treatments that are going on, all the management strategies, and all the medications, time and date stamping everything. By time and date stamping everything, we can see the order in which treatments are given and relate that to outcomes of care. So the overall model for clinical practice improvement studies is to measure outcome, patient, and treatment factors and determine for specific combinations of patient factors, what are those treatments that are associated with the best outcomes, focusing on clinical outcomes first.

CPI methodology goes beyond the usual outcomes research, because most outcomes research uses existing databases, often claims databases or administrative databases. These databases contain labels of diagnosis codes or treatment codes, age, gender, providers, etc. From these databases one can identify where one falls across some outcome continuum. However, these databases normally do not have enough details about the process steps that were done. They just tell if a procedure was performed, but we don’t know all the details underlying that procedure.

Also, they do not adjust for how sick patients are. They tell if a patient has a certain disease or combination of diseases, but not how abnormal the signs and symptoms are for each of those diseases. So we find we are some place across the outcome continuum, but often we have no idea how we got there. We don’t know if our treatments were different, we don’t know if our patients were different, or a combination of the two.

CPI methodology also goes beyond the usual approach to guidelines. Most of the time in the development of guidelines, there is not a lot of data available to tell what should be done explicitly for specific types of patients. Thus, for most guidelines, you will find that in the patient ‘box’, they often give a vague description of the patients for whom the guideline applies, e.g. people at risk of developing pressure ulcers using some kind of risk scale. And secondly, they often give a menu of process steps to follow for those kinds of patients, not telling you precisely for which patient

2003 General Practice and Primary Health Care Research Conference — Proceedings

35

characteristics, what treatments are associated with the best outcomes. Having a comprehensive database allows one to determine best practice based on analyses of data, not just expert opinion.

Finally, comparing clinical practice improvement to randomised controlled trials (RCTs), some of the major differences are as follows. First, for the patient ‘box’, the approach in randomised trials is to make sure the patients are similar or as similar as you can make them. Thus, there are many exclusion criteria for patients with a particular condition to be part of a randomised trial. RCTs do not want any patient factors to bias the outcomes, so people are eliminated who have, for example, certain co-morbidities or more serious forms of a disease. This makes the patient population homogeneous. In clinical practice improvement, for the patient ‘box’, we have to treat all the patients, so we don’t want to get rid of any of them. Instead, we measure all the factors the study team thinks make patients different, measuring multiple dimensions about the patient, including everyone of their diseases, all the relevant signs and symptoms related to severity, and any other factors that the study team thinks could influence the outcomes. Thus, CPI measures patient differences while, in contrast, RCTs exclude patients with differences. Regarding the process ‘box’, often randomised trials only look at one or maybe two or three explicitly-defined treatments and have a very specific treatment protocol to follow. In Clinical Practice Improvement, as you will see in some examples later on, we were surprised about some of the treatments that turned out to be significantly associated with better or poorer outcomes. Thus, rather than hypothesising apriori what the treatment effect should be, we measure what is actually going on in the practice of care, in order to let the data tell us what treatments are associated with best outcomes. Again, we measure multiple dimensions of treatments rather than fixing one or two treatments as in RCTs.

Even though we include many more patients in a Clinical Practice Improvement study — sometimes one or two thousand patients — CPI studies still cost much less than RCTs. CPI studies cost in the thousands of dollars because they use information abstracted from patient charts or sometimes from easy-to-fill out additional documentation forms to know exactly what treatments were done and when. In contrast, RCTs usually cost in the millions of dollars, because of the screening that has to be done to select a homogeneous patient population that does not have the exclusionary criteria and also to make sure that the patients are following the specified treatments in the protocol. Thus, in randomised trials we discover what is best based on very controlled circumstances, whereas in Clinical Practice Improvement we discover what is best in the actual practice of care. In summary, a Clinical Practice Improvement study endeavours to create a comprehensive picture of patient factors and treatment factors to discover what is associated with the best outcomes based on everyday treatment circumstances.

As we heard yesterday, randomised trials are considered the highest-grade evidence. In contrast, almost any form of observational study has been viewed as having less validity, mostly because it has been reported that observational studies over estimate treatment effects. Recently, researchers have studied this in a systematic way. Of course, there are anecdotal situations where a randomised trial said one thing and an observational study said the opposite. But when researchers actually looked at the differences in study findings, as reported in three recent studies, two published in the New England Journal of Medicine and the other in the Journal of the American

Medical Association, they found that the average results from observational studies are actually quite close to those of randomised trials on the same topic, and that well-designed observational studies do not systematically over-estimate the magnitude of the effects of treatments as compared to those of randomised trials on the same topic. I suggest this is a complementary situation. We clearly need randomised trials when new technologies come onto the market to see to what extent they can be beneficial or possibly harmful. On the other hand, once technologies get into practice, the question then becomes for whom are they best and that is where a CPI study can discover, for specific types of patients, which of the multiple treatments are associated with the best outcomes. Also, there are situations where findings from CPI studies suggest new randomised trials. For some of these findings, we would never have hypothesised them and no institutional review board or ethics committee would have approved such a study until we had some evidence to question what we were actually doing and whether there was some different way to improve care. Thus, the two study methodologies — RCTs and CPI — are complementary methodologies: we need both of them, because sometimes findings from randomised trials do not

36

2002 General Practice and Primary Health Care Research Conference — Proceedings

translate into best practices in actual practice of care. The big picture of the CPI methodology is to connect outcomes with detailed process steps, adjusting for how sick patients are.

For the first twenty years of my research career, I was on the faculty of The Johns Hopkins

University and developed severity of illness measures for every one of about 19,000 ICD-9 codes.

Here are the criteria to measure severity for adult patients with pneumonia. This is the list of codes that get mapped to the pneumonia criteria set, because they have similar signs and symptoms that indicate how sick the patients are. You see on the left-hand side the various body systems that get involved with patients with pneumonia. Within the body systems there are criteria and their signs and symptoms, levelled from normal to mild, up to catastrophic or life threatening. Thousands of these criteria sets have been created, so that no matter what disease or disease combination a patient may have in a CPI study, we have every one of their diseases recorded, all the relevant signs and symptoms — the more abnormal the signs and symptoms are, the higher the severity scores. Sometimes we analyse the data using an overall severity score; other times, we look at the criteria level to see how abnormal the criteria are, such as, does the child have apnoea or cyanosis. Whenever clinicians say a child is different or a patient is different because of certain criteria, we know that we are going to have the criteria because of the comprehensive way we have measured how sick patients are.

That is a description of the overall CPI framework. Let me now share with you some surprises from past CPI studies. We have applied CPI technology in ambulatory settings, inpatient settings, and long-term care settings. Some of our current studies are in the areas of hospice care and in poststroke rehabilitation, but those data are not yet analysed. I am going to start with the end of the story and tell you about one of the biggest surprises from the analyses of a very large ambulatory

CPI study we performed. We were looking for factors associated with better cost control in ambulatory care. One of the biggest surprises was that the more access to medications to treat very common ambulatory diseases was curtailed, rather than saving money we found restricted access associated with more office visits, more emergency department visits, more hospitalisations, more concomitant medication use, and increases, not decreases, in cost of health care.

Let me share with you the findings in a little more detail from the study that you see here. This was our study question. When one looks across multiple managed care organisations at a year’s worth of actual data on the care of thousands of typical patients treated by their regular doctors

(this is the actual practice of care), how is the amount of health care services used associated with certain cost-containment efforts by managed care organisations. In the United States, managed care is responsible for the total care of patients: all of their ambulatory care, their hospital care, their emergency room care, and their medications. So we can put together a comprehensive picture of all utilisation. Normally, patients are in a managed care plan for at least a year’s time — that’s the sign-up interval. In this study the managed care plans and patients came from all across the United States, both the east and the west; half the plans were for profit, half were not for profit and they had various cost containment practices in place. The patients were selected for study if they had at least one of five very common ambulatory diseases: ear infections (usually younger people), asthma (a mixture of both younger and older), arthritis, hypertension, and stomach ulcers. About 70% of ambulatory visits in the United States will have one of these diseases recorded. In total, across the 6 HMOs we followed the care of 13,000 patients over the year. They had over 99,000 office visits, for which we measured the Comprehensive Severity

Index score at every visit, over 1,000 hospitalisations, and used over 240,000 prescriptions. So this is a large database to try to discover to what extent the cost containment practices actually contain costs and utilisation.

In the analyses we have adjusted for many possible confounding variables, such as severity of illness of the patients. I know you will not be surprised to see that sicker patients use more of everything. We expect this, but we wanted to see what happens after controlling for severity of illness, age and gender, number of months in the study (for about 3% of the patients we did not have a full year’s data, so we controlled for this), number of different providers (to address continuity of provider), various HMO characteristics that we did not consider cost containment strategies, and finally this list of cost containment strategies. Strategies like 1) requiring a second

2003 General Practice and Primary Health Care Research Conference — Proceedings

37

opinion before having a surgery done, 2) having a gate keeper such that you couldn’t see a specialist until the primary care provider approved it, 3) how high were the drug co payments, 4) how high were the visit co payments, 5) how limited were providers in terms of the drugs that could be prescribed without prior authority, and 6) what was the extent of use of generic drugs.

Most factors turned out as expected, for example higher drug co-payments were associated with less drug use, and higher visit co-payments were associated with fewer visits. But the biggest surprise was that for each of the five study diseases (which we analysed separately because of their different patterns of utilisation), we found that greater limitations on access to medications for that disease — having fewer drugs available that a doctor could prescribe without going through a prior authority process — was associated with more patient visits to physicians, more emergency room visits, more hospitalisations, greater cost of prescriptions no matter what discounts were used, and greater numbers of prescriptions over the year. Although this is derived from big regression equations, it is hard to visualise a regression equation, so I tried to draw it for you here and show you a few examples.

Here are patients with asthma and we are looking at the outcome of number of prescriptions over the year. All the other variables on the previous slide have been controlled for except for two of them. Here I show you what happens with severity of illness. Severity is really a continuous score because it includes all diseases, all their signs and symptoms, the more abnormal the signs and symptoms, the more severity points at a visit. We then add up the severity scores for all a patient’s visits over the year and end up with a severity sum score between 4 points at the low end to 500 or 600 points at the high end depending upon what happened to the patient over the year. Although the continuous severity sum score goes into a regression for that patient, what I have done to draw severity here is to take, e.g., the asthma population (there are about 3,000 asthma patients), and group together the bottom 30% of the continuous severity sum distribution under low severity here, the middle 40% under medium severity, and the highest 30% under high severity. So if you take any site and go from low severity to medium severity to high severity for that site, you see severity does exactly what you think: sicker patients have more prescriptions.

But after controlling for severity and all the other variables, when we order the sites from least restrictive for asthma drugs to most restrictive for asthma drugs, we find this pattern of utilisation: greater numbers of prescriptions for more drug-restricted asthma patients over the year. When we first found this, we thought maybe there are more prescriptions but if they are less expensive prescriptions, then the cost could be the same. But that was not what happened.

I will show you another disease, arthritis, with cost of prescriptions as the outcome, and now ordered from least restrictive to most restrictive for arthritis drugs. You see a similar pattern of lower cost at the site that did not restrict and highest cost at the site that restricted most. Here is the outcome of visits for another disease, ulcers, again ordered from least restrictive to most restrictive for ulcer drugs, and you see a very similar pattern. This was a great surprise to the managed care organisations because, of course, the hope was that if they restricted access to medications, and emphasised use of less expensive medications, then they would save money and not hurt quality. What we found was that restricting access to medications did not work in that manner. And in fact in the United States today about 80% of the managed care organisations have given up on the practice of requiring prior authority before certain drugs are prescribed, because they have found it has not been cost effective in saving money. What is the most cost effective thing to do is to discover what medications and other treatments are associated with getting a patient better as quickly as possible, because when you get a patient better, that is when all the costs go down. It doesn’t mean that we have to look at the cost of each item separately and minimise the cost of each item. The goal is capture the overall picture, and strive for the lowest total cost. Sometimes we have to spend more on a component or two to save overall.

We also separately studied data from elderly patients (>65 years old) to see if they were at even greater risk if you limited access to medicine and found yes, they could be. For example, here are patients with hypertension. This is the non-elderly group (<65 years old), and this is the elderly group. We asked what happens when we limit the access to fewer loop diuretics. This is what I want you to note, the significance level. Big numbers here mean not statistically significant. So limiting loop diuretics was not statistically for the non elderly population, but for the elderly

38

2002 General Practice and Primary Health Care Research Conference — Proceedings

population it was highly statistically significantly effecting greater drug use, which may be due to elderly being more frail.

Another area in the United States where people try to save funds is in mental health services. The thought here is that people may not be using mental health services appropriately, so managed care restricts what drugs can be used and how many mental health services a patient can have over a year. We studied this with the help of a number of psychiatrists from various institutions across the United States. We found that when you limit mental health services, you actually can end up spending a lot more. Let me share with you some of the results. First, in our population of

13,000 patients, we found about a third had either a psychiatric diagnosis or were taking a psychiatric drug at sometime throughout the year, although they got into our study because they had medical conditions. In other words, there is a lot of psychiatric co-morbidity in this medically ill population. We looked at the subset of patients who had a diagnosis of depression or were on an anti-depressant and asked the question: To what extent did they get mental health services? What you see is that fewer than 15% of them, whether they were non-elderly or elderly, ever saw a mental health provider. Many saw primary care providers, they saw their medical specialists, and they saw surgical specialists, but even though they had a diagnosis of depression or were on an anti-depressant, they did not have much access to mental health providers, because of the restrictions that are in place.

Again for patients with depression, we asked what kinds of medications they were on. Because of restrictions on psychiatric medications, you see much higher utilisation of the older tricyclics and benzodiazepines, and much lower utilisation of the newer SSRIs. So Dr Steven Bartels, a psychiatrist from Dartmouth University, asked about the rest of the care. We looked at all the medication use of patients in our study over the year. The yellow bar represents patients who have no psychiatric co-morbidity and no psychiatric drug use over the year. The next bar is for patients with a psychiatric diagnosis, and the last bar is for patients with a psychiatric drug. Notice the utilisation of those with a mental health problem. Utilisation is quite similar whether we identified patients by a psychiatric diagnosis or by a psychiatric drug. Also notice how much greater their drug utilisation is compared with patients with no psychiatric comorbidity, even after removing all psychiatric medications. Thus, patients with a psychiatric problem that, according to Steve, is not being treated optimally by limiting mental health services and limiting optimal medications, use about 50% more non-psychiatric medications, they had about 50% more non-psychiatric visits, they had about 50% more non-psychiatric medication costs, more emergency room visits, and more hospitalisations. In other words, we think we are saving money by limiting access to mental health services, but in the end what we are really doing is spending much more on the rest of their care. What we have to do is to find out what are those treatments that get patients better as quickly as possible. Because when we try to limit access, such as limiting access to medications or limiting access to mental health services, we find that such practices are associated with higher utilisation overall, not lower utilisation. This is really a systems problem. We cannot manage health care optimally in silos: Silos of mental health visits, silos of medications, silos of general practitioner visits, silos of hospitalisations. It is only by considering the total picture that we are going to achieve not only better quality of care but lower cost overall.

A large portion of health care in the United States is funded by employers for their employees. One thing that happened over the last ten years, as managed care grew in the United States and limitations were put in place to save money, was that employers began to notice that their employees were getting less productive. Researchers at MIT developed definitions of worker productivity, which they call ‘presenteeism’ measures: you are at work but you may not be functioning very well. Previously they could measure when you were not at work, that’s absenteeism, but they began to measure a presenteeism concept because people, by not receiving the treatments that would get them well as quickly as possible, were becoming less productive.

Some of the research published in 2000 and 2001 looked at restrictions beyond their impact on the health care system to examine the direct impact on the economy. This research found that newer drugs were associated with more active and productive employees (based on the new productivity measures), less staff absenteeism, and lower employee turnover.

2003 General Practice and Primary Health Care Research Conference — Proceedings

39

What is it that we should be doing? An example of a possible solution comes from a study to address the question: Is it best to limit access to antibiotics? Antibiotics was one of the first drug classes to be controlled by formulary, because administrators felt that if providers were limited to only a sub-set of antibiotics, that they could both control costs and also control possible over utilisation. The study compared outcomes in 1988, when the facility had a formulary for antibiotics that could be prescribed without prior authorisation, with outcomes in 1994, when a patient-level algorithm was created and implemented based on patient characteristics to prescribe antibiotics.

The algorithm for antibiotic ordering had the following characteristics. The first thing was, since we know that antibiotics can be associated with toxicity, one wants to make sure someone does not have kidney function problems. So the algorithm checks a patient’s creatine clearance and selects the antibiotic that is best to not cause toxicity based on the creatine clearance level. Then the algorithm looks at other relevant vital signs, what other drugs the patient is on, what diet the patient has, etc. The only way antibiotics could be ordered in 1994 in this facility was by the physician going to a computer program, answering the above questions (the creatine clearance and other relevant factors about the patient) and up on the computer screen would come the recommended antibiotic to use, the dose, the route, the duration, that was based on data analyses that were done over the previous years. If the physician did not want that drug, the physician could press another button and up on the screen or subsequent screens would come all the drugs the computer program had considered with a short message as to why it was not selected first.

Then the computer asked the Doctor, what do you want to do? There was no compulsion for the doctor to use the first drug that was selected, but the selection was based on multiple dimensions about the patient that probably the clinician could not have processed in his or her head as well as the computer could process these multiple dimensions to select the optimal drug to use.

This is what happened. The percentage of patients receiving antibiotics between the two time periods went from 32% up to 53%. That is probably because many patients, who may have been less sick back in 1988, are no longer being admitted to hospital. Regarding receipt of broadspectrum antibiotics, when there were restrictions, they made up 24% of antibiotics, but with the algorithm this almost doubled to 47% of antibiotics. Looking at the case mix index, an index we use in the United States to measure how sick a population is, notice that the population was actually sicker in 1994, since higher numbers means sicker patients. So we have a sicker patient population, using more broad spectrum and expensive antibiotics, and many more patients are receiving antibiotics in 1994 compared with 1988. However, the total antibiotic acquisition cost went from just under a million dollars in 1988, down to just over six hundred thousand dollars in

1994, or three hundred and seventy-five thousand dollars lower cost. Not all drug categories went down because they only had an antibiotic algorithm in this facility, but antibiotics went from 25% of the total pharmacy budget in 1988 down to 13% in 1994. Per treated patient, costs went down by $70. Researchers looked at appropriate utilisation for prophylactic antibiotics, empiric antibiotics, and therapeutic antibiotics. Here are some data on prophylactic antibiotics. This institution prides itself on making sure that all patients get their prophylactic antibiotic preoperatively within half an hour before incision. This occurred 99% of the time in 1994, compared to 40% in 1988. However, because the algorithm recommended the antibiotic, the dose, the route, and the duration, patients received an average of 5.3 prophylactic doses in 1994 compared to 19 doses in 1988. Antibiotic-associated adverse drug events decreased by 30%, and antibioticassociated mortality went down by a full percentage point. We need more of these algorithms.

Unfortunately, the best drug decision isn’t as easy as yes, you can use this drug, or no, you can’t use that one without prior authority. If it were that easy, it would be very nice. But we have shown that by putting together patient-level algorithms to make decisions based on patient criteria, we can get the appropriate medications when we need them, and have better quality along with lower cost.

Let me switch over to the long-term care setting and share with you briefly some findings applying a Clinical Practice Improvement methodology in that setting. These data come from 109 long-term care facilities all across United States, and include data on the care of 2,500 very frail elderly residents all at risk of, or having a pressure ulcer. Many variables were significant when we examined the outcome of developing a pressure ulcer, if the residents did not have a pressure ulcer at study start. We studied factors about residents and their treatments that were associated with development or prevention of pressure ulcers, things like incontinence, pressure relief, etc.

40

2002 General Practice and Primary Health Care Research Conference — Proceedings

What you see under the heading general assessment are resident characteristics: positive factors were associated with significantly greater development of pressure ulcers, and I don’t think any of these would surprise us. The very oldest residents were more likely to develop pressure ulcers, also males, sicker residents, residents who had a history of a pressure ulcer, etc. You can’t do anything about these risk factors, that’s who you are treating in this situation, but at least it tells us who are at higher risk and who we want to be sure to give the best treatments to prevent pressure ulcers.

Let me share with you what we discovered about the best incontinence treatments for residents.

We discovered that using disposable briefs, or, if the residents could do this, being on a toileting program, were associated with significantly fewer pressure ulcers developing. That was a surprise because many thought that the best way to keep residents as dry as possible was to catheterise them. Providers thought if residents were catheterised, then their skin would be dry and there would be less skin breakdown. But what we found was just the opposite. Mechanical catheterisation was associated with significantly more pressure ulcers.

Another aspect that we studied was nurse staff time. How does nurse staff time either by registered nurses (RNs), certified nursing assistants (CNAs), or licensed practical nurses (LPNs) affect development of pressure ulcers. Notice the negative coefficients here. These are thresholds of minimum amounts of nursing care that were associated with fewer pressure ulcers. I will show you more about that in a moment, and also about medications, but with regard to nutrition, we found that if residents were either dehydrated or had weight loss, they were more likely to develop pressure ulcers. However, if they had a fluid order or oral nutritional supplements or enteral supplements, they had significantly fewer pressure ulcers.

Let me show you some of the bivariate associations. For example, I know you will not be surprised to learn that enterally fed residents were the sickest. Consequently, you would expect them to develop more pressure ulcers. What we discovered was they actually had the fewest pressure ulcers developing; that’s what the negative regression coefficient was telling us. Also, the residents taking oral standard medical nutritionals had fewer pressure ulcers developing compared to residents who had no nutritional risk or those who were given the standard of care ‘food first’. The words ‘food first’ mean: when the resident comes in, put a nice meal before him/her; they will eat when they are hungry and that’s the way to make sure they have the best nutrition. Frankly, we found this did not work well. We really needed to supplement residents almost from the start: within 24 hours one needs to assess whether or not this resident is going to eat enough, because if you do not recognise an eating problem early enough, more pressure ulcers develop.

Here are data on toileting programs and disposable briefs. Notice the disposable briefs users had

27% pressure ulcers developing compared to residents using re-usable briefs, where 35% developed pressure ulcers. Notice almost 54% of the residents using mechanical catheters developed pressure ulcers. Again, we need data to understand for specific kinds of residents, which of the many treatments are associated with the best outcomes.

Another area that we examined in this data set was the affects of different medication use. One of the most difficult groups of residents to treat in long-term care are residents with agitation and dementia. These residents are often screaming, they can be violent, and they are really a challenge to the long-term care staff. There is a guideline in place in the United States, developed by experts using whatever data that they had available. The guideline in Australia to handle this situation is the same and this is how the recommended guideline reads. First, do not use chemical restraints on these residents, so the first message is ‘use the fewest number of medications possible.’ In particular, in the United States before providers can give these residents any kind of medication, they have to document sufficiently the behavioural problems to justify medication, otherwise our government criticises them. If a provider documents specific behaviours, the recommendations are to minimise the use of benzodiazepines, and to use the newer atypical rather than typical antipsychotics if giving an antipsychotic. If a provider gives an antidepressant, use an SSRI, and finally, avoid combination therapy. Now, this is a typical guideline and the one in

Australia reads the same way. It is not very precise about what to do, but I think we get the overall picture: we don’t want to over medicate and we don’t want to use poly pharmacy.

2003 General Practice and Primary Health Care Research Conference — Proceedings

41

The people who developed this guideline heard that we had a long-term care data set that included all the medications along with all the other treatments, and asked us to see if this guideline was working in actual practice. This is what we found. First, we had 800 residents who qualified with the criteria of having dementia and agitation. About one third of them got no psychiatric medications. When we asked the long-term care facilities why, they said, frankly, they didn’t have enough staff to document the behaviours, so it was just easier to give them nothing, so the

Government wouldn’t criticise them. That’s what happened in practice. Then, because there really wasn’t any data to support one medication over the other, about a third of them received antipsychotics, about a third received anti-depressants, and about a third received anti-anxiety agents, so naturally we had some patients with combination therapy. To everybody’s surprise, after much analysis, we found that what was recommended against was actually associated with the best outcomes. Combination therapy of any two or more psychiatric medications compared to mono therapy, was associated with significantly fewer hospitalisations and emergency room visits, significantly less urinary incontinence, and significantly fewer pressure ulcers. Particularly if the combination included an SSRI, there was even lower utilisation and fewer complications as well as less suffering. The person who led this analysis from Dartmouth University, at the time the study was being done, had just been elected the president of the Association of Geriatric Psychiatrists in the United States. His inaugural address had the following title: ‘Combination therapy for residents with agitation and dementia: Bad practice with better outcomes’. He has shared these results with many colleagues and has told me he continues to get calls from colleagues across the country, both nursing colleagues in long-term care as well as physician colleagues, saying they are finding it remarkable that by using combination therapy, these residents have become so much better controlled and are not suffering all the extra hospitalisations and pressure ulcers. Dr Bartels noted that this is an example of a study that could not have been done as a randomised trial, because everybody would have thought combination therapy was bad. It wasn’t until we had the data that we could even question combination therapy. If he had proposed a randomised trial to look at combination therapy, no Institutional Review Board would have approved it, because they would have thought it might harm residents. Thus, we really need various kinds of study methodologies, including CPI studies from the actual practice of care, to discover best practice.

Let me show you one other cost saving mechanism that has not worked in the United States. You may have similar problems here in Australia. The question is what is the affect of nurse staffing.

The picture is that nurses are expensive, and consequently a way to save money is to have fewer registered nurses (RNs) and more lower level staff. In our long-term care database we had nurse time and the staff type for each of 2,500 residents. What I am showing you here is the registered nurse time per 10 minute increments. The average in the United States is 12 minutes for RN time/resident/day, so we had a lot of people down here at less than 10 minutes per resident per day, and at 10–20 minutes, with fewer at 20–30 and 30–40 minutes. Notice the rate of pressure ulcer development. The greater the amount of RN time per resident per day, the fewer pressure ulcers developed. You saw the negative regression coefficient before. This is what’s underlying that negative coefficient. More RN staff time was associated with significantly fewer pressure ulcers developing. Also with the cut point of 45 minutes here, more LPN staff time was associated with fewer pressure ulcers, and finally more CNA time was associated with fewer pressure ulcers developing. When we put it altogether in a big regression equation to see the effects of nursing times: RN, CNA, and LPN along with all the other variables, we found significant effects. What we see looking at the level of significance here of the Chi squared variables is that the biggest effect comes from 30–40 minutes of RN care per day. Remember, the average in the United States in 12 minutes. We also found the greater the RN staff time, the fewer the residents that were hospitalised. There was no effect from LPN or CNA time here. When we put it altogether in regression analyses you see again the effects of the RN staff.

The long-term care facilities said they thought they were saving money by reducing RN staff time, but let’s look at the bigger picture. Suppose we changed residents getting an average of less than

10 minutes of RN time per day and we gave them 30–40 minutes of RN time per day; that would be adding 30 minutes of nursing care a day per resident, which is expensive. In fact, a nurse’s salary in long-term care has a median of $48,000 and a high of $60,000 in the United States.

Depending upon which of these values you use, the cost of an extra nursing 30 minutes a day per

100 residents is between $415,000–526,000. However, by preventing pressure ulcers (comparing

42

2002 General Practice and Primary Health Care Research Conference — Proceedings

the pressure ulcer rate at less than 10 minutes/day with the rate at 30–40 minutes/day), which cost an average of about $7,000 to treat, we save more than $800,000 per year. By preventing those hospitalisations we save more than $500,000 per year. When you finally add these up, the net savings to the system is between about $850,000–950,000 per 100 residents. In other words, by increasing RN time to 30–40 minutes per resident per day we have a net expected annual savings of about $900,000. That’s what we have to look at. We must look at the total picture, not just that RNs are costly and hence we can’t afford them.

In closing, an example I want to share with you are some findings from an inpatient surgery study.

We studied about 1,000 patients with abdominal surgery and some of the significant factors were surprising. For patients with bowel surgery we found using nutrition, total perenteral nutrition

(TPN) was initially associated with poorer outcomes, in particular, the outcome of length of stay.

That was a surprise. We thought nutrition support should be beneficial after controlling for many other variables. Another big surprise was using patient controlled analgesia (PCA pumps). That was associated with significantly more wound infections. Let me show you what we have discovered after researching these in greater detail. First for PCA pumps. For patients with bowel surgery in our study, 12% developed post-operative wound infections if they used a PCA pump compared to 4% if they had the identical drugs but by the nurse giving them an injection. In particular, for deep wound infections it was 4.2% for PCA pumps compared to 1% for the injections. These both were statistically significant. Where we found the biggest difference was as follows. If we had a patient with a clean contaminated wound, in other words their bowel had not ruptured in this situation, the rate of infection for PCA pumps was 8.6% compared to 3.8% for injectables. That was significant. But here is what we think is going on. Notice for patients that had either a rupture or a contaminated wound, the rate of infection for PCA pump users was 24% compared to 3.5% for the same drugs as injectables. What we think may be underlying this (and several randomised trials have been proposed to study this) is there may be something related to

PCA pump use, lack of pain, and the immune system. Where you need your immune system really reved up is when you have a rupture. There is some suggestion in the immunology literature that if we don’t have at least a certain amount of pain, the immune system doesn’t rev up enough and hence can’t fight the ‘bugs’ from the rupture. All of us thought, no pain, that’s the way it should be. In fact in the United States, our joint commission makes providers measure pain as the 6 th vital sign to be sure that patients don’t have pain. If I knew these data and I was having bowel surgery, I would want a little pain in order to avoid an infection. These results will appear in the journal Surgical Infections in July.

Let me just share with you what we found about the TPN variable. When you think about giving nutrition support you would think that providers would give enough to meet calorie goals for a patient. What we found was that providers were giving nutrition support but too late and they weren’t giving even 60% of the patient’s protein and calorie needs. So we looked at the data and saw the following picture. We asked what is the optimal time to start giving nutrition support and how much should you give. We discovered the optimal time was within 48 hours after surgery. If you give nutrition support within 48 hours after surgery, that’s what’s we called ‘early’, and if you give them enough (we determined the threshold was 60%), that’s what we called ‘sufficient’. We found patients who were fed early and sufficiently, compared to the other three possibilities here, were the sickest in terms of our disease severity score. Also if we looked at the nutritional component of the disease severity score, they also had the highest nutritional severity. However, we found those that were fed early and sufficiently had a day and a half to three days shorter length of stay, even though they were sicker, and between $2,000 and $5,000 lower cost. You would think if they were sicker they would have longer length of stay and higher cost. Again, we need data to discover what are the best things for specific kinds of patients and when we discover what those best treatments are, we are able to have enormous savings on cost.

Something that we discovered in a very large study of infants with bronchiolitis, the average age being about 3 months old, was for infants who stopped breathing, it was almost always equally good to tickle their feet as it was to incubate them. We can explain more of that in the question and answer period! But that was a big shock to those physicians who were trained when a little baby stops breathing, the first thing you do is to incubate them. They said it was not possible that those infants were the same as the ones who got out of the episode with tickling the feet, but we

2003 General Practice and Primary Health Care Research Conference — Proceedings

43

were able to show them with the data that they were the same and you got just as good clinical outcomes and spent about $8,000 less in cost when you tickled feet compared to incubating them.

To close, I think you see there is a lot of data involved in the Clinical Practice Improvement approach. There is every detail any provider on the study team can think of for patient variables they want to control for, all the treatments that are going on, and the outcomes. To be able to support such studies, clinical information systems (CIS) are critical. Consequently, if you have a

CIS that gathers these patient, treatment, and outcome variables in a rigidly structured format so that you can get all the levels of the detail, you will definitely recoup your investment. Clinical

Practice Improvement studies help to construct such details for a CIS. I tell people in the United

States, whatever you think the CIS will to cost you to implement, the savings you will get after implementing it to do two things: more quickly discover what are the best treatments for specific patients and more easily implement change in practice because as soon as specific patient characteristics are found, up on the computer screen can come what are the best things to do, will far outweigh the cost of the CIS. Following the Clinical Practice Improvement discovery phase for a study, we have shown that after implementing the newly-discovered optimal processes of care consistently, sites have demonstrated savings between 30–50% of what they were spending to treat that condition previously. That’s the return on investment, but the cost is in gathering the data first and analysing them.

In summary, to reduce costs and simultaneously improve outcomes, we need comprehensive patient, treatment, and outcome data from the actual practice of care. If such data were in clinical information systems, CPI studies would be much easier to conduct and ultimately information systems will pay for themselves. Society will benefit in the end, the more we can do these kinds of projects and share our results together. For those of you who might be interested, we have written a textbook on how to do CPI studies and I know the Health Department here is going to have several copies if you would like to borrow one.

The PowerPoint slides for this address are available at: www.phcris.org.au/events/conference_frameset.html

44

2002 General Practice and Primary Health Care Research Conference — Proceedings

More than just political correctness: making consumer participation in primary health care research a productive reality

Hilda Bastian

International Convenor, Cochrane Consumer Network

I am not going to put up any data today, but I thought I would bring in a ‘heavy-weight’ to help me along today, so I picked on Einstein — figuring he would be an accepted expert on science, to balance some of the things I wanted to say. He said: ‘Problems cannot be solved at the same level of awareness that created them’. He was talking about people broadening their minds and involving other people in the activities they were involved in, to try and find other solutions to things. Einstein’s idea was that there is something valuable in involving not just the ‘usual suspects’ in any kind of activity around thinking. After twenty years of supporting consumer participation in health research I suspect this is one of the main things that consumers do.

Among other things, what consumers bring to the research process is passion and interest in the issue, because it affects their lives so much. (This isn’t a negative comment about randomised trials I feel strongly that people should not experiment with treatments without having a very good idea that they work, and that takes randomised trials). Another thing consumers contribute is the individuality of the experience. The perspective you get when you’re involved with research really does depend a lot on individuals. What happens and what’s important might not be what people expected, and it’s really useful that consumers are there to help investigators keep a perspective on reality. For example, I have just been with consumers talking about research and research results with researchers in the Cochrane Collaboration. The discussion was about research into sound medical advice for people with low back pain. In particular whether the advice should be to go back to work, or stay in bed and take a few days off. The research had concluded that it was much better to tell people not to take some time off, but the outcome that they had chosen for this research was return to work. My point is that this research had proceeded for a very long time without anyone looking at the irony of the principal outcome measure chosen, which could just be the result of not giving people a medical certificate. After all, people often go to the doctor just to get the medical certificate they need to have some time off work. If the intervention is advice to return to work, and no certificate for time off is forthcoming, of course some people will be back at work sooner. This may say nothing though, about the state of their backs. Somehow this managed to get all the way to Level 1 Evidence without anybody querying the logic underpinning the choice of outcomes. I think that’s the kind of classic theme that a consumer, who hasn’t participated in the spiral of thinking, can comment on by saying: ‘What are we actually doing here?’, ‘Why and what does this mean?’, and ‘In the real world something different is happening to me’.

Einstein said that: ‘Science can only ascertain what is, not what should be, and outside of this domain value judgements of all kinds remain necessary’. There are value judgements at every single step within a scientific process as well, and I think that this is one of the things that sometimes people lose sight of. It’s important to have a range of different values involved in decision making at all sorts of key points, because, very often, the outcomes chosen, the population chosen, or whatever, has potential to influence the outcome of the research.

Consumer participation, whilst important, can sometimes be a very unpleasant experience. Those of us consumers, who have been involved in health research for a long time, find that at times it can be a damaging and disempowering experience. It’s not enough to just involve us in token ways. We all have a lot of other things that we could be doing, as many of you doing research have other things that you could be doing. What is important is that participation isn’t just about a politically correct process; it should be done in a way that allows for appropriate changes to happen. In other words, consumer participation shouldn’t be a WOMBAT, which, as well as being the name of an animal, stands for Waste-Of-Money-Brains-And-Time. It’s very important to all of us that in every aspect of health care we are not wasting money, brains and time. Unfortunately a lot of the kinds of activities that I have personally been involved in, and that I have watched consumers be drawn into, actually were WOMBATS, in quite profound ways.

2003 General Practice and Primary Health Care Research Conference — Proceedings

45

Some of the ways that consumer participation can go off the rails, and become a waste of time from a consumer point of view, is often because of just plain poor communication. For instance, when we are not really listened to, or people haven’t had the experience or developed the skills or know-how to convert the kinds of concerns and issues which consumers bring to the table into something that changes the research. I have sat in on discussions between consumers and researchers and heard consumers make comments about things that I recognise should fundamentally alter the entire course of the research project, but see researchers still sit there and go: ‘Well that was all very interesting, but I cannot see that changes anything.’ Because consumers aren’t talking like researchers and they are not converting their concerns and issues about the health care topic into suggesting a sub-group analysis or whatever — their input isn’t converted into concrete changes to the research.

I think there is both a level of people needing to see that there is some value in consumer participation, and also of researchers learning some skills about how to actually incorporate consumers’ knowledge and real life experiences of particular diseases into research. I don’t think there has been enough work at all on these areas.

Participation can also be a WOMBAT for us [consumers] because the research itself simply wasn’t worthwhile, or it was not of interest to us. Other times consumer participation doesn’t work because people have got the wrong consumers involved — those with the wrong motivations, the wrong reasons or whatever. There may be a whole variety of reasons why enough effort wasn’t put into finding the right people, or simply that the right people weren’t there, but what resulted was that a lot of effort was put into something that was actually never going to necessarily work.

Another reason why these relationships come adrift is the assumption that consumers have similar expectations to researchers, including shared knowledge and values. Researchers, who are often immersed in their own culture, cultural expectations and processes to do with research, often don’t appreciate the extent to which people coming from the community don’t share in these understandings or have the same expectations. These inadequate understandings it may be as basic as not understanding contract time-frames (that a person they [consumers] are putting a lot of effort into developing a relationship with is just a short term project officer and will be gone the next year), through to critical things about what’s going to happen because of the research, what results can be published or made public, and so on.

Another significant thing that can make consumer participation a bit of a WOMBAT for us

[consumers] (that I have seen a lot of in areas where communities were involved — community development and health projects in developing countries, in research in indigenous communities and in consumer participation in research) is participation that effectively (not intentionally) becomes an exploitive relationship. In these situations it’s very clear what’s in it for the researchers, and maybe, if the research ends up having a really big impact there will be a clear benefit down the line for the community, but very often one side gets PhDs, publications, etc., and the other side doesn’t even get to hear the results, let alone their names acknowledged, or their time paid for.

These issues are really important to effective processes of engagement. The very way that something is done can be empowering or disempowering to the people involved. If researchers want consumer and community people to stay involved, and to pass on to others’ in their community the process is worth working with, the consumers have to have good experiences. If they don’t, this is the knowledge that will be shared, and pretty soon we all start to tell each other: ‘Look out for them, it’s not going to be really worth your while’.

In a lot of ways the relationship between researchers and consumers in research is very similar to the one that happens when you start to have more patient-centred consultations. If you are genuinely going to do that, it changes some things. You need to start to reconsider things like time-lines and what your budgets are for and so on. This is important if you want to genuinely talk to consumers, involve them and bring them along in a project in a way that helps them really feel comfortable enough to say things. This is when the really valuable input can come. What needs to be understood is that it takes a lot of confidence for consumers to get to the point where they feel like they can say things that are challenging, this takes time and feeling safe.

46

2002 General Practice and Primary Health Care Research Conference — Proceedings

Another thing, is that what results from consumer participation may not be what was expected.

For example, when people started spending more time talking to patients, informing them more and not just making treatment decisions on their behalf, one of the things that happened was that sometimes people didn’t make the choices that the experts expected them to. For example, when they researched what happens when you tell women of low socio economic background the evidence about mammography screening (the idea being that it would increase the uptake rate of mammography) what they found was the reverse often happened, that people said: ‘Well no, I don’t want to do it’, which was not the intention of informing people. It can be the same with research; getting involved in a genuine engagement with community groups and consumers can take you into places that you absolutely never expected to go, which means it might actually be quite difficult to what you expected.

Even when researchers want to work with consumers, and do so effectively, its actually going to take quite a long time I suspect until consumers really start to come into the picture. When they do, it will mean some quite fundamental things will change in research, particularly about what is going to get researched, what the priorities are and so on. It also means that some things that people have just totally overlooked, may turn out to be the most important thing of all. We all need to recognise that just as it’s taking a long time and a really large process to try and make consumers the centre of the health care system, it’s also not going to be an easy road in research.

In the interests of getting on with it, we need to have as many good experiences as possible, to make sure there are as few cynical and disillusioned around saying: ‘Don’t worry about being involved, they don’t really want to hear what you think’ as possible. Some of the things that I would have expanded on here are going to be in the, soon to be issued, new Statement between the Consumers’ Health Forum of Australia (CHF) and the National Health and Medical Research

Council (NHMRC) about the relationship between researchers and research participants and consumers in the community. The Statement is a start in setting out some understandings and expectations to do with good practice when involving consumers in research.

Arnstein’s classic ladder of citizen participation is relevant to this process in research (Arnstein SR

1969. A ladder of citizen participation. AIP Journal, USA: 261):

Citizen control

Delegated power

Partnership

Placation

Consulting

Informing

Therapy

Manipulation

What I would say is that traditionally most of the involvement of consumers in research, once you take out some areas like AIDS and Indigenous research and now breast cancer and a few others, largely hovers down around the lower rungs, e.g. at the informing/consulting levels of participation, with maybe a bit at the even lower placation level. To get higher up the ladder, to the levels where it is actually genuine partnership, or in the case of areas like Indigenous health, genuine citizen control requires a whole lot more effort. This is a much, much, larger agenda.

Generally speaking when consumers and community come into a research process and we hear words like partnership or whatever, we are thinking of something really quite high up on the participation ladder. In reality what we often experience is a very, very, passive role usually being involved as objects for data extraction. I think there is a lot that researchers can learn, and apply from what has been learnt through dealing with Indigenous communities and in working in developing countries. The processes are very much the same as what will amount to genuine engagement with consumers and the community. Some of the things that are important are relationships, including long term relationships and also that in research there are going to be shared benefits and so on.

Guideline documents for involving consumers in research that Indigenous communities, and the

CHF/NHMRC have developed are saying that it’s about sharing skills and whatever and that it isn’t

2003 General Practice and Primary Health Care Research Conference — Proceedings

47

just about money. It is also about people who are leaders in their communities walking away with a bunch of skills that they didn’t have before. This a really important thing that the research community has to offer people in the community, because a lot of these skills to do with how you work through things systematically, and work out what really works and what doesn’t, are very important for people who are activists in their communities and will be exerting lobbying force in all kinds of ways of trying to change systems to improve their lot. This is another way consumers can benefit from involvement in research.

When consumers are involved in research in the end what is actually important to them is to know that there is going to be some action come out of research. Just doing research for its own sake is not usually going to be satisfactory to consumers or community groups. They usually try and help actually make some changes happen because making people better or finding out what can they do about a problem is really important to those who are actually experiencing the problem/s.

Often in research it seems to me it’s almost as though people have got exhausted by the end of it, the money has run out and there’s all sorts of other things that have happened, which mean that the actual real value that research could have can’t be put to use. My point is that change as a result of research doesn’t happen as much as it should, and that involving people from a community who are already trying to make changes, is one of the ways you can do that.

In the literature the Cochrane Collaboration, and our Consumer Network, has been used as a good example of how consumer participation in research can work. A lot of the work we have done has been quite important, and changed the direction of individual pieces of research. With people in research currently wrestling with setting priorities for what research is going to be done, consumers are putting onto this agenda issues like diet in multiple sclerosis and prostate cancer, and really pushing non-medical forms of care that they want to see happen.

Also being drawn into the agenda of the Cochrane Collaboration are issues about supporting making something happen with the research at the end of it. In the last issue of the Cochrane

Library there was a review about whether it was worth your while to set aside some of your own blood for a blood transfusion to avoid risks of infection if you were having an operation. These findings were really important for consumers, very important actually, but we found out that the

Red Cross Blood Transfusion Service didn’t know that any of this research that had been going on.

This is an example of how something we were doing to get the word out there to communities about this information started to alert other important stakeholders to the important issues that were emerging from research.

What helped consumer participation in the Collaboration was that from the beginning consumers were given an explicit position and political representation, and some power to influence the way the organisation was set up. It took quite a few years but, what it meant was an increasing commitment and expectation within the organisation that people were going to wrestle with this issue about how to meaningfully involve consumers. This commitment is one of the things that’s really important because while people are marginal outsiders and don’t actually have much power or funding or anything, it’s very, very, hard to make any kind of substantial input or make much of a difference. Having good experiences over a period of time is another important thing. What was also important was that researchers started hearing about benefits and advantages of consumer involvement from other researchers. Things like: ‘You know I really didn’t think this [involving consumers] was going to make much difference, but it radically altered everything’.

Making the effort to ensure there are good experiences all round is important. It is also important that there is support for consumers, including training and skills development and, increasingly we have started to realise, the other way around as well. Researchers need training and skills to help them apply some of the knowledge and experiences that consumers bring through their involvement. In this area consumer advocates like me have a lot to contribute. We have skills that we teach each other, that we can actually start to teach to researchers as well. I think that this is going to be an important linking role for us to play because researchers often never actually reach the community in any really direct way. In the beginning at the Cochrane Collaboration we recognised we needed to construct something that makes it worthwhile for consumers to be there.

So we make sure they directly benefit from the research, and go away with research results to

48

2002 General Practice and Primary Health Care Research Conference — Proceedings

share with their community in ways that can be understood. We also provide them with opportunities to influence future research agendas. An area the community really doesn’t have much of chance to have a role in currently. Consumer participation/involvement in research reminders us that: Again, to quote Einstein, ‘not everything that counts can be counted, and not everything that can be counted counts’. It’s important to remember the limitations of research, and the realities of peoples’ lives.

To finish off, I’ll return again to Einstein. One of the important things that Einstein proved, apart from teaching us that things were relative, and a lot depends on your point of view; was that it’s both important and possible to change the way that scientists think. He said: ‘To raise new questions, new possibilities, and to regard old problems from a new angle requires creative imagination, and marks real advances in science’. I think one of the things about finding new partners is that in doing so we can actually start to achieve this. Consumer participation in research helps researchers to look at old things in new ways, and identify new and different possibilities. Einstein proved it could be done, and that in doing so you can change the face of science.

2003 General Practice and Primary Health Care Research Conference — Proceedings

49

Plenary paper abstracts

Designing research to maximise impact

Professor Paul Glasziou (Facilitator)

Professor of Evidence-Based Practice, Centre for General Practice, University of

Queensland Medical School, Brisbane

X

X

X

X

Primary care research

Identify a problem

Researchable question

Research methods appropriate to question

Change

Primary care research

X

X

X

X

Problem

Smallpox a big killer

1723 — Lady Montagu had introduced ‘variolation’

Researchable question

Jenner observed diary maids didn’t get smallpox

Does cowpox protect against smallpox?

Research methods

1796 — Gives James Phipps cowpox; then smallpox

Change

1797 — paper rejected by Royal Society

Does more extensive research; self-publishes Inquiry

X

X

X

X

Primary care research

Identify a problem

Researchable question

Research methods appropriate to question

Change

Margaret Sanger

X

X

X

X

Problem

Fertility control

Researchable question

Can an oral contraceptive pill work?

Research Methods

Laboratory work on 200 substances: George Pincus

Field trials: Brookline, Puerto Rico

Change

Searle market OC in 1956

The conventional picture of the research worker is that of a rather austere man in a white coat with a background of complicated glassware. My idea of a research worker, on the other hand, is a man who brushes his teeth on the left side of his mouth only so as to use the other side as a control and see if tooth-brushing has any effect on the incidence of caries.

Sir Robert Platt (1900-?)

British Medical Journal I:577, 1953

50

2002 General Practice and Primary Health Care Research Conference — Proceedings

How can research turn clinical general practice questions into clinical practice?

Professor Chris Del Mar

Professor of General Practice, Centre for General Practice, University of Queensland

Medical School, Brisbane

Abstract

Before this question can be answered, we need a re-think about the very definition of the discipline general practice itself. This will have to expand beyond general practice processes, and lay claim to actual illnesses. But we also need to focus on the purpose of research. When does basic research become applied, or clinical? When does it become audit? Why do we do it, and what does society value?

General practice research is at a watershed. No longer at the parochial end of addressing purely general practice processes, including health services research, it must start to take on the wider research agenda of mainstream clinical practice. This will inevitable mean collaboration, not only outside the walls of the institution, but outside the citadel of the discipline.

Implications are that we will need to move away from traditional funding sources, join up with new partners, and move down the applied — basic research continuum into unfamiliar territory.

X

X

X

X

X

X

We must grow up!

Do clinical research

Remain focused

Cooperate

Break down trade barriers

Stride the national stage

what is general practice?

Care

Continuous

Comprehensive

Community

Communication

what

isn’t

general practice?

Diseases?

Illness?

Primary?

Palliative?

Preventive?

2003 General Practice and Primary Health Care Research Conference — Proceedings

51

what is general practice?

ARIs / URTIs

insomnia

diabetes

osteoarthritis?

skin cancer

etc

etc

what

isn’t

general practice?

Diseases?

Illness?

H

Y

A

C

E

R

H

I

what is general practice?

general practice / primary care specialties (paediatric neurosurgery) basic sciences (eg physiology) more basic sciences (eg biochemistry) even more basic sciences (eg physics)

A

T

I

C

P

L

A

P

I

O

N

52

2002 General Practice and Primary Health Care Research Conference — Proceedings

What is general practice?

Process health services communication

Content clinical quality

Great GP researchers

Fry

Bridges-Webb

Kamien

Watt

Stott

Mant

Jungfer

Grol

Baker

Howie

Mant

Knotternerus

Pickles

Howie

Bain

Jones

Blomberg

(not complete) epidemiology epidemiology records inequity access prevention prevention quality quality quality consultation length prescribing diagnosis infectious hepatitis

Bornholm’s disease

ARIs

ARIs dyspepsia low back pain

For example workforce efficiency

For example treatment diagnosis aetiology prognosis etc

2003 General Practice and Primary Health Care Research Conference — Proceedings

53

Know why you are doing it...

Doing things cheaper

‘efficiency’

Doing things right

Muir Grey 97

Doing the right things right

Doing things better

‘Quality assurance’

1970s

QA

1980s

Doing the right things

‘effectiveness’

EBM

1990s

Diffusion trials

2000s

Frequency of questions

40

20

0

140

(n=124)

120

100

80

60

(n=23)

(n=12)

(n=11)

(n=10)

(n=5) (n=5)

Therapeutics

(n=3)

(n=0)

Harm

Diagnosis

General

Organisational

Aetiology Guidelines Screening Prognosis

Classification category

Brassey et al. BMJ 2001;322:529–30.

54

2002 General Practice and Primary Health Care Research Conference — Proceedings

Know why you are doing it

X data for that hunch:

X

X

X

(organised) curiosity they need to know… career/general intellect etc

‘can I join your team…’?

Know why you are doing it...

… data for that hunch

… (organised) curiosity

they

need to know…

… career/ general intellect etc

Funding

prestige join a group…?

perverse incentives funding > publishing

Know why you are doing it

research

funding

Wills’ principles curiosity prestige type fun applicability

Responsive

?

?

X health services

(x)

(x)

Investigator driven

x x

?

clinical

?

?

2003 General Practice and Primary Health Care Research Conference — Proceedings

55

types of research

audit epidemiology systematic review health basic sciences methodological stringency (generalisability) —> different study types answer different Qs…

treatment prognosis

I

I diagnosis

I I aetiology phenomenon

levels of evidence

II III

III

III

Types of research

II

II

III III III III

IV

III

RCT cohort case -

sectional

X

X

X

X

X

X

What I said

We must grow up!

Do clinical research

Remain focused

Cooperate

Break down trade barriers

Stride the national stage

56

2002 General Practice and Primary Health Care Research Conference — Proceedings

Researching new models of care: the impact and economics of the

Glasgow Heart Failure Nurse Liaison Service

Assoc Professor Simon Stewart

Division of Health Sciences, University of South Australia,

Australia and Clinical Research Initiative in Heart Failure,

University of Glasgow, United Kingdom

Abstract

Background: Hospital activity represents the major cost component of health care expenditure related to heart failure (HF). We have recently shown (in Australia) that a post-discharge, specialist nurse management program designed to limit HF-related morbidity and mortality is extremely effective. This presentation describes the baseline costs and potential economic benefits of applying the Glasgow Heart Failure Liaison Service, a citywide program predominantly based on our research, on a UK-wide basis.

Methods: Using an accurate estimate of the current level and cost HF-related hospital activity in the UK, we determined the threshold at which the actual cost of applying this service would be equal to the ‘cost’ of bed utilisation associated with preventable hospital readmissions. It was assumed that a caseload of up to 122,000 patients discharged from UK hospitals would be managed in the year 2000.

Results: We estimated that 47,000 of these 122,000 patients would normally accumulate a total of

594,000 days of hospitalisation from 49,000 readmissions (for any reason) within one year of hospital discharge. The cost of these hospitalisations was calculated at £166.2 million. Taking into account other costs (e.g. primary care and hospital out-patient visits) each 10% reduction in recurrent bed utilisation equates to a saving of approximately £18.0 million. The cost of applying a

UK-wide program of heart failure management based on the Glasgow model of care was estimated to be £69.4 million per annum. The relative threshold at which the cost-savings associated with reduced bed-utilisation for recurrent hospitalisation would equal the cost of applying this service on a UK-wide basis was, therefore, 39%. If, as expected from research and actual outcome data from Glasgow, it reduced recurrent bed utilisation by 50%, annual savings equivalent to £169,000 per 1000 patients treated would be made.

Conclusion: These data suggest that specialist nurse management programs designed to optimise the post-discharge care of patients with chronic HF not only have the potential to improve quality of life, reduce hospital readmissions and prolong survival, but also release resources to increase the efficiency of the health care system.

Background

X

Heart failure is a significant burden on the health care system.

X

Frequent hospitalisation in a subset of patients.

X

Otherwise, complex community-based management.

X

Preventing expensive hospital-based care the key to cost-effectiveness.

2003 General Practice and Primary Health Care Research Conference — Proceedings

57

International cost of heart failure

Sweden (1996)

UK (1991)

France (1990)

USA (1989)

NL (1988)

UK (2000)

SEK 2579m (74%)

£UK 360m (60%)

FF 11.4b (64%)

$US 9b (71%)

NLG 444m (67%)

£UK 1042m (70%)

0.0

0.5

1.0

1.5

Percentage of total health care expenditure

2.0

Stewart et al. Eur J of Heart Fail 2002

Randomised trials of nurse-led management of heart failure: effect on readmissions

Less

Education alone (Jaarsma NL)

Discharge planning (Naylor US)

Discharge planning (Laramee US)

Home + (Rich US)

Home + (Stewart Au)

Telephone F/U (Riegel US)

More

Home (Stewart Au)

Home (Blue UK)

Home (Moser US)

Home + (Krumholz US)

Home + (Kimmelstiel US)

Home + (Atienza Sp)

Home + (Tofler Au)

Home + (Sindone Au)

Clinic (Cline SE)

Clinic

+

(Doughty NZ)

Primary care (Weinberger US)

Australian experience

X

X

X

CHF pts (n = 98) in randomised study of HBI in chronically ill pts (n = 762):

– initial follow-up for 6 months

– prolonged follow-up to 18 months.

CHF pts (n = 200) prospectively recruited to a study of HBI (n = 200):

– range of follow-up: 6–22 months.

Combined analysis of 298 pts in these trials:

– median follow-up 4.2 years (3–6 years).

58

2002 General Practice and Primary Health Care Research Conference — Proceedings

Long-term survival

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

0

Independent predictors:

HBI -

LVEF -

RR 0.72 (0.54 - 0.97) *

RR 2.00 (1. 29 - 3.11) ***

Hypoalbuminaemia - RR 1.65 (1.09 - 2.69) *

Charlson index RR 1.79 (1.13 - 2.85) *

Pulmonary oedema - RR 1.46 (1.08 - 1.98) *

HBI (n = 149)

UC (n = 148)

6 12 18 24 30 36 42 48 54 60 66 72

Month of follow-up

Long-term event-free survival

0.4

0.3

0.2

0.1

0.7

0.6

0.5

1.0

0.9

0.8

0.0

0

Independent predictors:

HBI RR 0.70 (0.49 - 0.91) **

Beta-blocker - RR 0.71 (0.51 - 0.99) *

Past AMI - RR 1.56 (1.14 - 2.12) **

HBI (n = 149)

UC (n = 148)

6 12 18 24 30 36 42 48 54 60 66 72

Month of follow-up

2003 General Practice and Primary Health Care Research Conference — Proceedings

59

Long-term readmissions

500

450

400

350

UC (n = 148)

HBI (n = 149)

300

250

200

150

Cost of admissions:

UC = $AU 1.9 million

HBI = $1.1 million

100

50

0

0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72

Months of follow-up

Cost–benefits over 3 years

$3,000,000

$2,500,000

$2,000,000

- $883,411

(- 34%)

- $727,039

(- 34%)

$1,500,000

2353 days

(- 21%)

2837 days

(- 22%)

$1,000,000

Usual care

- $738,271

(- 47%)

$500,000

$0

Hosp

D/C

3m

Home-based intervention

6m 9m

1378 days

(- 37%)

1 year

15m 18m 21m 2 years

27m 30m 33m 3 years

60

2002 General Practice and Primary Health Care Research Conference — Proceedings

Scottish experience

Blue et al, BMJ 2001;323:715-718

Usual care outcomes at 90 days

60

%

50

%

40

%

30

%

20

%

10%

0%

Event-free

AU (n = 148) Stewart

UK (n = 75) Blue

USA (n = 141) Rich

Readmitted Multiple events Died

Stewart et al. Eur J of Heart Fail 2001

Aims

To examine the economic impact of a specialist nurse directed multidisciplinary service to optimise the post-discharge management of heart failure within the whole UK population:

X

X

X home-based clinic-based hybrid home and clinic-based using accurate UK-based data.

2003 General Practice and Primary Health Care Research Conference — Proceedings

61

Trends in heart failure admissions in Scotland (1990–1996)

Men

6,500

6,000

5,500

5,000

4,500

Admissions

Patients

4,000

3,500

3,000

2,500

First-ever

90 91 92 93 94 95 96

Women

6,500

6,000

5,500

5,000

4,500

Admissions

Patients

First-ever

4,000

3,500

3,000

2,500

90 91 92 93 94 95 96

Stewart et al Eur Heart J 2000

More malignant than cancer?

1.0

0.9

0.8

0.7

0.6

0.5

0.4

W o m

0.3

0.2

0.1

0.0

0 6 121824303642485460 e n

Breast

MI

Bowel

Ovarian

Heart Failure

Lung

Month of follow-up

1.0

0.9

0.8

M e n

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

0 6 121824303642485460

Month of follow-up

MI

Bladder

Prostate

Bowel

Heart Failure

Lung

Stewart et al Eur J Heart Fail 2001

62

2002 General Practice and Primary Health Care Research Conference — Proceedings

The cost of heart failure in the UK

2.4 m GP consultations (£44.9m)

650,000 GP referred OPD visits (£35.2m)

2.1 m prescriptions (£128.6m)

988,000 individuals with CHF

Direct *

£715.7 million

135,000 primary admissions (£489.7m)

300,000 post discharge OPD visits (£17.3m)

Stewart et al. Eur J of Heart Fail 2002

X

X

X

X

Potential savings when applying national heart failure program in the UK (2000)

X

Expected caseload = 122,000 patients from total of 244,000 (50%) > 55 years and discharged to home in 2000

47,000 patients readmitted at one year

594,000 days of readmission

Total cost of readmission = £166.2 m + primary care and outpatient visits

Each 10% reduction in hospital stay saves £18m

X

X

X

X

X

X

Key assumptions

UK population = 60 million

60 distinct service centres required

Based on ‘Glasgow model’ — 8 FTE heart failure nurses per service area

Case-load of 200–250 pts/annum/nurse

Two additional GP visits per patient

50% increase in drug treatment

X

X

X

X

X

X

X

Components of cost

Establish and maintain service area, including training etc…(£75,000 pa)

Each heart failure nurse (£33,000 pa)

Heart failure coordinator (£40,000 pa)

Office management (£18,000 pa)

Increased pharmacotherapy (£86/pt/pa)

More GP visits (£36/pt/pa)

+ Specific program costs…

2003 General Practice and Primary Health Care Research Conference — Proceedings

63

X

X

X

X

X

X

X

Home-based program — annual costs per annum (Blue/Stewart)

Infrastructure costs = £5.6 million

Heart Failure nurses = £15.8 million

More pharmacotherapy = £10.7 million

More GP visits = £4.4 million

Nurse coordinator = £2.4 million

Home visits + phone follow-up = £30.5m

Total expenditure per annum = £69.4m

X

X

X

X

X

X

Clinic-based program — annual costs per annum (Cline/Doughty)

Infrastructure costs = £5.6 million

Heart Failure nurses = £15.8 million

More pharmacotherapy = £10.7 million

More GP visits = £4.4 million

Clinic visits (6 per patient) = £36.6m

Total expenditure per annum = £73.1m

X

X

X

X

X

X

X

Hybrid program — annual costs per annum (Rich/Stewart*)

Infrastructure costs = £5.6 million

Heart Failure nurses = £15.8 million

More pharmacotherapy = £10.7 million

More GP visits = £4.4 million

Nurse coordinator = £2.4 million

Clinic + home visits = £33.6m

Total expenditure per annum = £72.5m

Cost impact of nurse-led management of heart failure in the UK

9

80

70

60

50

40

30

20

10

Home Clinic

50%

40%

30%

20%

Stewart et al Eur Heart J 2002

64

2002 General Practice and Primary Health Care Research Conference — Proceedings

X

X

X

X

Summary

Cost of heart failure in UK equivalent to that of other developed countries

Little difference between cost of three models of post-discharge care

Approximate cost of implementation = £1.2 million/annum (population of 1m)

Total cost of readmissions per service area = £2.77 million/annum

X

X

X

X

X

Conclusions

Burden of heart failure enormous due to costly admissions

Heart failure programs remarkably cheap

Studies suggest 40–50% reduction in recurrent hospital stay

Potential for ‘cost savings’ of £169,000 per 1000 patients treated

More ‘cost effective’ than majority of pharmacological agents

Divisions and researchers: developing viable and sustainable partnerships for quality systems of care in general practice

Dr Peter Del Fante

GP and Medical Director, Adelaide Western Division of General Practice, Vice-Chair,

Australian Divisions of General Practice

(Presented by Don Allan, Program Director, Adelaide Western Division of General Practice as

Dr Del Fante was unable to attend.)

Abstract

Since their inception in 1993, the majority of Divisions of General Practice have been supported to some extent by external research or evaluation expertise (eg SERUs and Academic General

Practice). A local case study is presented showing how the Adelaide Western Division of General

Practice has forged strong links with researchers in both the development and evaluation of quality systems of care in General Practice. These links are now being further enhanced by adapting some of key learnings from recent overseas developments in this area. An outline will be provided of our

‘work in progress’ in developing a local ‘quality institute’ and Quality Peer Support Groups in general practice. These new initiatives aim to not only integrate research and health care delivery to improve health outcomes, but also to create a culture of continuous quality improvement in health care.

Outline of presentation

X

The Adelaide Western Division (AWDGP) experience with researchers.

X

X

Quality systems in general practice.

Building on Quality (BoQ) project

– core elements of BoQ model for AWDGP.

X

AWDGP, researchers and BoQ — next steps

– quality collaboration concept

Quality Peer Support Groups and CQI.

X

Summary.

2003 General Practice and Primary Health Care Research Conference — Proceedings

65

Adelaide Western Division of GP (AWDGP)

X

Location: North Western suburbs of Adelaide.

X

Population: 210,000

– high level of NESB and socio-economic disadvantage (including 1.2% ATSI).

X

X

General practices: 164 (53% Solo) — 240 GPs.

Collaboration with researchers: since 1993

– field support; SERUs; Uni of Adelaide (GP, Public Health); Uni of SA (IT and OT); TQEH

Clinical Epid Unit; DHS Health Outcomes Unit; DATIS; PHCRIS etc.

X

X

X

X

X

Example 1: RCT Respiratory Mini Clinics

TQEH Clinical Epidemiology Unit researchers.

Asthma and COPD clinics in GP.

RCT Trial — value added component.

Program roll-out and RCT design con-current.

Outcomes: not significant — redesign; Paper for publication.

X

X

X

X

X

X

Example 2: Economic Evaluation of Obstetric Shared Care Program

GPs accredited in a CQI based program.

Researchers incl Health Economist — Uni of Adelaide Dept of General Practice.

Cost effective high quality care by GPs.

Huge savings to State Health identified.

Working party to review Models of OSC and State wide roll-out.

Opportunities raised for Measure and Share.

X

X

X

X

Quality systems in General Practice

Adapted from Boston Institute for Healthcare Improvement

Service: accessibility.

Care: patient interaction.

Clinical outcomes: reliability.

Vitality: sustainable practice.

X

X

X

X

X

X

Service: accessibility

After hours care — co-location model/GP tele-triage — 24/7/365.

Practice facilities (one stop shop).

Point of Care Testing (PCT).

Nursing and allied health services.

Appointments scheduling to handle demand, DNAs, urgent/routine.

F to F; Phone; Email/Web interactions.

Care: patient interaction

X

X

X

Individualised/customised care

– patient-centred interaction.

Culturally sensitive.

Team and shared care approach.

66

2002 General Practice and Primary Health Care Research Conference — Proceedings

X

X

Group interactions.

Empowerment — self care.

Clinical outcomes: reliability

Proactive systemic and population based management of practice patients underpinned by Quality

Information Systems and Care Coordination [Preventing ‘Sins of Omission’ and Reducing variation in care as shown on any clinical indicator eg HbA1c].

Vitality: sustainable practice

X

Reflection:

– practice as a learning organisation; quality peer support groups (QPSGs).

X

Development:

GP and staff CME, up-skilling, professional health.

X

Innovation:

– incorporate effective new technologies (e.g. smart BP devices, PoCT, spirometry).

X

X

X

X

Quality systems in general practice

Service: accessibility.

Care: patient interaction.

Clinical outcomes: reliability.

Vitality: sustainable practice.

Quality Indicators are required for ALL of these components of quality in general practice — a key role for researchers

Building on Quality

The Building on Quality (BoQ) project is about developing better ways to organise and reward

General Practice within a ‘quality framework’ that creates sustainable and systemic improvements in both patient health outcomes and professional satisfaction.

BoQ — setting the scene:

X

X

X

X

X

X

X

X

X

6 Divisions — AWDGP,BNDGP, BDGP, CDGP, HUDGP, CBDGP — six BoQ models

250 page literature summary — Monash Uni

AWDGP BoQ — Adelaide Uni GP/PH Dept

60 days of study tours — US, UK, NZ & CN

~5000 hours of grassroots GP consultation consumer & health care provider workshops

4 national stakeholder workshops timeframe: Feb–Sept 2001 synthesis of six BoQ models — Monash Uni.

BoQ model aims to create:

X

X

X

X

X integrated models of primary care in general practice and Divisions development of systems and capabilities which can be applied to practice populations continuous quality improvement culture improved management of health system resources improved patient health outcomes, including better equity and access to services.

2003 General Practice and Primary Health Care Research Conference — Proceedings

67

Patient Health Outcomes

Finance

GP

Incentives

Practice payments

Continuous Quality Improvement Model

Population health approach

Small group education and peer review

Chronic disease management

Practice self analysis

Resource management

Division data analysis

IM flow

Practice

Capability

IM/IT

Practice

Nurses & Allied

Health Team

Planning

Practice

Management

Support and

Redesign

Adelaide Western Division and Quality Institute – Support and R & D

AWDGP — ‘Quality Collaboratives’

IHI Boston — Quality Collaboratives

X

US Minneapolis Collaborative

Mayo Clinic/Healthpartners/Health Systems Minnesota

Institute for Clinical Systems Improvement — 1999

– funds: US$3 million/3 years for ‘Cultivating quality’

– support for clinical improvement — peer review.

X

UK National Primary Care Collaborative

Dr John Oldham and Team

– funding: UK11 million pounds

PDSA cycles and peer review.

X

AWDGP Quality collaborative with:

University of Adelaide Departments of General Practice and Public Health, and

TQEH Clinical Epidemiology Unit

Other potential partners:

DHS Health Outcomes Unit

DATIS (NPS) — SA

X

Activities:

– research, innovation and development role

– facilitation of Quality Peer Support Groups

– applied EBM and EBP

CME/CPD for GPs

GP leadership and research training (PHCRED).

68

2002 General Practice and Primary Health Care Research Conference — Proceedings

X

Funding:

PHCRED MoU;

PHQIC application with DoHA.

Divisional GPs and

Practice staff

QI ACTIVITIES

Quality Peer Support Groups

Quality Systems of Care - R&D

CQI support & culture change

Clinical Audit

EBM Guidelines & Support

Register/Recall systems

Data analysis

Education & training -

CME/CPD

Leadership training & support

PHCRED Program

Other Research support

AWDGP and Consumer

Reference Group

QI STEERING COMMITTEE

T of R

Quality Institute

Program Manager

(based at Division)

GP Division / Academic positions

Access to other academics and researchers: eg health economist, statisitician, policy analyst

Department of

General

Practice

Adelaide Uni

Dept of Public

Health

Adelaide Uni

Other Partners

TQEH Clinical

Epidemiology

Unit; DHS Health

Outcomes Unit;

NPS (DATIS)

AWDGP: Quality Peer Support Groups

X

Develop QPSG model:

– small group learning and peer review of aggregated data

– pilot with 10 GPs — meet 5 times per year (2 hours each).

X

Plan Do Study Act (PDSA) cycles:

– integrated systems of care — Diabetes CDM

– antibiotic prescribing.

X

CQI:

– create a Continuous Quality Improvement (CQI) culture.

X

GP benefits:

RACGP 5pph CPD

Medical Defence Organisation discount

– payment for participation (protected time).

X

Quality collaborative:

– facilitation for QPSGs

– applied EBM and EBP; data analysis and feedback

– monitoring and evaluation of processes and outcomes.

X

Data sources:

HIC Clinical Diabetes Project Data

DATIS (NPS) — Antibiotics PBS data.

2003 General Practice and Primary Health Care Research Conference — Proceedings

69

X

Division services:

Practice management support (clinical business flow)

IT/IM support — eg age sex disease registers/recalls.

X

Funding:

HIC, plus in kind support — AWDGP and DATIS.

Target outcomes (negotiable):

X

X

Increase percentage of diabetics treated according to best practice from current HIC baseline of 25% to 50% within 12 months, and to 75% within 18 months

10% reduction in PBS antibiotic prescribing

– including a shift towards equivalent but less expensive antibiotic groups.

Summary

X

Build on current collaborations between Divisions & Researchers (eg Quality Collaborative).

X

X

X

Focus on integrated models of primary health and quality systems of care (eg BoQ).

Grassroots driven process to create a CQI culture (eg QPSG).

Viability depends on sustainable funding and achievement of desirable health outcomes.

Involvement in the research process of those who can use the research is the best indicator of it’s uptake.

Jonathon Lomas

Professor of Clinical Epidemiology and Biostatistics

McMaster University

Collaborations for relevance, depth and insight: meeting the challenge of different language and perspectives between research disciplines

Dr Rosemary Aldrich

Executive Officer, Newcastle Institute of Public Health (NIPH)

Abstract

General practitioners rarely work in isolation, and an increasingly multi-disciplinary approach to patient care is reflected in primary health care research. Two divisions of general practice, the

NSW Rural Doctors Network, and the Discipline of General Practice at the University of Newcastle have joined more than 10 other public health and health services research groups to form the

Newcastle Institute of Public Health (NIPH). On any one NIPH project team researching primary health care issues there might be medical practitioners, registered nurses, social and behavioural scientists, ethicists, health economists and biostatisticians.

Established as a partnership between the University of Newcastle and the Hunter Area Health

Service, NIPH’s mandate is to provide the structure and resources to help multi-disciplinary collaborative research happen. However, this is sometimes easier said than done: a multidisciplinary approach can lead to tensions at project design and implementation stages, and result in project failure. This can be due to the differences in language, perspectives, values and agendas each collaborator or collaborating organisation brings to the process. Using examples I will illustrate how language, values and perspectives can differ between disciplines, and how the same

70

2002 General Practice and Primary Health Care Research Conference — Proceedings

words can mean different things, but how these differences can be harnessed to enhance research designed to benefit communities and consumers.

X

X

X

Words, meanings, values and beliefs

Different perspectives and their impact on the multidisciplinary research process

Barriers to collaboration and how to overcome them

X

X

X

X

Four steps of the research process

problem researchable question research method to answer the question change as a result.

Words, meanings, values and beliefs

crook

X

X

X

Different perspectives and their impact on the multidisciplinary research process

At every step.

Can enhance or derail.

Increasingly important.

NIPH

X is the public health and health services research arm of the Hunter Medical Research Institute

X is a partnership between

– the University of Newcastle and

– the Hunter Area Health Service

X comprises more than 20 groups:

– epidemiology, biostatistics, health promotion, law and ethics, academic and clinical units in medicine, nursing and allied health (including divisions of GP and NSW RDN), health economics, health sociology and social sciences, health research centres.

X

X

X

X

We do research in

health service quality and delivery ageing health inequalities behavioural science in relation to medicine and health.

X

X

X

Staff and team leaders promote collaboration

internal to NIPH other institutions and groups locally, nationally and internationally.

X

X

X

Different perspectives and their impact on the multidisciplinary research process

At every step.

Can enhance or derail.

Demands action.

2003 General Practice and Primary Health Care Research Conference — Proceedings

71

X

X

X

X

Barriers to collaboration …

Different priorities and timeframes.

Inflexibility in funding arrangements.

Challenges of strength and change.

Practical considerations: geography, infrastructure, schedules.

X

X

X

X

… and how to overcome them

Respect your colleagues: trust.

Have mutually beneficial goals.

Be prepared to modify your ideas.

Establish mechanisms for setting out roles and responsibilities.

X

X

X

Take-home messages

Researchers do not always share the same language and perspective.

Shared values more important than exactly the same language.

Remember, collaboration enhances your profile and yields better information.

Challenges of community-based interventions to improve nutritional status of Indigenous people in remote communities

Professor Kerin O’Dea (presenter)

Menzies School of Health Research, Darwin, NT

Kevin Rowley

Department of Medicine (University of Melbourne), St Vincent’s Hospital, Fitzroy, Vic

Karen Skinner, Michelle Skinner

Looma Community, via Derby, WA

Gwyneth White

Derby Aboriginal Health Service, Derby WA

Abstract

Type 2 diabetes is a serious health problem in Aboriginal and Torres Strait Islander communities.

Not only does it occur at much higher prevalence, but the age of onset is much younger than in the general Australian population. In many communities there is very high awareness of the health problems posed by diabetes, in particular, renal disease and premature cardiovascular disease. In this presentation we will describe an intervention we evaluated over a number of years with the

Looma Aboriginal Community in the West Kimberley region of WA (at their invitation), to reduce the risk of diabetes and its complications. After a baseline screening, widespread discussions were undertaken with community representatives on the nature of the intervention: improving the quality of the food supply, increasing the levels of physical activity across the community, and smoking reduction. Over time the community became very involved at a number of levels:

X the Council employed a Sports and Recreation Officer using CDEP funds, and a range of sporting programs were supported

X the Council provided a community vehicle for ‘bush tucker’ trips at least once per week

72

2002 General Practice and Primary Health Care Research Conference — Proceedings

X a community person was appointed to the position of Store Manager, and initiated comprehensive improvements to the food supply

X the School developed a ‘Healthy Canteen’ policy and transformed the quality of the lunches that could be purchased

X community nutrition workers completed a TAFE course in Nutrition and established a breakfast program at the School.

Baseline screening of those over 15 years was repeated after 2 and 4 years. Significant improvements were observed in cardiovascular disease risk factors (lipids, smoking, homocysteine, insulin) and biochemical markers of fruit and vegetable intake increased. These changes were backed up by store turnover data. Unfortunately, the prevalence of overweight and diabetes did not fall, which was a great disappointment to the community nutrition workers.

These results will be discussed in terms of what are achievable goals in relation to communitybased chronic disease prevention programs. The barriers to, and facilitators of, health promoting changes in diet and lifestyle will also be discussed, along with the challenges to achieving sustainability over the long term.

Looma Healthy Lifestyle

Chronology

X

Late 1992: invited to visit Looma by Council Chair to discuss possible diabetes intervention

– community meeting agreed to a baseline screening.

X

Screening conducted in early 1993

203 out of 230 eligible adults (>15yr)

..

..

25% people had diabetes, 42% of those over 35 yr vascular disease risk factors also high

– appointment of a project officer 3 days/week, and 2 part time community diabetes workers

(from NHMRC funded project grant).

X

X

Looma Diabetes Program commenced

– initially targeted people with diabetes, or at risk

..

..

60 people participated in diet + exercise

6 monthly monitoring for 2 years.

Repeat of full community screening after 2 years

– program broadened to target whole community — Looma Healthy Lifestyle

..

..

regular exercise: sport, hunting, walking store: major change in food supply

– program extended to children (primary prevention)

..

..

school lunch program — canteen breakfast program — Looma Healthy Lifestyle team.

Community initiatives

X

Looma Diabetes Program/Looma Healthy Lifestyle

– appoint community members as diabetes workers

– regular hunting trips

– nutrition education, cooking classes

– store tours to identify healthy food choices.

2003 General Practice and Primary Health Care Research Conference — Proceedings

73

X

Community Council

– appointment of a sport and recreation officer

– appointment of community member as store manager with mandate to improve nutritional quality

– non-smoking policy in public buildings

– provision of vehicle for Looma Diabetes Program

– health promotion — sport/art competitions.

X

Community store

– improved quality and quantity of fresh produce

– promotion of wholemeal bread and flour

– healthier fat sources — oils and spreads.

X

X

School

– healthy canteen policy — school lunch program

– health education integrated into curriculum.

Community members

– regular family walking groups

– activation of sporting teams.

30

25

20

15

10

5

0

Prevalence of obesity, 1993 to 1997

p = 0.237

p = 0.378

men women

1993

1997

74

2002 General Practice and Primary Health Care Research Conference — Proceedings

35

30

25

20

15

10

5

0

Prevalence of obesity, 1993 to 1997

p = 0.802

p = 0.712

men women

1993

1997

14

13

12

11

10

9

8

7

6

40

35

30

25

20

15

10

5

0

Prevalence of hypercholesterolaemia, 1993 to 1997

p = 0.037

men p = 0.057

women

1993

1997

Plasma homocysteine, 1996 to 1997

p = 0.001

Looma '96

Looma '97

mean + 95% ci

2003 General Practice and Primary Health Care Research Conference — Proceedings

75

3

2

5

4

1

0

No change in plasma lycopene levels among Looma adults

10

Men Women

9

8

7

6

1996 1997 1996 1997

Baseline Feb 1996. Follow up Nov 1997. Tomatoes are the major source of lycopene

Increased lutein in plasma in Looma adults (Feb 1996 to Nov 1997)

Men Women

7

2

1

4

3

6

5

0

1996 1997 1996 1997

Baseline Feb 1996. Follow up Nov 1997. Major source of lutein green leafy vegetables

76

2002 General Practice and Primary Health Care Research Conference — Proceedings

Different uptake of healthy diet in men and women at Looma — cryptoxanthin

Men

Women

6

5

4

3

2

1

0

1996 1997 1996 1997

Baseline Feb 1996. Follow up Nov 1997. Citrus fruit was the major source of cryptoxanthin

Different uptake of healthy diet in men and women at Looma — ß-carotene

10

8

6

4

14

12

Men

Women

2

0

1996

1997 1996 1997

Baseline Feb 1996. Follow up Nov 1997. Carrots were the major source of ß-carotene

2003 General Practice and Primary Health Care Research Conference — Proceedings

77

25

20

15

Plasma carotenoids, 1996 to 1997

Looma '96

Looma '97

Reference

* *

10

5

0 lutein/zeax. Green leafy veg cryptoxanthin Fruit lycopene

Tomatoes b-carotene

Carrots, leafy veg

Improving nutrition at Looma School canteen

Aims: (initially)

X

To promote healthy eating habits among students, thereby preventing excessive weight gain in early adulthood;

X

To provide nutritious lunch, in line with NHMRC recommendations for children and adolescents;

(later)

X

To provide healthy breakfast for students, many of whom previously attended school without having eaten breakfast.

Outcomes:

X dietary quality, growth rates, clinical events.

An integrated program of improved diet, health education, regular exercise

X

X

X food supply sport health education in curriculum

– diet

– exercise

– smoking

– alcohol and other drugs.

78

2002 General Practice and Primary Health Care Research Conference — Proceedings

Typical lunches, 1996 vs 1998

tomato sauce; whole apple; slice of cake;

total

:

:

470g food + 250mL drink

energy:

3.7 megajoules wholemeal roll with meat, cheese, low

-

fat yoghurt; fruit salad: banana, apple , , pineapple, strawberry , canteloupe ;

;

200

200

total:

615g food + 200

energy:

3.0 mL megajoules drink

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

Calcium

Nutrient intakes provided by typical lunches

1996

1998

Iron

Zinc

Magnesium

Folate

Thiamin

Riboflavin

Niacin

Retinol eq.

School breakfast program

This was initiated by the Looma Healthy Lifestyle team

X

X

X

X

X recognition that primary prevention meant early intervention many children coming to school hungry simple: cereal, dried fruit, milk, fruit juice funded by parents in advance on ‘pay day’ — but not all paid eventually failed through lack of funds.

2003 General Practice and Primary Health Care Research Conference — Proceedings

79

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

Nutrient intakes provided by breakfast plus lunch, 1996 vs 1998

1996

1998

Iron

Zinc

Calcium

Folate

Thiamin

Riboflavin

Niacin

Retinol eq.

Looma healthy lifestyle

X

Identified as a priority by the community.

X

Directed at the population (community)

– genuinely community-based and controlled

..

Council, Store, School, Clinic

– involvement of community members at all levels and phases

– focus on diet, exercise and smoking cessation at the individual and community levels

– need for ongoing technical support/infrastructure

..

clinical, nutrition, health promotion, community development

– evaluation of both process and outcomes

..

regular feedback through community reports.

Interventions to reduce the prevalence of non communicable diseases

When targeted at adult population

X reduction in CVD risk factors achievable

– lipid levels (change in type rather than amount of fat)

– smoking

– leisure time physical activity

– blood pressure.

X reversal of trends towards increasing obesity and Type 2 diabetes not demonstrated

– realistic goals should highlight risk reduction.

Primary prevention of obesity and Type 2 diabetes should begin in early life

X intrauterine

– diabetes in pregnancy

– low birthweight

80

2002 General Practice and Primary Health Care Research Conference — Proceedings

X establishing diet and physical activity patterns with educational and environmental approaches

– food supply (low fat, energy density)

– opportunities for physical activity

..

incidental, structured.

Challenges

X

Community participation

X

Absolutely essential

..

Support for consultation.

Sustainability

Infrastructure

..

Opportunities for exercise

..

Food supply — access, affordable, high quality

Technical support

..

Health, nutrition, community development, evaluation

Funding over the long term.

X

Transferability

Need to develop mechanisms and fund them.

Too risky not to rethink: research and policy

Professor Mark Harris

Professor of General Practice, School of Public Health and Community Medicine,

University of NSW

There has been much discussion about the value of linking research with policy. Evidence based policy is an as yet unattained goal of policy makers and researchers alike. It is viewed as a risky process both for researchers and policy makers. This is because there are many barriers to transfer of research into policy and because policy makers and researcher have different agendas creating few drivers for policy related research.

Priority setting

The first opportunity for policy linked research begins with the selection of a focus area for research.

Policy driven prioritorisation sits uncomfortably with the goal of investigator driven research prized by the Australian research community. Allowing policy direction to set research priorities is seen as potentially stifling creativity and innovation. Even where policy makers and health services researchers agree on the focus of research, they may disagree about the particular research questions which are most important to address and in what time frame. What is of most interest and relevance to primary care researchers may not appear to be of most importance to policy makers and vice versa. Both may be seen as irrelevant to the acute difficulties faced by primary care providers, their patients and GP professional organisations.

These conflicts between policy, researchers and practice and between long and short time frames represent an evolving tragedy for Australian general practice. I would like to explain this by using examples based on the critical characteristics of primary care as defined by Starfield and others —

2003 General Practice and Primary Health Care Research Conference — Proceedings

81

accessibility, comprehensive care, longitudinality and coordination of care across the health system.

1

Primary care is first contact health care. As such it has been seen as a fundamental human right — thus the slogan — ‘Health for All’. This implies that it is accessible to all population groups. Not just those with incomes over a certain level. Not just those who live in better class parts of our cities.

We know that 87% of the population see a GP each year. But how does this relate to need? What is the impact of the growing move away from bulk billing especially in rural areas? What will the impact of the medical indemnity crisis be on access to care? What quality of care do different groups get in general practice? We have seen at this conference that low socio-economic groups are less likely to have access to long consultations or preventive care. Why is this? We know that certain referral services are less accessible for those who cannot pay. What difference does this make? What difference do the various models of GP funding make to access?

Researchers in Australia have begun to research such policy questions in relation to access to after-hours care and the rural workforce. We have done some work on these questions in relation to diabetes. However, to date there has been little attention more broadly. Co-payments are again being suggested with few answers to questions about how these might affect not only access to primary care but also the roles of GPs in population health and linkages with the rest of the health system. By contrast access has been a major focus of evaluation and quality improvement in UK general practice especially in the development of Primary Care Trusts

2

.

Comprehensive primary care includes not only providing care that is patient centred but also population centred. In the context of our aging population with complex needs this means practices working together within Divisions and across disciplines as part of a team

3

. Our own experience with research into integrated primary care for chronic conditions especially diabetes is that this relationship between policy and research needs to be broad and take a long term view.

For example at any particular point in time, policy makers may be concerned with particular models of care (eg coordinated care, budget holding etc). We need to see research into specific models of care as informing a broad field of research comprising an array of interlocking questions

— for example not just those relating to diabetes care but more broadly the structured and multidisciplinary care of patients with chronic disease in primary care. Also the substance of the research here requires development activities — for example building organisational structures, information systems, re-orienting the health system by providing new incentives etc. Researchers need to involve themselves in trialing and evaluated these developments in an active way in order to find answers to a broader field of research questions. Those who were involved in the coordinated care trials know, however, that this is a difficult process. Although research has been done overseas these often lack the specificity to the Australian context to be directly transferable.

Continuity of primary care is associated with greater satisfaction and lower costs

4

. However, the ability of primary care providers to take a longitudinal perspective on care has suffered from attacks on many fronts — the commodification of health care in the minds of consumers resulting in increased doctor shopping, problems with access especially after-hours, the continued poor linkage between primary care and increasingly specialised acute care services, corporatisation of primary health care. While continuity of care is area that has received some attention by primary care researchers in Australia — this has not been strongly linked with policy initiatives. What impact have initiatives such as vocational registration, PIP, GPII, EPC, practice amalgamation,

MAHS, nurse practitioners and trends corporatisation and work force changes had on the longitudinal component of general practice? What are the costs and benefits of the increasing fragmentation? How would patient linkage affect this?

5

Lastly there is the coordination or leadership role which primary care providers can play along side patients in the rest of the health care system. This role has been recognised as being increasingly important both in Australia and overseas especially in the interface between primary care and acute care. It has been a focus of some policy oriented research in Australia including some of our own demonstrating the importance of capacity and team building within primary care to sustain this role even with incentives.

6

Again this research often involves developmental work within

82

2002 General Practice and Primary Health Care Research Conference — Proceedings

health care systems developing information and communication systems to make research and evaluation possible.

It is difficult to communicate the relevance of this research to coal face providers, with many seeing it as just being associated with that great bug bear of GPs — more paper work. There are, however, many critical unanswered questions of direct relevance to GPs — what impact does this coordination role have on GP workload, what role can practice nurses play in this, what impact would different models of primary care organisation and funding have? What is the impact of the plethora of liaison positions and services being developed by Commonwealth and State governments?

The fact that there has been relatively little policy oriented research in these areas is surprising given their centrality in general practice. This is in contrast with the UK and New Zealand. This research is difficult and requires a longer term view than has been possible with short term project funding or evaluation of short term policy initiatives. However careful evaluation is important. The spectre of the large scale policy reversals that have occurred in the UK regarding GP fund holding should be enough to convince of this.

Evidence-based policy

The second opportunity is in the application of research finds to the generation of policy.

There are problems here as well. Elliott and Popay described models of research use by policy makers within the UK National Health Service.

7

Financial and time constraints as well as the policy makers own personal attitudes and experiences influenced the uptake of research into policy.

Research takes time and research questions may be overtaken by events. There has been much debate about the evidence block — preventing or delaying application of findings in clinical practice. The block with policy may be even worse — with policy makers often having little access to research findings that may have a great bearing on their policy development.

Research and evaluation may also produce answers which policy makers do not like — especially if there is political commitment to a policy course that is at variance with the research finding.

Machiavelli warned, ‘The innovator makes enemies of all those who prospered under the old order and only lukewarm support is forthcoming from those who would prosper under the new’.

8

Conversely researchers may become alarmed when their findings are interpreted by policy makers who draw over simplistic conclusions with dire consequences. Researchers are unwilling say anything until all the information is analysed and caveats made; while policy makers will make best use of information to answer their current policy need.

The obvious answer to these problems is for researchers to take a more active role in the communication of research findings to policy makers as well as practitioners. Some research will be so ground breaking that everyone will want to know about it. However the majority is published or presented without policy makers becoming aware of it unless someone is specifically tasked to conduct a review of literature or similar. We have been involved in a number of activities which have attempted to review not only the literature but experience in the field on issues such as integration between GPs and hospitals in order to address specific policy questions. Informally we also have presented these and other research findings to policy makers. For policy makers attempting to bring about system wide change, questions related to adoption and feasibility may be as important as those concerned with effectiveness.

Lomas has identified a series of challenges for building better linkage between research and policy in both research and policy environments.

9

This included for researchers having time and capacity to meet; understanding of the timeliness of policy demands; finding and dealing with multiple decision makers and changes in staff. For policy makers challenges included time and other priorities, understanding of research process and the ways researchers communicate findings, lack of control over findings, and lack of continuity of staff within policy bodies.

2003 General Practice and Primary Health Care Research Conference — Proceedings

83

So what is the way forward?

Evidence related policy seems an ‘obvious’ approach. The potential benefits for policy makers are a greater likelihood of getting it right. The potential benefits for researchers are that their work gets to have some impact. For providers and the community there is a better chance of addressing the very real problems with primary care without costly and disruptive policy back flips.

There are risks. As alluded to before, policy makers may find the ‘evidence’ inconclusive, untimely, politically unpalatable and may find research too expensive. Researchers may find it difficult to separate their role from that of policy developer with resultant compromise to their work and ability to produce recognised academic outputs. For primary care providers it may seem too removed from their problems they face and the community may be unconvinced that their interests will be protected.

There are a number of strategies designed to reduce these risks.

Firstly we need to recognise that supporting research designed to inform future policy is only one part of the picture. We also need to have opportunities for research that is driven by curiosity and relevance to the delivery of primary care in the current health system.

There also needs to be continued efforts to build capacity for policy relevant research at all levels.

In particular their needs to be capacity for longer term streams of research linked to short term policy issues but with broader long term goals such as those related to the four areas that I have outlined. This is difficult and requires creativity and imagination not only on the part of researchers but also policy makers — perhaps weaving together disparate funding opportunities to support development of capacity for longer term program of work.

We need to negotiate a shared agenda which is designed to address the priority policy issues likely to face primary care in Australia over the next decade. This requires both a longer term view than we take at present and greater trust among policy makers, researchers and representatives of practitioners. The priority setting process of PHCRED was an attempt to do this. However it is has at various at times been bedevilled by variable commitment and institutional and personal jockeying from the stakeholders, disconnection from the policy setting agenda and anxiety by researchers who see it as a threat to continuation of their stream of work.

We need to establish better partnerships to exchange information and build trust among researchers, policy makers and provider organisations. There need to be more opportunities for policy makers and researchers to interact and to identify common ground and opportunities for research or research transfer. This may involve researchers acting outside their comfort zone. We have participated in a number of policy structures which have helped to form the basis for ongoing research and development activities by ourselves and others. For example in diabetes we have participated in structures such as the NDDWG, NDSG, NIDP working groups. This has helped facilitate research transfer and to help build an infrastructure for primary diabetes care and health systems research. It has required considerable investment of time over a decade and a long term commitment that was based not on seeking possible funding opportunities but rather a vision of a possible direction for primary care in Australia. This has allowed us, in particular, to be aware of windows of opportunity to provide input from research into policy development along with other stakeholders.

Just as we need to develop more opportunities for researchers to be involved in policy development, we also need to provide more opportunities for policy makers and clinicians to be involved in research. This is difficult and it can be easy to slip from the agenda of policy makers and researchers. For researchers this can not only ensure that research effort remains relevant to the policy issues but also helps policy makers to understand the research process and provide a key bridge for the transfer of evidence into policy.

Why now?

I would like to conclude by injecting some urgency into this endeavour.

84

2002 General Practice and Primary Health Care Research Conference — Proceedings

We currently face a deepening crisis in primary care in Australia. GPs are increasingly frustrated and anxious about their future. The litigation crisis is only the latest of many. The workforce crisis has extended to urban areas. Health system and market forces present serious challenges to each of the four key roles of general practice in Australia — access, comprehensiveness, continuity and coordination.

In the face of rapid change and profound uncertainty in the primary care environment, researchers, policy makers and professionals need to develop flexible, adaptable, and responsive approaches. Unfortunately, many of the pressures on primary care, Universities and government impede such a rethink, and change often fails to materialise unless it is precipitated by crisis.

10

We need to start focusing on finding, communicating and implementing some answers to the key issues facing primary care. It is not an ‘academic’ exercise.

References

1. Starfield B. Primary care: is it essential? Lancet 1994; 344: 1129–1133

2. Wilkin, D; Dowswell, T; Leese, B. Modernising primary and community health services. BMJ 23

June 2001 322: 1522–1524

3. Wilkin, D; Dowswell, T; Leese, B. Modernising primary and community health services. BMJ 23

June 2001 322: 1522–1524

4. Freeman G, Hjortdahl P. What future for continuity of care in general practice? BMJ

1997;314:1870–1873

5. General Practice: Change the future through partnerships. Report of the General Practice

Strategy Review Group. Commonwealth of Australia 1998. ISBN 0 642 36715 9

6. Blakeman T, Harris MF, Comino EJ, Zwar NA. Evaluating general practitioners’ views about the implementation of the Enhanced Primary Care Medicare Items. Med J Aust 2001; 175: 95–98.

7. Elliott H and Popay J. How are policy makers using evidence: Models of research utilisation and local NHS policy-making. J of Epid and Community Health 2000; 54: 461–468.

8. Machiavelli N. The prince. Harmondsworth: Penguin, 1975.

9. Lomas J. Using Linkage and Exchange to move research into policy at a Canadian foundation.

Health Affairs 2000; 19: 236–240.

10. Davies, H T O; Nutley, S M. Developing learning organisations in the new NHS. BMJ 2000;

320: 998–1001

The PowerPoint slides for this address are available at: www.phcris.org.au/events/conference_frameset.html

2003 General Practice and Primary Health Care Research Conference — Proceedings

85

Summary session

Moving PHC research forward — summary

Libby Kalucy (facilitator)

Co-Director & Senior Research Fellow, Primary Health Care Research and Information

Service

Where do we go to move primary health care research forward in Australia?

David Wilkin

Synthesis of the work that we are doing is essential if we are going to ensure that the research we do actually does make a difference to health and health care.

1.

2.

Conference displayed excellent primary care research covering a wide range of subject materials, methods, disciplines.

We need to look to bodies of knowledge, bodies of theory, bodies of evidence that are outside our own particular field of health care research.

3.

4.

Researchers should be aware of the problems caused by silos within research — the randomised controlled trial silo, or the qualitative research silo, or the observational study silo — as well as in health policy, health funding, and the delivery of health services.

We all need to think and find ways of better synthesising the knowledge that we can derive from a variety of different cultures, traditions, disciplines, designs, and methods.

Hilda Bastian

1.

Better priority setting to reduce fragmented, duplicated, poor quality flawed research.

2.

3.

Increasing diversity amongst researchers for true primary health care rather than general practice focus.

Consumer participation — invest in restoring the mechanisms which created the long tradition of significant consumer and community participation in Australia

Paul Glasziou

PHC RED — Now and future

Identify problems

X

NOW: Mix of health services and clinical.

X

FUTURE: DALYs burden + systems analysis/data.

Answerable research question

X

X

NOW: Use more than one method & one discipline resulting in increased number of bright ideas.

FUTURE: Has it been answered? (WOMBAT).

Appropriate research methods

X

NOW: different methods for different questions.

X

FUTURE: unethical not to randomise patients (patients do better: TROUT review).

86

2002 General Practice and Primary Health Care Research Conference — Proceedings

Change

X

NOW: dissemination is part of the problem, leading to information overload instead.

X

FUTURE: organisational change so teams access the best research/practice.

Margaret McDonald

Signs of success of this conference

X

More sophisticated level of debate.

X

X

More diversity in the audience

Participation through sharing, developing a common language and experiences

X

X

X

Trust between people shown by conversations between different groups

Greater willingness to combine our efforts to serve the needs of the health consumers.

More diverse program representing different research traditions

Challenges

X

To find and develop forums to work together to gain insights and deal with the language problems.

X

X

To open ourselves to different methodologies

To look at the utility of methodologies in different settings.

X

X

X

Policy makers able to talk about research in a sophisticated way

Culture of partnership.

More constructive positive feel about the debate.

Vision: essential elements for success

X

Strategy.

X

Priorities.

X

Know the problems we want to solve.

X

Garner the effort and contributions of various people.

X

Systems and opportunities for people to get together at conferences and events that are more focused around the topics and issues.

X

Involve researchers, policy makers, funders and other participants getting together and deciding priorities and issues.

X

X

X

More sensitive funding systems.

RED program — secure funding contributes to more strategic thinking in terms of research.

Achieving optimal workforce combinations, with diversity of methodology, diversity of effort, succession planning.

I would like to see in the future

X

If I have got a question I know the people I can ring and say, can you give me some answers to go on with, where do you think we need more research.

X

Clinical evidence, references and shared information about optimal workforce combinations and the best way to use staff — powerful information which gives clearer ideas about what we want to do.

2003 General Practice and Primary Health Care Research Conference — Proceedings

87

Concurrent paper abstracts

What are GPs’ attitudes to research?

Results of a quantitative and qualitative assessment

D Askew, Centre for General Practice, The

University of Queensland, P Glasziou,

A Clavarino, C Del Mar

Objectives

Although there is national and international recognition of the need to strengthen the research culture in general practice, little is known about the attitudes of Australian general practitioners (GPs). This study investigated GPs’ attitudes and involvement in general practice research, and the perceived relevance of research to their everyday practice.

Methods

We undertook (i) a postal survey of all GPs in four Divisions of General Practice (n = 631) and (ii) semi-structured interviews with 18 selected GPs who had been involved in a project run by the Centre for General Practice at the University of Queensland.

Principal findings

467 (74%) GPs completed the survey. The majority of respondents had positive attitudes towards research in general practice, but less than a third wanted to increase their involvement in research. One in five respondents thought that research evidence was at least as important as their clinical experience when making decisions for individual patients. Respondents younger than the mean age of 45, and those graduating within the last 20 years consistently had more positive attitudes towards research. The majority of respondents (90%) had access to the

Internet, but less than half were aware of access to Medline, provided free on the

Internet by the US National Library of

Medicine’s database ‘PubMed’. Perceived barriers to participation in research included lack of adequate remuneration, time, skills, personnel, and access to appropriate information sources. An additional barrier to involvement was lack of opportunities to participate in relevant, non-commercially oriented research. Interview data found GPs use words such as ‘quantitative’, ‘reliable’,

‘reproducible’, ‘generalisable’ or ‘statistical significance’ to describe research. They consider there is a need for more general practice research, but feel the general

88

practice environment itself militates against involvement. The lack of an evidence base for much of what occurs in general practice was discussed, and clinical topics were perceived as the main areas of need for general practice research.

Discussion

Although GPs consider that research is important in everyday general practice, few have research skills, are involved in research or desire to be more involved. GPs are unaware of available information resources, although access to appropriate resources was cited as a barrier to increasing involvement.

GPs tend to be limited in their perception of research, thinking mainly about large randomised controlled trials rather than secondary research such as systematic reviews of existing evidence.

Implications for policy, delivery or practice

Australian general practice does not have a strong research culture. This study suggests increasing the awareness and ability of GPs to access available resources could strengthen the research culture. Additionally,

GPs may willing participate in research that is relevant to general practice and structured to minimise the barriers to participation inherent in the general practice environment.

Development of the North East

Victorian Koorie Health Partnership

Committee

V Atkinson, Department of Rural Health,

The University of Melbourne, G Atkinson,

M Atkinson, P Atkinson, R Patton,

D Webster, J Murray, K Dwyer,

D Tournier, N Atkinson, A Appo,

D Simmons, L Bourke

Objectives

The Koorie Health Partnership Committee

(KHPC) and the Department of Rural Health,

The University of Melbourne, Shepparton have over the past two years worked on projects which have been directed by the

Koorie Community to do research and also gain funding to improve better health outcomes.

Methods

Regular meetings were established with the following organisations:

2002 General Practice and Primary Health Care Research Conference — Proceedings

X

Local Aboriginal Education Consultative

Group (Shepparton)

X

Bangerang Cultural Centre (Shepparton)

Rumbalara Medical Service (Mooroopna)

X

X

X

X

X

Cummeragunja Medical Service

(Cummeragunja)

Swan Hill and District Aboriginal Co-op

(Swan Hill)

Mungabareena Aboriginal Co-op

(Wodonga)

Njernda Health House (Echuca)

Allinjarra Aboriginal Co-op (Kerang)

X

X

Rumbalara Football Netball Club (RFNC)

(Shepparton)

X

Koorie Economic and Employment

Training Agency (Shepparton)

Principal findings

Guidelines were established to ensure that appropriate and culturally sensitive research would be undertaken.

The KHPC and DRH has received funding for

Healthy Life Style Project with the RFNC, the recruitment of a GP for the Cummeragunja

Medical Service and a Diabetes Project with the Rumbalara Medical Service. We have applied for funding for a Bush Tucker Project with Allinjarra Aboriginal Co-op (Kerang) and a Cancer Research Project with the Njernda

Health House (Echuca).

Discussion

The overriding concern of the partnership activities has been an emphasis on Koorie ownership, self-determination and control of the ethical aspects of the studies. The research was made possible by the existing community-based primary health care networks established by the local Koorie communities.

Implications for policy, delivery or practice

Such collaborations between existing Koorie primary health care services and research institutions guarantee that research in these settings is acceptable to the community, culturally sensitive and responds to the health needs of Koorie communities throughout the North East of Victoria.

Culture, illness and self management: raising awareness in the primary health care setting

L Bailey, Western Australia Office, The Royal

Australian College of General Practitioners,

S Roach

Objectives

To develop an understanding of the issues related to the health beliefs of those from culturally and linguistically diverse (CALD) backgrounds as they impact on potential self management.

To identify potential means for increasing awareness among primary health care providers of health beliefs as they relate to illness and the therapeutic relationship.

To identify an area of research that can be expanded upon within an existing environment of developing self management.

Methods

Extensive research was conducted in order to develop clinical practice guidelines for chronic condition self management through a literature review and delphi consultation with various experts. This research and the resulting guidelines highlight the need for a greater awareness of the meaning of illness to the patient. This is evident to a greater extent when the patient is not from the same cultural background to the health care provider.

Principle findings

Research indicated the absence of any existing tools or guidelines to assist practitioners to increase awareness of the health beliefs of those from a culturally and linguistically diverse (CALD) background.

Discussion

There is a growing need to focus on the differing health belief systems of those from

CALD groups in primary health care in

Australia. In the current climate of a diversifying population, it is necessary for health care providers to rapidly become aware of and adapt to the need for change in service delivery.

The Commonwealth’s Enhanced Primary Care

Initiative has acknowledged the needs of older Australians and those with a chronic condition. As a part of this initiative, varied approaches to chronic condition self management (CCSM) are currently being trialed around the country. The clinical practice guidelines for CCSM developed by the Royal Australian College of General

2002 General Practice and Primary Health Care Research Conference — Proceedings

89

Practitioners identify the need to take into account the illness experience and health beliefs of the patient in order to develop a partnership for management. The guidelines provide a framework for the development of self management approaches and how these can be promoted in clinical practice.

Implications for delivery and practice

The purpose of this paper is to highlight the need for a concurrent development of an

approach to recognising the health beliefs of

those from CALD backgrounds along with the approaches to self management to enable a true partnership to be formed. It is suggested that a useful approach is that developed by Kleinman et al in their cross cultural and anthropologic research on culture and illness. They utilise a series of questions to elicit discussion related to illness and health beliefs. This is integral to the patient-centred approach advocated by the

CCSM guidelines.

Disseminating research results:

What does this mean?

B Beacham, Primary Health Care Research and Information Service (PHC RIS),

Department of General Practice Flinders

University Adelaide, M Hale, B Lowcay,

E McIntyre, L Kalucy

The General Practice Evaluation Program

(GPEP) funded 248 research projects between 1990 and 2000. GPEP aimed to: evaluate changes in general practice, determine the effects of these changes upon quality and economic efficiency, and examine how general practice reforms have affected the Australian health care system.

Objectives

This study aimed to understand and describe the roles and attitudes of GPEP chief investigators in relation to informing others and implementing (using) research results.

Methods

Semi-structured telephone interviews (taped) were conducted with chief investigators of

GPEP projects submitted to PHC RIS between

January 1999 and July 2001.

Principal findings

Chief investigators of 29 projects participated in the study. Investigators were from general practice (n=14, 48) and non-general practice

(n=15, 52%) backgrounds.

Eighty-six per cent (n=25) of participants had actively informed others of their project

90

results. Their activities were helped and/or hindered by factors at both system and practitioner level. Investigators agreed that research teams should play a significant and very active role in informing others about their results, but only when results were established to be credible. In relation to the researchers’ role of informing others, a number of issues were raised particularly the implications of a broader interpretation of dissemination activities than are traditionally observed.

When asked about implementation of results,

19 investigators (66%) knew whether their results had been used. The majority of this group played a role in implementation, which was usually related to further research work

(n=12), followed by influencing changes in policy/practice and building knowledge.

Activities supporting usage of results in further research included networking and disseminating widely. Activities supporting usage of results to influence policy/practice included collaboration with stakeholders, representation on decision-making committees and continuity in the field. Lack of access to resources, particularly time, money and experience/expertise, and lack of knowledge about how results are used were highlighted as hindrances to taking a role in implementation. Investigators agree that researchers should play some role in usage of research results however; implementation activities were viewed as additional to those of informing others about results, with the scope and researchers’ responsibilities requiring clarification.

Discussion

The term ‘dissemination’ has traditionally defined activities related to information provision and exchange (informing others).

More recently understandings include activity related to changing practices

(implementation). The results of this study suggest that in disseminating research results, the roles and attitudes of GPEP investigators are guided by a traditional understanding of dissemination, which is supported at a system level, rather than the more active purposeful process defined by recent interpretations. These differing understandings, together with their respective implications, require further discussion and formal clarification.

Implications for practice, policy, research

Practice and policy initiatives in primary health care research, evaluation and development would benefit from clarification of the term ‘dissemination’, as this would assist definition of the scope of required

2002 General Practice and Primary Health Care Research Conference — Proceedings

activities, as well as associated roles and responsibilities.

Grief, shame and pride. A study of the impact of family violence and the strength of the community to act against it

M Binell, Weena Mooga Gu Gudba and

Ceduna Koonibba Health Service, I Gentle,

Spencer Gulf Rural Health School/SACRRH,

J Taylor

Overall objective of the presentation

To share information regarding a community based study into family violence in an

Aboriginal community with regard to issues in conducting community-based research in

Aboriginal communities.

Objectives of the research

Weena Mooga Gu Gudba Inc (a women’s group in an Aboriginal community) commissioned research to undertake a community wide study in order to

X identify the impact of and issues surrounding family violence

X establish community strengths to deal with the violence

X identify workable solutions to strengthen families and reduce violence

South Australian Centre for Rural and

Remote Health in partnership with the

University of SA were the researchers

Methods

A community owned qualitative study over a period of 6 months was conducted using interpretive, participatory action research. A research assistant was employed for the period of the research that was a member of the community

X demographic data was obtained for relevant key stakeholders

X interviews were conducted with community members;

X interviews were conducted with

Aboriginal and non-Aboriginal key stakeholders

X focus groups were held with Elders, and work based groups.

Principal findings

This presentation will discuss the new understanding of family violence and holistic model to address the issues in the community.

Discussion

This presentation will discuss the key issues in undertaking research in Aboriginal communities

X community ownership and management of research

X

X researching highly sensitive issues issues in conducting a community-wide study

X

X

X ethical issues and confidentiality the credibility of the researchers issues in disseminating research findings.

Implications for policy, delivery and practice

The holistic model for addressing family violence will be discussed. This involves a

‘whole of Government’ response supported by appropriate policy development.

Evaluation of Queensland’s General

Practice and Psychiatry Partnerships

Initiative

I Blignault, Queensland Divisions of General

Practice, G Groom, R King

Objectives

The General Practice and Psychiatry

Partnerships Initiative (GPAPP) is a four-year project that commenced in late 1999 and is due for completion in June 2003. Its principal aim is the development and implementation of collaborative ways of working between GPs and mental health service providers, which are cost effective and sustainable and result in improved quality and continuity of care for mental health consumers in Queensland.

GPAPP is funded by Queensland Health, with some additional funding for the evaluation coming from the Commonwealth Department of Health and Ageing, and coordinated by

Queensland Divisions of General Practice

(QDGP). With a total budget of $2.3 million, it is the largest project of this type in

Australia.

External evaluation is being conducted by Dr

Robert King (from The University of

Queensland) and Prof. Len Bickman

(Vanderbilt University). The aim was to

2002 General Practice and Primary Health Care Research Conference — Proceedings

91

investigate the effectiveness of all major aspects of this complex project of including, where possible, cost effectiveness and sustainability.

Methods

GPAPP involves three separate pilot projects: a remote pilot centred on Central Western and Central Queensland, a provincial pilot centred on Rockhampton, and a metropolitan pilot which covers the Brisbane–Gold Coast corridor. In addition, there is a statewide component of GPAPP that includes every

Division of General Practice and every mental health service in Queensland.

Outcome evaluation was planned for the metropolitan pilot only. This pilot is based on the CLIPP model, developed in Victoria. It has two components: a consultation-liaison component targeting high prevalence low severity disorders; and a shared care component targeting low prevalence high severity disorders, which involves transfer of stable chronic patients from the mental health service to GP care.

A qualitative evaluation of implementation and process was planned for all three pilot projects.

Principal findings

Preliminary data from all aspects of the evaluation will be presented.

Discussion

Implementation of the pilot projects has not been straightforward for a number of reasons. This has affected the evaluation. In addition, as the projects have progressed, thoughts about what are the most important and relevant questions have changed. In the busy, demanding world of health service provision, it can be difficult to apply rigorous research methods as planned. As GPAPP enters its final year, we look forward to completing the evaluation and to disseminating findings. Also to sharing lessons learnt about evaluation of such a complex initiative involving multiple projects, multiple sites and partnerships at different levels.

Implications for policy and practice

Evidence from the GPAPP evaluation will guide the future development of general practice and psychiatry partnerships in

Queensland. In addition, it will provide invaluable information for other States and

Territories wishing to expand their approaches to intersectoral collaboration in the mental health field and to primary mental health care. Separately, lessons from each of

92

the pilots will provide Divisions, mental health services and other stakeholders with information on models of collaboration, and enabling and constraining factors, relevant to their geographic region.

Are guidelines useful? The impact of clinical practice guidelines on the management of dementia in general practice

C Bridges-Webb, Royal Australian College of General Practitioners, NSW Projects

Research and Development Unit,

Y Zurynski, N Hiramanek

With the ageing of the Australian population the prevalence of dementia in the community can be expected to rise considerably and these increases are likely to have considerable impact on general practice.

Guidelines for the Care of Patients with Early

Dementia in General Practice (RACGP and

NSW Health) have been developed using the best evidence currently available. We plan to obtain information about the practical usefulness of these guidelines to suburban

GPs and what impact they have for the patients and their carers.

Objectives

The study is designed to

X give insight into the health status of a sample of patients with early dementia and their carers

X document the medical and psychological care and social support provided by GPs

X determine the impact of the guidelines on the patient, the carer and the GP

Methods

Forty GPs will recruit 200 patients and collect the following information:

Patients

Age and gender

Dementia diagnosis and how arrived at

Co-morbidities and their severities

Carers

Age and gender

Relationship to patient

Zarit Caregiver Burden interview

MMSE score

Geriatric Depression

Scale

Activities of daily living usually completed

Living situation

List of aged care services used within the last month

Basic assessment of the home environment in terms of general organisation and safety

2002 General Practice and Primary Health Care Research Conference — Proceedings

General practitioners will be asked to list services used and any services that are NOT available that would have been useful for the management of the current patient

One group of GPs will be given a copy of The

Guidelines and will be offered an academic detailing visit from the research nurse. Six months later, at follow-up, all of the variables will be measured again for both patients and carers.

A questionnaire to measure GP acceptance of the guidelines and the perceived benefits to their patients will be administered and analysed at the end of the study.

Implications for policy, delivery or practice

This project has the potential to improve the medical care and psychosocial support provided by general practitioners to sufferers of dementia who are still living in the community. By providing better care to sufferers of dementia and by supporting the carers of patients with dementia, these patients may remain living in the community for longer periods of time, therefore reducing the burden on already stretched residential care resources for the elderly.

Asthma 3+ Visit Plan Initiative — collection of qualitative baseline data

T Bubner, Department of General Practice,

The University of Adelaide, J Beilby

Objective

To gather qualitative data on asthma management within general practice prior to the implementation of the Asthma 3+ Visit

Plan initiative.

Methods

In order to gain a picture of current asthma management within general practice a number of focus groups were held across

Australia (SA, NSW and ACT). The process involved collecting qualitative data through focus groups with general practitioners (GP) and patients in both urban and rural settings.

In addition GP and patient questionnaires were developed to gain information about the characteristics of each participant and to further explore their asthma management practices.

A total of 64 GPs and 42 patients participated in the focus groups. Of these, 48 were urban

GPs and 16 rural GPs. Of the patients 17 were from a rural area and 25 from an urban area. 11 of the urban patients were in fact parents of children with asthma. All other participants were over the age of 18 years with asthma.

The focus groups were conducted by a facilitator and scribe and all meetings were recorded and subsequently transcribed.

NUDIST, a computer-based qualitative software package, was used to assist in analysing the data.

Principal findings

Many of the same themes were evident in both the GP and patient focus groups and there was little variation between the rural and urban setting.

Both GPs and patients were supportive of the

Asthma 3+ Visit Plan initiative and whilst a number of key issues around implementing the initiative were raised, recommendations to address these were also put forward.

Key findings around the following areas will be presented at the conference in more detail:

X spirometry use

X

X written action plans education for patients and GPs

X

X

X utilisation of register and recall systems different models within general practice team approach.

Discussion

For successful implementation of the Asthma

3+ Visit Plan a two-pronged approach directed at GPs and patients is needed. GPs require further information and education around the creation of appropriate written action plans, the steps of the Asthma 3+

Visit Plan, spirometry use and current medications. They also need to be encouraged to utilise register and recall systems and have access to appropriate support systems. Patients need information about the seriousness of asthma and encouraged to manage their asthma in a proactive manner. Educating the community about the Asthma 3+ Visit Plan will give people an understanding of the process, validate the GPs intentions and may prompt patients to initiate the approach with their

GP.

Implications for policy, delivery or practice

The findings of this research will be used as baseline data for the Asthma 3+ Visit Plan initiative and to inform the debate around

2002 General Practice and Primary Health Care Research Conference — Proceedings

93

developing strategies to ensure its successful implementation.

Quality improvement in diabetes at practice and Division level

JR Burns, Centre for General Practice

Integration Studies, University of New South

Wales, G Zingarelli, PG Powell Davies,

MR Harris

Objectives

To develop a model of quality improvement in diabetes management which uses data collected at practitioner level and aggregates it to Division, state and national levels and invites peer comparison amongst practitioners and Divisions.

To test whether a quality improvement activity aimed at GPs and Divisions of

General Practice around diabetes management could assist with uptake of the data and prompt changes at practice and program level.

Methods

The NDDP Data Collation Project collated aggregated Division data on clinical indicators from 27 Divisions of General Practice using a paper-based proforma. Unit level data on the clinical indicators was collected electronically from seven Divisions who use the same register and recall system (CARDIAB). The results were analysed using SPSS and Excel and descriptive statistics generated.

Participants from the 27 Divisions were invited to a small group learning workshop aimed at GP Program Managers and Diabetes

Program Officers separately. Eleven from 27

GP Program Managers attended and 19

Diabetes Program Officers attended. A structured workshop was conducted where

Divisions reviewed results and developed interventions at practice or program level. A pre- and post-workshop evaluation was conducted on the day of the workshop. A follow-up study was conducted six months later as a phone survey using a structured set of closed and open ended questions.

Principal findings

The quality assurance workshop was successful according to participant ratings.

There was an increase in the discussion and usage of the results of the Data Collation

Project which participants attributed to the

QA workshop. Interventions were developed and implemented. Improvements to programs were made. GPs reported changing their own practice behaviour in managing patients with diabetes.

Discussion

The NDDP Data Collation Project of 1999–

2000 gathered data on a number of clinical indicators relating to clinical management of people with diabetes. These data were aggregated to Division level to give a comparative picture of diabetes management at Division level. The results were used by both practitioners and by Divisions for different purposes. A conference where the results were released initially suggested that a more intensive workshop style session was necessary to enable both GPs and Divisions to further explore the data with their peers in a small group learning environment. The follow-up phone survey confirmed that further learning and usage of the results would probably not have taken place without such an opportunity. Having provided the opportunity, it was clear that both practitioners and Divisions used the results to make changes to their practice and programs. Participants have indicated an interest in participating in this study when it is repeated later this year.

Implications for policy, delivery or practice

This study highlights the importance practitioner and practice level data can have at Division and state and national levels. It also highlights that the quality improvement cycle can operate at practitioner/practice,

Division and state/national levels.

Conclusions

Research where data is collected from practitioners, practices and Divisions requires a feedback loop to those levels in order for change to result. Small group learning is a useful way of doing this. Given such an opportunity, GPs and Divisions have shown a willingness to make changes in practice which they believe will improve their practice or program.

A review of models of consumer participation in primary care

A Butler, L Allwell, National Resource

Centre for Consumer Participation in Health

The National Resource Centre for Consumer

Participation in Health has examined the literature on existing models of consumer and community participation in primary care in Australia, focusing on four areas:

Community Health, Divisions of General

Practice, the Australian Coordinated Care

94

2002 General Practice and Primary Health Care Research Conference — Proceedings

Trials, and the Chronic Disease Self

Management Program.

Objectives

The objectives of the review were to identify the issues which affect consumer participation in primary care, how these might differ from those which are relevant to other sectors, and to point to lessons which have been learned to date.

Method

Literature review.

Principal findings and implications for policy and practice

On the whole, the strongest models of consumer and community participation in primary care have emerged from the community health field. By comparison, the prevailing models of participation in the

Divisions of General Practice and the first round of Coordinated Care Trials have tended to be those of information seeking and provision, with some consultation. The

Chronic Disease Self Management Program is still at an early stage, but offers the potential for strong models of collaboration and partnership at both a project and individual level. The primary care sector has become much more diverse over the past 30 years, and that this has resulted in a diversity of models of consumer and community participation.

Conclusions and lessons from the literature include:

X

Everyone is a primary care consumer, therefore the theoretical potential for consumer and community involvement is high. However, while this potential has been realised in some situations, it can be difficult for consumer groups to respond and reorganise when primary care is restructured and regionalised.

X

Models from one part of the primary care sector do not necessarily or automatically translate or expand to other parts.

X

Despite the overall trend towards greater consumer involvement in primary care, it is still possible for new initiatives in the sector to reach quite an advanced stage of development without any consumer or community input.

X

Formal support for consumer and community involvement in primary care from the highest level of policy making is essential if gains made and lessons learned are to be transferred and built upon.

X

Peak consumer bodies require resourcing and support to allow them to plan to meet recent and emerging challenges in the primary care sector.

X

Consumer bodies at all levels require support and training to enable them to articulate and strengthen the connections between individual consumers and consumer groups, and between local and peak groups.

X

The experience of community development in health demonstrates the importance of encouraging the emergence of new groups and coalitions.

X

Since there is a tendency for different types and models of community and consumer involvement to reinforce one another, it is important to encourage participation at all levels of the primary care sector, from that of the individual client/practitioner relationship through to decision making at the national and state level.

The Commonwealth Department of Health and Ageing is funding the NRCCPH Primary

Health Care Project.

Results from two years of research capacity building in a remote area

E Chalmers, Mount Isa Centre for Rural and

Remote Health, James Cook University,

J Hollins, P Entwistle

Objectives

To describe the outcomes of two years of research capacity building activities in a remote area

Methods

This paper reports on a process, already described, of research capacity building in a remote setting (1). The factors considered relate to: national or external issues; the individuals concerned — their academic qualifications, research skills and motivation: and departmental factors such as time for research, research culture and subject specialisation. Recommended evaluation (2) reports on the number of papers published and presented at conferences, recruitment of skilled workforce and number of health professional conducting research projects

2002 General Practice and Primary Health Care Research Conference — Proceedings

95

Principal findings

A variety of process activities have been successful: Research interest groups of staff and local clinicians meet fortnightly; two research skills training workshops have been held and visiting researchers have delivered smaller workshops; interdepartmental links within the university are established with six joint submissions for funding developed.

There are two postgraduate research students in the Centre

The program was enhanced by two events — the Centre becoming part of James Cook

University and the PHC RED program;

Out puts include the following: a total of four papers have been published, thirty-one conference papers presented; fourteen grants applications submitted, five of which have been funded. The Centre has recruited two new research staff with postgraduate research qualifications and experience. Some four local health professionals and seven

Centre staff are currently engaged in health research.

Discussion

The MICRRH research program had three initial questions — Is it possible to conduct high quality research in remote areas? How do we go about it? and What should the focus of research be? The advent of the PHC-

RED program and the move to a university environment have boosted the scope of both the range of activities and the expertise of staff recruited. The limitations are that the program remains linked to sustainability, which remains unknown at this stage. The focus of research is the local health workforce and local health services. Planned expansion of the JCU School of Medicine to develop a remote clinical training centre in the near future may enhance the research program

Implications for policy, delivery or practice

The focus of the Centre’s research program on local health services and professionals as being key to the MICRRH research program has implications for recruitment and retention of remote health staff. In addition, the types of research conducted should strengthen the newer methodologies for evaluating complex care for complex health conditions in remote practice.

References

1. Smith D, Chalmers E, Hollins J. Research

Capacity Building in a University Rural

Department in press proceedings of

ACRRM Annual Scientific Meeting 2001.

2. Trostle J. Research capacity building in international health: definitions, evaluations and strategies for success.

Social Science and Medicine

1992;35:1321–1324.

The ‘NACCHO Ear Trial’ — Aboriginal community controlled clinical research into chronic suppurative otitis media

S Couzos, National Aboriginal Community

Controlled Health Organisation, T Lea,

M Culbong

Objective

To seek better ways to treat chronic suppurative otitis media (CSOM) in Australian

Aboriginal children using a research process where Aboriginal health services take control and are the lead agencies.

Methods

Double-blind multi-centre randomised controlled trial involving eight Aboriginal

Community Controlled Health Services commenced in April 2001. The research team comprised 13 Aboriginal Health Workers

(AHW’s) already employed in the services.

Principal findings

Over 130 Aboriginal children have been treated for CSOM representing the largest

CSOM study ever conducted in Australia.

Children assessed longitudinally over a 12week period for clearance of initial otorrhoea, recurring discharge, tympanic membrane healing, hearing status and school attendance. Discussion will use the Trial to exemplify how clinical research affecting

Aboriginal and Torres Strait Islander health should be conducted.

Discussion

Most research involving the Aboriginal population is descriptive, clinical trials are rarely national, often small-scale, local capacity is not fostered, and often results are inapplicable to the primary health care setting. Aboriginal ‘participation’ (primarily as subjects) rather than ‘partnership’ in the research process is the norm.

In 1998, the peak Aboriginal health body in

Australia, the National Aboriginal Community

Controlled Health Organisation (NACCHO) convened and led a research consortium to investigate the role of ototopical ciprofloxacin and Sofradex in treating CSOM in Aboriginal children. Until this time, no Australian clinical trial had reported on the impact of ototopicals in CSOM in Aboriginal children.

96

2002 General Practice and Primary Health Care Research Conference — Proceedings

Australian Aboriginal children suffer CSOM at rates similar to those seen in children from developing countries. This infection, a disease of poverty and deprivation, should not exist in Australia. The World Health

Organisation reported in 1998 that a CSOM prevalence > 4% in a defined child population indicates a massive public health problem needing urgent attention. Many

Aboriginal communities in rural and remote

Australia have rates exceeding this.

Unfortunately, there is much clinical practice uncertainty regarding treatment and a high level of community tolerance.

Building community capacity was a key feature of the NACCHO Ear Trial research process. Governing community boards selected AHW’s to undertake a training program. This team incorporated locally specific community health promotion, recruitment of children into the study, administration of treatment, undertook all assessments, follow-up over 12 weeks and data collection. Fostering the research partnership with other institutions through

NACCHO ensured that leadership came from the Aboriginal organisation and not the other way around.

Implications for policy and practice: This Trial is the first time research had been led and conducted by an Aboriginal organisation in

Australia. The conduct of quality research addressing research questions asked by

Aboriginal communities and the translation of these findings into practice requires a true partnership between research agencies and community representative bodies. Funders of research should recognise this interdependence when commissioning and appraising research proposals as policy makers depend on quality research to direct resources into effective, acceptable and much needed programs for disadvantaged populations.

An evidence-based literature search service for general practitioners.

A feasibility study

C Del Mar, Centre for General Practice,

University of Queensland, C Silagy,

P Glasziou, D Weller, A Spinks,

V Bernath, J Anderson, D Hilton

Objectives

To test the feasibility of running an evidencebased clinical literature search service to help answer general practitioners’ (GPs) clinical questions.

Methods

Two services were established for 9 months:

GPs could send in requests for information, based on questions arising during clinical consultations. A search was undertaken for each request aimed at finding the best available empirical evidence to support the clinical decision. Clinical consultants provided a brief interpretation of the evidence. The inter-reliability of the answers provided was assessed for a sample of questions. GPs were also surveyed at the conclusion of the study about the service.

Principal findings

58 GPs asked 160 questions (half asked only one; 11 asked five or more), which were about treatments (65%), aetiology (17%), prognosis (13%), or diagnosis (5%). It took a mean of 3 hours 50 minutes of personnel time to answer questions (SD = 3 hours 26 minutes). Nine questions took longer than 10 hours each to answer, the longest taking 23 hours 30 minutes. There was evidence of suitable quality for the given question type to provide a sound answer for 79% (n=126) of questions submitted.

42 GPs responded to a feedback survey, providing data for 84 (71%) questions answered. GPs appreciated the service and felt questions altered clinical care, but found it difficult to formulate questions.

The inter-centre reliability study found many minor differences between the answers from the two centres to questions, with some different studies being identified. In 4/14 questions, there were substantial differences in the evidence found through the search strategy, perhaps reflecting differences in the way questions were reformulated. However conclusions were largely similar, with no or minor differences for all questions.

Discussion

An evidence-based literature search service is feasible. The number of questions asked by

GPs was surprisingly modest. This may reflect several factors, including the novelty of the process, fear of being seen to ask apparently naïve questions, lack of skills and appropriate attitudes in asking such questions, and time constraints.

Implications for policy

Funding such a service might provide benefits for improving the access of GPs and other primary care clinicians to the best available research evidence, along with other such interventions directed at evidencebased care. Further research to establish how

2002 General Practice and Primary Health Care Research Conference — Proceedings

97

effective it might be and whether the cost is worthwhile, would be expensive.

Multiple pathways to building capacity: a comparison of frameworks

N Dunbar, School of Public Health and

Community Medicine, University of NSW,

E Comino, M Harris

In recent years there has been increased interest in capacity building within the broader health system. In NSW initiatives include the development of a comprehensive framework for building capacity within the health promotion sector, as well as the development of indicators to measure change. Part of the NSW Government Action

Plan for Health also proposes to build primary care capacity at a regional level through the establishment of primary care networks within Area Health Services.

At a national level, the Commonwealth

Department of Health and Ageing launched a new Primary Health Care Research Evaluation and Development Strategy in 2000, which had a major focus on building research capacity in general practice and primary health care. This program is being implemented through 18 university departments of general practice and rural health nationally. In NSW, the university departments which have been funded under this program have formed the NSW Primary

Health Care Research Capacity Building

Program, which aims to increase the capacity for primary health care research and to provide leadership in developing a research culture in Australian general practice and primary health care. This is being achieved by focussing on developing and delivering education, establishing and supporting research networks and identifying potential research collaborations. Universities are working at both a state level and with Area

Health Services, Divisions of General Practice and individual GPs at a local level.

This paper will discuss these different initiatives, particularly in terms of the way they conceptualise capacity building, the focus of their implementation and their impact potential on primary health care in general.

The National Prescribing Service: translating research into practice across Australia

M Fitzgerald, National Prescribing Service

Ltd, J Bowman, M Kingsland

Objectives

To influence prescribing behaviour of all general practitioners through the implementation of an evidence-based, quality use of medicines program across Australia.

Methods

Applied research was undertaken to describe implementation models used by Divisions of

General Practice in October 2000. A mix of qualitative semi-structured interviews, selfreported data and quantitative data were gathered to identify strengths, weaknesses and factors that facilitate or act as barriers to implementation.

Six divisions of general practice were selected from across Australia which were thought to represent different models of implementation with regard to the NPS program. These divisions were also a mix of urban and rural areas and had been funded by the NPS to run a program for at least

9 months.

Quantitative data was derived from centralised data on actual participation in

NPS activities at the GP level. Data was aggregated to divisions of general practice using postcode of the GP which approximates to a division using a formula from the

General Practice Branch of the Department of

Health and Ageing.

Principal findings

Factors that were highlighted as essential for the success of the program were the dedication and skill of the facilitator, staff and management support and a productive steering group. Divisions nominated educational visits (academic detailing) as being the most successful component of their

NPS program. Factors that were viewed to contribute to the perceived success of this intervention were the high GP participation rates, personal interaction between the facilitator and GP and cost-effectiveness.

In this sample of divisions few had undertaken clinical audits with feedback.

Small group case studies were popular.

Continuing medical education events, were commonly mentioned as being the least successful despite satisfactory participation rates.

98

2002 General Practice and Primary Health Care Research Conference — Proceedings

The main factors, described by divisions, impeding local delivery were the competition with pharmaceutical companies, perception of the NPS as the government, limited facilitator time and lack of GP interest and participation.

Discussion

Translation of evidence into practice must be achieved with partnerships from many stakeholders, especially divisions of general practice. For this reason the NPS has been flexible in the local delivery of the national program. The national program available to divisions include educational visits, clinical audit with feedback, and small group case based discussions.

Educational visits are an evidence-based intervention that has been demonstrated to influence prescribing behaviour change.

Variation in the vocational training of the visitor, structure of the program in the division and resources allocated may all modify the degree to which the intervention may succeed.

At this time, prescribing data from either the health insurance commission or from the

GP’s own records are not readily available to provide quantitative evidence of success for this intervention.

Implications for policy, delivery or practice

The local delivery of the NPS program is influenced by many factors including the environment in which divisions operate. The

NPS has modified program delivery and training to be dynamic to the changing environment in which divisions operate. Most recently the domiciliary medication review program has commenced in divisions using a similar model to the NPS program. Useful prescribing data remains a challenge for all who work in this area.

Individualised medication effectiveness tests may improve osteoarthritis management and decrease NSAID expenditure

J Nikles, P Glasziou, Division of General

Practice and Community Health, The

University of Queensland, C Del Mar,

C Duggan, M Yelland, J King

Objectives

To improve decision making about long-term medications for osteoarthritis by using

Individualised Medication Effectiveness Tests

(IMETs) (Single Patient Trials or n-of-1 trials).

To estimate the economic impact of conducting IMETs on Australian PBS NSAID expenditure.

Methods

Self-selected patients with a clinical diagnosis of osteoarthritis with pain of sufficient severity to warrant consideration of longterm NSAID use, undertook a within-patient randomised, double-blind, placebo-controlled cross-over comparison of ibuprofen with paracetamol. Using a double-dummy design, participants underwent three four-week treatment periods. Pain and stiffness were assessed daily using visual analogue scales, with side effects and a global assessment self-assessed on a weekly basis.

Principal findings

97 patients commenced and 47 completed the IMET to date. There were 8 probable and

4 definite responders to ibuprofen. Of the 15 who were on NSAIDs initially, either alone or in combination with analgesics, 3 continued and 12 ceased taking them immediately postIMET. A clinically useful decision has been made for all participants, with the IMET changing medication management in 31 out of 42 (74%) where results are available.

Discussion

IMETs may have significant and lasting impact on clinical management of chronic conditions such as osteoarthritis.

Implications for policy, delivery and practice

In 2000–01 there were over 6.1 million prescriptions dispensed in Australia for antiinflammatory and anti-rheumatic products, at a total cost to the PBS of over $163 million.

If 30% of patients discontinue

NSAIDs/antirheumatics for at least a year after doing an IMET, NSAID/antirheumatic expenditure by the PBS would decrease by a total of $54 million dollars per year in

Australia.

Non-insulin dependant diabetes mellitus and Indigenous communities: a research partnership

D Gorman, Faculty of Sciences (Nursing),

University of Southern Queensland, O Best

Non-insulin dependent diabetes mellitus

(NIDDM) in Indigenous peoples of Australia is one of the most pressing and significant health problems in contemporary Australian society. Although NIDDM can be controlled and managed through lifestyle changes, such an approach has not yielded results in

Indigenous people.

2002 General Practice and Primary Health Care Research Conference — Proceedings

99

This paper reports on research that particularly emphasised a partnership with the Indigenous participants to explore the importance of traditional beliefs and health practices of Indigenous people related to

NIDDM.

Objectives

Determine the issues of primary concern to

Indigenous individuals with NIDDM.

Develop a support network for the people experiencing NIDDM in this community and facilitate an ongoing sharing of information productive for managing NIDDM Aboriginal

Health Service.

Methods

This exploratory study was planned with the an Aboriginal Health Service and they were involved in all stages of the project to ensure that it identified their areas of concern.

Initial contact was made with the Community

Controlled Primary Health Care Service whose approval and assistance was sought to involve the local health workers as Field

Workers from amongst the community. Field

Workers, who were experienced Aboriginal health workers’ were given training in data collection by the researchers to undertake interviews and recording of information.

Members of the Health Service Community who with an interest in NIDDM were invited to a public meeting.

The researchers held bi-weekly face-to-face meetings with the field workers and additional contact by phone at regular intervals.

The research participants kept a record of their health/illness behaviours in a way that was culturally appropriate for them.

The field workers met with the participants on a daily basis and recorded information for a period of three months.

The analysis of the data included statistical analysis of demographic data of participants and of health records from the Health

Service. Qualitative analysis using NUD IST was carried out to identify the themes of the interview data and health diaries.

Principal findings

That Indigenous people are concerned about their diabetes and willing to make lifestyle changes if it is introduced in a culturally appropriate way and if they are involved in the development of strategies that are relevant to them.

Discussion

While the lifestyle changes achieved are commonly accepted as directly related to diabetes the process of implementing it with

Indigenous people is critical to its success and needs to be different from that used with mainstream Australian. The Indigenous people need to be directly involved in the development of the strategies to ensure not only ownership but relevance.

Implications for policy, delivery or practice

Community involvement is essential for success. This means that the traditional strategies of treatment and advice to individual sufferers is insufficient. A broader social approach is necessary to bring about the lifestyle changes necessary to manage the individual’s problems associated with

NIDDM which go beyond the symptoms.

Bells and whistles (and carrots): exploring strategies to improve GP response rates

C Harris, Department of General Practice,

University of Melbourne, N Day, D Young,

D Southern, D Dunt, G Elsworth

Objectives

To assess the acceptability and feasibility of widespread uptake of the ‘GP Integration

Index’ by Divisions as part of their ongoing quality enhancement programs.

The GP Integration Index is a tool to measure GP integration within the wider health care system. Achieving good response rates in wide scale GP surveys is notoriously difficult. This pilot project investigated strategies to maximise returns

Methods

Engaging Divisions: 25 Divisions were randomly selected based on proportional representation from all states and the rural/urban split. Several strategies were employed to encourage and facilitate divisions’ participation.

Engaging GPs: ‘Bells’ (Getting their attention): Two strategies were outlined — the traditional approach of mailing surveys directly to GPs or sending individually addressed letters for all GPs in a practice to the Practice Manager for distribution, collection and return.

‘Whistles’ (Reminders): A detailed reminder schedule was recommended.

100

2002 General Practice and Primary Health Care Research Conference — Proceedings

‘Carrots’ (Incentives): Each division was provided with $5000 towards the costs of implementing the survey and many used it to provide incentives to GPs.

Principal findings

Almost 4000 questionnaires (3924) were distributed to GPs in the participating

Divisions and the average response rate was

65%.

General observations suggest that response rates were better in smaller Divisions; those

Divisions who sent their questionnaires to the

Practice Managers rather than directly to

GPs, particularly if they made personal contact; and Divisions where a cash or gift payment was used. Reminders improved response rates — ‘letter plus phone call’ was better than ‘letter alone’ and response rates improved with additional reminders

Discussion

Several strategies have been identified to increase response rates, however they are difficult to implement in large Divisions. It is interesting to note that one large Division

(over 200 GPs) surveyed a randomly selected group of 80 GPs, achieving an 86% response rate. They received a similar number of responses to the amount they would have received with the 30% response rates that

Divisions of similar size attained by surveying all their members. The sampling method resulted in a similar total number of responses, but has the added advantage of reducing the response bias if the sample is selected at random.

Implications for policy, delivery or practice

These results have implications for future wide scale surveys of GPs. Surveying a large number of GPs is costly in money and time and strategies that improve response rates are more difficult to implement with very large numbers. Perhaps large divisions or others administering wide scale GP surveys should consider whether sampling is a feasible option. If the information can be obtained from a specific group e.g. GPs with a special interest, group with similar demographic characteristics, etc, then focusing resources and applying the various strategies to the identified group rather than the population as a whole will increase the response rates.

Illicit Drug Shared Care Project: education and training needs of general practitioners providing treatment for people who use illicit drugs

SL Purcell, Centre for General Practice,

School of Community Medicine, University of

New South Wales, MF Harris, G Powell

Davies, J Copeland, E Comino

Objectives

The Illicit Drug Shared Care Project is funded by the Commonwealth to develop, trial and evaluate a model for the development of shared care for primary health care providers in an urban and rural area. Part of this project involves the identification of education and training needs for primary health care providers, predominantly General

Practitioners and Drug and Alcohol workers.

Methods

The identification of education and training needs for these groups relate to illicit drug treatment issues and to systems and mechanisms for working successfully in shared care arrangements. Pre-trial questionnaires were conducted with 104 service provider participants in an urban and a rural area. The questionnaires consisted of four main components: illicit drug treatment and prevalence; confidence, knowledge and qualifications of providers; current levels/involvement in shared care; and a section devoted to identifying the education and training needs of providers. These identified education/training needs were analysed in terms of service provides in general and then broken down to identify any differences in the needs of urban and rural providers, and also any differences between

General Practitioners and Drug and Alcohol workers

Principal findings

Results indicate different needs for urban and rural service providers with those in rural areas being less frequently confronted with some illicit drug problems and less reliant on support from specialised services. Those in urban areas were more likely to be dealing with a range of problems and interventions but also more concerned about difficulties with support from specialised services. The results also confirmed the findings of other research where GPs indicated that they didn’t feel adequately equipped to deal with certain illicit drug treatment issues.

2002 General Practice and Primary Health Care Research Conference — Proceedings

101

Implications for policy, delivery or practice

GP education and training needs to be incorporated at a policy level, as there is a need for better qualified and confident GPs working with people who present with illicit drug concerns. Linking in Enhanced Primary

Care Items (EPC) for care plans and case conferences are important for delivery and practice issues. Utilising these items can provide GPs with reimbursement for the extra time taken for coordinating care plans, and care plans can provide mechanisms of support for GPs as there are then other services and providers involved in treatment.

How effective is primary care-based cardiac rehabilitation in a rural setting?

A Hawkes, Northern Queensland Rural

Division of General Practice, K Battye,

M Harris, G Powell Davies

Objectives

An innovative general practice-based cardiac rehabilitation (CR) program (Outreach

Program) was designed for rural and remote patients who had no access to hospital or facility-based CR. The objectives of this study were: to compare patient outcomes of the

Outreach Program with the outcomes of standard facility-based CR, and to compare these results with usual care (patients receiving no CR); to investigate the patient outcomes further in focus groups; and to investigate the general practitioners’ (GPs) perspective in focus groups. The National

Innovative Funding Pool, Australian Divisions of General Practice, Commonwealth

Department Health and Ageing funded this study.

Methods

Cardiac surgical patients were recruited predischarge from Townsville hospitals. Patients completed a self-administered questionnaire in hospital (baseline), and at 2, 6 and divided in to 3 groups. Group 1 included patients who were enrolled in the Outreach

CR Program. Group 2 included patients who had attended a facility-based CR program in

North Queensland. Group 3 included those patients who had received usual care.

Changes in coronary heart disease risk factors were assessed at each time point, and comparisons were made between the 3 groups. A series of focus groups were also run with GPs in North Queensland rural communities, as well as with cardiac patients.

Principal findings

604 patients were recruited (68% response).

At 2-month follow-up, patients indicated whether they had participated in facility based CR (26%), Outreach CR (34%), or usual care (40%). The main reasons for patients not participating in CR included remoteness, a lack of interest, or returning to work. At 2-month follow-up, compared with baseline, there was a significant decrease across the 3 groups in the proportion of patients with hypertension, elevated cholesterol, chest pain, shortness of breath, who were inactive, who drank more than the recommended level of alcohol, who were overweight, and who had higher than recommended fat intake. There was an increase in the number of patients who had ceased smoking, and a decrease in mean body mass index. However, there were no further significant changes from 2 to

12 months follow-up. Comparison of the 3 groups revealed that at 2-month follow-up, a higher proportion of CR (Outreach and facility based) than usual care patients, were active, and limited alcohol intake to recommended levels.

Discussion

Participation in CR does not appear to be associated with patients’ risk factor status rather; accessibility and availability appear to be the most important predictors of CR attendance in North Queensland. The current models of CR are effective in promoting activity and reducing alcohol intake in cardiac surgical patients, but do not promote further coronary heart disease risk factor changes above the behaviour changes observed with usual care.

Implications

This is the first study of its kind in Australia and the lessons learned may influence the future direction of CR in rural and remote areas.

Screening for chlamydia in general practice

C Heal, James Cook University

Objectives

To determine the prevalence of genital

Chlamydia trachomatis infection in a general practice population in Mackay.

To evaluate criteria for selective screening, and determine whether screening on the basis of age alone would be viable.

102

2002 General Practice and Primary Health Care Research Conference — Proceedings

To determine whether men should also be screened for chlamydia.

Methods

Design: Cross sectional survey.

Setting: Six general practice clinics, and a

Youth Clinic (run by General Practitioners) in

Mackay, North Queensland.

Subjects: 508 men and women aged between 18–24 attending their GPs for any reason during the testing period.

First catch urine sample was tested for C

trachomatis using Ligase Chain Reaction

(LCR) testing.

Principal findings

A total of 745 subjects were eligible to participate, of whom 508 did. 127 candidates refused or failed to return a sample. A further 110 were not invited to participate through oversight.

Of 508 samples, 25 were positive, with an overall prevalence of 4.9%.

Youth Clinic attendance and recent change in sexual partner were found to be significant risk factors for infection. Age under 20yrs tended towards significance as a risk factor.

The prevalence was slightly higher in men than women. No other potential predictive factor was shown to be significant.

Discussion

Estimated overall prevalence of 4.9% is comparable with similar primary care studies in the UK.

Applying NHMRC guidelines to women did not appear to significantly improve detection rate.

Screening for chlamydia satisfies the Wilson and Junger criteria. The cost of diagnosis and treatment should be balanced in economic terms in relation to total expenditure on medical care. Recent computer modelling of

Victorian data has estimated that screening would become cost-effective at a prevalence of 2.1%.

This project helped to build research capacity within the general practice community in a rural centre (Mackay). The researcher learnt how to develop a research question in a clinical setting and how to conduct a research project. Ten General Practitioners collaborated to collect data. The research findings have been disseminated locally to enable practitioners to apply them to their practice. It is possible for a clinician in private practice to develop effective research.

Implications for policy, delivery or practice

Implications of this study for the GP:

X chlamydia screening is generally well accepted by the younger community

X evidence supports screening all under

25yr olds

X teenagers, youth clinic patients and patients with a recent change in sexual partner are at increased risk of infection

X men should be screened as well as women.

Measuring the effectiveness of community health services

G Jolley, South Australian Community

Health Research Unit, S Masters, F Baum,

J Duffy, D Bament

Performance and outcome measurement has become evident throughout industry and public services in most developed countries.

These systems of measurement have also been applied to the primary health care sector. While this paper will argue that it is important to know if these services are effective and of high quality, it will also raise many challenges and issues associated with developing appropriate systems for the assessment of effectiveness and quality in primary health care. Research examining current evaluation and performance methods in South Australian community health services (CHS) and investigating the potential for performance and outcomes measurement, will be presented.

Objectives

Within the community health sector, the objectives are to:

X conduct a literature review of economic analyses of health promotion and the feasibility of these methods for community health.

X assess the application of systematic reviews

X assess the use of performance measures

X describe current data systems and analyse their potential for use in measuring effectiveness.

2002 General Practice and Primary Health Care Research Conference — Proceedings

103

This paper focuses on a critical assessment of the use of performance data as a way to measure the effectiveness of community health services.

Methods

Research was conducted through literature reviews, a consideration of the national agenda for performance indicators in primary health care and a local project to develop performance indicators for the sector.

Interviews were conducted with ten senior health department bureaucrats and CHS executives on the use of current data systems and what data might be needed to address issues of effectiveness.

Principal findings

While methods of economic analysis, systematic review and performance measurement have been applied to selective health promotion interventions, most typically those using a behavioural framework, there are few examples of their use in the community health setting. This research suggests that these models are limited in their applicability to assessing the outcome effectiveness of community health services. Currently collected performance data are limited to activity and outputs, with some quality measures such as accreditation status. Little data are available by which to judge effectiveness or outcome.

Discussion

The nature of much community health work is such that demonstrating causal links between interventions and outcomes is very difficult. CHS play only a small part in the determination of health status, and have little control over the many factors which impact on health and well-being in their communities. Much of the potential benefit is long term and is about prevention rather than incidence of illness or disease. These challenges are especially evident when assessing capacity building approaches used by CHS.

Evaluations of effectiveness require recognition of qualitative experiences as well as quantitative measures. Measurement of outcomes remains outside the scope of data systems and the capacity of agencies.

Frameworks for assessing qualitative evaluation are likely to be of more assistance in judging quality of community health services.

Implications for policy, delivery or practice

It seems unlikely that effectiveness in the community health sector can be demonstrated by quantitative data alone.

104

Policy makers, funders and practitioners should be aware of the need for a long-term, realistically resourced research effort to enable further development of rigorous methodologies for evaluation of the community health sector.

A longitudinal cohort study of 1990

South Australian final year medical students

G Laven, Department of General Practice,

The University of Adelaide, J Newbury,

J Marley

Objectives

Determine the factors affecting career choice among South Australian medical graduates by analysing career intentions and background influences reported in serial surveys of the cohort of 1990, final year medical students.

Methods

In 1990, final year medical students from

The University of Adelaide (N=82) and

Flinders University (N=51) were invited to participate in a prospective survey of factors influencing their career intentions with particular reference to rural practice. A questionnaire was developed to record demographical data, educational background, factors influencing their choices for rural or metropolitan practice, their career plans and preferences. The questionnaire was completed by 76.7% (102/133) of the cohort.

In 1996, the 1990 cohort completed a second postal questionnaire. Accurate addresses were obtained from all the Australian state medical boards for 102 members of the cohort. Completed questionnaires were returned by 89 doctors giving a response rate of 87.3% (89/102). In total 55.6% (74) have participated in both surveys. Two of the original cohort (1.5%) are deceased.

Anonymity is being maintained by a unique numerical identifier throughout the 12-year study period. In 2002, 110 of the original cohort were registered as Medical

Practitioners within Australia.

Registration and contact details could not be found for 15.8% (21/133) of the cohort.

Contact with the two universities determined that four of the 21 members of the cohort for whom no registration details could be found, were enrolled as International students.

2002 General Practice and Primary Health Care Research Conference — Proceedings

In March 2002 a third questionnaire, together with a brief summary of the project to date, was sent to 131/133 members of the cohort — the 110 identified medical practitioners and for the 21/133 members of the cohort for whom no Australian medical registration details could be found, the last known address was used.

Principal findings

Interim analysis will be presented.

Discussion

Final analyses will explore career intentions and career choices over the 12-year period

1990 to 2002. Analyses will also explore influences on career choices particularly associations between urban/rural background, spouse/partner influences and current practice location.

Implications for policy, delivery or practice

Longitudinal cohort studies of medical students, comparing their career intentions with career choices may facilitate workforce planning. This study has the potential to influence recruitment policy for medical schools and scholarship schemes throughout

Australia.

GASP (GPs Assisting Smokers

Program): findings from the pilot

JCB Litt, Department of General Practice,

Flinders University, T Shelby-James,

D Edwards

Objectives

To improve the practice-based approach to identifying and assisting patients who smoke and wish to quit through:

X viewing the practice as the target population

X taking a long term view with repeat cycles of activity and review

X developing networks, acknowledging expertise and respecting autonomy

X focus on implementation and sustainability.

Methods

GPs from 2 urban and 2 rural divisions of general practice in South Australia were invited to enrol in the study. The interventions were:

X an interactive training workshop for both

GPs and practice staff which promoted proven strategies and raising the issue of smoking, and:

X practice visits to tailor the strategies to the unique needs of the practice.

Pre and post evaluation consisted of GP and practice staff questionnaires as well as a survey of 50 patients per GP to validate GP self-report.

Principal findings

Workshop: Measurements taken before and immediately after the workshop indicate an increase in GP confidence and willingness to undertake key strategies.

3-month follow-up: There was an increase in

GP self-reported ratings of skills, confidence

(59.2% pre – 68.4% post) and effectiveness

(43.7% pre – 54.7% post). GPs identified a number of barriers (e.g. patient factors, time) to providing smoking cessation activities in general practice, these did not change as a result of the intervention. Selfreported changes in GP practice were validated by the patient surveys, in the 3month follow-up period the rate of enquiry about smoking increased 10%, discussion about smoking increased 20%, this led to a

17% increase in attempts to quit.

Discussion

This study has developed strategies that are acceptable to GPs and practice staff working in general practice in South Australia. We have tailored the strategies to address the main barriers facing GPs and provided long term skills that can be used for other lifestyle diseases.

Implications for policy, delivery or practice

GASP is supported by a broad consortium of organisations with an interest in smoking cessation. This will allow the strategies that have been developed by GASP to be incorporated into a number of organisations therefore ensuring the sustainability of GASP.

The consortium is currently pursuing further funding to implement a second phase study examining the effect of GASP on quit rates and GP performance.

2002 General Practice and Primary Health Care Research Conference — Proceedings

105

Rhetoric or reality? Stumbling blocks in the partnership between government, doctors and allied health

D Malone, Australian Physiotherapy

Association

Objectives

Australian health policy and services would be stronger if there was a more effective three-way partnership between the Federal

Government, the medical profession and the

90,000-strong workforce of allied health professionals. This paper looks at obstacles to building this partnership, and what is needed to put more reality behind the rhetoric of multi-disciplinary health care.

Methods

The paper is based on the experience of the

Australian Physiotherapy Association and the peak allied health organisation, the Health

Professions Council of Australia, in negotiations and liaison at a national level with the Federal Government and medical organisations such as the Australian Divisions of General Practice.

Principal findings

The Federal Government needs to be encouraged to move from rhetoric to real action, in relation to multi-disciplinary health care. In recent years the Government has been overly focused on the role of medical practitioners and nurses in the delivery of primary care, whilst paying lip-service to the importance of allied health. The new Federal

Health Minister, however, appears to have a strong appreciation of the role of allied health, and there are hopeful signs that the three-way partnership between doctors, nurses and allied health can be substantially strengthened.

There are encouraging signs of a more effective liaison between Divisions of General

Practice and allied health. The peak allied health organisations are working closely with the Australian Divisions of General Practice to foster multi-disciplinary approaches to health service provision.

Discussion

There is tension between the Federal

Government’s growing recognition of the need for multi-disciplinary health care, and its long-standing funding focus on doctors and nurses. The extraordinary complexity of health funding in Australia, with funding fragmented between Federal, State and local governments, works against the delivery of seamless health services; and the approach which says that doctors and nurses are

‘Federal’ and allied health is ‘State’ is not in the interests of Australian health consumers.

There are also particular issues in relation to health services in rural and remote areas, and for Indigenous communities.

National organisations such as the Australian

Physiotherapy Association and the Health

Professions Council of Australia are increasingly working with governments and other decision-making bodies to find ways to address this funding impasse.

Implications for policy, delivery or practice

A stronger partnership between the Federal

Government, medical practitioners and allied health will deliver better clinical services for consumers, and provide a sure foundation for research and evidence-based practice.

A stronger allied health focus would also make federal funding go further.

Physiotherapy and other allied health services rarely involve the use of expensive equipment, and tend to be very cost effective. Their strong focus on preventative therapies can actually reduce health expenditure.

Specific programs such as Enhanced Primary

Care and ‘More Allied Health Services’ need adjustment to ensure the real involvement of allied health professionals.

‘Take your partners by the hand …’

Engaging members of a GP special interest group as co-investigators

Z McCallum, Centre for Community Child

Health, Royal Children’s Hospital, Melbourne,

C Harris, J Gunn, M Wake, E Waters,

L Sanci

Objectives

To engage, recruit and train GPs from a child health special interest group as coinvestigators in a RCT; to participate in the research process; undertake research activities in their practices and to deliver an intervention.

Background: The ‘Health for Kids Network’

(HFKN) is a partnership between six Divisions of General Practice and two hospitals committed to improving child and adolescent care in the community. The HFKN agreed to collaborate with two academic units (Centre for Community Child Health, Royal Children’s

Hospital Melbourne and Department of

General Practice, University of Melbourne)

106

2002 General Practice and Primary Health Care Research Conference — Proceedings

undertaking an RCT to evaluate an intervention to reduce overweight in Victorian primary school aged children.

Overweight and obesity is now one of the most prevalent chronic childhood health problems in Australia. Previous research has shown that GPs feel they have insufficient training and lack the resources to manage childhood obesity confidently. This project addresses these issues.

Methods

Engagement: HFKN became a formal partner in the research collaboration and an article about childhood obesity was published in

HFKN newsletter. HFKN GPs were sent a personalised letter inviting them to participate as co-investigators.

Recruitment: Information evenings were held and followed up by personal contact from the project manager. GPs were encouraged to contribute to the research process and a

Memorandum of Understanding was developed.

Training: Research methods were piloted in the practice setting and participants completed questionnaires. This information was used to develop three 2½-hour training modules covering both research methods and management of childhood obesity

Principal findings

Personalised letters were sent to the 336 members of HFKN and 57 (17%) expressions of interest were received. Thirty-nine GPs attended the information evenings and five had a practice visit. Thirty GPs were recruited.

The sample size required 35 GPs so the same process was repeated with an additional

Division who had expressed interest in the project. Letters were sent to 262 GPs with 26

(10%) responding. Fourteen attended the information evening and eight were recruited. Three GPs joined later having heard about the project, resulting in a total of 41 participants.

Thirty-five GPs attended the education sessions. Evaluation of the education modules revealed GPs were very satisfied.

There have been no dropouts following the education program.

Discussion

This project demonstrates that membership of a common interest group is a useful base for recruiting GPs to a research activity.

Engaging the group (HFKN) as collaborators in the project appears to have been a successful strategy in a number of ways — achieving the required number of participants, increasing GP research skills and informing the research process.

Implications for policy, delivery or practice

Encouraging GP special interest groups to sponsor research projects and engaging GPs as co-investigators rather than the traditional

‘data gatherers’ has potential for facilitation of research in primary care.

GP attitudes towards case conferences and their ideas about making them work

G Mitchell, Centre for General Practice,

University of Queensland, I de Jong,

M Lewis-driver, R Kennedy, C Del Mar

Research objectives

To study case conferences with specialist services, especially to:

X identify barriers and facilitators to implementation;

X develop a strategy for their implementation;

X focus on palliative care.

Study design

Focus groups were conducted with GPs

(some selected randomly from Divisions) in urban, regional, and rural settings.

Principal findings

33 GPs participated from a total of 207 approached (16%). GP practice characteristics and work patterns mitigate against participation in the case conference process as traditionally conducted.

Commonwealth bureaucratic processes required to claim a fee are onerous and comprise another major disincentive to participate. Routine use of case conferences will fail to occur if it is left for GPs to initiate the process. They will have to be initiated in the first instance by specialist services, who must be convinced that that there is benefit in this degree of interaction with GPs. A process for conducting case conferences that takes both work practices and commonwealth requirements into account must be identified. Such a program has led to the conduct of over 20 case conferences in two locations in Queensland in the context of a project in Palliative Care.

2002 General Practice and Primary Health Care Research Conference — Proceedings

107

Conclusions

Case conferences can be conducted successfully, with a clearly defined strategy to minimise the barriers and take into account work practices. They must be initiated in the first instance by specialist services, who must see major benefits to their patients in engaging GPs in this way.

Implications for policy, delivery or practice

Strategies to encourage specialist services to undertake this or similar process and engage

GPs should be developed. The

Commonwealth should consider simplifying the administrative requirements of case conferences, in consultation with GPs, to design less complex compliance requirements as a matter of urgency.

The social determinants of

Indigenous health in Australia

M Morrissey, Centre for Remote Health,

Alice Springs

Objectives

This paper aims to examine in a critical fashion the current ‘state of the art’ of social epidemiology in relation to Indigenous

Australians. It will examine both empirical and theoretical work with a view to suggesting a future research agenda and an improved methodology.

Methods

A critical review of recent Australian work in the field of social epidemiology, set in the contexts of the health status of the

Indigenous population, recent international literature and wider social theory (particularly health social science).

Principal findings

The paper will conclude that the vast bulk of recent research in social epidemiology is inadequate as a guide to action that can effectively confront the pandemic of chronic disease which is currently threatening the very existence of Indigenous society. This is largely due to:

X lack of theoretical rigour, exemplified, in particular, by indifference to social theory;

X a failure to confront the multi-layered and complex aetiology of chronic disease with an appropriately multi-layered and holistic research method;

X a failure to engage Indigenous communities as active and equal research partners; and

X the implicit victim-blaming inherent in epidemiological approaches which concentrate exclusively on proximate risk factors.

The conclusion is that future work should be, to the greatest possible degree, outcomeoriented, transdisciplinary and inclusive of

Indigenous communities as equal research partners.

Discussion

This paper addresses, in the context of

Indigenous people, all four of the conference’s main themes since it:

X deals with relevant research questions;

X is concerned with methodology and epistemology;

X argues for partnerships between research institutes and Indigenous communities; and

X argues for capacity-building to facilitate research by Indigenous people themselves.

Implications for policy, delivery or practice

It is argued that good policy, delivery and practice are all dependent on the quality of knowledge (and hence, ultimately, the quality of research) on which they are based.

Using medical education to develop referral pathways from GPs to the community

A Mutch, School of Population Health, The

University of Queensland, J Dean, F Boyle

Chronic disease represents a significant and costly challenge to the Australian health system that is expected to increase with an ageing population. General practitioners play a central role in assisting people to manage chronic illness, yet people requiring high levels of ongoing care often need more support than is possible within the organisational confines of general practice.

As a service complementary to general practice, self help organisations (SHO) have a potentially important but, to date, largely untapped role in improving and supporting chronic illness care. However, in order to take advantage of community-based resources, the referral pathways linking the

108

2002 General Practice and Primary Health Care Research Conference — Proceedings

formal health system to the self help sector must be developed so that the health consumers who stand to benefit from community-based resources can gain access to them.

Objectives

This paper outlines an innovative program of community-based medical education for students completing a general practice rotation. The Building Partnerships Program is laying the foundation for the development of referral pathways between future primary medical practitioners and the SHOs. The

Program allows third year medical students to spend time learning in non-clinical community-based settings. The overall purpose of the program is to promote awareness and understanding of the ways a range of self help and community-based organisations contribute to chronic illness management and to consider ways to develop partnerships between general practice and community-based health providers.

Methods

Students and community organisations participating in the program are evaluated at the end of each 8 week rotation. Students are asked to complete a questionnaire, and organisational information is collected by questionnaire, telephone interviews and at a workshop for participating organisations.

Principal findings

The level of student engagement with the program has increased markedly with time.

Evaluation of the Building Partnerships

Program indicates that students report an increased awareness and understanding of the range of community resources, their roles and the services they provide. They also describe the benefits of experiencing wider viewpoints on health issues in the community, including an understanding of the ‘patient behind the disease’.

Discussion

To establish the benefits for consumers of

SHO-GP partnerships in the longer term, research evaluating the impact of the

Building Partnerships Program is essential.

We must investigate whether the changes in student attitudes observed following a community attachment extend into future medical practice, and whether students are more likely to establish referral pathways with SHOs and engage in collaborative models of patient care.

Implications for policy, delivery or practice

The Building Partnerships Program seeks to improve health care by acknowledging the contribution of SHOs to chronic disease management. By investing in medical education we can help students identify ways to establish referral pathways and draw on the under-recognised skills and resources that exist in the community.

Evaluative action research into primary care: a multi-method

Victorian study

G Whiteford, B Neumayer, Charles Sturt

University, L Peady, Upper Hume Primary

Care Partnership

In the north-east region of rural Victoria a model of primary care called the Primary

Care Partnership (PCP) is currently being piloted. This partnership model aims to enhance consumer participation, general practitioner and service provider interaction, and has as its ultimate goal coordination of care. It is expected this model will have a positive impact on reducing hospital admission rates and increasing the health outcomes of a range of user groups in the region.

Given that the PCP model is a new approach to coordinated care in the state, an evaluative research project was undertaken to determine the effectiveness of primary care strategies. The project tracked 70 older adults through the local health care system for up to 12 months. This multi-method evaluative study utilised both quantitative and qualitative strategies aimed at generating in-depth data as to the experience and impacts of the PCP model from both consumer and provider perspective.

This presentation will present some of the findings from the study and, based on these, posit some recommendations for future practice and research in the area of primary care.

2002 General Practice and Primary Health Care Research Conference — Proceedings

109

Effectiveness of a community-based

‘No Anaemia For Kids’ program in a remote Australian Aboriginal community

HT Nguyen, Top End Division of General

Practice, NT

Objectives

To implement community-based, community initiated multisectorial health promotional activities in order to improve awareness of relative contributions to anaemia and to reduce anaemia prevalence rate in infants and preschoolers of a remote Aboriginal community with a 50% prevalence of anaemia.

Methods

A needs analysis and analysis of local contributory factors to anaemia in children was presented to Health Clinic staff and other community organisations.

Health promotional activities were undertaken in almost all sectors and all organisations in the community. These organisations include the Women Centre, school, health clinic, general store, Respite

Centre, Homelands Service and Community

Council.

The main health promotional activities included: young mothers with young babies days at the Women’s Centre; school anaemia art competition, vegetable garden and physical activity days; educational sessions for young mothers run by Aboriginal Health

Workers (AHW) and a community health promotional fun day.

Evaluation is by monitoring biannual data of nutritional status collected by the community clinic for Growth, Assessment and Action

(GAA), an initiative of the Northern Territory

Department of Health and Community

Services.

Principal findings

Based on GAA data there has been no change in nutritional status parameters

(wasting and stunting). There has been a

38% reduction in anaemia rates (from Feb

2001 66% to Oct 2001 41%) in the first

8 months of the project. There appears to be no seasonal variation in anaemia prevalence rates in this community. If seasonal variation was taken into consideration, however, there would be at least 18% reduction in anaemia prevalence (from Oct 2000 50% to Oct 2001

41%).

Discussion

This project did not have the scope to evaluate the level of awareness and behavioural changes that may have taken place during and after the implementation of the project. Anecdotal observation from the

Women’s Centre, however, suggest a high uptake of health promotional message on anaemia and nutritional intervention but a lower internalisation of such knowledge and level of behavioural change. The outcome that matter in the end is a reduction in anaemia prevalence. There may other reasons for a fall in anaemia prevalence besides uptake of nutritional message. An increase in compliance with oral or intramuscular iron therapy may result from increase awareness of the consequence of childhood anaemia.

Implications for policy, delivery or practice

Aboriginal Health Workers initiated and contributed to this project. Community understanding of need for the project ensured success and participation.

Targeting all organisations in the community ensures that health promotional message reached all sectors of the community. Some sectors of the community, clans or families tend to gravitate to different organisations for example only certain families attend the

Women Centre.

Involving key organisation or services and influencing their programs, activities or curriculum reduces funding needs in implementing community health promotion activities.

ADHD a model of shared care

K Pedlow, ‘Youth Mental Health’, Midwest

Division Of General Practice, K Whiting,

G Bresland

Objectives

Prior to this project, funded through the Mid

West Division of General Practice, management of ADHD in Mid West Region of

Western Australia was less than satisfactory.

Parents were forced to travel to Perth for follow-up and ongoing care. Local GPs were not involved in the ongoing care of ADHD children.

The development of the share care model aimed at enhancing GPs’ education and information about ADHD in order to facilitate its diagnosis and management. It also aimed at increasing links and communication between paediatricians, GPs and health and

110

2002 General Practice and Primary Health Care Research Conference — Proceedings

educational professionals. Co-prescribing, referral protocols, local diagnosis and management by GPs and fast-tracking of therapy were specific objectives.

Through the share care co-prescriber arrangements between GPs, visiting paediatricians and the Stimulants Committee of the WA Health Department, GPs took on a case manager role. The specialists transferred their ‘mantle of authority’.

Method (N=±70)

Evaluation of the patient sample from the period 1998 to 2002

X parental satisfaction with the share care model of service delivery.

X a quantitative study of stimulant therapy efficacy and predicative factors for longterm adherence to therapy.

Qualitative measures: A questionnaire developed by the team seeks to gauge parental satisfaction with the project and the care they received by their GP and specialist.

Specifically we wish to sample if the information and support received by parents from diagnosis through to long-term management were satisfactory.

Quantitative measures: Concurrently, we also seek to ascertain the efficacy of stimulant therapy. A pre-post test design (using the

Massachusetts ADHD rating scale) will be used to gauge if symptoms have decreased

(paired t-tests) and if improvement has occurred on such variables as social interactions, academic achievement, parent– child relationship and the like.

Correlational studies and a regression analysis will investigate possible factors (age at diagnosis, medication regime, biological factors, adverse reactions to medications, gender, comorbidity etc.) predicating outcomes and medication compliance.

Principal findings

We have begun the data gathering phase of the study. We expect preliminary results by the time of the conference in May.

Discussion

Studies testifying to the efficacy of stimulant therapies in the short and midterm are many. Recently, studies have sought to assess which factors will predicate long-term medication use. Theruchelvan, Charach &

Schachar (2001) found that absence of oppositional disorders at school, higher levels of ADHD at school and diagnosis at a young age best predicted continuation of medication

> three years. The present study wishes to address these questions in a unique dataset in a rural/remote setting.

Implications for policy, delivery or practice

We expect a positive outcome relating to parental satisfaction with the model of shared care. The implications are for this model to be used as a blueprint for other regional areas where access to specialist’s services is restricted. This model allows GPs to delve deeper into sub-specialties and forge networks with other health professionals within their communities.

The gold standard applied to a complementary therapy — methodology for the PAV trial

M Pirotta, Department of General Practice,

University of Melbourne, J Gunn,

P Chondros

Objectives

The aim is to conduct a randomised doubleblind placebo-controlled trial in general practice to test whether oral and/or per vaginum lactobacillus will prevent postantibiotic vulvovaginitis (PAV).

Methods

The trial is a factorial design, requiring 124 women in each of four groups, to detect a clinically significant reduction in PAV from 23 to 11.5%. Eligible women aged 18–50 years are recruited through their local general practitioner. Women are requested to take both oral powder and per vaginum pessaries of lactobacillus (or placebo) for ten days.

They provide a self-effected low vaginal swab and complete a survey at recruitment and again at day fourteen. The measurable endpoint of cases will be by microbiological evidence plus women’s self-reported symptoms.

Progress

This multi-centred trial has ethics approval from the RACGP. Seventeen clinics have agreed to be involved in subject recruitment.

Funding has been obtained from the

Shepherd Foundation, which covers the costs of microbiological sampling, courier costs etc.

A vaginal form of lactobacillus has been sourced in Canada and both oral and vaginal lactobacillus and their placebos have been supplied. Block randomisation in groups of eight has been performed and a comprehensive database designed to deal with the data generated. The pilot phase is

2002 General Practice and Primary Health Care Research Conference — Proceedings

111

complete. Many process problems, in particular relating to the transport of specimens, have been addressed. Telephone interviews of the subjects reveal that the women who chose to take part did not find the study protocol to be too onerous and enjoyed their participation.

Discussion

The use of complementary therapies is common in Australia among patients who attend general practitioners. Many of those therapies commonly used are untested. Postantibiotic vulvovaginitis has become a problem largely hidden from general practitioners, since anti-fungal agents have been available from pharmacies over-thecounter. Anecdotally, general practitioners hear that many women use lactobacillus and other complementary therapies to prevent and treat PAV. This RCT builds on our previous work which, using HIC data and a survey of 1300 women in general practice, found that PAV is a relatively common problem for women, that they are concerned about it and that this concern may affect their decision making about concordance with antibiotic regimes. We also found that 41% of our sample had used the complementary therapy lactobacillus or yoghurt to try to prevent PAV. As there is currently no evidence that lactobacillus is effective in preventing post-antibiotic vulvovaginitis, this randomised controlled trial is needed to answer a common clinical question in general practice.

Implications

It is a logistical challenge to mount a trial of any therapy, based in private clinics in the community. However, with adequate funding, support and energy, it is possible. The results of this trial will inform women and general practitioners as to whether the use of lactobacillus either orally or vaginally is effective to prevent post-antibiotic vulvovaginitis.

Impacts of retrenchment over time: comparing retrenched steelworkers to employed controls

D Pond, Discipline of General Practice,

University of Newcastle, E Harris, P Magin,

A Sutton

Objectives

To compare a group of retrenched men on a number of health, social and psychological parameters, with an employed control group at baseline and 12-month follow-up.

Methods

In September 1999 the BHP steelworks in

Newcastle closed down. The Discipline of

General Practice secured funding from GPEP to follow-up a cohort of men retrenched by the closure.

A controlled longitudinal cohort study design was implemented. Retrenched study group participants were interviewed three times over a 12-month period following the BHP closure. An employed control group from

OneSteel was also recruited, providing comparison data for the baseline and 12month follow-up period. Data collected included demographics as well as physical, psychological and social parameters.

Principal findings

This presentation will a) summarise the comparison between study and control groups at baseline (previously presented), b) present the results of the comparison between study and control groups at 12month follow-up, and c) compare change over time in study and control groups using a mixed (time by group) analysis of variance design.

At baseline, the retrenched study group showed more stress and financial strain compared to employed controls after controlling for demographic differences, but no differences on other outcome measures.

By 12-month follow-up, those differences were no longer evident, but differences in body mass index (BMI) and social interaction emerged with the study group having higher

BMI (were heavier relative to height) and less social interaction than controls. A group effect was found over time for systolic blood pressure, BMI, social support (total score) and social interaction with the retrenched study group showing poorer outcomes on social support measures and BMI, but having lower systolic blood pressure over time when compared to employed controls. Significant group by time interactions will also be presented.

Discussion

Initial results show fewer than expected differences between groups. The interpretation of these findings needs to be considered in the light of overseas studies on the health effects of unemployment and qualitative data from this and other studies.

A number of factors may have buffered this population, including large pay-outs and a culture in Newcastle of respect for BHP workers.

112

2002 General Practice and Primary Health Care Research Conference — Proceedings

Implications for policy, delivery or practice:

The findings have been incorporated into educational seminars for GPs and an audit activity on the management of unemployed patients in the Hunter.

Delivery of mental health care in general practice

J Richards, University of Ballarat,

M McCabe, P Ryan, I Hickie, G Groom

This paper will report on the results of the

National Study into the Management of

Depression in General Practice, which was funded by beyondblue in 2001–02.

Questionnaires were sent to a cross-section of 54 urban and rural Divisions of General

Practice throughout Australia. The Divisions then distributed the questionnaires randomly to their members, in all, to 608 general practitioners. The response rate was 69% with a total of 420 questionnaires returned.

Respondents estimated that 1 in 5 patients present with significant anxiety and/or depression, with somewhat less meeting full criteria for a specific psychiatric diagnosis.

Urban GPs were more likely to use newer assessment and treatment methods, and those who had undertaken mental health training in the past five years or who had postgraduate mental health qualifications, were more likely to identify problems of depression and anxiety and to use nonpharmacological methods of treatment. Rural

GPs identified more barriers to the effective management of these patients. There was a strong stated need for more educational activities about the identification and management of depression and anxiety in general practice, although rural GPs identified more barriers to accessing it. These and other results from the survey will be discussed.

The heart of country medicine: retaining GPs in rural WA

S Roach, WA Research Unit, The Royal

Australian College of General Practitioners,

M Delane

Objectives

This research project aimed to identify the needs of partners of rural and remote doctors in areas that the Rural Medical Family

Network (RMFN) in WA currently offered support, as well as to determine areas and ways in which the RMFN could provide other services.

Methods

Focus groups comprised of partners of rural doctors in WA were conducted in the

Kimberley, Pilbara, Central Wheatbelt and

Upper Great Southern Divisions of General

Practice. A separate focus group was also conducted for male partners. The data were analysed qualitatively and the information gained was used to develop a questionnaire.

The questionnaire was mailed to all 413 partners of rural doctors in WA. Analysis of the questionnaire was undertaken using

SPSS 10.

Principal findings

Results from the qualitative analysis of focus group data identified five major issues and twelve recommendations for the RMFN to apply their resources. The questionnaire asked respondents to indicate how important each of the issues and recommendations were to them. The main issues were found to be on-call requirements, lack of locum cover, ability to pursue own professional and personal development and poor housing. The most highly supported recommendations were that the RMFN provide an information pack upon arrival to a town and that another

GP family welcomes them. The RMFN newsletter and the provision of local social events that bring GP families together were also strongly supported.

Discussion

It is important to note how the results varied between regions, even when those regions were geographically close. This emphasises the need to seek solutions at a local and individual level. A common remark was that if the partner and the family were happy then this flowed on to the doctor. Similarly, if the doctor was happy, the partner and family were more likely to be content. The importance of the family in recruitment and retention issues of rural medicine must not be underestimated.

Implications for policy, delivery or practice

Recruitment and retention of doctors for rural and remote WA is an ongoing issue of paramount importance. State workforce agencies must continue to be supported to provide this valuable service to rural doctors and their families. Other GP organisations must work with the RMFN and local rural

Divisions of General Practice to support these programs.

2002 General Practice and Primary Health Care Research Conference — Proceedings

113

From efficacy to effectiveness: bridging the great divide between research findings and clinical practice in relation to alcohol and drug problems

AM Roche, National Centre for Education and Training on Addiction, Flinders University of South Australia

Objectives

To overview key alcohol and drug research findings and their potential significance for clinical practice. To identify the extent of translation of these research findings into clinical practice. To present an analysis of factors that impact on this nexus. To outline methods demonstrated to facilitate uptake of research findings of demonstrated efficacy.

Methodology

This paper presents a review of the evidencebase established over the past 10–15 years in relation to alcohol and other pyschoactive drug interventions in general practice and primary health care settings. It reviews the extent of uptake and utilisation of interventions of known efficacy and delineates the range of potential barriers.

The evidence of effective continuing medical education and other professional practice behaviour change strategies are then reviewed. These two disparate literatures are then synthesised to identify matched and mis-matched strategies.

Primary findings

A sound evidence base exists for primary care interventions in relation to alcohol and drugs problems. However, uptake of these well-established interventions has been consistently low, in spite of evidence to indicate growing need and an expanding burden of illness. While there is a less definitive evidence base for continuing medical education and clinical practice behaviour change strategies, there remains untapped potential to utilise strategies of known efficacy.

Discussion

The successful translation of research into clinical practice remains one of the greater challenges for the health care field. Over recent years, strategies and technologies available to facilitate this process have improved significantly, but are less than optimally utilised. This paper takes the case illustration of a health care problem with a substantial burden of illness, and a sound evidence base in regard to interventions of known efficacy and yet low levels of uptake or translation of research finings into clinical practice. The presentation provides a background overview of the evidence base and then critiques issues related to the translation of research into practice. In addition to examining the process of translating the clinical practice evidence base, the paper will also address the evidence and related issues that pertain to professional practice behaviour change. It is argued here that the successful translation of research findings into clinical practice requires the successful interface of two distinct, and at times disparate, disciplines — i.e. the content area in question (in this case alcohol and other drugs) and generic behaviour and systems change.

Implications for policy, delivery or practice

Clearer identification, articulation and application of effective research dissemination strategies are required. This can be achieved by adoption of the conceptual framework of practitioners and policy makers, rather than researchers, and through integration of research findings with day-to-day practical considerations and overarching structural factors.

How to improve access to diabetes care for people of low socioeconomic status: GPs’ views

V Rose, Department of Public Health and

Community Medicine, University of NSW and

Centre for Health Equity Training Research and Evaluation, M Harris, T Ho

Objectives

The aim of this study was to explore GP and patient attitudes to health care, including, attitudes towards the health system and each other, within the context of diabetes care.

Material from this study will be used to inform the development of further research on the health care attitudes of people of low socio-economic status and health care utilisation practices, and develop ways to improve service use by low socio-economic groups.

Methods

Focus groups were held with GPs from the

Fairfield Division of General Practice and people with diabetes living in a disadvantaged area in Sydney. The focus groups were audio taped and a content analysis was performed on the transcribed data to identify major themes. Results from the GP focus group are presented.

114

2002 General Practice and Primary Health Care Research Conference — Proceedings

Principal findings

GPs who participated in the focus group were culturally diverse and practicing in areas of low socio-economic status. Major themes to arise from the discussion included:

X how GPs identify the socio-economic status of their patients

X the main causes of diabetes for people of low socio-economic status

X the relationship between patterns of health service use and health status for patients of low socio-economic status

X barriers to prevention, early detection and effective management of diabetes among low socio-economic groups

X improvements and priorities for diabetes prevention, early detection and effective management for low socio-economic groups

X the relative importance and interaction of cultural factors with low socio-economic status in diabetes

Discussion

Fairfield is a culturally diverse area in Sydney that has high numbers of people of low socioeconomic status. The Fairfield Division of

General Practice had an active diabetes program that needed to better access low socio-economic groups. The issues raised in the focus group discussion suggest that patient socioeconomic status is an issue of relevance to GPs in how they provide primary care to people with diabetes and promote access for all community members to their services. As primary providers of care to people with diabetes, GPs have a role in informing research, policy and the development of service strategies to improve the utilisation of health care services, including diabetes services, and the quality of care provided to patients of low socioeconomic status.

Implications for policy, delivery or practice

This study informs the development of further research and methods to improve service use by low socio-economic groups.

The residential care, general practice and hospital interface — a service plan to improve afterhours primary medical care

D Ruth, Department of Community Health,

The Royal Melbourne Hospital, Melbourne

Health

Objective

This paper describes a partnership approach to improve the service system of after-hours primary medical care (AHPMC) for residents of residential care facilities (RCFs).

Aims were to:

X

Assess needs, barriers and potential solutions for AHPMC to RCF residents.

X

Develop a service plan that will:

– meet identified needs of residents of

RCFs in the northwest Melbourne area

– minimise demand for usage of hospital emergency departments

– be part of the continuum of care provided by GPs

– be part of general AHPMC services.

Methods

A Service Plan to improve AHPMC to RCF residents was developed, based on:

X

Consultation strategy with the key stakeholders as collaborative partners.

X

Qualitative needs assessment with residents and their relatives; GPs; RCF staff; medical locum services; ambulance service; and hospital staff providing emergency, acute and aged care services.

X

Quantitative data collection and analysis of service utilisation.

X

Service mapping to assess current

AHPMC to RCFs in the region.

X

Literature search of issues and interventions relevant to improving afterhours medical care in RCFs.

Principal findings

The paper describes the decision pathway when a resident in a RCF needs medical care after hours, and factors influencing care that impinge before a person needs AHPMC, at the time of the acute episode and afterwards.

AHPMC service utilisation data will be reported for a population of 2598 residents in

66 RCFs in the northwest Melbourne area.

2002 General Practice and Primary Health Care Research Conference — Proceedings

115

Stakeholders selected interventions perceived to have a high chance of success in a short period of time and capitalise on partnerships developed. The Service Plan outlines the stages, activities, time-lines and risk management for implementing 7 key interventions to improve the appropriateness and timeliness of AHPMC service delivery:

X identification of residential care clients throughout the service system

X key performance indicators and standards

X client/carer evaluation form

X patient information sheets

X acute care plans

X clinical pathways for symptom and disease management

X telephone triage.

Discussion

RCF residents form a large and growing population group with complex medical needs related to dementia, chronic illness and physical disability. Neither the residential care reforms nor the general practice reforms over the past decade address these needs.

Our findings confirm that improving access generally to quality care in RCFs requires system change at national level. However, significant improvements to AHPMC for RCF residents may be achieved through regional partnerships of service providers to address interface issues through improved information management and care coordination. This would facilitate continuity of care, reduce hospital presentations and bed days, and avoid the morbidity associated with movement of frail elderly patients from their familiar environment.

Implications

Our locally developed Service Plan may be generally applicable elsewhere in Australia by

AHPMC service providers wanting to address specific needs of people in RCFs.

Acknowledgments

This paper is based on work funded by the

Department of Health and Ageing, and conducted collaboratively by Melbourne

Health with North West Melbourne Division of

General Practice, Carers Association Victoria, the North West Melbourne Aged Residential

Care Network, and Campbell Research and

Consulting.

Problem solving issues that impact upon Indigenous students in a NT remote primary school

J Sherwood, Centre for Remote Health,

Alice Springs, Pilawuk

Objectives

To immerse students in curriculum that is meaningful and culturally relevant.

To support teachers in a collaborative partnership to enhance the students sense of self worth and self determination.

To provide students with the opportunity and the skill development to problem solve issues that they have raised, which impact upon themselves, their class, their family and their communities.

For students to utilise this model of participatory action research to promote positive health outcomes for themselves and their community.

Methods

Utilisation of participatory action research model.

Researcher and teachers, discussed model, developed it further to meet the needs of the class, as well as incorporating the model in the curriculum.

Students have directed the process, and have become immersed in the language of research also.

Principal findings

Students are actively enjoying this process, and it has become a great focus to develop literacy, numeracy and community participation.

Issues have been worked upon that have both health and social justice perspectives.

Student confidence has increased and they are relating the process back to their families and community.

Discussion

The project was undertaken to provide evidence of what can be achieved if we start at the grass roots, in a collaborative manner, to improve health outcomes for Indigenous communities. The model has developed in meeting the needs of the students, staff and community, and hence all have ownership of the process. This process of community development and self-determination is

116

2002 General Practice and Primary Health Care Research Conference — Proceedings

imperative for any further partnerships in research.

Implications for policy, delivery or practice

Indigenous health is more then simply the absence of disease, we know this, so our practice and research must reflect this.

The way the research is done is what fosters partnerships, and ensures sustainable and relevant outcomes.

Keeping our GPs healthy: initiatives to address Australian GPs’ wellbeing

J Sims, University of Melbourne, Department of General Practice, S Davidson, C James,

D Young, P Schattner, B Hartwig

Objectives

A number of factors have been identified that adversely impact upon the health and wellbeing of general practitioners (GPs). The aim of the study was to describe and analyse current responses to GP well-being in the

Australian primary health care sector.

Methods

To determine where initiatives to address GP well-being existed, 100 general practice related organisations at national,

State/Territory and regional level were purposively sampled and surveyed using a standard pro forma. To obtain further details,

95 telephone interviews were conducted with informants. Respondents discussed systems the agency had in place to address the factors that adversely effect GP well-being.

They provided feedback on the barriers and enablers to their endeavours to assist GP well-being. The resolution of factors that impact upon GP well-being was analysed in terms of personal, professional and organisational strategies.

Principal findings

Feedback was obtained from a range of stakeholders, including the Department of

Health and Ageing; peak bodies representing

GPs and the medical profession; state-based organisations e.g. rural workforce agencies and registration boards; divisions and academic departments of general practice.

Current responses to factors adversely impacting upon GP well-being were varied.

Initiatives provided assistance to GPs via education, mentoring, counselling, organisational and practice support (e.g.

locums), media campaigns and policy development activities. There is established national support for psychosocial and

2002 General Practice and Primary Health Care Research Conference — Proceedings psychiatric problems e.g. Doctors Health

Advisory Service and regional programs to enable physical health promotion e.g. DrDoc.

Many programs originated from the need to address workforce challenges e.g. the Rural

Medical Family Network. However, there was limited strategic planning to explicitly address GP well-being over time. The findings highlighted that the range of activity that addressed GP well-being, whilst varied, had limited reach.

Discussion

To maximise the impact of current initiatives, several strategies are required. To assist in overcoming general practices’ cultural ambivalence towards personal health support programs, well-being needs to have a more overt profile across stakeholder organisations. The availability of relevant programs should be extended and fully evaluated to determine their applicability to different types of GP.

Implications for policy, delivery or practice

The primary and secondary data gathered in this study provide the basis for the construction of a range of research questions to be addressed that have relevance to consumers (the GPs) and policy makers alike. Many contemporary policy and organisational initiatives can be expected to positively influence well-being. Their impact will require monitoring and evaluation in this regard. In particular, the influence of GP well-being upon the quality of health care delivery remains to be determined.

Educating consumers and GPs about evidence-based primary health care: recruiting for the WISDOM study in

Australia

N Stocks, Department of General Practice,

The University of Adelaide, A MacLennan,

J Marley, P Ryan, B Paine,

P Komersaroff, D Pond

Background

One objective for the PHCRED strategy is to increase the number of practitioners who know about and participate in research.

Another is encouraging the use of critical analysis skills and the application of evidence-based care in clinical practice.

Consumer participation in this latter process is essential and it is likely that success in applying research evidence will be enhanced if consumers understand the reasons why it is appropriate to undertake research in the primary care setting.

117

Method

The Women’s International Study of Long

Duration Oestrogen after Menopause

(WISDOM) hopes to recruit 34,000 women to a 15 year trial of hormone replacement therapy in the UK, Ireland, New Zealand and

Australia. This randomised placebo controlled trial seeks to assess the efficacy of taking estrogen for 10 years (with an additional five years of follow-up) with respect to cardiovascular disease, stroke, fractures due to osteoporosis and breast cancer.

Recruitment of general practices and women in Australia commenced in July 2001.

A particular feature of the recruitment strategy — for both women and GPs — has been the emphasis on providing an accurate picture of the current levels of evidence from available observational trials and the need for a large ‘gold standard’ randomised controlled trial. The recruitment strategy informs both

GPs and potential participants about the current levels of evidence for the long term use of oestrogen after menopause. The advantages and disadvantages of different types of trials, the strengths of different levels of evidence and how the results of studies might affect clinician behaviour are all discussed.

Results

To date 30 practices and 220 GPs have been recruited in South Australia. Over 18,000 letters of invitation have been sent to women aged 50–69 and 28 information evenings held for interested participants.

Approximately 500 women have been screened prior to trial entry and we hope to achieve our recruitment target of 1,250 women by the end of the year. New trial centres in Newcastle and Melbourne have recently commenced recruiting using the same methods. A recruitment rate of 3–4% of all eligible women aged 50–69 is being attained.

Discussion

The recruitment strategy and its potential impact on doctors and patients attitudes towards research/evidence-based care in the primary care setting will be discussed. It appears that the study is being well received by general practitioners with high recruitment rates. The novel recruitment strategy of inviting women to educational sessions about the research and empowering them to become part of a research team and screen themselves for the inclusion and exclusion criteria — despite there being few incentives to join — is also working. Women’s altruism is being successfully encouraged by introducing them to the concept of evidence-

118

2002 General Practice and Primary Health Care Research Conference — Proceedings based medicine and the need for participation whatever therapy they randomly are allocated.

Implications for practice

Increasing the body of evidence for general practice and primary health care is very important. To achieve this well conducted randomised controlled trials will need to be undertaken in the primary care setting.

Educating GPs and consumers about evidence-based care is also important so combining the two makes sense, is practical and can be done.

Asthma education intervention trial in GP setting: baseline information

N Sulaiman, Department of General

Practice, University of Melbourne, T Liaw,

C Harris, C Barton, T Maksimovic,

J Byron, C Hogan

Objectives

To implement and evaluate the effectiveness of locally adapted paediatric asthma Best

Practice Guidelines (BPG), administered alone or with an educational package, on the management of asthma by GPs in North West

Metropolitan Melbourne.

Methods

The project is collaboration between the

Department of General Practice, the Royal

Children’s Hospital (RCH), Northern and

Northwest Melbourne Divisions of General

Practice and Royal Australian College of

General Practitioners (RACGP)

The National Asthma Council and RCH best practice guidelines were adapted for this study. Existing instruments were examined to inform the measurement of GPs and patient’s asthma knowledge, attitudes and practices, at baseline and after 3 and

6 month. The education package, guidelines and questionnaire are available for review on: www.asthma.unimelb.edu.au.

The project was promoted through the

Divisions and RACGP weekly Friday Fax and email circular. Each GP on the division’s membership list was sent a personalised invitation with a fax-back reply sheet, followed by a telephone reminder one week later. GPs who accepted the invitation were sent the information sheet, consent form and baseline questionnaire.

Participating practices were randomised into three groups: (1) asthma education plus guideline group, (2) asthma guideline only

group and (3) ENT guidelines and education group.

Participating practices were assisted to search their practice database and identify patients, aged 2–14 years, with a diagnosis of asthma or who were prescribed asthma medications.

Randomly selected patients were sent an invitation, a plain language statement, a consent form and a questionnaire for children and their parents. The questionnaire included items on quality of life, asthma action plan and knowledge, attendance at ED, admissions to hospital, unscheduled visits to

GP and other patient outcome measures.

Preliminary findings

Thirty-two practices were recruited: 9 were solo practices, 13 with two GPs and 9 with 3 or more GPs (Mode = 3 GPs). Sixty-three

GPs completed the baseline questionnaire, identified patient details from their computer system. Invitations were sent to 980 guardians of children, aged 2–14 years, with

GP-diagnosed asthma or prescribed asthma medications. The computerised packages used were MEDICAL DIRECTOR (22/29),

GENIE (2/29), MIMS (1/29) and other (4/29) software. Participating GPs were predominantly male (60%) and trained in

Australia (62%). They work between 2 and

11 sessions a week (mode=10 and median=9 sessions) and have practiced in

Australia for 15 years (m 8) and overseas for

3 years (m 4). Fifteen of 21 GPs in the asthma intervention group (71%) attended the first two-day education workshop.

Discussion and implications

It is feasible to assist GPs to search their computer systems to identify patients with asthma or prescribed asthma medications.

GPs are willing to participate in research relevant to their day-to-day practice, in accordance with privacy and confidentiality laws. They preferred to use their practice letterheads to invite their patients.

Mentor mothers: a randomised community intervention trial to reduce depression and partner abuse, strengthen health and wellbeing and mother child attachment

A Taft, Centre for the Study of Mothers’ and

Children’s Health, La Trobe University,

K Hegarty, R Small, J Lumley, L Watson in collaboration with Women’s Health West

Objectives

The study’s primary aims are:

X to reduce partner abuse among women pregnant or with children under 5 whom

GPs identify as distressed

X to reduce depression in women pregnant or with children under 5 whom GPs identify as distressed

X to strengthen the health and well-being and attachment of at-risk women to their children

Its secondary aims are:

X to determine the effectiveness of mentor mothers’ support to women at risk through a randomised trial design, including process, impact and health outcome evaluation

X to enhance GP case management of family members living with partner abuse

X to strengthen infrastructure support for

GP management of partner abuse, by enhancing effective intersectoral collaboration between general practice and community-based family violence networks

Methods

The proposed study is a randomised controlled community intervention trial. The study is in four phases and is currently being piloted:

X

The GP continuing professional

development (CPD) phase. This phase follows a Memorandum of Agreement signed with 40 recruited practices from four Victorian urban divisions. All GPs in the practices will be offered a 30pt CPD up-skilling session to detect symptoms of partner abuse and depression in potential subjects and strategies for intersectoral collaboration. Following this the practices will be publicly randomised.

2002 General Practice and Primary Health Care Research Conference — Proceedings

119

X

Recruitment and training of the

mentor mothers. The mentor mothers’ coordinator will recruit trusted community mothers, who will be offered training about partner abuse, depression and parenting skills and associated services and introduced to GP intervention practices for which they will be a resource.

X

Data collection. Following identification by GPs over a 40-week period, 700 female participants will be surveyed by researchers using the Edinburgh

Postnatal Depression Scale, SF36,

Composite Abuse Scale and Parenting

Stress Index. Following the 12 months mentor mother intervention period, all women (intervention and control practices) will be surveyed again.

Qualitative process evaluation will be conducted through interviews and focus groups with all stakeholders.

X

Analysis. Analysis will be by intention to treat. The main analysis will compare the primary outcomes post-intervention

(depression, abuse, general health and parenting attachment/stress).

The development phase is

X scoping/adapting the goodness-of-fit of the Mentor Mother model within the existing service system

X piloting levels of resources and support necessary for the sampling strategy (GP recruitment of female subjects)

X piloting the survey instrument in English and Vietnamese (a sub-study)

Principal findings

Findings from the pilot phase will be presented.

Discussion and implications for policy and practice

The proposed study will

X build on a pilot study with support from key national and state government and community stakeholders, maximising the sustainability of the intervention, if effective

X provide model protocols and processes for intersectoral collaboration between

GP practices and community-based family violence services

X provide the world-first empirical trial of the effectiveness of an intervention to

120

2002 General Practice and Primary Health Care Research Conference — Proceedings reduce partner abuse and depression in a

GP population

The rationale, proposed study methods, pilot findings and potential contribution will be discussed.

The general practitioner and the control of sexually transmissible infections

MJ Temple-Smith, School of Health

Sciences, Deakin University

Objectives

Sexually transmissible infections (STIs), one of the major preventable health problems in

Australia, are often asymptomatic and, if undetected, can cause sub-fertility, infertility and chronic morbidity. Although General

Practitioners (GPs) are well-placed to have a significant impact on STI transmission by diagnosing and treating both asymptomatic and symptomatic disease, there are a number of barriers to achieving this. This project aimed to identify strategies to enable

GPs to more effectively contribute to STI prevention and treatment.

Methods

The study used both qualitative and quantitative methods. Information from key

Australian and Victorian policy makers and stakeholders was gathered to determine the criteria for ideal practice in relation to STI prevention and management by GPs. Data from a range of sources was then examined to determine GPs’ actual practice in this area.

The comparison of ideal practice and actual practice highlighted a number of discrepancies, which contributed to the development of recommendations for strategies to improve prevention and management of STIs by GPs. This analytical approach was drawn from the field of action science and has not been widely used in general practice research.

Principal findings

Ideal and actual practice was compared in the following areas: sexual history-taking,

STI testing and screening, STI treatment, notification, contact-tracing and acquisition of

STI knowledge. Identification of a number of discrepancies led to the development of recommendations: for education needs (3), for specific practical aids (5), for the establishment of a single institutional base and for novel strategies for GPs with a low

STI caseload to manage their patients more effectively (2). Opinions on these recommendations were sought from over 30

GPs and as a result of their suggestions minor modifications were made.

Recommendations which were most strongly supported were those which would assist GPs in very practical ways. Overall GPs strongly believed that all GPs should share a minimum level of competence in managing sexual health consultations, and in preventing and managing STIs. They highlighted an urgent need for STI advice and epidemiological information in a format which is relevant to general practice.

Discussion

Over the last 100 years GPs have always been viewed as treating a significant proportion, if not the majority of STIs. Their management practices in this area have been consistently criticised, yet health authorities and medical educators have expended very little effort to date to ensure that Victorian

GPs are appropriately trained to manage STIs effectively. This project has identified a number of very practical ways in which GPs can be supported in this difficult and sensitive area of general practice.

Implications for policy, delivery or practice

STI management and prevention policy, as it relates to GPs, needs to be cognisant of the constraints existing in the real world of general practice.

The work of registered nurses in the context of rural and urban general medical practice settings in

Australia: toward a model of effective practice

H Tolhurst, University of Newcastle,

I Madjar, L Schulz, A Schmidt

Objectives

The purpose of this presentation is to present a description of a research project being undertaken by a team of nurses and general practitioners, and the preliminary findings arising from the project.

The aims of the research project are to:

X describe current work of registered nurses employed in general medical practices in selected locations in NSW

X clarify focal and contextual issues that arise from their employment and impact on their current practice, or have relevance for development of new models of practice

X develop and pilot test a survey instrument for a national study to be undertaken in the future.

Methods

Individual and group, semi-structured, indepth interviews are being conducted with

GPs and practice nurses from 20 practices.

Maximum variation sampling is being used so that a wide variety of practices in both rural and urban locations, with differing practice structures and of different sizes are included.

Issues being explored with both nurses and

GPs include:

X

current

qualifications, roles and practices of nurses employed in general practice;

X

attitudes

toward an extended role for nurses in the area of patient care;

X

expected

effects of the extended role on patients’ access to, satisfaction with, and outcomes of care (especially in relation to health promotion and health education);

X

expected

effects of the extended role on the doctors’ workloads, and on financial management and viability of general practices;

X

the

resources, including education, needed to support nurses in extended roles.

X

possible

funding models for ongoing employment of registered nurses working in extended roles.

Tape-recorded interviews are being transcribed to form text for content and thematic analysis. The aim is to produce both a rich description of the current practices and views of registered nurses and general practitioners, and to identify relevant questions for inclusion in a national survey to be undertaken in the future.

Principal findings

Preliminary findings related to the nature and scope of nurses’ practice will be presented and the congruence between nurses’ and

GPs’ views and preferences for any future extension of nurses’ role explored.

Discussion

The discussion will focus on the implications of this research for the development of new models of nursing in the context of rural and urban general practice.

2002 General Practice and Primary Health Care Research Conference — Proceedings

121

Implications for policy, delivery or practice

In 2002, the DHA has started to provide funding for the employment of nurses in general practice engaged in specified activities or projects. The results of this study will assist in informing policy development in this important area of primary care, as well as providing baseline data for further research, including testing of different models of nurses’ work in this context.

Nurse practitioner role implementation in Victoria

C Trasancos, Department of Human

Services, Victoria

Objectives

To gain an understanding of the policy, legislative and practice frameworks underpinning the implementation of the nurse practitioner role in Victoria.

To identify general practice and primary health care nurse practitioner models of practice being developed and evaluated in

Victoria, the barriers to implementation, and the processes adopted for the efficient functioning of the role in a variety of settings.

Methods

Twenty-nine models of nurse practitioner practice have been funded to date to develop, implement and evaluate their models and to examine the role in terms of feasibility, safety, effectiveness, quality and cost. The models are focusing on the extended practices of the role including limited prescribing authority, ordering diagnostics, referral to medical specialists and admitting privileges. This work will provide valuable local data relating to the nurse practitioner role in specific contexts of practice, including general practice and primary health care, and will inform the ongoing implementation process.

A range of information is being captured from the triangulation of data from several complementary sources and methods including observation; structured and semistructured surveys; minutes of meetings; reflections on practice; comments by nurse practitioner candidates and medical officers during interludes of care; the use of case histories; in-depth interviews; focus group discussions; and the use of a common data set across all sites.

Principal findings

Evidence from the projects to date indicate that nurse practitioner candidates are feasible, safe and effective in their roles and that they provide quality health services in the range of settings researched.

Discussion

The Victorian Nurse Practitioner Project is in its second phase of role implementation with work having begun in July 1998.

Internationally, the Victorian implementation of the nurse practitioner role is unique because:

X

High skills and education levels have been set as requirements for endorsement.

X

The implementation process is not restricted by sector, geographic location or specific contexts of practice (such as acute care).

X

The government has supported an ongoing evaluation of the implementation process where a variety of Nurse

Practitioner models of practice are being developed and implemented.

Identifying research questions relevant to consumers, practitioners, managers and policy makers and utilising rigorous qualitative and quantitative methodologies that are appropriate to the purpose and setting of the research have been key aspects of the work of each of the funded models of practice. Developing effective and sustainable partnerships between researchers and stakeholders, and building the capacity of the research workforce to undertake research and the capacity of managers, practitioners, and policy makers to use research findings, have been crucial steps in establishing the viability of the models.

The nurse practitioner candidates involved in the demonstration projects have had a high level of acceptance by their clients, colleagues, and a variety of health care professionals. Their colleagues perceive them as assisting the professional roles of their coprofessionals. Client perception surveys gave overwhelmingly positive comments about the nurse practitioner candidates’ level of care, attention and expertise.

Implications for policy, delivery or practice

The potential benefit of multi-disciplinary approaches to primary health care in general practice settings suggest that the development of the nurse practitioner role may lead to improved service access, greater

122

2002 General Practice and Primary Health Care Research Conference — Proceedings

diversity in services provided and increased flexibility in mode of health care delivery.

The art and science of clinical decision making in general practice: the development of a decision aid for colorectal cancer screening decisions

L Trevena, Department of General Practice,

The University of Sydney, A Barratt,

L Irwig

Objectives

To describe a study in progress that uses a

‘decision aid’ to combine evidence and patient preferences into clinical decisions on screening for colorectal cancer.

Methods

Australian cancer registry data was used to calculate 10-year cumulative risks of bowel cancer for people of average risk by age, sex and family history. Risk reduction from FOBT screening and information about FOBT accuracy, such as the chance of having a false alarm, or a missed polyp or cancer was also compiled. This was combined into a simple workbook using ‘1,000 faces’ diagrams and explanatory text/illustrations along with a ‘values clarification’ exercise that assists people to weigh up the harms and benefits of each piece of evidence for themselves. The decision aid incorporates the major components of the Theory of Planned

Behaviour which is strongly predictive of screening behaviour.

Evaluation is by randomised trial with patients randomised to receive either the decision aid or the NHMRC consumer guidelines for colorectal cancer screening. It is hypothesised that those in the decision aid arm of the trial will have a higher rate of informed participation in FOBT screening than the standard guidelines group.

Principal findings

Preliminary evaluation with patients from a range of socioeconomic backgrounds showed that patients’ perceived risk of cancer was higher than their actual risk and that their perceptions of the benefits of screening we higher than is actually the case. There was also a poor understanding of the potential benefit of polyp removal, a relatively high tolerance of false positive results provided they were warned of this possibility, and a lack of knowledge about the possibility of a false negative result. Similarly, discussions with general practitioners considering participation in the trial revealed a desire for

2002 General Practice and Primary Health Care Research Conference — Proceedings readily accessible information about test accuracy and patient risk status like that contained within the decision aid.

Discussion

Recent studies have shown that general practitioners identify two particular barriers to the application of evidence — a lack of time and an inability to tailor populationbased research to their individual patients.

Legal, ethical and social imperatives also drive a need for informed participation in health care decisions. This is not always straightforward, often involving the weighing up of harms and benefits by practitioner and patient to reach (ideally) a shared informed decision. Its operationalisation requires new tools and methods of evaluation that combine empirical methods and behavioural theory reflective of the general practice context. This paper, therefore will describe the development of such a tool, its evaluation and potential application in general practice.

Implications for policy, delivery or practice

This study breaks new ground in applying evidence within a behavioural framework. Its successful evaluation potentially provides

Australian general practice with a new tool to facilitate the uptake of evidence within an ethical framework of shared and informed clinical decision making.

Why rural GPs stay

R Hays, S Wynd, C Veitch, School of

Medicine, James Cook University,

L Crossland

Objectives

The primary focus of most recent initiatives to address the rural medical workforce problems have focused on increasing the recruitment of appropriately prepared and motivated medical graduates. While these initiatives are sound, relatively little has been done to increase retention rates or to understand why doctors stay in rural practice for longer periods of time.

Methods

We re-interviewed a group of 15 rural doctors after a period of almost ten years to explore why they were still there.

Principal findings

Thirteen of the original group were still in rural practice, eleven of them in the same communities. Interviewees had strong local community connections, supportive local and distant personal and professional networks,

123

and locally shared after-hours arrangements that eased pressure on themselves and their families. Although most seemed to have coped, several participants still experienced similar negative feelings to rural doctors interviewed in a previous study. Two seriously considering leaving and one appeared to be close to ‘burnout’.

Discussion

Doctors who successfully remain in rural practice for 10 or more years appear to have developed ways of coping with the pressures of living and working in their small communities. However, more could be done to facilitate after hours and vacation arrangements as legitimate retention strategies.

Clinical psychology in rural general practice: progress report on a national pilot research program

R Vines, Charles Sturt University,

D Thomson

The aim of this national pilot research project

(funded by the Commonwealth Department of Health and Ageing for $347,000 over two years: 2001 and 2002) is to test the efficacy of introducing clinical psychology services into the rural primary care setting. Currently running in a number of regional general practices in New South Wales and Victoria, the Project is also piloting video-conferencing to general practices in more remote areas.

The research is being run in liaison with the local Divisions of General Practice, and highlights the benefits of collaborative mental health service delivery in regional, rural and remote settings.

The Project is significant not only to the local communities it is serving, but to the development of the DPych and MPsych programs at three regional universities (CSU,

Ballarat and UNE) as it provides local placements for postgraduate students in clinical psychology. The Project also provides continuing medical education ‘in situ’ for

General Practitioners involved in this new generic model of collaborative mental health service delivery.

The research team has been requested to provide funding models to the

Commonwealth Department of Health and

Ageing for allied mental health service delivery (and specifically psychological services) in the primary care setting in

Australia.

The paper will present interim results into the second year of the project.

Promoting partnerships in peer-led self-management of chronic disease

C Walker, Chronic Illness Alliance Inc,

J Belfrage, E Demetriou, J Furler,

H Swerissen, C Peterson, M Carter,

B McAvoy, A Fox, J Best

Objectives

This project involves a collaboration between consumer organisations, primary care providers, universities and professional colleges to test a model of peer-led selfmanagement. The aim of peer-led selfmanagement is to develop the consumer’s capacity to self-manage their condition.

This project replicates similar successful collaborations of ‘lay-led’ self-management in the US and the UK. However, it aims to implement culturally appropriate and generic self-management programs in culturally and linguistically diverse (CALD) communities of lower socio-economic status, in Melbourne’s north central suburbs. It also aims to thoroughly evaluate the benefits of peer-led self-management in comparison to usual care for consumers and for the health system in general.

Methods

The research will be undertaken as a randomised controlled trial to compare peerled self-management for people with chronic illnesses with usual care.

The design employs an experimental pretest/post-test control group design with 800 subjects randomly allocated on a 1:1 ration to one of the two groups.

Discussion

Peer-led self-management has been found to improve consumers’ personal health outcomes, result in lower emergency department admissions and improve personal levels of pain. However, its uptake has largely been by consumers who are relatively well educated where the improvements may be the result of a number of factors, so that the evidence for the contribution of selfmanagement programs is not comprehensive.

In addition, the relationship between peerled self-management is not well developed or well evaluated. Neither have selfmanagement programs been evaluated against usual care.

124

2002 General Practice and Primary Health Care Research Conference — Proceedings

Implications for policy and practice

Funded by the NHMRC, this project brings together a range of health organisations and disciplines to work together. In itself, the project ushers in a phase of new partnerships with the aim of producing better outcomes for people with chronic illness. The results of the project will provide a basis for implementing and evaluating similar projects in the future, should they be demonstrated to be successful in terms of benefits to consumers, themselves and to the health system as a whole.

Developmental evaluation of partnerships

R Walker, School of Public Health, La Trobe

University

The formation of partnerships between agencies is a complex, and often subtle, process that occurs over a period of time. In

Victoria, some Primary Care Partnerships are commissioning process evaluations that review progress on partnership development and identify obstacles. In this presentation some key issues in this kind of evaluation are explored.

Objectives

The purpose of this paper is to explore some of the key issues in the developmental evaluation of partnerships identified by the author during the evaluation of partnerships for primary health care in Victoria.

Methods

A participatory model of evaluation was used and the specific questions and data collection tools were negotiated with partnerships.

Principal findings

The issues in partnership formation that lead to success and to difficulties form, on the one hand a pattern that recurs in different places and at different times, and on the other hand patterns that vary immensely depending on local history, experience and resources. Prior knowledge of likely issues, found in the partnership research and evaluation literature, is of great value when designing particular evaluation projects. Similarly, processes to identify issues specific to the partnership need to be used.

Discussion

There are now a growing number of evaluation tools for use in the evaluation of partnership development. Some are brief and pragmatic, others are lengthy and exhaustive. In practice, partnerships tend to have specific concerns that go beyond the limits of specific tools. Specific selection and/or adaptation of tools is important for meeting the needs of particular partnerships, and in some cases development of specific tools is required.

Implications

There is a library of partnership evaluation resources available, many of them through the web. It is usually preferable to use established tools for these kinds of evaluation because they typically are designed to identify the common patterns that lead to success or difficulty. However, variation between partnerships means that alone they usually miss information that is important locally.

Translating research into practice in after hours primary medical care service provision

L Wood, Queensland Divisions of General

Practice

Objectives

X

To improve after hours primary medical care service provision in Queensland by supporting the integration and partnerships between existing services and identifying and creating opportunities for future collaborative arrangements.

X

To increase the capacity of managers and practitioners involved in after hours care to incorporate best practice findings into services.

Methods

X

Undertake a situational analysis of the after hours environment in Queensland regions identified as having particular needs.

X

Commence the dialogue regarding the issues, expectations, barriers and opportunities in after hours with the region, including the Division, GPs,

District Health Service Manager,

Emergency Department Director, private service providers, Community Health

Services and local community groups.

X

Build the relationship between service providers and the local Division.

X

Develop locally owned solutions through stakeholder workshops, introducing best practice components and experiences

2002 General Practice and Primary Health Care Research Conference — Proceedings

125

from models trailed elsewhere in

Australia.

X

Facilitate and support the process of implementation of any new service.

Principal findings

X

Integration remains key opportunity and key barrier to enhancement of after hours primary medical care service delivery.

X

Communication key issue with key success factor identified as the relationship between service providers.

X

Sustainable solutions require buy-in from all stakeholders.

X

Improving service delivery requires significant change management support and capacity building at the local level.

X

Ongoing facilitation and support is required to ensure best practice and quality practices are incorporated.

Discussion

The project has demonstrated the importance of facilitation and support in the development of a partnership approach to integration in the primary health care system. The role of the Division of General

Practice in the primary health care environment can be seen as the appropriate body to undertake this facilitation role.

However, many Divisions do not currently have the capacity or expertise to take on this role and require support to build their capacity in facilitating, networking and discussion at the local level.

The project findings identified that in order to progress improvements in the primary health care sector, in particular in the after hours primary medical care environment, a change of focus is necessary for GPs from their practice to the community, requiring significant change management support.

Implications for policy, delivery or practice

With the continuing changing environment in the health system and increasing need for a primary health system to support an ageing population and struggling tertiary sector, the need for facilitation, support and increasing capacity is of utmost importance. The challenge for government and for local

Divisions who are concerned about the ensuring a quality primary health care sector, is how to facilitate change management, incorporating both the primary and tertiary sectors exploring comprehensive systems that best supports their community translating research into practice.

An innovative approach in managing hypertension in general practice

T Ziaian, Department of General Practice,

The University of Adelaide, J Beilby,

D Turnbull, J Marley, F Qureshi,

M Nelson

Objectives

To assess the effectiveness of an innovative disease management program, used by a general practitioner to improve the management of hypertension.

Methods

This cluster Randomised controlled trial included 72 general practices in rural and urban South Australia and Victoria with 5–8 patients per practice. 373 patients who were aged between 18–75 years with mild to moderate hypertension were selected. A comprehensive disease management program (ADAGE) which was purposely designed for the management and control of a complex chronic disease, hypertension was used. This program, was a communication/information technology package with written and audiovisual patient education materials, utilising nonpharmacological interventions. The program with the assistance of a variety of health care professionals, aimed to maximise benefits from medication. The study was comparing cardiovascular risk in patients treated according to evidence-based, best practice via a purpose designed disease management program with that of current practice.

Principal findings

Six-month data will be presented. The pattern of blood pressure change in each arm of the study will be summarised. The effect will be assessed in terms of its effectiveness in reducing cardiovascular risk and controlling blood pressure.

Information on the use of anti-hypertensive medication and role of allied health staff particularly the dietician.

Discussion

This paper explores patterns of interventions that improve quality of health care and management of hypertension and proposes a unique Disease Management Model involving information-communication technology and dietician services.

126

2002 General Practice and Primary Health Care Research Conference — Proceedings

Implications for policy, delivery or practice

The quality of information and education for patients with hypertension is of high importance in improving the management of hypertension when it is guided and coached by a doctor and intervened by a dietician.

Including dietician’s intervention in the disease management of hypertension, can optimise the use of pharmaceuticals and streamline the management of patients’ hypertension.

2002 General Practice and Primary Health Care Research Conference — Proceedings

127

Concurrent poster abstracts

Development: the missing link in

PHCRED

E Albert, Department of Rural Health,

University of Tasmania, C Cook

Objectives

The objectives of this discussion paper are to:

X provide a cyclical model of research, evaluation and development that can underpin capacity building, research and evaluation activities, and provide a link to research priorities and policy making;

X describe the range of practical uses of the ‘RED Cycle’ model within the context of PHCRED capacity building activities;

X demonstrate a rigorous approach to development and capacity building based on learning organisation theory, and;

X give examples based on learning organisation theory of how a PHCRED capacity building program could overcome system challenges.

Discussion

Development is a crucial component of the

PHCRED program. It is the cluster of activities that actually make things happen on the ground. To develop is to ‘bring to a more advanced or more highly organised state’. It is the realisation of the potential of research and evaluation and the subsequent growth of health care provision. It is a very practical and pragmatic activity. Without

‘development’, all research and evaluation activities can only exist in isolation, and therefore lack substantial, practical meaning.

Effective development provides the raison d’etre for applied research and evaluation.

Yet, whilst the art and practice of research and evaluation receive much attention and consideration, development is largely ignored. Whilst research and evaluation have well recognised and finely honed theories and methodologies, development tends to progress in a more ad hoc fashion. Theories, methods and tools for development are under-utilised in comparison and the acquisition of skills left to chance or the fallacy of experience.

The RED Cycle model of research, evaluation and development is a Systems Thinking based model that integrates research, evaluation and development with priorities and policy making. It demonstrates the cyclical nature of these activities and clearly defines ‘development’ within this context.

Furthermore, the model allows one to define the content of capacity building. The active flowing nature of the model lends itself to the easy definition of the skills that are required.

It may be used as a template for program logic models and therefore act as a project management or evaluation tool.

Having defined and contextualised

‘development’, it is then necessary to formulate a rigorous framework for such activities. Learning organisation theory and practice is ideally placed to be one such framework for PHCRED. A learning organisation-based approach to the PHCRED program is described with examples of its potential benefits and effectiveness.

Implications for policy, delivery or practice

A rigorous approach to the ‘development’ side of PHCRED using the RED Cycle model and learning organisation theory will pay dividends at all levels. It will improve the effectiveness of the PHCRED program, ensuring innovative approaches to overcoming system challenges, a capacity building program that is appropriate to stakeholders needs and the development of viable and sustainable partnerships with other primary health care organisations. A systems-based approach to PHCRED will increase the likelihood of closing the gap between research and practice.

ROBIn (Reduction Of Breast

Infection) Trial: methodology

L Amir, Centre for the Study of Mothers’ and

Children’s Health, La Trobe University,

J Lumley, S Garland

Objectives

The aim is to conduct a trial to reduce mastitis in lactating women.

Methods

In this double-blind placebo controlled trial of flucloxacillin, two groups of 130 eligible women, being positive for S. aureus from their cracked nipples, will be recruited from

128

2002 General Practice and Primary Health Care Research Conference — Proceedings

postpartum wards, breastfeeding clinics and through domicilary midwives. The ROBIn study sample size should detect a reduction in mastitis from 30% to 15%. Both the researcher and the participants will be blind to allocation of participants in treatment or control groups. During analysis the groups will also be blinded.

Progress to date

A multicentred trial is required in order to recruit an adequate recruitment. Ethics applications have been submitted to the university and six hospitals. Multiple funding applications have been made to gain adequate resources, principally for the microbiological assessment. A placebo capsule identical to the flucloxacillin was obtained, and randomisation conducted.

Databases have been set up to collect data on potential participants (including reasons women are ineligible or reasons they refuse to participate) and tracking of participants

(including results of tests and dates when follow-up interviews are due).

Discussion

Twenty per cent of lactating women are treated for mastitis in the first six months postpartum. One study has found that women with a cracked nipple colonised with

S. aureus have a 30% risk of mastitis in the following week (7/23), compared to 5%

(1/19) of women who received oral antibiotics (Livingstone and Stringer 1999).

Although the authors of that study have advocated aggressive treatment of cracked nipples with systemic antibiotics, it is necessary for an adequately powered trial to be conducted in order to ascertain the possible benefit of antibiotics in this situation.

Assessment of nipple damage will be by review of digital photographs by lactation consultants, in addition to clinical assessment, thus allowing a more independent measure of nipple damage and change over time, than relying on clinical assessment alone.

Implications

Important clinical questions can be answered by well-designed RCTs. Although there are many logistical barriers to conducting a clinical trial, it is possible for a single researcher to conduct a multi-centre trial.

Defining the role of the care promoter in diabetes for the 21st century

J Artridge, School of Rural Health,

University of Melbourne, D Simmons,

M Summers

Objective

Complications of diabetes can be significantly reduced by tight metabolic control. Metabolic control can be improved within primary care, by stimulating access to care and improving disease self-management skills. We investigated the qualities and functions of ‘a care promoter’ for diabetes

Method

Using face-to-face semi structured interviews, fifty-two people were interviewed.

Of these, 29 had diabetes (21 with Type 2, 7 with Type 1 and 1 with Gestational

Diabetes), 4 were lay carers, and the remaining 19 were either directly or indirectly involved in the clinical care and/or education of people with diabetes. The aim of the survey was, through discussion of barriers and enablers to self-care, to develop a description of the role of the diabetes care promoter.

Principal findings

The average age of the participants with diabetes was 56 (range 28–88). The average age at diagnosis was 42 (16–70 years), and on average had lived with diabetes for 15 (1–

62) years.

The ‘care promoter’ is a philosophy, not a new position in the diabetes care team. It involves working closely with the person with diabetes, guidelines for best practice and local clinicians. The role has case management characteristics, but all participants with diabetes believed expertise in the management of diabetes was more important. Although most clinicians and all participants with diabetes believed that the

Diabetes Educators are the best people to promote diabetes care, Diabetes Educators felt this proposal was idealistic. They felt uncomfortable about adding this to their current work routine which they describe as overwhelming, and pointed out that ‘no-one would pay you for that sort of work even though that’s exactly what we think we should be doing and would love to be doing.’

Discussion

Currently the promotion of diabetes care within primary care is lacking consistency, but a concept that many people with diabetes thought would be very helpful. Further

2002 General Practice and Primary Health Care Research Conference — Proceedings

129

interest in this holistic strategy is required from within primary care providers, and it may help many people with diabetes to become independent and proficient in self managing their condition and avoiding complications of diabetes.

Implications for policy, delivery or practice

The findings are suitable for immediate translation into primary care practice and policy. Where care consumers are able to contribute to a strategy that stimulates access to care and improves disease selfmanagement skills, the strategy has great potential to be successful. The qualities and functions of a care promoter are eminently incorporated in current primary care clinicians. Adding structure to the integration of care may increase work satisfaction, and level out workloads. Recent Commonwealth incentives are supportive of such structure and processes, perhaps providing a means of reimbursing protected time for diabetes care promotion.

Research and GPs’ training: a comparison with physicians, surgeons and public health physicians

D Askew, Centre for General Practice

University of Queensland, P Glasziou,

C Del Mar

Objectives

To assess how research is incorporated into the training of GP registrars, and to compare this with the training of physicians, surgeons and public health physicians.

Methods

The training programs of GPs, physicians, surgeons and public health physicians were reviewed. Data extracted included the incorporation of research in the curriculum objectives, the existence of research training, financial support for undertaking research, allocation of time during training for research, examination of research activity, and whether research is compulsory in training.

Principal findings

Research knowledge and skills were included in the stated curriculum objectives for the four disciplines included in this study.

Physicians can do up to six months of research during basic training, and 100% of advanced training (36 months) can be spent undertaking a PhD or MD, but research involvement is neither compulsory nor

130

examined. Research is compulsory in the surgery training program, and must result in either a higher degree in surgery, reviewed publication or conference presentation, or a relevant dissertation. Surgical trainees can receive exemption for up to one year of advanced training for completion of research.

Public health physicians receive formal research training (through the Masters of

Public Health program), and research is an expected and examinable component of their training. Research is not incorporated into the training of GPs in any manner, although limited research exposure may occur if the registrar does an academic or rural term. In general, undertaking any research activity would prolong the training period for GP registrars.

Discussion

Despite the stated curriculum objectives of the RACGP training program, minimal research training, support, or assessment occurs. Time involved in research prolongs the training period. This contrasts with the other disciplines investigated that incorporate research as an integral component of the training, particularly for surgeons and public health physicians where it is compulsory and examined.

Implications for policy, delivery or practice

Current training of GP registrars does not value research. To strengthen the research culture in general practice and primary care, research must be fostered in the training program to be perceived as important, relevant and interesting. This could have a positive influence on the attitudes of GPs to research, the knowledge-base of general practice, GPs’ self esteem, and later behaviours in relation to using and participating in research.

Can n-of-1 trials be used in general practice for optimising prescribing?

D Askew, Centre for General Practice,

University of Queensland, A Clavarino,

P Glasziou, C Del Mar

Objectives

To investigate general practitioners’ (GPs) attitudes after using n-of-1 trials to optimise prescribing for 4 different clinical conditions: osteoarthritis, menopause, insomnia, and

Attention Deficit Hyperactivity Disorder.

Methods

We undertook (i) a post-card survey to evaluate the attitudes of GPs to n-of-1 trials;

(ii) treatment related pre- and post-

2002 General Practice and Primary Health Care Research Conference — Proceedings

questionnaires completed by GPs for each patient referred for an n-of-1 trial; and (iii) semi-structured interviews conducted with a sample of referring GPs.

Principal findings

The postcard survey achieved a 81% (56/69) response rate — 50 GPs and 6 specialists. Nof-1 trials were considered clinically useful by

76% (33/42) of GPs, and worth the extra time involved by 79% (27/34). Seventyseven per cent (27/35) of GPs considered nof-1 trials worth doing again, and 41%

(15/37) considered their prescribing changed due to the n-of-1 trial. The pre and postquestionnaires found certainty about treatment increased in 17% (7/41) of cases.

Interview data indicate GPs were mainly positive about their involvement in n-of-1 trials. GPs consider n-of-1 trials useful for optimising medication management for some patients, and empowering patients to be active in medication decision making. In some instances, GPs theoretical knowledge was confirmed by the outcomes of the n-of-1 trials, e.g. the effectiveness of stimulants for treating ADHD. However, the GPs considered the n-of-1 trials time consuming to explain to patients, and impractical in routine general practice. The n-of-1 trials’ research questions were not always relevant to patients’ clinical problem, and some GPs had concerns about the risk/benefit ratio of the n-of-1 design and process.

Discussion

GPs consider n-of-1 trials useful in optimising prescribing for some patients, particularly if the clinical questions are important, the processes are simple and administered externally. N-of-1 trials can be a ‘light-bulb’ experience for GPs when their theoretical knowledge is robustly confirmed in practice.

GPs also appreciate the potential for patient empowerment through active participation in medication decision making.

Implications for policy, delivery or practice

N-of-1 trials are useful for optimising prescribing when the outcome of the medication decision has potentially serious consequence, e.g. serious adverse effects or high cost. N-of-1 trials have much potential for removing uncertainty about the efficacy of particular medications for individual patients, but they must be packaged appropriately for the general practice environment.

Stress incontinence in pregnancy — a prospective study

L Bailey, Western Australia Office, The Royal

Australian College of General Practitioners,

D Cordell, R Guibert, N Tsokos, E Seah

Objectives

Identify the prevalence of stress urinary incontinence (SUI) amongst women during late pregnancy (36 weeks), the early post partum (8–12 weeks), and the late post partum periods (6 months).

Identify risk factors associated with the onset of SUI during late pregnancy (i.e. family history and obstetric experiences including birth trauma from previous childbirth) and following childbirth.

Identify obstetric factors associated with the onset of SUI following childbirth and those factors associated with the persistence of SUI up to 6 months post delivery.

To ascertain the validity of a battery of questions used to diagnose SUI and urge urinary incontinence (UUI) in pregnant women by comparing relative prevalences of these conditions with those in the literature.

Methods

Using a self administered questionnaire at 36 weeks gestation, 8 weeks post partum and 6 months post partum, the prevalence of stress urinary incontinence (SUI) and the causes of the condition post delivery will be determined. A sample of approximately 1000 women are being recruited at their 36 week medical review, with the subsequent questionnaires mailed out to them post delivery. This will be combined with birth data from a neonatal registry. A battery of questions used to identify the presence of

SUI and urge urinary incontinence (UUI) previously used in older women is being used in this population for the first time to ascertain its validity.

Principle findings

Preliminary data analysis of questionnaires 1 suggests estimates of the prevalence of SUI and UUI using a battery of questions are comparable to those found in the larger population studies of pregnant women.

Discussion

One of the outcomes of this study is to compare the results of a battery of questions designed to accurately diagnose SUI and UUI in older women presenting for surgery (but now being used with pregnant women) with estimates of the reports of urinary

2002 General Practice and Primary Health Care Research Conference — Proceedings

131

incontinence in pregnant women being reported in the literature. The preliminary analysis suggests this battery estimates prevalence similar to the more rigorous of the studies presented in the literature.

Further work using this battery on pregnant and post partum women will clarify its potential usefulness as a diagnostic tool.

Implications for delivery and practice

The purpose of this paper is to demonstrate the use of a standardised battery of questions to determine the prevalence of SUI in a population in which it has not previously been used — the perinatal and postnatal population.

Enhancing the interface between hospitals and general practice: experiences from an electronic datasharing study

L Barclay, WA Research Unit, The Royal

Australian College of General Practitioners,

S Roach, P Bouhalis

Objectives

This ‘virtual amalgamation’ project aims to establish a secure, electronic, patient datasharing framework between a tertiary hospital and local general practices. It is anticipated that the provision of a mechanism for the instant transfer of medical data will result in a more efficient bidirectional referral process. Long-term costsavings are expected due to a reduced amount of assessment duplication, particularly for those patients who present to multiple health-care providers.

Methods

The Royal Australian College of General

Practitioners, in conjunction with Health

Industry Technology professionals, is coordinating the implementation of electronic links between Fremantle Hospital and three local general practices. Participating staff have been trained in the use of Public Key

Infrastructure (PKI) technology, the latest

Commonwealth Government sponsored electronic data encryption method.

Medical data of consented patients is transmitted in the form of a summary health information sheet. A clinical audit of all practices involved has been conducted to ensure that these summaries are adequately completed. Health summaries are encrypted and transferred via email between participating sites. Received patient data is then downloaded as a text (HTML) file or into patient files within Medical Director software as an XML file.

As an innovative study, qualitative analysis is ongoing in an effort to inform future uses of the technology. Outcomes of the study include barriers and enablers to the process, and GP, other medical staff, and patient satisfaction.

Principal findings

The project has been under way since

October 2001 and its completion is expected in November, 2002. Implementation of the project has to date identified several technical issues relevant to the use of such technology elsewhere. Software incompatibility has been identified as a potential barrier, although simple methods exist to overcome this. A further issue is that information sent to an individual GP cannot be accessed by other staff in the practice in the GP’s absence. Where necessary, this issue is also overcome by the provision of both individual and ‘site’ licences for PKI technology.

Discussion

The electronic exchange of patient information represents a culture-change that, in our experience, has support from GPs and patients. Several enablers and barriers to the successful implementation of electronic patient data links have been identified.

Simple methods to overcome the major barriers have been trialed with success.

Implications for policy, delivery or practice

The implementation of this project within a specified geographical area immediately surrounding a hospital, has progressed smoothly. Adequate staff training and infrastructure is required to support similar networks elsewhere. Secure patient datasharing between authorised medical staff allows better continuity of care, potentially substantial cost-savings, and a more efficient referral process.

Mixed method research design in primary care research

C Barton, Department of Epidemiology and

Preventive Medicine, Monash University,

N Sulaiman, M Abramson

Objectives

To report challenges and benefits of mixed method (quantitative and qualitative) research designs that we encountered while developing a protocol for our study of management of acute asthma attacks by

132

2002 General Practice and Primary Health Care Research Conference — Proceedings

parents following an asthma education intervention delivered to their child’s General

Practitioner.

Methods

A case-control study nested within a larger

Randomised Controlled Trial of the Utility and effectiveness of locally adapted best practice guidelines. Our case-control study will utilise an innovative mixed method (quantitative and qualitative) research design. Initially, a quantitative study employing reliable and valid questionnaires will test our hypotheses.

Following this, in-depth interviews will be utilised to explore issues of care important to the participant.

From participants completing the larger RCT a sample of 60 cases (children with uncontrolled asthma) and 60 controls

(children with controlled asthma) will be selected. Parents will complete the questionnaires. Cases and controls will be matched by severity of asthma, age of child with asthma and arm of the RCT. Of these,

20 cases and 20 controls will be followed up for in-depth interviews. We will use random

purposeful sampling, quota sampling and

theoretical sampling to select participants for the qualitative interviews.

Methodological triangulation will be used to investigate the research question from two perspectives: 1 - Quantitatively: What coping style do parents use to manage stress associated with managing their child’s asthma? What is the outcome of this coping style (biomedical — asthma control, psychosocial — anxiety/depression)? 2 -

Qualitatively: Does the parental relationship with the child’s GP influence coping and ultimately asthma management? How do the experiences of parents affect coping? Why are participants coping or not coping and what are the resources they use/need to be able to cope? Global analysis of quantitative and qualitative data will highlight important social issues influencing parents’ coping and illuminate ways that GPs can address these issues in consultations with parents of young asthma patients.

Discussion and implications

The best research design must be employed to answer a research question and should not be limited to either quantitative or qualitative methods. We have presented a proposal utilising both quantitative and qualitative research methods to investigate management of paediatric asthma by parents. Traditionally, qualitative research is conducted independently of quantitative studies or in the design of survey instruments, or to generate testable hypotheses. However, we have proposed a research design that reverses traditional practice. We propose to use a quantitative survey initially and then a qualitative interview subsequently on a subset of participants. The benefit of this study design is that while we will initially test a hypothesis, our study design allows exploration of the ‘why’ and ‘how’ the hypothesis was supported or not supported.

We can then make more accurate inferences about how the intervention has influenced parent’s ability to cope with managing asthma that are ‘grounded’ in the patient data.

Mixed-method research designs can be of particular use in Primary Care research were psychosocial and cultural influences are potentially important considerations in care.

How relevant are recently completed

GPEP projects (1999–2001) to priorities for primary health care research, evaluation and development?

B Beacham, Primary Health Care Research and Information Service, Department of

General Practice Flinders University,

L Kalucy, B Lowcay

The National Health and Medical Research

(NHMRC) Project Grants in Primary Health

Care (announced in December 2001) aim to support primary health care research within research institutions. Eligibility criteria for

NHMRC Project Grants in Primary Health Care are the same as those for other NHMRC

Project Grants, however these proposals are required to supply additional information in several areas, including other work that relates to the project. The information required in this area includes a discussion of existing research/knowledge, together with a search of the relevant General Practice

Evaluation Program (GPEP) Database.

Objectives

This study explores the contribution of recently completed GPEP projects to future research into the priority areas identified to guide the Commonwealth Primary Health

Care Research, Evaluation and Development

(PHC RED) Strategy.

Methods

The reports of 52 GPEP projects completed between January 1999 and September 2001 were thematically analysed independently by two research officers.

2002 General Practice and Primary Health Care Research Conference — Proceedings

133

Principal findings

All projects in the study researched at least one of the seven PHC RED priority areas.

Twenty-eight projects (54%) covered two areas, nine (17%) covered three areas. Four priority areas were researched more frequently than others: quality of care

(n=20), evidence-based practice (n=11),

models of organisation and delivery of care

(n=9) and integration (n=3). Fifty-seven per cent of projects (n=32) had implications for the field of further research. Implications for practice and policy were identified in fewer projects (45% (n=25) and 39% (n=22) respectively), with the smallest proportion of projects having implications for education and training (16%, n=9). In five priority areas: quality of care, models of organisation

of care, integration, economic issues and

illness prevention GPEP projects were more likely to have implications for further research.

Discussion/conclusions

Recently completed GPEP projects have high content relevance to the PHC RED priority areas. Research conducted through these projects interacted across priority areas, as many projects covered two or three areas.

The results also indicate these projects provide a strong foundation for future research, evaluation and development in four of the identified PHC RED priority areas —

quality of care, evidence-based practice, models for organisation and delivery of care,

and integration. The foundation for research in the priority areas of illness prevention and

health promotion, health inequalities and

health economics is less strong. Projects in the study were most likely to have implications for the field of further research in the five priority areas of: quality of care,

models of organisation of care, integration,

economic issues and illness prevention and

health promotion.

Implications for practice

In designing future PHC RED research proposals an examination of the designs and methods used by relevant GPEP projects may be useful, enabling expansion of previous work. Challenges for researchers moving from the GPEP evaluative framework to the

PHC RED priority framework include: broadening the scope of research projects to encompass primary health care, with an emphasis on health promotion and inequalities as well as primary medical care; and increasing capacity as a research workforce to conduct research on health inequalities, health promotion and health economic issues.

Establishing a method for performing targeted 75+ Health

Assessments

K Berrisford, Department of General

Practice, Adelaide University, J Newbury,

C Laurence, J Beilby

Objectives

Initial studies of models of care similar to

Australian 75+ Health Assessments (75+ HA) produced inconclusive results.(1) However a secondary analysis(2) of a Californian randomised controlled trial(3) suggested that the elderly with Instrumental Activities of

Daily Living (ADL) impairment might be most likely to benefit from this model of care. A subsequent study found the elderly at high risk of nursing home admission did not benefit from this model of care compared with the rest of the elderly population.(4)

No Australian data was available on the percentage of the elderly population who had

Instrumental ADL and/or Basic ADL impairment. It was decided to test telephone screening on sample practice population to form risk cohorts prior to an evaluation of delivering appropriately different models of care to each risk cohort.

Methods

X

Step 1: Pilot conducting the

Instrumental and Basic ADL questionnaires over the telephone.

X

Step 2: Determine the size of the following cohorts:

(A) No ADL impairment

(B) Only Instrumental ADL impairment

(C) Basic ADL impairment (with or without Instrumental ADL impairment)

Principal findings

The acceptance rate among eligible patients contacted was 74.3% (342/460). There was no difference in age between those consenting and those not consenting

(81.0±4.1 years and 81.5±4.6 years, respectively).

Table 1 Size of the cohorts from the practices included in the trial

Cohorts* G (%)

A 36 (37.9)

B

C

36 (37.9)

23 (24.2)

Total 95 (100)

*see step 2 above.

B (%)

36 (57.1)

14 (22.2)

13 (20.7)

63 (100)

P (%)

48 (61.6)

15 (19.2)

15 (19.2)

78 (100)

M (%)

58 (54.7)

Total (%)

177 (51.8)

24 (22.6)

24 (22.6)

91 (26.6)

74 (21.6)

106 (100) 342 (100)

134

2002 General Practice and Primary Health Care Research Conference — Proceedings

Discussion

It was found the Instrumental and Basic ADL questionnaires were a quick and simple method of screening the 75+ population and forming the three cohorts.

This initial pilot study did not aim to determine if each risk group benefited from a

75+ HA. No trial has screened patients using the Instrumental and Basic ADL questionnaires and targeted services to each risk cohorts. Therefore a further trial needs to be conducted to establish this.

Implications for policy, delivery or practice

At present this screening method offers busy

General Practitioners, lacking the time or resources to perform 75+ HA on all their eligible patients, a method of conducting 75+

HA on those who are most likely to benefit.

This method has the potential to save money should a future trial find positive outcomes of performing a 75+ HA on those with only

Instrumental ADL impairment and a Care

Plan on those with Basic ADL impairment.

References

1. van Haastregt J, Diederiks J, van Rossum

E, de Witte L, Crebolder H. Effects of preventive home visits to elderly people living in the community: systematic review. BMJ 2000;320:754–758.

2. Bula C, Berod A, Stuck A, Alessi C,

Aronow H, Santos-Eggimann B, et al.

Effectiveness of preventive in-home geriatric assessment in well functioning, community-dwelling older people: secondary analysis of a randomized trial.

J Am Geriatr Soc 1999;47:389–395.

3. Stuck A, Aronow H, Steiner A, Alessi C,

Bula C, Gold M, et al. A trial of annual inhome comprehensive geriatric assessments for elderly people living in the community. N Eng J Med

1995;333:1184–1189.

4. Stuck A, Minder C, Peter-Wuest I,

Gillmann G, Egli C, Kesselring A, et al. A randomized trial of in-home visits for disability prevention in communitydwelling older people at low and high risk for nursing home admission. Arch Intern

Med 2000;160:977–986.

Research capacity building through a rural research network (EARIG)

A Bettess, Department of General Practice,

The University of Sydney, S Reid,

L Trevena, J Beard

Objectives

To develop research capacity in primary health care within the Northern Rivers region of NSW through a framework which:

X provides education about research and evaluation methodologies

X facilitates research partnerships and peer support between local primary health care providers

X establishes original and important research priorities

X provides mentoring for primary health care practitioners with evolving research ideas that are relevant to ‘grassroots’ primary care.

Methods

An Education and Research Interest Group

(EARIG) has been established for primary health care practitioners with interests in research and education in the Northern

Rivers region of NSW. The format of these monthly meetings over the next twelve months consists of:

X one hour of education on a topic pertaining to research skills followed by

X one hour as a forum for research ideas to be discussed.

Its membership comprises local general practitioners, population health workers and academics. The group acts as a ‘sounding board’ for the development of research ideas in primary care. It aims to provide a safe environment for peer and local academic support, mentoring and review, to facilitate the development of new research and the application of existing research knowledge.

An application for small group learning CPD points is anticipated to be successful.

Evaluation of the intervention is being conducted by pre and post meeting assessments of knowledge and attitudes to the research topic of interest each month.

There will also be a global measurement of knowledge and attitude to research and its application before and after the twelvemonth trial period (i.e. ten meetings). In addition the number and nature of new research ideas arising from this group will be

2002 General Practice and Primary Health Care Research Conference — Proceedings

135

recorded alongside a qualitative evaluation of mentoring and support processes.

Principal findings

Analysis of the self-reported knowledge and attitudes of the eight group participants at the commencement of the ten meeting intervention showed that all participants considered research in primary health care to be relevant but most indicated that their knowledge and perceived confidence in research skills were insufficient.

Discussion

It is hypothesised that EARIG will a) increase the knowledge of research methodologies in attendees b) increase the ability to apply research skills in everyday general practice and c) increase the intention to perform research in general practice. The EARIG research network will therefore assist in the establishment of partnerships in research and its application by engaging primary health care workers at the ‘grassroots’ level.

Implications for policy, delivery or practice

It is anticipated that the outcomes of this intervention and its evaluation will provide valuable information about a rural research network model and its potential role in research capacity building for primary care.

Such a model, we propose, will enhance research skills and interest and its application which will be reflected in the success of primary care-based research proposals, procurement of grants and the enhancement of an evidence-based approach to patient care.

Examining working relationships between GPs and mental health professionals: a journey through professional territories

K Boase-Jelinek, Department of General

Practice, University of Western Australia,

M Kamien, G Riley

Objectives

The project aims to examine collaboration between GPs and mental health professionals managing the health of consumers with a long-term illness, by:

X establishing the forms of collaboration that currently exist between GPs and mental health professionals

X determining the perceived effectiveness of this collaboration from the perspective of consumers, GPs and mental health professionals

136

2002 General Practice and Primary Health Care Research Conference — Proceedings

X identifying the historical, political, economic and social contexts in which

GPs and mental health professionals work together to provide services to consumers

X developing a process to assist GPs and mental health professionals to achieve successful collaboration.

Methods

GPs and mental health professionals are participating in individual in-depth interviews exploring ideas and experiences in working with each other. They are also taking part in a series of group discussions to develop strategies to improve working relationships.

GPs, mental health professionals and consumers form triads to participate in the case studies. The case studies are designed to track ‘day to day’ interactions between participants to manage the consumer’s health, as well as their experience of those interactions. Data is collected through interviews and an interaction record.

Principal findings

GPs and mental health professionals have variable levels of understanding of the key aspects of the tasks and functions of each other’s role. Mental health professionals tend to draw on personal experiences of seeing their own GP to build up a picture of the GP’s role. Almost all participants expressed the view that relationships were ‘better than they have been’. There were a number of cases where collaboration between the GP and mental health professional had contributed to a very successful outcome for the patient. A number of issues remain unresolved. These issues relate to access, consistency and form of communication, and an inadequate understanding each other’s role.

Discussion

This project is designed to enable GPs and mental health professionals to benefit throughout their participation in the project.

Reflection is a key component of almost all project activities. Involvement in individual and group activities encourages the development of self-awareness and the awareness of the roles of others. The length of involvement of GPs’ and mental health professionals’ participation in research activities provides them with the opportunity to integrate their learning into their practice and to reflect on the resulting changes.

Implications for policy, delivery or practice

Policy guiding the delivery of mental health and primary care services places an expectation on GPs and mental health

professionals to work together to manage the health of consumers who have a long-term mental illness. However, the nature of working relationships and the extent to which

GPs and mental health professionals collaborate seems to vary with individual practitioners. Recurring themes of problems related access, communication and understanding roles continue to adversely effect delivery and practice. At the same time the pockets of successful collaboration are growing and all those involve appear to benefit consumers. The challenge is to enable and motivate all GPs and mental health professionals to collaborate successfully.

Relationship between house dust and children’s blood lead levels in

Broken Hill

F Boreland, Department of Rural Health,

Broken Hill, University of Sydney, DM Lyle,

J Wlodarczyk, B Balding

Objectives

To determine the influence of indoor lead dust on children’s blood lead levels in Broken

Hill.

Methods

Indoor lead (measured as the amount of lead settling on surfaces per 30 days) was measured in the homes of 74 pre-school aged children, and compared with their blood lead levels. Both blood lead and dust lead data were found to be highly skewed, and so were log

10

transformed for the analysis.

Ecological analysis was used to determine whether findings for individual children were consistent across the whole population. Agesex standardised geometric mean blood lead level was estimated from routine surveillance data for each of ten districts in Broken Hill and compared with the geometric mean indoor lead level of each district.

Principal findings

Children’s blood lead level increased as the lead level in their home increased — children living in homes with high indoor lead levels were five times more likely to have a significantly elevated blood lead level than children living in homes with the lowest lead levels. The data indicate that increasing indoor lead by 90% increased blood lead levels by 30–40%. Findings for individual children were consistent with results of the ecological analysis.

Discussion

This study indicates that high indoor lead levels are a potential hazard to young children in Broken Hill. If this reflects a causal linkage, it indicates that a dramatic reduction of lead levels in the home is required to reduce blood lead levels by a moderate amount. This has implications for the lead management program, given that home remediation has not shown a dramatic effect on blood lead levels to date. The data will allow better targeting of remediation, help provide realistic targets for reduction in blood lead level following such interventions, and provide a benchmark against which the effectiveness of interventions can be measured.

Implications for policy, delivery or practice

Reducing indoor lead levels by maintaining homes in good condition is an important component of a multi-faceted strategy for reducing blood lead levels among Broken Hill children. It is particularly important that homes in areas of high soil lead levels be adequately maintained.

Effect of remediation on indoor lead levels in Broken Hill homes

F Boreland, Department of Rural Health,

Broken Hill, University of Sydney, D Lyle

Objectives

To determine if home remediation (removing or stabilising sources of lead dust) reduced indoor lead levels in Broken Hill, a silverlead-zinc mining town in outback Australia.

Methods

A study of the effect of remediation on indoor lead levels was embedded into a randomised control trial of the effect of home remediation on elevated blood lead levels for children aged six to 60 months. Lead loading

(micrograms of lead per square metre) was measured in 103 homes by wiping areas of known size with Black and Gold moist

Towelettes. Samples were collected from floors and windowsills of at least three rooms

(kitchen, lounge room and child’s bedroom) and immediately in front of the most used entry to the house. Homes were sampled twice before remediation, immediately after, and then at two-monthly intervals for ten months. Where possible, samples were collected at the same location on each visit.

In some homes it appeared that the first sample included a large ‘historical component’ — that is, it was obvious that the surface had not been cleaned for many

2002 General Practice and Primary Health Care Research Conference — Proceedings

137

months. To avoid bias due to the historical component, analysis was based on the second pre-remediation sample. Data were log

10

transformed for the analysis.

Principle findings

Remediation reduced average indoor lead levels by approximately 50%, and lead levels remained low for the duration of the followup period (10 months). Homes with the highest initial lead levels showed the greatest gains, with indoor lead levels reduced by 80–

90%; homes with low pre-remediation lead levels showed little or no benefit. Before remediation, homes with poor dust proofing had higher lead levels than those with medium or good dust-proofing; these relative differences remained after remediation. This was in part due to their location, and may also reflect the difficulty of adequately dust proofing a house which was in poor condition to begin with. There was no evidence that lead loading was reduced by increased opportunity to become aware of lead issues.

Discussion

This study indicates that the benefits of remediation are not uniformly observed across all houses. Given that the home remediation program to date has not shown a dramatic effect on children’s blood lead levels, and that data from another study suggests a dramatic reduction in indoor lead levels may be required to achieve a modest reduction in blood lead levels, this study suggests home remediation is most likely to reduce blood lead levels for those children who live in homes with high indoor lead levels. As well as helping to identify homes which are likely to benefit from remediation, this study will allow realistic targets to be set for reductions in lead levels following remediation.

Implications for policy, delivery or practice

Remediation is an effective strategy for reducing lead exposure to children living in homes with high indoor lead levels. The results suggest that lead awareness alone is not enough to reduce lead levels in some homes, implying ongoing need for targeted remediation.

Desktop clinical audit

D Brookman, Discipline of General Practice,

Faculty of Health and Population Science,

University of Newcastle

Definition

The process of conducting a computer-based audit of General Practitioner activity using

138

the data collected by the GP in their normal daily activities and providing results that can be used for Interpractice quality of care performance comparisons, and/or aggregated for assessment of adoption of evidence-based practice, policy changes, and in some circumstances cross sectional studies.

Objectives

X

Primarily

X

– to develop a system of clinical audit that did not impose additional work upon rural GPs

– to provide a data collection system that GPs could use for research either as a solo practice, or as a group(Divisions of General Practice)

– to provide a system which GPs could use to audit their application of evidence-based medicine.

Secondarily

– to provide a system for overall quality of care assessment

– to comply with privacy regulations

– to provide a tool for estimation of selected disease prevalence in identifiable communities.

– to provide a tool which identified to the practitioner patients who needed to be reviewed, either because they were at risk and had not been checked, or because their disease

Methods management was suboptimal.

Programs have been written to audit the quality of hypertension management, and diabetes care (funded by the University of

NSW). The hypertension program assesses the recording of various anthropometric data against diagnosis, and comorbidity. The quality measures chosen were.

X proportion of patients screened for hypertension

X success in lowering blood pressure

X use of ACE inhibitors in diabetics with hypertension

X use of Beta Blockers in hypertensives with ischaemic heart disease

X use of dihydropyridine calcium channel blockers on hypertensives with ischaemic heart disease.

The diabetes project has extended along similar lines additional modules to scan the electronic pathology results for results such as HbA1c and incorporate them into the

2002 General Practice and Primary Health Care Research Conference — Proceedings

quality of care assessment, and produce cardiovascular risk estimates based on the

NZ dyslipaemia tables.

Principal findings

The programs are written in MS Access

©

, and use ODBC (open database connectivity) to read Medical Director

©

(MD) data files

(without altering them) and aggregate them into tables which can then be crosstabulated for analysis. The analyses provide an audit of the entire active patient data set

Discussion

The extent of recording of diagnostic, therapeutic and anthropometric information by General Practitioners in Australia into their computer database is unknown it may be very poor. It is hoped that provision of this computer-based method of auditing will promote better data recording by GPs as well as provide them with a tool for ongoing research and audit.

Implementation

Divisions are the ideal distribution point they have the networks to set up small groups for self review, and provide additional education as needed, and in some cases provide computer support.

The clinical audit software can be used immediately by those who already have good quality data in Medical Director.

Policy implication

This provides a simple, cheap, easily adopted means of total quality of care review.

Quality indicators for diabetes programs in Australian Divisions of

General Practice

JR Burns, Centre for General Practice

Integration Studies, University of New South

Wales, PG Powell Davies, MF Harris

Objective

To outline recent initiatives and issues arising around the development of quality indicators for diabetes programs in Divisions of General

Practice, develop a framework and make recommendations for selection and use of indicators.

Methods

Current literature in the area of quality improvement and quality indicators within

Australia and overseas was reviewed as was the literature around best practice care in diabetes and current clinical management guidelines.

The framework and examples of indicators were presented to the General Practice Policy

Advisory Committee and to other key stakeholders.

The Data Collation Project results from 1999–

2000 represent the feasibility of collecting and using the indicators for quality improvement.

Principal findings

Recent initiatives in the area of diabetes are reviewed and key issues identified. A quality framework was developed showing the flow of quality information in Divisions.

Critical concepts include the quality assurance role in Divisions, strengthening links and acting on results, the use of different indicators for different purposes, the use of simple and sustainable indicators for general practice, adopting a positive approach and issues around privacy, consent and data ownership.

Key elements of structured care including register and recall systems, clinical management guidelines, diabetes education in general practice, quality assurance in diabetes programs and communication problems between providers are discussed as well as future directions.

Quality indicators for diabetes programs are described at practice, Division, state and national levels with likely sources and a guide to interpretation of the indicators and what change in the indicator may mean.

Discussion

The introduction of the National Integrated

Diabetes Program by the Commonwealth government with its focus on quality of care of people with diabetes has meant that the description of indicators which are well defined and where a guide to interpretation is provided is timely. It is also important to recognise that the same indicator can be used for different purposes. Likewise, different indicators are often appropriate for different purposes.

One critical aspect of the collection of this information in the current Australian health scene is the importance of such information at different levels of the system. In particular, information collected at provider level has important implications for other levels of the system (practice, Division, state and national) and aggregation of such

2002 General Practice and Primary Health Care Research Conference — Proceedings

139

information in a de-identified form means that it can be used for purposes such as quality assurance, service planning and continuing education. Importantly, it can also be used to inform policy development particularly at state and national level. There are also important issues around who should collect and aggregate such information.

Implications for policy, delivery or practice

Description and definition of quality indicators is important. However describing how that information can appropriately be used and at what level and for which purposes in the health system can make it more applicable, more relevant and easier for policy makers to take up and use.

Conclusions

To take research into policy, it is necessary and desirable to not only present the results, but to demonstrate its feasibility and describe how it can appropriately be used and for what purposes.

Barriers and facilitators of research in GP setting: experience from asthma education trial in north-west

Melbourne

N Sulaiman, J Byron, Department of

General Practice, University of Melbourne,

T Maksimovic, T Liaw, C Harris, C Barton,

C Hogan

Aim

To examine the barriers to and facilitators of education intervention trials in general practice. Specific questions are:

X

What are the main barriers to research in general practice in the North West

Melbourne?

X

Why do GPs participate in education trials?

Methods

This paper discusses the experience accumulated from conducting a trial of a GP asthma and ENT educational package. The methods used included:

X

Building collaboration between the

Department of General Practice, Northern and Northwest Melbourne Divisions of

General Practice, Royal Australian College of General Practitioners (RACGP) and the

Royal Children’s Hospital (RCH).

X

Setting up a local guidelines development group (GDG) to adapt best practice

140

guidelines on asthma.

www.asthma.unimelb.edu.au

X

Promoting the project through two divisions of general practice and RACGP and recruiting GP based on the membership list of the two divisions.

X

Randomising the practices into three groups: (1) asthma education plus guideline group, (2) asthma guideline only group and (3) ENT guidelines group.

X

Visiting each practice at least twice to engage the practices and assist them to identify eligible patients in their practice computer systems.

X

Recruiting and engaging patients.

Findings

The main barriers to research, identified during the asthma/ENT study are presented according to the following framework:

X

GP factors, including lack of time, cynicism of research, fear and isolation of

GPs, perception of data collection role, and lack of financial incentives.

X

Practice factors, including privacy and confidentiality, getting practice commitment, level of computerisation, time and resistance of the practice manager/nurse, availability of diagnosis

(misdiagnosis), changing status such as retirement or moving practice, and knowledge of software

X

Patient factors, including wrong address, inability to follow patients because telephone is not working or wrong address, denial of some diseases such as asthma, lack of incentives, wording and length of the questionnaire, language difficulty, perception of the diseases and value of research to patient and the community. Finally reply-paid postal services may not be reliable.

X

Investigator factors such as extensive time commitment to engage GPs, waiting time in their practices and difficulty of establishing contact with GP.

Continuing professional development points and asthma education workshops were found to be effective in attracting 63 GPs from 32 practices, who are currently engaged in the study. Other facilitators include personal contact, persistence, enthusiasm and providing assistance to GPs to identify patients and facilitate research.

2002 General Practice and Primary Health Care Research Conference — Proceedings

Discussion

Despite numerous research, evaluation and development initiatives in general practice, research is not considered to be an important aspect of general practice. To participate in research, GPs need incentives, motivation, time compensation, interest, passion and acknowledgment. The research question should address practice-based problems, be relevant to their patients and have a potential to make a difference to their patients’ health care and health.

Transferring the care of survivors of childhood cancer to the patient’s own GP

L Cavenagh, Hornsby Ku-ring-gai Ryde

Division of General Practice, R Starrett

This project has developed and implemented a protocol which will facilitate the seamless transfer of medical care of patients who are survivors of childhood cancer from the Late

Effects Oncology Clinic at the Children’s

Hospital at Westmead NSW to the patient’s own general practitioner in the community.

Patients are selected for transfer to their GP using a specially developed protocol. Patients are given an individualised Patient Passport and an individualised copy of patient guidelines, which explain to the patient the possible late effects of treatment.

Their GP is contacted by the Late Effects

Oncology Clinic at Children’s Hospital at

Westmead and sent a copy of the GP

Guidelines, which are more comprehensive and also contain information about the patient’s actual treatment to date. They also contain a complete list of drugs used in treatment of childhood cancer and the likely late effects. In addition the GP is given a chart with recommended monitoring and screening procedures for the individual patient.

The transfer of care from hospital to GP is recommended through a case conference involving the patient, GPLO from the Late

Effects Clinic, the patient’s oncologist and the

GP.

The poster outlines the process of the project and gives results of GP and patient compliance and satisfaction.

Health status and general practitioner services: preliminary results from an Australian health production study

LB Connelly, Brisbane Graduate School of

Business, Queensland University of

Technology, DP Doessel, S Strong

Objectives

This study is concerned with measuring the direction and magnitudes of influence of a range of social, economic, cultural, and health care resource variables on mortality rates in the Australian population. The objective of the study is to quantify the impacts of marginal variations in the levels of each of these variables, including variations in expenditures on services of a general practitioner (GP) kind (i.e., ‘unreferred attendances’), on health states (as proxied by mortality rates).

Methods

Using multiple regression techniques, treating health as a latent variable, and controlling for endogeneity between the health status and health care variables, we estimate health production functions on data collected at the level of the statistical local area (SLA) for all Australian SLAs. Data on economic, social, cultural and demographic variables, e.g. data on income, education, non English-speaking background (NESB),

Aboriginal and Torres Strait Islander (ATSI) status, age and gender distributions, and so on, are taken from the 1996 Australian

Census of Population and Housing (ABS

1996). Data on health care expenditures, numbers of services and health personnel are derived, by broad service type (e.g.

‘unreferred attendances’, ‘assist at operations’, ‘anaesthetist services’, and so on) are derived from the Health Insurance

Commission (HIC) records for the same year.

1

Data on mortality are derived from detailed, de-identified, unit record data, that were supplied by the Australian Bureau of

Statistics (ABS 2000) for five different years,

viz. 1994, 1995, 1996, 1997 and 1998.

Principal findings

The empirical results are expected to reveal that variations in key economic and social variables, along with variations in primary and other levels of health care are associated

1 We are grateful to Mr Ross Saunders and the

Medicare Statistics Branch of the Commonwealth

Department of Health and Ageing for supplying the

Medicare data described here.

2002 General Practice and Primary Health Care Research Conference — Proceedings

141

with variations in mortality rates, at the SLA level.

Discussion

The results of this economic study are pertinent to the formulation of national health policy. They emphasise the multivariate nature of the health production process and indicate, empirically, that increased expenditures on health services constitute one of many possible ways of improving the health status of the Australian population. The results provide an insight into the marginal productivity of health inputs of a kind that has not previously been available to Australian policy makers.

Research as practice: using individualised medication effectiveness tests to improve prescribing

J Nikles, P Glasziou, C Del Mar, Division of

General Practice and Community Health, The

University of Queensland, C Duggan,

G Mitchell, V O’Connor, J Price,

M Yelland, P Coxeter, D Askew,

T Janamian, A Spinks, A Clavarino,

P Schluter

Objectives

To evaluate the uptake and impact of ‘clinical trials’ for individual management decisions in medical practice, using Individualised

Medication Effectiveness Tests (IMETs) (also known as Single Patient Trials or n-of-one trials)

Methods

The study subjects were patients who were uncertain about the effectiveness of their medication, in any of the 4 clinical areas, meeting appropriate inclusion/exclusion criteria. The design was a series of randomised double blind placebo-controlled therapeutic drug crossover trials, with 3 pairs of treatment periods, in single patients.

Principal findings

IMETs have been developed in four clinical areas: chronic pain, ADHD, insomnia and menopausal symptoms. 210 patients of over

140 doctors have commenced IMETs and 126 patients have completed IMETs to date.

25% of the OA patients and 35% of the

ADHD patients were responders. Of 47 out of

74 patients where data is available, management plans changed in 47 (64%).

Discussion

Doctors and patients anywhere in Australia now have a practical and effective method to rigorously assess whether an individual patient is a responder to a particular medication in certain chronic conditions. This allows tailoring of medication management to the individual case.

Implications for policy, delivery or practice

IMETs are a practical tool allowing GPs anywhere in Australia to undertake clinical research on a day to day basis to solve clinical problems, and the results of the research can be used to inform prescribing decisions.

Rethinking trials of antibiotics for acute respiratory infections: 1) taking account of recruitment exclusion of most ill, and 2) the inadequate estimates of the quantity of illness

C Del Mar, Centre for General Practice,

University of Queensland, J Doust

Objectives

Trials of antibiotics for symptoms of ARIs have developed in an environment in which antibiotic use was already established. Thus trials have been of ‘no-antibiotic’ vs usual care (antibiotics). This might influence recruitment (less severely affected vs more).

If severity influences response to treatment, then excluding more severely affected patients would yield a smaller effect size.

Similar concerns centre on the measurement of illness (whose dimensions include severity as well as duration) by duration alone will seriously under-estimate differences (and therefore power) in trials of effectiveness.

Method

We reanalysed the trials selected for the

Cochrane meta-analysis of ‘antibiotics for acute otitis media in children’, looking for evidence of patients excluded on the basis of being too ill. We also looked for evidence of severity influencing outcome.

Principle findings

We could extract ‘non-recruitment because the child was too ill’ out of the total recruited from only two of the seven trials. They were

52/232 and 27/240 respectively, or 17%.

In none of the trials were estimates of severity measured.

142

2002 General Practice and Primary Health Care Research Conference — Proceedings

Discussion

Both these effects have the potential to result in under-estimates of the effects of antibiotics for acute otitis media.

Implications for research

It is necessary to consider both these effects in the design of future trials of effectiveness in areas where ‘non-treatment’ is being tested against established treatment.

Enhancing software development by understanding prescribing decisions: a user-centred approach to design

E Deveny, Department of General Practice,

University of Melbourne

Objectives

Adverse events, Quality Use of Medicines and medical informatics literature argue that electronic decision support, embedded in clinicians’ computers, will improve the quality and ‘appropriateness’ of prescribing. This paper explores the decision-making process of general practitioners (GPs) in relation to prescribing and then identifies the features of electronic resources, such as a pharmaceutical decision-support system, that might better facilitate quality prescribing.

This work forms part of a larger project that aims to develop an integrated approach to the development of an electronic guideline production system and process for

Therapeutic Guidelines Limited.

Methods

Twenty GPs participated in semi-structured interviews that focused on four themes — prescribing, decision making, information resource use and computers. Eighty general practice consultations were also observed.

The interview and observation transcripts were coded and analysed both by hand and using Atlas ti, a qualitative software analysis package.

Principal findings

The analysis of interview and consultation material indicates that each decision to prescribe is a highly complex event, involving an amalgam of factors. The professional identity of each general practitioner is shaped by their personal identity, their practice style and philosophy. This professional identity, along with the GP’s experience of the patient, the disease and the drug, is reflected in each prescribing decision. The identity and experiences of the patient also critically inform the decision to, or not to, prescribe a therapeutic agent.

Discussion

The use of computers by General

Practitioners, particularly when prescribing, has increased significantly over the last five years, and it continues to do so. Government policy and incentive schemes are both encouraging and obliging GPs to computerise their work. This computerisation brings benefits but also generates a financial cost and creates a significant change in GPs’ work practices and their interactions with patients and computers. Electronic resources are often designed with a tendency to focus on the clinical aspects of the doctor–patient encounter, and so, are of limited use to GPs and their patients. It is vital that research is undertaken that investigates the work practices of GPs with a view to better informing those developing medical software.

Electronic resources based on human-centred design principles, which take into account the professional, organisational, and sociocultural aspects of prescribing decisions, would enhance application use and usefulness.

Implications for policy, delivery or practice

Consumer perspectives on health care, patient-centred medicine and evidence-based medicine are just three orientations to medicine that are currently being reflected in the kinds of consultations, decisions and relationships that patients and GPs now share. Software that can support interactions of varying natures will not only be more useful but also provide an opportunity to improve the quality of the doctor–patient– computer relationship.

Medical emergencies in general practice in south-east Queensland: prevalence and practice preparedness

C Johnston, M Coulthard, P Schluter,

M Dick, Centre for General Practice,

University of Queensland

Objectives

To determine: (i) the type and frequency of emergencies encountered in general practice; and (ii) the extent to which general practices are equipped to appropriately respond to emergencies.

Methods

A random sample cross-sectional questionnaire survey was distributed to 900

2002 General Practice and Primary Health Care Research Conference — Proceedings

143

GPs currently working in different general practices in Southeast Queensland.

Preliminary telephone calls alerted GPs of the forthcoming questionnaire, and clarified contact details. A questionnaire was posted to each GP seeking a range of information regarding their experience of medical emergencies. Telephone reminder calls were made to non-responding doctors at 2 and 4 weeks after posting the questionnaires. The

SAS statistical software package was used to analyse responses to the questionnaire.

Principal findings

512 GPs (response rate, 57%) reported managing a cumulative total of 5,640 emergencies over the preceding 12 months, with GPs seeing a median number of 8 emergencies/year that potentially required resuscitation. Non-metropolitan GPs saw about 30% more emergencies than their metropolitan counterparts (P = 0.02). The three most commonly seen emergencies

(seen by more than 48% of all GPs) were acute asthma, psychiatric emergencies, and convulsions, whilst thyroid crisis, immersion and asphyxia were the least commonly seen emergencies (seen by less than 4% of GPs).

Most GPs (77%) stocked 15 or more of the

16 emergency Doctor’s Bag drugs but a smaller proportion (67%) had all of the basic emergency equipment items considered essential.

Discussion

This study reports findings from a large population-based study of GPs, the first of its kind in Australia, designed specifically to determine metropolitan and rural GPs’ experiences in dealing with medical emergencies. Key findings are that GPs are required to manage a substantial number of patients with potentially life-threatening emergencies, and that whilst Doctor’s Bag emergency drugs are reported to be available in most general practices, the availability of basic emergency equipment is suboptimal.

This study demonstrates that general practice is a vital part of the health system’s response to medical emergencies, and it provides information to support health service planning and development. The suboptimal availability of basic emergency equipment calls for action, such as the development of national guidelines specifying the minimum requirements of emergency equipment and drugs in Australian general practice, and the adherence of these guidelines by all GPs.

Implications for policy, delivery or practice

This study’s findings have important implications for general practice accreditation

144

both in Australia and internationally, and it provides information to support health service planning and development.

General practitioners’ confidence and perceived competence in dealing with medical emergencies

M Dick, Centre for General Practice,

University of Queensland, P Schluter,

C Johnston, M Coulthard

Objectives

To determine GPs’ (i) past training in emergency skills; (ii) level of confidence in dealing with medical emergencies; (iii) perceived level of competence in emergency skills; and (iv) interest in attending an emergency skills update course.

Methods

A random sample cross-sectional questionnaire survey was distributed to 900

GPs currently working in different general practices in Southeast Queensland.

Preliminary telephone calls alerted GPs of the forthcoming questionnaire, and clarified contact details. A questionnaire was posted to each GP seeking a range of information regarding their experience of medical emergencies. Telephone reminder calls were made to non-responding doctors at 2 and 4 weeks after posting the questionnaires. The

SAS statistical software package was used to analyse responses to the questionnaire.

Principal findings

Whilst only ~20% of GPs recalled training in all the listed skills, more than 90% recalled past training in the more basic skills. Most reported their skills training as occurring after medical school.

GPs’ mean levels of comfort in managing the listed medical emergencies were at least moderate only for acute asthma and hypoglycaemia. Higher levels of comfort were significantly associated with male gender, full-time work status, and having managed that emergency in the preceding 12 months.

GPs’ mean perceived skills competencies were at least reasonable for 10 of the 16 listed emergency skills, with highest skills competence levels being reported for venesection, external cardiac compression, cannulation and IV infusion. Higher levels of skills competence were statistically associated with male gender, working in a rural or remote location, full-time work status, having essential items of emergency

2002 General Practice and Primary Health Care Research Conference — Proceedings

equipment and drugs, and having had past skills training.

Approximately 70% of GPs indicated an interest in attending an emergency medicine update course for GPs in Brisbane.

Discussion

This study sought to identify GPs’ selfreported confidence and perceived competence in dealing with medical emergencies, with a view to identifying whether deficiencies existed, and if so, possible means to improve the quality of health care delivery. The key findings were that greater skills competence and higher levels of comfort in managing medical emergencies in general practice are significantly associated with prior emergency medicine training and having well-equipped practices. Additionally, GPs identified a need for readily accessible emergency medicine update courses. The study therefore clearly identifies some practical means by which patient care may be enhanced. It supports the increased provision of courses designed specifically for GPs to enhance their opportunities for training and experience in medical emergencies, and the provision of well-equipped medical practices.

Implications for policy, delivery or practice

The findings from this study have important implications for the undergraduate and postgraduate training of general practitioners, and general practice accreditation both in Australia and internationally.

Assessing data sources for the monitoring of the Australian General

Practice Asthma Initiative: the

Asthma 3+ Visit Plan

A Duszynski, Department of General

Practice, The University of Adelaide, J Beilby

Objectives

Provisions introduced within the Australian

Federal Budget of 2001–02, enables general practice to improve the extent of clinical care to moderate to severe asthmatics; the

Proactive GP Management of Asthma. The

Budget initiative entails the Asthma 3+ Visit

Plan — asthma management at the general practice level through evidence-based measures. Assessing change encompasses a continuum of care; with the majority of the burden of the disease observed within general practice.

Methods

In evaluating the plan, it is important to assess current primary care data collection activities, including measures of asthma management through self-reported questionnaires and national data collection activities.

An extensive search of Australian sources of such data was conducted with relevant contact made with key stakeholders. Data items collected address the draft asthma indicators developed by the Australian

Institute of Health and Welfare (AIHW).

Results

In reviewing primary care data, it is apparent that key data providers exist to cover the gamut of the AIHW asthma disease indicators. Items relate to disease prevalence

(risk factors, co-morbidities and emergency attendances), patient management

(pharmaceutical interventions, health reviews and education) and morbidity data.

Although no single clearing-house collects all indicators, it is recognised that general practice collection and collation activities provide the majority of such indicators.

General practice derived data allows longitudinal and retrospective tracking of patients with asthma, existing in a format that allows introspection; including the

University of Adelaide’s MEDIC-GP electronic clinical casenote database and the Health

Communication Network (HCN) General

Practice Research Network (GPRN) clinical software casenote database. Further general practice data collection activities include the continuous cross-sectional BEACH surveys

(AIHW: Bettering the Evaluation and Care of

Health).

Activities at the state and national levels, include self-reported data collection through questionnaire-based ‘health surveys’; instrumental in gauging consumer perceptions of disease management, quality of life measures and the effects of education.

Finally, national data collections through the

Health Insurance Commission (HIC), monitor patient outcome data including asthma mortality and hospital presentation data.

Discussion

In monitoring general practice initiatives for moderate to severe asthmatics — the Asthma

3+ Visit Plan — it is apparent that general practice in servicing the majority of the burden of asthma, would be the source of current (and retrospective) data on patient management. It is thus proposed that the use of pre-existing casenote databases would

2002 General Practice and Primary Health Care Research Conference — Proceedings

145

provide the most complete picture of management and measures.

Computer-assisted telephone instruments

(CATI) utilised in the range of self-reported health surveys, provide the consumers’ perspective on asthma management and quality of life measures not typically collected within general practice. National data collection activities albeit limited, round out the picture of asthma in Australia today.

Implications for policy, delivery or practice

The use of general practice clinical casenote records which provide retrospective longitudinal data, is recognised as an important tool in not only assessing the management of asthma but chronic disease in general. No further collection activities within Australia collect a range of asthma management indicators as succinctly, with the use of such databases currently demonstrating the effects of general practice policy initiatives.

Why collaborate? State-based strategies for capacity building in primary health care to lead to a more influential and appropriate program

EA Farmer, South Australia PHC RED

Collaboration, A Jones, KM Weston,

J Fuller, J Taylor, N Stocks, T Burgess

Introduction

Eighteen university departments of general practice and rural health in Australia have been funded over a five-year period to undertake local activities that aim to build capacity in primary health care research. This initiative, the Primary Health Care Research,

Evaluation and Development (PHC RED) program is a major element of the

Commonwealth PHC RED strategy. This strategy evolved through a consultative process following a general practice strategy review in 1998 and a total of $50 million has been allocated to fund the program over the initial five-year period. There is a significant amount of work to be done to address current barriers to practitioner involvement in research, in training individuals to undertake research, and in supporting, and consolidating, the evidence-based approach in primary health care. Three principles underpinned the strategy:

X that decisions made in primary health care should be evidence-based;

X that primary health care is central to our health system;

X that the evidence base for primary health care needs to be strengthened.

State-wide collaboration

This presentation outlines the collaborative approach taken in South Australia. The three university departments in South Australia funded under the PHC RED capacity building program formed a state-wide collaboration in

2001. The collaboration comprises the

University of Adelaide Department of General

Practice, Flinders University Department of

General Practice and the South Australia

Centre for Rural and Remote Health

(SACRRH). The universities also decided to maintain a close link with the state-based organisation, South Australia Divisions of

General Practice Inc. (SADI), and so SADI was formally associated with the collaboration.

While there is a need to balance individual

Departmental and collaborative aims, the foundation of effective collaboration is based initially upon sharing expertise and creating a cohesive team rather than creating new knowledge. Exploiting this possibility strategically can enhance program development and lead to a more powerful and influential program in areas of mutual importance than could be achieved by each

Department working in isolation. At the same time the team is able to work towards developing joint training and research projects, which adds even more value to the potential of the collaboration in the longer term.

One key element of the current collaborative approach, a ‘Research Training Needs

Analysis’, is presented in detail to provide an example of the operation of the collaboration in 2002.

The poster presentation provides a forum for both discussion of the value and the barriers to a collaborative approach and sharing of experiences from similar strategies used elsewhere at the general practice and primary health care level.

146

2002 General Practice and Primary Health Care Research Conference — Proceedings

Conversation analysis applied to auditing GP consultations — what was old is new again

B McGrath, P Freebody, Faculty of

Education, Griffith University, J Freiberg

Objective

This project shifts conversation analysis (CA) from a research methodology remote from

GPs to an everyday tool to enhance GPpatient interactions, medical education, GP research knowledge and patient outcomes.

Methods

A research protocol utilising CA and action research is now the basis of a clinical audit option offered to all Australian GPs through the QA and CPD office of the Royal Australian

College of General Practitioners. The clinical audit focuses on the talk within consultations.

Audio-taped consultations of GPs will be transcribed and analysed using CA. Groups of

GPs will then meet to discuss the aggregated core findings of each GP’s analyses and provide additional information based on clinical perceptions of the interactions under scrutiny.

Principal findings

The essential elements of our Australian

Research Council funded project which examines the interaction within the GP consultation, forms the basis of the clinical audit. The study demonstrated the acceptability of the ethics approval process, the recruitment of and the gathering of informed consent for patients within a general practice setting. Participating general practitioners felt that the audio-taping of fifteen consultations had little impact on their consultation style. The lessons learnt from addressing the practical aspects of transcribing and analysing consultations will facilitate the establishment of a national team to provide state-based meetings of general practitioners who undertake the clinical audit.

Discussion

This clinical audit project embeds research into the day-to-day clinical work and ongoing medical education of Australian general practitioners. It provides a unique means to enable individual doctors to conduct research, using an innovative methodology and a collaborative process.

Implications for policy, delivery and practice

The audit takes place within the setting of the surgery and enables participants to meet with other doctors and academics to increase understanding of research and clinical practice. Most importantly, lessons learnt through the research aspects of the audit may be translated immediately into practice.

Encounters in general practice with patients from areas of high and low disadvantage: a secondary analysis of Bettering the Evaluation and Care of Health (BEACH) data

M Harris, J Furler, University of Melbourne,

L Valenti, E Harris, H Britt

Objective

To compare general practice management of patients from areas of high and low socioeconomic disadvantage

Methods

A secondary analysis of the BEACH data set for the period April 1998 to March 2001

(inclusive) was undertaken. The Socio-

Economic Indexes for Areas, Index of

Relative Socio-Economic Disadvantage

(SEIFA IRSD was used to assign patients to areas of relative disadvantage based on patient postcode. Data was grouped into quantiles from 1 (most disadvantaged) to 11

(least disadvantaged). Encounters with patients from the most and least disadvantaged groups were compared.

Multiple logistic regression was used to control for social, demographic and morbidity variables.

Principal findings

Of 303,000 patient encounters over this three-year period, 289,172 had a valid postcode that mapped to a SEIFA quantile.

SEIFA quantiles 1 and 11 contained

19,790and 37,608 encounters with 1522 and

1696 GP respectively.

There are significant differences in the patient characteristics, encounter type and management of problems between the two groups. Patients from areas of high disadvantage (AOHD) are more likely to be male, and hold a health care card, but less likely to be over 75 or hold a veterans card.

Encounters with patients from low disadvantage areas are more likely to be long.

In the univariate analysis, patients from

AOHD are more likely to present with skin, digestive and psychological problems, and are also more likely to have them managed, except for psychological problems, where the difference disappears. They are less likely to present and have managed female genital

2002 General Practice and Primary Health Care Research Conference — Proceedings

147

system problems. A medication prescription was a more likely outcome for the high disadvantaged group, while testing and referral were more likely for the least disadvantaged group.

Multiple logistic regression showed patients from AOHD to be significantly more likely to come from non-English speaking or

Aboriginal and Torres Strait Islander backgrounds, or be from a rural location and less likely to have an encounter with a female GP. They were more likely to have hypertension, diabetes, asthma, back complaints and osteoarthritis managed, but less likely to have an immunisation given or dermatitis managed.

Discussion

Equity of access to care is a key area of research in primary care. This involves both equity of access to care encounters, and to quality care. The higher burden of disease found in people from AOHD is not reflected in the lower number of encounters recorded with such patients over this time period, while it is reflected in the higher likelihood of such encounters including management of

1696hypertension, diabetes and asthma.

However, psychological conditions and depression are not managed at a significantly different rate, despite some evidence of a higher prevalence in disadvantaged patients.

The variation in the type of care provided

(less likelihood of preventive care offered

(immunisations), higher reliance on prescriptions and less testing and referral) may also reflect inequity of access to quality primary health care.

Implications for research, funding and practice

Matching service delivery to need requires attention to workforce distribution and gender balance, and an exploration of the factors contributing to variations in care provision.

The Servier Australia National

Diabetes Audit

A Gialamas, Department of General

Practice, The University of Adelaide,

A Duszynski, J Beilby, H McElroy,

T Burgess

Objectives

To encourage general practitioners (GPs) to evaluate their type 2 diabetes management; to provide GPs with an opportunity to improve their skills and knowledge in type 2 diabetes and to provide GPs with a mechanism for continuing medical education in diabetes.

Methods

The Servier Australia National Diabetes Audit involved GPs auditing 10–15 type 2 diabetes patients, receiving feedback from that audit, attending a workshop which discussed the key findings and 12 months later re-auditing those same patients. 176 GPs participated in the first audit with 113 GPs submitting second audit data and completing the audit cycle. The total number of matched audit patient records was 1325.

Results

The results indicated that patient management through appropriate testing improved between audits. Results from the first audit showed a high rate of HbA1c assessment (91%) which significantly increased by 4% to 95% in the second audit.

Further to this, there was a 14% increase in

GPs testing urinary microalbumin in the second audit (53% to 67%), an 11% increase in assessing nail/foot care (67% to

78%), 10% increase in assessing ulcer pressure points (66% to 76%) and a 9% increase in assessing pulses/circulation (72% to 81%).

The first audit showed that 76% of patients had never been reviewed by a podiatrist and

34% had never been reviewed by an ophthalmologist. Comparative analysis between audits showed a 4.8% increase in referrals (in the last 12 months) to the podiatrist increasing from 15% to 20%.

However, there was only a modest increase in referrals to ophthalmologists with a 5% increase between audits (53% to 58%).

Discussion

Results from the first audit showed that the majority of GPs were generally providing integrated care for their type 2 diabetes patients. However, there were still certain areas which required improvement, particularly foot care. The second audit data showed significant improvements in a many areas of diabetes care including patient management of peripheral circulation and associated nail/foot care and referral to the podiatrist. These results are particularly positive given that foot care tends to be overlooked in diabetes management.

Overall, the combined audit-workshop activity has been shown to be effective in changing practice. The audit results helped

GPs identify the areas important to improving care specific to their practice setting. The workshop allowed GPs to discuss issues

148

2002 General Practice and Primary Health Care Research Conference — Proceedings

raised by the audit process as well as compare practice with their peers and listen to expert opinion from podiatry professionals and GP/endocrine specialists.

The program also evaluated GPs’ perceptions of the audit process. The majority of GPs

(92%) returning an evaluation form stated that undertaking the audit had assisted them in their clinical practice. The two main areas of assistance identified by GPs were an increased awareness of the need to better manage their diabetes patients and recognition of deficiencies in the care they were providing.

Implications for policy, delivery or practice

Clinical audit is viewed as one method that can provide improvement in the quality of patient care. This study has demonstrated that the audit process is an effective tool in improving the management of type 2 diabetes. The majority of GPs participating in the audit also rated the monitoring process as very useful/useful.

What database should GPs use for asthma medication usage audit?

G Gill, GP North (the Division of General

Practice-Northern Tasmania), K Loone

Aim

To design a model that would allow GPs to determine their patients usage of Asthma medication especially the preventer/reliever

DDD ratio

Method

Over a six-month period in 1998 the numbers of asthma medication DDDs in

Northern Tasmania, supplied by pharmaceutical wholesalers, dispensed by local public hospitals, recorded as being dispensed under Pharmaceutical Benefits

Scheme (PBS) arrangements, dispensed by a sample of community pharmacists or prescribed by a sample of GPs who used computer prescribing packages, were calculated. Comparison was made between each source of data for the asthma medication preventer/reliever DDD ratios.

Evaluation

Public hospitals supplied less than 1% of medication. PBS data was incomplete. The higher proportion of concession cardholders in rural locations prevented comparison using

PBS data with urban patients. PBS data also recorded specialist prescribing which differed markedly from GPs. Pharmacy data showed a marked variation in over the counter (OTC) purchased relievers. It was not possible to access data from some community pharmacy dispensing computer programs. Local GPs did not write the majority of medication scripts dispensed by some pharmacies. GP prescribing software databases were not helpful as GPs often prescribed the maximum quantities of repeats.

Conclusions

It is not possible to utilise PBS, GP computer software or local community pharmacy prescribing data to provide local GPs with accurate feedback on the dispensing of their prescribed asthma medication. The best data was from the regional pharmaceutical wholesaler.

Similar database issues will impact on the

Enhanced Divisional Quality Use of Medicines program.

Long-term effects of cholesterol lowering in 9014 patients with coronary heart disease and average cholesterol levels: the LIPID trial follow-up

P Glasziou, RJ Simes, A Kirby, S Mulray,

W Hague, D Hunt, H White, P Alyward,

A Keech, A Tonkin, for the LIPID Study

Group

Objectives

The Long-term Intervention with Pravastatin in Ischaemic Disease (LIPID) study showed that pravastatin therapy over 6 years reduced mortality and cardiovascular events in patients with previous acute coronary syndromes and average cholesterol levels.

We aimed to assess the longer-term effects of the initial 6 years of treatment with pravastatin on further cardiovascular events and mortality, overall and within important subgroups, over a total follow-up period of

8 years.

Methods

9014 patients with prior myocardial infarction

(MI) or unstable angina and a baseline plasma cholesterol 4.0–7.0 mmol/L were randomised to pravastatin 40 mg daily or placebo and followed up for a mean of

6.0 years. All patients were then offered open-label pravastatin for 2 more years.

Major cardiovascular events and adverse events over 8 years were compared according to the initial treatment assignment.

2002 General Practice and Primary Health Care Research Conference — Proceedings

149

Principal findings

7680 (97% of those still alive) had 2 years of extended follow-up; 86% of those assigned placebo and 88% assigned pravastatin agreed to take open-label medication. During this extension period, patients originally assigned pravastatin had similar cholesterol levels to those assigned placebo, but a reduced risk of death from all causes (5.6% v

6.8%, P=0.029); coronary heart disease

(CHD) death (2.8% vs 3.6%, P=0.026) and

CHD death or nonfatal MI (4.5% vs 5.2%,

P=0.08).

Over the 8 years, the absolute benefits of treatment increased. Total mortality was reduced from 19.7% to 15.9 (P<0.0001);

NNT = 26. Stroke was reduced from 6.0% to

5.0% (P=0.015); NNT = 100. Pravastatin had no significant adverse effects.

The cost-effectiveness in the initial 6-year study was calculated to be $7842 per life year saved, but the extension results suggest this is an underestimate.

Discussion

The stronger evidence of sustained treatment benefits and safety of long-term pravastatin treatment reinforces the importance of longterm cholesterol-lowering treatment for secondary prevention in almost all patients with prior CHD events.

Implications for practice

Cholesterol-lowering treatment of patients with established coronary heart disease group and ‘normal’ cholesterol appears both clinically important and cost-effective.

Delayed prescriptions: evolution of an innovation

F Goodyear-Smith, Division of General

Practice and Primary Health Care, University of Auckland, B Arroll, D Thomas, N Kerse

Objectives

To better understand general practitioners’

(GPs’) use of delayed prescriptions to reduce unnecessary antibiotic use in treating upper respiratory tract infections and reasons why their use of this innovation might change over time.

Methods

Qualitative study conducted in general practice in Auckland, New Zealand involving thirteen GPs recruited from a previous crosssectional study looking at reported use of antibiotics. GPs were either high or low-users of delayed prescribing. They were recruited by a GP researcher and paid for their participation. The semi-structured telephone interview progressed from open to more focused questions with both hand-written and audiotaped recording. Interviewing ceased at data saturation. Transcripts were collated and analysed for emerging categories. Themes were double-coded as a consistency check with discrepancies resolved by adjudication. Main outcome measures were GPs’ experience of delayed antibiotic prescription use; their reasons for using or not using delayed prescriptions; to which patients they would give them, and the specific instructions they give to patients.

Principal findings

Only one invited GP declined to participate.

GPs’ rationale for delayed prescription use was that limiting antibiotics would reduce appearance of resistant strains of bacteria and patient side-effects. Factors in its favour included empowering patients to be more involved in decision making, giving them reassurance, meeting their expectations for antibiotics and saving them time and money,

Negative implications included perception of doctor incompetence and loss of management control. All GPs interviewed considered they had spontaneously developed the strategy to reduce unnecessary antibiotic use. Increase use of the innovation occurred in response to knowledge of its use by other GPs; reference to it in the literature and it becoming a legitimate topic of research. Utilisation subsequently diminished over time as patients became educated not to expect antibiotics for viral illnesses. There was no consistency in the use of this innovation with respect to which patients, which conditions and which instructions were considered appropriate.

Discussion

Delayed prescriptions are seen by some GPs as a mechanism for reducing antibiotic intake while preserving the doctor–patient relationship, although not all view it as a safe tactic to use. Unlike interventions such as new drugs, delayed prescribing has been generated spontaneously and independently by GPs and hence consensus in its use was lacking.

Implications for policy, delivery or practice

This study demonstrated rigorous use of qualitative methodology relevant to GPs in everyday practice (by GPs for GPs).

Successful GP recruitment could be attributed to invitation to participate made by a GP researcher; telephone interview at GP’s

150

2002 General Practice and Primary Health Care Research Conference — Proceedings

convenience with re-scheduling if required, and honorarium valuing their participation.

Use of qualitative methodology differentiated previously high-users who now limited use because the intervention had successfully their educated patients from those who had never used the innovation extensively. In a cross-sectional study both groups would be classified as ‘low-users’.

Delayed prescriptions are effective in reducing antibiotic consumption for conditions not clinically warranting antibiotics, although not without risk. The development of more formalised recommendations for use of this innovation is warranted, with appreciation that need for this strategy may decline over time.

First-hand experience establishing a research practice: key issues and a preliminary framework

E Hansen, Discipline of General Practice,

University of Tasmania, P Mudge

Objectives

As a preliminary step in developing

Tasmania’s first primary health care research network we are involved in establishing the first research practice in the state. This presentation discusses issues encountered during this process and presents a framework for establishing a research practice based on our ‘first-hand’ experiences.

Methods

The Discipline of General Practice, University of Tasmania has recruited two general practice clinics where several of the general practitioners have expressed the desire to conduct research. One of these practices is now involved in a collaborative research project with the Discipline of General Practice and the School of Medicine. While designing this project we have become aware of a number of key factors that need to be considered when establishing a research project in a general practice.

Principal findings

In terms of staffing, work routines, doctor/patient relationships and levels of research experience available, general practice differs from the hospital/laboratory location of the majority of clinical research.

This has practical implications for the design and implementation of research projects that go beyond the usual issues of research methodology. These include:

X the degree to which staff at the practice will be involved in research

X the role of patients/consumers in the research

X ethical issues associated with the nature of general practice

X the role of general practitioners in collaborative research teams

X establishing communication structures between researchers and stakeholders

X the relationship between research networks and research practices.

Discussion

Evidence from the United Kingdom demonstrates that research practices are an integral strategy for building the capacity of the primary health care research workforce.

Through close involvement in research, general practitioners gain research skills and experience. By instigating and conducting research into topics of relevance to their clinical work they also gain a sense of ownership that improves research uptake and enhances dissemination.

Despite frequent mention of the desirability of establishing research practices in the

PHCRED literature, little has been written about how to do this. On the basis of our experience establishing a primary health care research network and a research practice we have developed a preliminary framework to serve as a useful guide when establishing future research practices and/or designing/implementing a research project in a general practice setting.

Implications for policy, delivery or practice

The topics discussed in this presentation and the framework developed have considerable utility for other groups interested in conducting research in a general practice setting. In particular, those groups working in locations where such research has not previously been conducted. Awareness of these issues at the preliminary stage of establishing a research practice will assist all research teams involved in building capacity

‘from scratch’ and will also contribute to the long-term sustainability of the research practice being established.

2002 General Practice and Primary Health Care Research Conference — Proceedings

151

Assessing attitudes and improving health outcomes of asthmatics: using the Asthma Navigator in doctor-patient partnerships

RE Hicks, Psychology Department, Bond

University, G Harris

This paper describes the ability of a simple questionnaire to identify attitudes held by patients with asthma and it relates these identified attitudes to different ‘risk’ levels in terms of potential health outcomes for the patient. It also shows how over a period of more than nine months patient compliance and healthy lifestyle were improved.

Ten attitudes are identified through the

Asthma Navigator TM, a questionnaire with sound psychometric qualities developed in

Australia, and overall patterns and combinations of results indicate those patients most likely to be ‘at risk’. The paper discusses these attitudes and then indicates the results of one major study using the

Asthma Navigator. When the questionnaire was used in a partnership between some 130 doctors (GPs) and their patients, there was a marked increase at the end of the study period in both the attitudes themselves and associated knowledge shown, AND in several direct health indicators. These health indicators included lower rates of hospitalisation, better sleeping patterns, and better compliance behaviour (in terms of appropriate medication adherence).

The team of researchers included specialist medical, psychology and other health professionals.

Is it cost-effective to prompt parents to bring their children with asthma to the general practitioner for proactive review under the

Asthma 3+ Visit Plan?

C Holton, Department of General Practice,

University of Adelaide, J Beilby, N Glasgow,

R Yates

Objectives

The aim of this project was to perform an economic evaluation of the implementation of the Asthma 3+ Visit Plan with primary school children in the ACT, in parallel with a randomised controlled trial.

Methods

The trial involved 24 GPs and 174 children aged 5–11 years with moderate to severe asthma, cluster-randomised into two groups

152

2002 General Practice and Primary Health Care Research Conference — Proceedings by medical practice. The control group received usual asthma care, while the intervention group were reminded to attend by the researchers prompting practices to recall patients according to the recommended schedule.

The evaluation focused on the marginal costs of promoting the Asthma 3+ Visit Plan to GPs and patients, and compared the costs with the economic consequences of the plan on attendance rates, use of other health services and use of medication. The difference in costs of asthma care between the two groups was analysed over a twelvemonth period. A societal perspective was taken, with both direct and indirect costs being considered as they impacted on general practitioners, other health providers, patients and their families.

Principal findings

At the systems level, the costs of promoting and implementing the plan averaged approximately $400 per GP, with the prompts to practices costing $100 per GP.

The costs to the practices to recall the patients for proactive visits were minimal, averaging approximately $8 per patient.

Analysis and comparison of parent costs associated with the patients’ asthma care will be presented in detail at the conference.

While there was a statistically significant increase in the number of asthma-related GP visits by those in the intervention group, there was no significant difference in the total number of visits, nor in the associated travel and time costs for parents of the children. The number and cost of referrals to specialists and allied health professionals were lower for the intervention group, while the use and cost of diagnostic tests was higher, but neither were statistically significant. The proportion of patients incurring any medication costs was lower in the intervention group, with a significantly reduced cost per patient. A smaller proportion of children in the intervention group reported losing days from school as a result of their asthma, and incurred a lower average number of days lost — these differences were statistically significant.

Discussion and implications for policy, delivery or practice

As asthma has been recognised as a health priority, it is imperative that research is undertaken into not only the effectiveness of new models of delivery of care at the primary health care level, but also whether their promotion is cost-effective. With the recent introduction of financial incentives being

offered to GPs as part of governmental policy initiatives around asthma, it has been an opportune time to begin this investigation.

This study suggests that the asthma 3+ visit plan is cost-effective when implemented with children with asthma.

Access to, and use of, the Royal

Flying Doctor Service medical service in far western NSW

T Horsburgh, Department of Rural Health,

Broken Hill, University of Sydney,

G Oldman, A Wakatama, DM Lyle

Objectives

To determine community access to the Royal

Flying Doctor medical services in Far Western

NSW. To compare the patients who frequently attend the medical clinics or are evacuated by the RFDS with patients who attend less frequently.

Method

We used the Royal Flying Doctor Service

Patient Management Information System

(PMIS) to identify residents from remote communities in far western NSW who accessed services from 1 July 2000 to

June 2001. The PMIS records demographics for each person registered and clinical data from each consultation with the service. Frequent use of medical services was defined as greater than 12 clinic consultations, or three or more evacuations during the study period. The principal health problem or concern (diagnosis or issue) for the encounter was coded using ICPC+ for clinic consultations and using ICD9 for evacuations. National benchmarks for access to general practice services were derived from Australian Medicare data reported by

General Practice Branch of the

Commonwealth Department of Health and

Ageing. Analysis of patient demographics and patterns of use was conducted using

Microsoft Access and Microsoft Excel.

Principle findings

2,788 people (75% of the estimated resident population), used the RFDS medical clinic or evacuation service over a 12-month period.

These patients had fewer ‘GP’ clinic encounters with the RFDS on average than

Australian GP patients — 3.8 and 6.5

encounters per annum respectively. 240 people required medical evacuation during

2000–01, of whom 19 patients (8%) had 3 or more evacuations. There were also 113 frequent users of RFDS clinic services (3.5% of the clinic population), six of whom used evacuation services frequently as well.

Discussion

This paper presents work in progress. Further analysis of frequent users of the service will focus on describing the specific circumstances of those clinical interactions and comparing the demographic and principal reason for the encounter for these patients with other patients who use the services less frequently. It is anticipated that the results of this study will be used to develop possible quality indicators or audit flags for RFDS clinicians to use to monitor and improve the quality of care for patients who access services frequently.

Implications for policy, delivery or practice

People living in remote communities generally have reduced use of, and access to, health care. The role of general practice in coordinating care for patients with chronic and complex medical and health conditions takes on added importance where services and patient access is limited but their needs are not. The development of practical and manageable clinical audit and review systems should assist organisations such as the RFDS working in remote communities around

Australia to achieve better outcomes for their patients.

Developing an on-line suite of databases

B Jacobs, Primary Health Care Research and

Information Service, Department of General

Practice, Flinders University, V Sellick,

A Dunn

Objectives

The Primary Health Care Research and

Information Service (PHC RIS) has provided databases of Divisions’ activities, General

Practice Evaluation Program (GPEP) funded research and projects funded under Divisions

Projects grants for some time. This Suite of

Databases was previously distributed on CD-

ROM to relevant stakeholders but, due to protracted production times, the information was often out-of-date and did not provide optimum value to researchers and other relevant stakeholders. During 2001 and early

2002, PHC RIS undertook the development of an on-line version of the Suite of Databases, with the primary aims of making the information within the databases readily accessible to a broader range of stakeholders and decreasing production time to make the information more timely and relevant.

2002 General Practice and Primary Health Care Research Conference — Proceedings

153

Methods

The structure of information held within the

Suite of Databases meant that the existing database technology was inappropriate to support an on-line version of the Suite of

Databases. Java and Delphi programming languages were used to develop software programs to complement an XML-Database supporting context-sensitive full-text searching. In-house tools were developed to facilitate the conversion of the existing Suite of Databases and data supplied by Divisions via the OBF Template to the new database format. These tools allow incremental addition of information to the database as it becomes available, rather than ‘batching’ updates for periodic release on CD-ROM.

Principal findings

A combination of cutting-edge technologies has allowed the implementation of contextsensitive full-text searching — subsequently decreasing the overall search time, and increasing the relevance of search results.

These technologies are able to significantly decrease the time required to search the database — a fundamental requirement for an on-line multi-user delivery mechanism.

The new database is easier to maintain than the existing CD-ROM based products, and the time to incorporate new data is significantly reduced.

Discussion

With the wide acceptance of the Internet as a powerful research tool, the World Wide Web

(WWW) provides a powerful delivery mechanism for timely dissemination of information within the database, while simultaneously making the database accessible to a broader range of stakeholders. Preliminary evaluation shows that a broad range of users are increasingly using the database for investigating Divisions activities and GPEP-funded research.

Implications for policy, delivery or practice

The PHC RIS on-line Suite of Databases provides a valuable source of information for those interested in Division activities or

GPEP-funded research. The development of the on-line Suite of Databases now makes this information available to a much broader group of users. The increased timeliness and relevance of the information is more valuable to direct the activities of Divisions, policy makers, consumers and researchers.

Projection of future demand for primary medical care: influence of key factors

C Joyce, Department of Epidemiology and

Preventive Medicine, Monash University,

J McNeil, S Wimalaratne

Objective

To estimate future demand for primary medical care in Australia through quantification of key factors influencing demand

Methods

Three major factors influencing demand for primary medical care services are considered. These are: overall growth in population, ageing of the population, and trends in utilisation of services. Data from a range of sources including general practitioner (GP) Medicare claims data and

Australian Bureau of Statistics population data are analysed for trends over time to quantify the influence of these factors.

Parameters for projection of demand are thus defined, and level of future demand under a range of scenarios estimated. Workforce requirements associated with projected levels of demand can then also be estimated.

Principal findings

Trends in future demand for primary medical services under a range of scenarios are estimated. Key assumptions (and hence sensitivities) of the projections are outlined.

It is planned to use the results to define parameters for a new health workforce planning model that incorporates estimation of both supply-side and demand-side factors.

Discussion

The findings contribute to the development of more sophisticated approaches to workforce planning. Previous approaches in Australia have concentrated more on projection of supply rather than a combined approach estimating both supply and demand.

Development of methods for prediction of both the supply and demand sides offers the greatest potential for effective and accurate projection.

The study raises issues in relation to a range of questions including defining benchmarks for quality of care and access, and the influence on demand of other factors, such as community expectations, requirement for practitioners to undertake non-clinical activities, and use of non-medical primary health providers.

154

2002 General Practice and Primary Health Care Research Conference — Proceedings

Implications

In the context of rising health expenditure, long-standing geographic imbalances in the

GP workforce, and even a lack of consensus about the adequacy of the GP workforce overall, health workforce planning has never been more important. This study can contribute directly to this key policy area.

Government policy and programs in related areas such as workforce support and incentives for recruitment and (re)training must also be informed by sound information on expected levels of demand for primary medical services.

Relevant, available, timely data: the result of collaborative revision of the

Annual Survey of Divisions

L Kalucy, Primary Health Care and Research

Information Service, C Modra, R Forbes,

A Dunn

Research data are useful in policy making if relevant to policy interests, and available when needed. The Annual Survey of Divisions conducted by PHC RIS for the GP Branch potentially can provide useful data to monitor a wide range of activities of Divisions of

General Practice. In 2001 the ASD was revised with input from many people, to make the content more relevant to potential users, and the timing more appropriate for policy decisions.

Objectives

To increase the usefulness of the Annual

Survey of Divisions to Division staff, policy makers, and researchers by using a collaborative process to update its form, content and timing.

Methods

X

Divisions of General Practice, researchers and policy makers were consulted about the form and content of the survey.

X

A working group convened by PHC RIS considered the findings and recommended revision of the instrument to make it more current, and meet the needs of a wider range of potential users.

The Working group included representatives from GP Branch of

Commonwealth Department of Health and Ageing, Division of GP, the Divisions

Standing Committee of GPPAC (a consumer representative), State-based

Organisation, University and PHC RIS.

X

All sections of the GP Branch, Divisions

Standing Committee of GPPAC, ADGP and several university departments of general practice were consulted about possible items.

X

The working group developed and refined the survey instrument via teleconference and email links.

X

After piloting, the new instrument was administered as a Word form for electronic completion in November 2001.

The results will be available in report form in June 2002.

Principal findings

Survey items e.g. governance, aged care, consumer participation, and practice nurses were included in the survey as a result of the collaborative process.

Survey items were designed to meet specific requirements of different users.

The survey was well received by Divisions of

General Practice.

PHC RIS adapted the research timeline to meet the needs of stakeholders by providing subsets of the data on request.

Discussion

Members of the working group developed considerable understanding of each other’s interests and perspective. The collaborative approach raised awareness about the Annual

Survey and its potential among those consulted, increased their sense of ownership of the survey, and has led to further collaboration. Stakeholders want to obtain survey data while they are current, and analyse the data according to their own requirements.

Implications for policy, delivery or practice

Data from the revised Annual Survey of

Divisions are more likely to inform policy for the following reasons:

X items are more relevant to stakeholders because of the collaborative approach used in revising this survey in 2001;

X

PHC RIS has provided data subsets on request in an accessible and timely manner, to fit in with policy and planning timelines;

X increased awareness of the survey content and availability will further increase its usefulness.

2002 General Practice and Primary Health Care Research Conference — Proceedings

155

Smoking advice in general practice

— is the most appropriate advice being delivered?

F Lewi, Swinburne University of Technology,

N Crafti, L Turney

Objectives

The transtheoretical model of behaviour change has been well researched with smoking cessation advice. The importance of understanding and applying this theory when advising people to quit smoking is well documented. To date, little research has been done with General Practitioners (GPs) and their knowledge and use of this theory.

This project aims to begin to fill the gap in such research by surveying general practitioners in the Melbourne metropolitan area with respect to their current knowledge and usage of smoking cessation advice.

Methods

GPs belonging to the North West Melbourne

Division of General Practice (NWMDGP) were recruited by invitation. GPs were advised of the research by notification in the Friday Fax sent out by NWMDGP each Friday. This was followed up with a mail-out to each of the

410 GPs on the NWMDGP mailing list.

GPs were asked to complete a questionnaire including four case studies and return it in a reply paid envelope supplied. Eighty-five GPs returned questionnaires, a response rate of

21%.

Principal findings

Participants were 58% male, ranging in age from 24 to 76 years. GPs reported seeing a number of patients per week ranging from 10 to 300. The percentage of patients who smoked reportedly ranged from 1 to 60%.

Almost 50% of GPs reported asking patients about their smoking history only when it was relevant to their current medical complaint.

Almost 50% of GPs considered the patients’ presenting medical complaint to be the most important issue to be considered when considering advising about smoking. This compared with only 20% of GPs who considered the patients’ stage of behaviour change to be the most important issue when considering advising about smoking.

With each of the four case studies presented,

GPs selected the most appropriate stage of change advice in only 54% for case 1, 15% for case 2, 50% for case 3, and 32% for case 4.

Discussion

Given that, for the majority of Australians, general practice is the first point of contact with the health care system, GPs are critical to population health and its effective promotion. Clinical guidelines to assist GPs with health promotion and preventive medicine activities such as smoking cessation recommend incorporation of the transtheoretical model of behaviour change.

Until now little research has been conducted to ascertain how GPs are fairing with implementing these guidelines. The current research provides initial evidence that GPs’ knowledge about this model is limited, and that current practice in smoking cessation advice involves inappropriate matching with patients’ stage of change.

Implications for policy, delivery or practice

Inappropriate matching of smoking cessation advice with patients’ stage of change obviously limits the success of such a health promotion activity. The current research suggests that education of GPs with respect to the transtheoretical model of behaviour change may be appropriate to improve the practice of smoking cessation advice. Further research is also indicated for other health promotion activities in general practice.

Not another health information website!

V Sellick, Primary Health Care Research and

Information Service, Department of General

Practice, Flinders University, B Beacham,

B Lowcay, L Kalucy

To conceive of knowledge as a collection of information seems to rob the concept of all of its life...

Knowledge resides in the user and not in the collection. It is how the user reacts to a collection of information that matters.

C.W. The Design of

Inquiring Systems, Basic Books, New

York, NY, 1971, p. 10)

The Primary Health Care Research and

Information Service (PHC RIS) is funded by the Commonwealth Department of Health and Ageing to provide easy access to up-todate information about research, evaluation and development in general practice and primary health care to practitioners, researchers, policy makers, consumers,

Divisions of General Practice and Primary

Health Care Research, Evaluation and

Development (PHC RED).

156

2002 General Practice and Primary Health Care Research Conference — Proceedings

Objectives

This poster aims to raise awareness about how our website supports the management of knowledge in primary health care settings.

The PHC RIS website was first launched in

1996. When the website was redeveloped in

2001, the aim was not to maintain a website for technology’s sake, but to use the website as a tool to link, inform and foster the exchange of information between our stakeholder knowledge communities.

Methods

The redevelopment of the PHC RIS website was informed by a combination of approaches including: a formal website review, the PHC RIS Needs Assessment

Review, additional feedback (particularly during the redevelopment stage) and a literature search about website evaluation and knowledge management.

Principle findings

A great deal of information is generated by the research and development activities in general practice, primary health care and

Divisions of General Practice. PHC RIS collects much of this data, but simply providing this information on-line is not knowledge/information dissemination. The role of organising this information into a useful format is extremely important. An equally important finding is the greater appreciation of the specific needs of our various stakeholders and their preferred mode of communication.

So why use the web? Identified (website review, Needs Assessment) as a stakeholder preference, ease of accessibility and geographic spread (PHC RIS based in

Adelaide but are national organisation with stakeholders in rural/urban areas).

Discussion

Our core function is communication and dissemination. Without effective infrastructure to manage this, it will not occur. Our website is more than an on-line access point for information, but a tool to manage knowledge, because the needs, requirements and feedback from our stakeholders determines the content and format of our website. Fostering partnerships with our stakeholders and the data we collect from them ensures quality, accuracy, currency and continued use of the website both now and in the future.

Implications for policy, delivery or practice

X

Ability to target information to specific stakeholder groups.

X

Ultimately supports informed access

(portal) and use of information that is available.

Trends in blood lead levels among young Broken Hill children 1991 to

1999

D Lyle, Department of Rural Health, Broken

Hill, University of Sydney, A Phillips,

W Balding, H Burke

Objectives

To investigate trends in blood lead concentrations in Broken Hill preschool children between 1991 and 1999, as part of the evaluation strategy of the public health lead management program.

Methods

Since 1991, all Broken Hill children aged 1–4 years have been offered at least annual blood lead screening as part of a community-wide lead management program. Recruitment of children was promoted throughout the period using local media and distribution of promotional material from health care centres and preschool, childcare and educational facilities around the city. Venous blood samples were collected using standard procedures and analyses were subjected to internal and external quality control programs. Because the frequency distribution of blood lead levels were skewed, geometric rather than arithmetic means were used for comparative purposes. Trend analysis was based on age and sex standardised mean blood lead levels.

Principal findings

The number of 1–4 year old children screened ranged between 538 and 948 in any one year and response rates varied between 41% and 75%. The mean blood lead level decreased from 16.7 µg/dL to 8.4 µg/dL between 1991 and 1999. The rate of change decreased over time with several annual reductions of above 1 µg/dL noted prior to

1996–97, and levels consistently below that since. Overall, blood lead levels halved over the nine years. These reductions were consistently observed irrespective of age or where a child lived in the town.

Discussion

This evaluation indicates that real progress has been made in dealing with the lead problem in Broken Hill, although the rate of decline has slowed and work still needs to be done to bring the number of young children

(14%) with significantly elevated blood lead

2002 General Practice and Primary Health Care Research Conference — Proceedings

157

levels (>15 µg/dL) down to the NHMRC target of 5%.

Implications for policy, delivery or practice

Continued public health action is required to achieve the NHMRC community-based target for lead in young Australians. With a slowing of the rate of decline of lead levels now evident, further development and evaluation of the program are required if this target is to be reached.

A pilot project to monitor drug product promotion by the pharmaceutical industry to general practitioners

JM Mackson, National Prescribing Service,

Sydney

Medicine use is an important component of health care and the quality of use of medicines must be optimised to optimise health outcomes. The National Prescribing

Service (NPS) provides a range of educational interventions on prescribing to general practitioners (GPs), medical specialists and pharmacists, targeting selected therapeutic areas.

To establish whether NPS activity and interventions results in changes in drug utilisation (or lack of effect) it important that the environment in which drug use is occurring and any other factors which may affect this are understood. Monitoring of pharmaceutical industry promotional messages to GPs is proposed as one of the elements of this ongoing environmental monitoring.

Objectives

The aim of the study was to pilot a monitoring program to assess the extent and content of drug product promotion to general practitioners by the pharmaceutical industry.

The objectives were to develop and pilot the methodology for data collection and recruitment of GPs; to test the feasibility for and acceptability to GPs of the data collection; to pilot test the instrument for reporting representatives’ visits; to test the feasibility of data collection and to explore the utility of reporting of key messages from representatives’ visits.

Methods

A convenience sample of GPs was recruited by field workers in three Divisions of General

Practice in metropolitan and rural areas. An instrument for recording representatives’ visits was developed. GPs collected all

158

printed materials received from pharmaceutical companies for four weeks. A structured exit interview was conducted to determine acceptability.

Principle findings

Ten GPs in three Divisions of General Practice collected printed and other materials and reported on representatives’ visits over a four-week period. A mean of 1584g per GP of written material was collected. There were 63 visits from representatives recorded for 111 products, ranging in duration from 1–30 minutes. GPs recorded key messages for 105 products most of which provided therapeutic messages that could be further analysed. Of these 49 (47%) related to therapeutic areas and/or drugs included in NPS programs.

On interview all GPs estimated their collection of materials and recording of representatives’ visits to be 90% or 100% complete. Nine would be prepared to participate in a further study of similar duration.

Modifications should be made to the visit recording instrument to improve clarity and utility, mailed and hand-delivered materials separated and a longer data collection period used. Content analysis should be undertaken to establish the usefulness of the printed material collected.

Discussion

Recruitment for a larger study is feasible.

Meaningful qualitative data can be systematically collected by GPs where there is good communication from investigators and financial incentives are offered.

Implications

It is feasible to use the methods described to collect information on the extent, content and specific promotional messages from the pharmaceutical industry. A continuous crosssectional survey or an intermittent monitoring system may both be suitable for ongoing data collection in general practice.

2002 General Practice and Primary Health Care Research Conference — Proceedings

General practitioner and patient perspectives on depressive symptoms after acute cardiac events: informing the application of evidence in community practice

J Marker, Health Outcomes Unit, SA

Department of Human Services, F Cheok,

G Schrader, D Banham, L Rayson,

G Greenwood

Objective

To investigate the experiences of patients screening with depressive symptoms while hospitalised for a cardiac event and their general practitioners.

Method

Patients: From 1400 patients participating in the Identifying Depression As a Comorbid

Condition (IDACC) trial across four public hospitals in Adelaide, 27 ‘depressed’ patients were randomly selected for in-depth interview in their own home. Nine intervention patients (where GPs took part in a telephone case conference with a hospitalbased psychiatric liaison registrar and a cardiac rehabilitation nurse), nine alternative intervention patients (GPs took a 1:1 call with a psychiatric consultant [GS]) and nine

‘control’ patients (who received ‘usual care’) were interviewed. All patients had been identified as having ‘depressive symptoms’ on the Hospital Anxiety and Depression Scale

(HADS) and the Centre for Epidemiological

Studies Depression Scale (CESD).

General practitioners: A random selection of eighteen general practitioners who took part in the intervention program were interviewed at their surgery.

Interviews were taped verbatim for transcription and analysis. Questions were focussed but open ended.

Principal findings

Data was collated in the context of particular everyday experiences, relationships and the impact of a cardiac event. The analysis was presented in the form of recurrent themes and patterns. There was no difference between the themes from the different groups of patients.

For patients, five emergent themes included: the presence of stressors prior to cardiac event; fear; descriptions of feelings/emotional response; discussion of emotions with GP; lifestyle change following cardiac event.

Four themes appeared common with GPs, regardless of whether they participated in the case-conference or received only a one-toone discussion with a psychiatrist. These were: personal contact with other health professionals was welcomed; patients often deny depressive problems; depression presented as physical symptoms; and, a waiting and watching approach was generally adopted. Many patients and GPs thought depression after a cardiac event was a normal response to a serious illness.

Discussion

When patients return to the community, unforeseen dilemmas arise that are interdependent with patient social beliefs and health activities. These impact on the type and quality of care general practitioners may be able to offer patients and serve to highlight barriers to be overcome if evidencebased treatments for depression are to be translated into community-based practice.

This valuable qualitative feedback from GPs and patients informed the development of a quantitative survey for GPs and further qualitative focus groups for GPs, patients and hospital-based professionals involved in

IDACC. This synthesis of qualitative and quantitative methods involving a range of stakeholder perspectives in the delivery and receipt of health services will provide important information on the uptake, acceptability and effectiveness of case conferencing utilising the new Medical

Benefits Scheme (MBS) items.

Implications for policy, delivery or practice

This qualitative component of the IDACC project identified themes relevant to the interactions of consumers and practitioners as recipients/providers. The qualitative methodology presents useful information in itself. In combination with quantitative data it helped identify barriers and facilitators to evidence-based continuity of care out of the hospital and into the community.

Transdisciplinary synthesis through an Indigenous Knowledge Model in health care research and education

J Smylie, C Martin, Department of Family

Medicine, University of Ottawa

Objectives

An exploration of the contribution of transdisciplinarity in research and education approaches to the North American

Indigenous Knowledge framework. In order to integrate different approaches, particularly

2002 General Practice and Primary Health Care Research Conference — Proceedings

159

Western scientific and social science research models within an Indigenous Knowledge

Model, there is a need for synthesis of disciplines and research methodologies. The practice of imposing Western definitions on traditional ecological knowledge makes

Indigenous peoples vulnerable to Western definitions, decisions and control.

Transdisciplinarity refers to a blurring of the boundaries of types of knowledge creation in order to integrate and is achieved through a range of mechanisms including: a social constructionist approach; the development of new language and common theories; the development of greater understanding of the processes of knowledge creation; the acceptance of new types of methodological rigour and a commitment to knowledge linkages.

Methods

An analysis of knowledge construction in

Indigenous and Western systems.

Principal findings

Indigenous knowledge systems are transdisciplinary. Rather than breaking knowledge up into disciplines, Indigenous systems instead focus on the relationships between domains.

Both Indigenous and Western systems organise data to condense facts or beliefs into recognisable forms. Dion Stout’s concept is that the main difference is that Indigenous

‘myths’ do this through story telling, whereas

Western science do not.

2

However, science can be treated as a narrative without compromising its seriousness and truth. In this way, the universe is made up of stories, not atoms. Transdisciplinary and transcultural approaches with different disciplines working together offer mechanisms to develop new solutions to complex problems, synthesising different types of knowledge creation.

Discussion

It can be argued that Indigenous knowledge is dependent on the well-being of Indigenous communities and the environments in which they live and vice-versa. Indigenous peoples are highly aware of their unique knowledge base as well as specific threats the latter faces. Despite the fact that Western science is beginning to welcome non-Western

2 Dion Stout M., Kipling GD. Establishing a Leading

Knowledge-Based Organization. Discussion paper prepared for the National Aboriginal Health

Organization, 2001. http://www.naho.ca

traditions, Western science still determines the threshold of this acceptance.

Implication for policy, delivery or practice

Research and education programs must be constructed such that Aboriginal culture, professional clinical knowledge, and individual research disciplines are respected.

Transdisciplinary synthesis of both Western and Indigenous narratives about health must be achieved through an Indigenous knowledge model in order for it to be relevant for Aboriginal people. Such transdisciplinarity includes problem and question conceptualisation in the community; establishing a research team to implement appropriate methodologies that reflect different ways of knowing; analysis; synthesis; and appropriate dissemination of research findings as well as publication in scientific journals.

Home-Based Intermediate Care — an acute hospital substitution program with family physicians, nurse practitioners and telehomecare

C Martin, W Hogg, J Lemelin, Department of Family Medicine, University of Ottawa

Home-Based Intermediate Care (HBIC) — an acute hospital substitution program to be delivered through Family Medicine — is an admission avoidance and facilitated early discharge program for people with common conditions that have been shown to be safely and effectively managed in the home. Acute hospital substitution is the delivery of care and services which, without the program, can only be provided by admission to hospital.

Victorian Hospital in the Home program provides a leading international example located in a hospital institution base. Other programs have been based in community care institutions.

Internationally, an ageing population with chronic conditions is placing increasing strain on the acute care sector. An intermediate care sector based in the home and community linking primary, secondary and community care provides opportunities to improve health outcomes and patient and carer satisfaction, and reduce unnecessary pressure on the acute care sector.

Objectives

A demonstration project and a pilot study to test the feasibility of the HBIC approach (in the Ottawa context) and the telehomecare technology operated by nurse practitioners

160

2002 General Practice and Primary Health Care Research Conference — Proceedings

associated with family medicine in order to lay the groundwork for future research and implementation.

HBIC aims to provide a model of managing patients at home of comparable quality with in-patient hospital care and other hospital substitution programs that is sustainable and likely to be sufficiently cost efficient in the long-term. The institutional base is a general practice/primary care — the Family Health

Network.

Methods

An literature review of international homebased hospital substitution programs as the basis for an evaluation of a demonstration project using case control methodology.

Principal findings

The outstanding question in the literature is not whether home-based acute hospital substitution can provide care that is of high or higher quality than hospital. Similar or improved quality has been consistently demonstrated through RCTs and metaanalysis. The key question is how to implement hospital substitution programs in the local setting that are efficient, sustainable and appropriately located in an institution.

Discussion

Assessing cost implications for the implementation of a particular institutional model and geographic setting, using metaanalysis and RCTs, is limited by problems of external generalisability across different hospitals and health systems and short term studies. There are also the issues of flow on and indirect costs and savings, and the medium and long-term impact. While clinical trials have demonstrated the effectiveness of

HBIC programs, there is need for broader based economic evaluations to inform the international literature. In addition there is a need for greater implementation research and comparability in program evaluations.

Implication for policy delivery or practice

Partnerships linking GPs and nurse practitioners using telehomecare technologies provide new policy directions to shape primary care and intermediate care.

Determining the quality of intermediate care programs such as HBIC is easier than determining their true costs. Thus in setting up new models while there is a need to demonstrate quality of care, the key questions are efficiency and institutional sustainability.

Evidence-based medicine and decision-support tools — what do

‘bag carrying’ GPs think?

B McAvoy, Universities of Melbourne,

Monash and Queensland, the Practical

Implementation Taskforce for the Prevention of Cardiovascular Disease

Objectives

To obtain the views of a sample of general practitioners across 5 states of Australia on:

X relevance of evidence-based medicine

X methods of accessing EBM guidelines

X use of decision-support materials

X a new decision-support tool for managing patients with cardiovascular disease.

Methods

A questionnaire survey of 335 general practitioners who attended evidence-based medicine seminars in 5 Australian states

(New South Wales, Queensland, South

Australia, Victoria and Western Australia).

Following 2 presentations (one from a specialist and one from a general practitioner) participants were invited to complete an 18 item questionnaire seeking their views on EBM in general and a on a new

EBM decision-support tool for managing patients with cardiovascular disease in particular.

Principal findings

Questionnaires were completed by 259 general practitioners giving a response rate of 77%. Compared to HIC figures for nonspecialist medical practitioners billing

Medicare in 1998–99, there was an overrepresentation of men, older general practitioners and solo and small practices.

Over two-thirds (66%) stated that they liked to be aware of EBM, then make up their minds on a case-by-case basis, with nearly

30% agreeing that EBM is the basis for ‘gold standard’ clinical practice. Over 80% admitted it was difficult to continuously be up-to-date with the latest EBM. Sources used to access the latest EBM included clinical journals (75%), pharmaceutical company representatives (58%), clinical meetings

(42%) and the Internet (22%). Various suggestions were made as to how to improve the decision-support tool, with nearly 50% requesting that the paper-based tool be linked in some way to a computerised format. Forty-eight per cent stated that they had previously used decision-support

2002 General Practice and Primary Health Care Research Conference — Proceedings

161

materials. Other clinical areas/disease suggested for a similar decision-support tool were diabetes (85%), osteoporosis (74%) and asthma (71%).

Mechanisms suggested for the dissemination of the decision-support tool included small workshops (51%), seminars similar to the one they had attended (42%) and through

Divisions of General Practice (26%).

Discussion

Although this was a self-selected sample of general practitioners it represents a significant number of doctors drawn from 5 different states around Australia.

Their views on EBM will be compared to those reported from other Australian and overseas studies. This initiative is an example of trying to develop a viable and sustainable ongoing partnership between researchers and stakeholders to enhance the dissemination of research findings, and also suggests how practitioners can be encouraged to make use of research findings.

Implications for policy, delivery or practice

While there is broad support for EBM among general practitioners, translating research findings into practice has proved to be difficult. Engaging general practitioners in the process of developing decision-support tools before they are disseminated should improve their acceptability and uptake. Further work is required to develop viable and sustainable ongoing partnerships between researchers and stakeholders to enhance the dissemination and uptake of research findings and to develop computerised formats of decision-support tools for general practitioners.

Sustainable mentoring partnerships with GP researchers

S Mickan, Centre for General Practice, The

University of Queensland, D Askew

Objectives

This paper describes the development and implementation of Primary Care Novice

Researcher Fellowships. These fellowships offered primary care clinicians protected time and mentoring support to develop a research idea into a PhD proposal or grant application.

Specifically, they are the equivalent payment of one day per week, for nine months, at a suitable academic level, e.g. Senior Lecturer.

These novice fellowships focus on enhancing skills of critical appraisal and research design. It is anticipated that the development of mentoring partnerships with active researchers will facilitate the fellows conducting research that is relevant to their practice.

Methods

Widespread advertising of these fellowships attracted 36 expressions of interest from

GPs, nutritionists, physiotherapists and occupational therapists, and 13 applications were submitted. Each application was content analysed against five selection criteria to present an overview of applicants’ expectations and aspirations towards research. These five selection criteria included:

X motivation to develop a career in primary care research,

X suitability for research training,

X demonstrable ability to fulfil the aim of the fellowship,

X academic record, and

X potential to inform clinical practice and contribute to the evidence base of primary care in Australia.

The four successful applicants were also asked about their specific intentions and attitudes to design and participate in research.

Principal findings

This paper presents detailed qualitative and quantitative summaries of applicants’ expectations and aspirations towards research, across all five selection criteria.

Applicants were highly motivated towards integrating research in their careers. Several clinicians were aiming towards academic posts and greater involvement in medical education. Applicants had a diversity of exposure to and involvement in research.

One applicant had completed a PhD, and six applicants had completed or were in the process of completing Masters degrees. All applicants were keen to improve their research knowledge and skills. Most wanted to conduct research in specific areas of interest, and they highlighted their desire to integrate research outcomes into their current clinical practice.

Discussion

Primary care clinicians appear keen to become involved with research, when the commonly acknowledged obstacles to research, of time, money and skills, are

162

2002 General Practice and Primary Health Care Research Conference — Proceedings

specifically addressed. Applicants’ abilities, aspirations and expectations are congruent with the infrastructure of researcher mentoring and support through the fellowships. Evaluation of all applicants will be undertaken at the end of 2002. This evaluation will describe the research outcomes of the four novice fellows and compare them with the 9 unfunded clinicians who were interested in research.

Implications for policy, delivery or practice

Primary Care Novice Researcher Fellowships are aimed to encourage primary care clinicians to design and conduct research in their areas of clinical interest. It is expected that novice fellows will be well positioned to integrate and disseminate their research results with current practice.

Building research teams

S Mickan, Centre for General Practice, The

University of Queensland

Objectives

Primary health care research requires collaboration between clinicians, academics and policy makers for widespread dissemination and uptake. This paper summarises recent research about health care teams to propose strategies for building research teams.

Methods

The way in which repertory grid interviews, a clarification questionnaire and a Teamwork in

Healthcare Inventory were used to develop an inductive model of effective health care teams will be summarised. An interpretive framework of four iterative stages will be briefly described, along with the integration of both qualitative and quantitative analyses.

Principal findings

Six characteristics that distinguished effective from ineffective health care teams will be described and applied to a multi-disciplinary research environment. Initially, research teams need to clarify their purpose, in accordance with team members’ organisational affiliation and funding parameters. Collaboratively, team members need to negotiate and define goals that align their purpose towards potential achievements. To ensure their effective functioning, research teams need to clarify their leadership and communication structures and processes in such a way that all members acknowledge and utilise them. It is also important for research team members to feel a sense of cohesion around a specific identity. Finally, at an individual level, team members need to develop a mutual respect for each other in terms of each individual’s personal and professional contributions to the team.

Discussion

Teamwork is often referred to as a common sense requirement of multi-disciplinary functioning. This paper presents research that has evaluated teamwork in detail in health care environments. It is important to disseminate these results across various health care environments.

Individual and team-based critical reflection of teams’ comparative strengths and weaknesses across the six key characteristics provides useful insights into ways of maximising team function.

Research environments can be particularly problematic for team functioning, because team members come from a diverse range of professional and organisational backgrounds, where communication and leadership practices often vary. While often unspoken, there is a need to identify and share similarities of purpose to promote effective team function that achieves the ultimate goal of successful research.

Implications for policy, delivery or practice

It is timely for managers and research team members to apply research findings to the structures and processes of their own teams.

As a consequence, they can enhance their ability to work together and achieve high quality research outcomes.

Dementia: reasons for admission to a care facility

F Millard, General Practitioner, Mackay,

Queensland

Background

As the population ages, there will be more people with dementia. As people grow older, they fear the loss of independence more than they fear death. People with dementia may lose their independence and rights as an individual, particularly when they are admitted to a care facility. This loss of status and the associated discomfort of a new environment cause stress for both the person with dementia and their loved ones. I am a general practitioner whose mother has dementia and is in a care facility. This study was conducted for my thesis for Master of

General Practice, Otago, NZ.

2002 General Practice and Primary Health Care Research Conference — Proceedings

163

Objective

To study a cohort of people with dementia,

12 in a care facility and 12 still at home, to identify factors affecting their ability to cope at home and reasons for admission to a care facility.

Method

The cases studied are patients of general practitioners attending a local care facility.

These are the most recently admitted patients to a rest home with a dedicated dementia facility, and patients with dementia who are still at home. Data is gathered from the patient, caregivers, family and general practitioner using taped interviews, MMSE and doctor’s notes.

Principal findings

Attitudes, belief systems and access to support are important issues in whether a patient with dementia stays in the community. There is a shortage of beds available for dementia patients in rest homes creating pressure on families to accept a suitable placement immediately one becomes available.

Discussion

It is possible for rest homes to offer facilities that are attractive to both the patient and their families.

Families can feel bereaved when a member suffering from dementia is removed from the family home. The separation can affect the health of both the family and the patient.

Education, planning and access to support help reduce stress and grieving.

Recommendations

The material in this study gives some insight into what is happening at present and how we can improve management of dementia in the future, with benefits across our ageing community.

Using a randomised controlled trial to assess the effectiveness of case conferences in the care of palliative care patients: methodological issues

G Mitchell, University of Queensland Centre for General Practice, C Del Mar,

Z Clavarino, T Richards

Objectives

To assess the effect on patient outcomes of case conferences between GPs and specialist

Palliative Care teams, on patients.

To assess the effect of case conferences between GPs and specialist Palliative Care teams on GP clinical behaviour in Palliative

Care.

To describe methodological problems and solutions in conducting a randomised controlled trial in a palliative care population.

Methods

All patients referred to two specialist

Palliative Care teams in Queensland were eligible for inclusion. Exclusion criteria were:

GP refused consent to approach patient; carer refused consent to approach patient: life expectancy less than four weeks; inability understand written or spoken English, altered mental state precluding informed consent.

Randomisation of consenting patients to either case conference or normal care at first team meeting after referral.

Interviews of patient and primary carer before case conference or first team meeting, then at 1 week, and fortnightly intervals till death. Primary carer interviewed 6 weeks after death. Validated Quality of life instruments used, plus semi structured interview at each interview stage. GPs completed questionnaire after death.

Medicare data and hospital chart data accessed. Analysis by intention to treat.

Primary outcome measures were: for patients, Global Quality of life score 1 week after case conference; for GPs, evidence of

GP contact with bereaved relative after patient death; and for carer felt, prepared for the death.

Principal findings

Data collection is ongoing until December

2002. Two hundred patients are required.

Randomisation rate for the study from July

2001 to December 2001 was 26% (50 of 190 referrals) Major difficulties relate to recruitment of patients. Different units see patients at different stages of the illness, with one unit seeing patients closer to death than the other. External factors have reduced referral numbers to one unit. Primary carers tend to protect their relatives from the potential stress of participation in the project.

A third Palliative Care unit is being included to boost data collection, and Ethics approval has to be sought for each site. There is variability in the requirements for each site.

No GPs have refused to participate in case conferences, but considerable administrative time is spent in organising them.

164

2002 General Practice and Primary Health Care Research Conference — Proceedings

Although some carers have been concerned at the potential stress of participation on patients, most patients are happy to take part.

Discussion

Case conferences between GPs and specialist units are being promoted as a means of improving patient outcomes. Their effectiveness in Palliative Care has not been established. This project reflects universal problems in conducting controlled trials in

Palliative Care patients. Secondary outcomes describing the process and qualitative dimensions of the effect on care may still yield useful information, should the required number of patients not be recruited.

Implications for policy, delivery and practice

Attempts to conduct controlled trials in this setting are warranted, even if difficult. Major expenditure by governments should be committed on the basis of evidence of effectiveness, or with a commitment to gather such evidence.

Using evidence to develop a continuing medical education program for a Division of General

Practice

G Mitchell, Centre for General Practice,

University of Queensland, C Del Mar,

H Eastwood, L O’Keefe, R Toalster

Objectives

X

To use best evidence to identify a CME program for the members of a Division of

General Practice.

X

To describe the differences between a

CME program generated by this method and the CME delivered by the Division.

Methods

Chart audit of 20 randomly selected patients seen in the previous month by of participating GPs drawn from the membership of one Division of General

Practice. Transcription and analysis of the notes of the last consultation. Recorded in each chart. Identification of the ten most common clinical reasons for encounter from the cohort. Assessment of clinical best practice recorded in literature for these presenting complaints, drawing evidence from systematic reviews and controlled clinical trials. Ranking of the conditions according to the greatest discrepancy between actual clinical practice recorded and best practice. Comparison between the derived CME program and the CME program offered by the division in 1998–2000

Principal findings

The study was conducted through the

Ipswich and West Moreton Division of

General Practice; Forty of 130 GPs in the

Division (31%) participated. Eight hundred consultations were recorded, comprising 993 reasons for encounter. The 10 most common reasons for encounter, ranked in order of greatest need for CME were: diabetes, hypertension, acute upper respiratory infection, earache/wax, mental health, red/sore eyes, urinary symptoms, headaches, low back pain and nausea/vomiting. The commonest CME activities offered by the division were:

Practice management, (17 events) cardiovascular (9), miscellaneous topics (4) and Mental Heath (3) and infectious diseases

(3).

Discussion

The effectiveness of a continuing medical education program (CME) is related to the extent to which individual GPs underperform in the discipline selected for study. Many general practitioners (GPs) elect to undertake

CME in areas in which they are comfortable or have a prior interest. It is possible to develop a CME program based on evidencebased best practice. The content of this program has major differences to a program dependent on either self-identified need or that delivered by the division.

Implications for policy, delivery, or practice

If CME is targeted to areas of lesser performance, it is likely to improve the quality of health care delivered to patients.

Using evidence to identify these CME subjects is a logical way of identifying the content of CME likely to be of most benefit to practitioners.

Individual medication effectiveness tests of stimulants in the treatment of attention deficit hyperactivity disorder

G Mitchell, Centre for General Practice,

School of Population Health, University of

Queensland, J Nikles, P Glasziou,

C Del Mar, P Schluter

Objectives

Attention deficit hyperactivity disorder

(ADHD) is the commonest behavioural disorder in children. The mainstay of treatment is stimulant therapy, which causes

2002 General Practice and Primary Health Care Research Conference — Proceedings

165

concern in parts of the general and medical communities. It also has a significant placebo response rate. Individual medication effectiveness tests (IMETs) are within-patient double blinded placebo controlled trials of a therapy. Our objectives were:

X

To use IMETs to determine the efficacy of stimulant therapy in patients with ADHD.

X

To assess IMET as a means of supporting treatment decisions by doctors and patients.

Method

Participants were school aged children with a clinical diagnosis of ADHD. Three cycles of randomly allocated 1 week placebo and stimulant treatment pairs. Evaluation of treatment with Connor’s rating scales, and purpose designed questionnaires. Semistructured interviews of participants.

Principal findings

There were 42 participants: 34 completed the trial. Twelve (35%) children were found to be responders. For participants for whom

3-month follow-up data is available, 6 responders of 8 (75%) continued stimulants, and 10 of 14 (71%) non-responders ceased stimulants. Parents and patients found benefit from the trials.

Discussion

The study population was not representative of all ADHD patients, as they self selected on the basis of concern about the efficacy of stimulant treatment.

Implications for policy, delivery and practice

The IMETs provide strong evidence for the appropriateness or otherwise of stimulant treatment in these individuals, and result in soundly based treatment decisions being made.

Doctors for doctors — facilitating access to general practitioners for junior medical officers

N Shadbolt, Hornsby Ku-ring-gai Ryde

Division of General Practice, L Mors

Introduction

There is a great deal of evidence that doctors are not good at looking after their own health. The consequences of this include increased rates of stress related illness, drug dependence, depression and suicide. These are powerful indicators of poor self-care.

Statistics indicate that junior medical officers are a particularly ‘at risk’ group.

Objectives

This work, which has received an Innovative

Funding Grant, documents the health care behaviours of hospital medical officers and explores barriers to appropriate health care and to seeking the help of a general practitioner. We identify ways in which general practitioners can facilitate access to their services for other doctors.

Methods

General practitioners often have difficulty seeing other doctors as patients. Junior medical officers are a particularly ‘at risk’ group. The objectives are explored through focus groups and questionnaires. The aspects of education, counselling and practice arrangements, which would assist general practitioners to take on the role, are explored.

Results

The results suggest ways in which hospitals, area health services and divisions of general practice might work together to promote appropriate health care behaviours among junior medical officers and facilitate access to services.

Conclusions

The data can be extrapolated to the wider medical community.

Evaluating an alternative model to screening for psychosocial issues in pregnancy

C Nagle, Department of General Practice,

The University of Melbourne, K Hegarty,

J Gunn, S Brown, D Forster, J Lumley,

J Collette, S Nicholson

Objectives

X

To enhance the knowledge and skills of midwives, GPs, and hospital medical staff to identify and appropriately manage women presenting with psychosocial issues in pregnancy.

X

To improve the degree to which pregnant women report they feel comfortable disclosing psychosocial issues to health professionals involved in their antenatal care.

Methods

This study is presently in progress and involves health professionals at the Mercy

166

2002 General Practice and Primary Health Care Research Conference — Proceedings

Hospital for Women undertaking an innovative, multifaceted educational program over 4 months. In small group learning environments, interactive workshops on advanced communication skills will be conducted using simulated patients. Skills will be developed, reinforced and supported using a modified Balint group technique.

Outcomes will be assessed by women booking at the Mercy completing a survey before and after the educational program. A questionnaire has been developed that incorportaes validated scales to assess psychosocial issues such as partner abuse and depression.

The sample size estimate of women participants was based on increasing the rate of disclosure of partner abuse by women from 1% to 5% (80% power, 0.05

significance level). This will require 333 women to complete surveys for the baseline findings and a different group of 333 women to complete post intervention surveys.

The participating GPs, midwives and hospital doctors will complete pre and post surveys and patient logs to measure changes in their knowledge and skills. The training program will be offered to one group of Shared Care

GPs (n=10), two groups of midwives (n=20) and one group of hospital doctors (n=10).

Rationale

In performing psychosocial risk assessment

GPs and other health professionals are confronted with a number of practical problems:

X the inclusion of diverse issues such as homelessness, violence from an intimate partner, depression, substance misuse, lack of social support, serious mental illness under the umbrella of psychosocial risk

X the reluctance of many women to disclose their experience of psychosocial issues and their fears of future events

X the reluctance of GPs and other health professionals to ask women directly about psychosocial issues.

The result has been the development of psychosocial risk assessment tools and the implementation of screening. This response is problematic as screening tests lack the level of sensitivity and specificity needed, to avoid an unacceptable level of misidentification. In addition, there is limited evidence to support the use of interventions with women identified by screening tests as ‘at risk’ of adverse psychosocial outcomes

In providing quality evidence-based pregnancy care GPs and other health professionals require the knowledge and skills to identify and appropriately manage women presenting with pyschosocial issues in pregnancy. This project seeks to implement and evaluate an alternative approach to screening.

Implications for practice

Evaluation of this model will inform moves to improve the emotional and social care provided by GPs and other health professionals to women in pregnancy.

After ANBP2 — ASPREE (ASPirin

Reducing Events in the Elderly)

M Nelson, Department of Epidemiology and

Preventive Medicine, C Reid, C Johnstone,

H Krum, J McNeil

Objectives

To present a proposal for a general practicebased double-blind placebo controlled trial to determine the effect of low-dose aspirin for the primary prevention of major cardiovascular events, mortality, cognitive decline, quality of life, major bleeding episodes and cost-effectiveness in elderly subjects.

Methods

This study will utilise the unique resource established for ANBP2 to conduct large-scale trials in general practice

1

. This study has recently been completed after successfully participation of 2681 GPs and 6083 subjects.

Principal findings

The effects of low-dose aspirin on cardiovascular outcomes has been tested in randomised trials in a variety of populations including, patients with a history of myocardial infarction (MI) or stroke, patients with evolving acute MI and in patients free of cardiovascular disease (CVD). Meta-analyses of these studies have shown that in all patient groups studied aspirin therapy reduces the subsequent incidence of cardiovascular death, non-fatal MI and nonfatal stroke by approximately 25%. In patients with established vascular disease the absolute benefit conferred is usually sufficient to outweigh concerns about adverse effects and as a result low dose aspirin therapy is widely recommended for the secondary prevention of myocardial infarction and stroke. In primary prevention

2002 General Practice and Primary Health Care Research Conference — Proceedings

167

the available data has been drawn from five major morbidity/mortality trials, whose participants were predominantly middle-aged males. Amongst such individuals the proportional reduction in vascular risk has been similar to that found in secondary prevention trials. However the absolute benefit to be gained from aspirin is generally less and therefore the trade-off between the benefits of reduced vascular risk and the burden of adverse effects is more finely balanced. Unlike the case with secondary prevention no primary prevention study has yet identified a benefit in terms of all-cause mortality.

Discussion

Outcomes measured are those important to an elderly cohort and reflect National Health

Priority Areas. They recognise that prevention of morbidity may be more important than end of life events in a group with a truncated life expectancy. Therefore

ASPREE includes cognitive, disability and quality of life measures.

All subjects will be aged > 70 years with hypertension, elevated total cholesterol, or smoking and without established CVD.

Sample size of 6000 subjects to detect over a

5 year follow-up period (on average), at the

5% level of significance and with a power of

90%, a 10% difference in total cardiovascular events (including cardiovascular deaths) between the subjects randomised to aspirin and the subjects randomised to placebo. This will also have power to detect significant differences in cognitive scores.

Implications for policy, delivery or practice

CVD is the commonest cause of mortality in the Australian population and dementia is second in Years of Life lived with Disability in females and the 6th leading cause of Burden of Disease.

Reference

1. Wing L, et al. Second Australian National

Blood Pressure Study (ANBP2).

Australian comparative outcome trial of

ACE inhibitor- and diuretic-based treatment of hypertension in the elderly.

Clin and Exper Pharm Physiol 1997;

19:(5&6) 779–791.

Cardiovascular risk factors in adults with diabetes: data from the

National Divisions Diabetes Program

Data Collation Project

J Oldroyd, School of Community Medicine,

University of New South Wales, J Burns,

D Penn, F Infante, G Powell Davies,

M Harris

Objectives

Cardiovascular disease (CVD) is a leading cause of death in adults with type 2 diabetes.

Despite intensive lifestyle intervention, adults with type 2 diabetes have been found to have high levels of modifiable CVD risk factors

1

. We investigated CVD risk factors in adults with diabetes using national diabetes aggregation data.

Methods

The National Divisions Diabetes Program

Data Collation Project

2

was a cross sectional survey of Australian divisions of general practice, from1999–2000, that included piloting an electronic collation of diabetes variables from those divisions using the

CARDIAB

3

(Version 1) register recall database.

Principal findings

In total, data was collated from seven divisions on 4359 individuals with diabetes

(mean age 62.2) of whom 3851 (88.3%) had type 2 diabetes. The mean duration of diabetes was 6.5 years (range 0–72 years).

In those for whom smoking data are available (n = 1209),163 (13.6%) smoked and 1039 (86.4%) did not smoke. The mean

(SD) body mass index was 30.2 (6.1) kg/m

2 with the prevalence of obesity 46.0%. In total, 2283 (52.4%) had systolic blood pressure (BP) >130 mmHg, and 835 (19.1%) had a diastolic BP > 85 mmHg. Mean (SD)

HbA1C was 6.9 (1.1%), 7.7 (1.5%) 8.2

(1.4%) for individuals treated by the diet, tablets and insulin, respectively. Mean (SD) total cholesterol and serum triglycerides were

5.3 (2.1) and 2.2 (1.5) mmol/l, respectively.

In logistic regression, obesity predicted elevated BP after controlling for age and sex.

Smoking was associated with elevated Hba1c and serum triglycerides.

3

CARDIAB was developed by the CARDIAB Alliance

(InTouch Consultancy, Macarthur Division of GP,

The Pharmaceutical Alliance, UNSW and the

National Heart Foundation). Version 2 is now available www.intouch.com.au/cardiab

168

2002 General Practice and Primary Health Care Research Conference — Proceedings

Discussion and implications

These data confirm a high level of risk for

CVD in adults with diabetes. Importantly, they demonstrate the potential for risk factor monitoring using a divisional electronic register recall system (CARDIAB).

References

1. Egede LE & Zheng D. Modifiable cardiovascular risk factors in adults with diabetes: prevalence and missed opportunities for physician counselling

Arch Intern Med 2002 162 (4) 427–33.

2. Burns J, Carter S, Bonney M, et al.

National Diabetes Program Data Collation

Project. Volume 2: Divisions of General

Practice — Diabetes Profiles. Divisions and Program Description. Sydney: Centre for General Practice Integration Studies,

School of Community Medicine UNSW,

2000.

What do general practitioners think about structured care for chronic disease management: a plan for focus group analysis

J Oldroyd, Centre for General Practice

Integration Studies, University of New South

Wales, G Powell Davies, F Infante,

J Burns, M Harris

Objectives

Chronic diseases are increasing in prevalence and a high proportion of consultations with general practitioners are for chronic diseases.

The way that general practitioners organise structured care and set priorities for chronic illnesses not well understood. We aimed to investigate the views and perceptions of general practitioners for the implementation of structured care. We report findings from initial focus groups, which aimed to establish a conceptual map for investigating chronic disease.

Methods

General practitioners were purposively selected from divisions to represent a variety of practice characteristics.

Principal findings

Three focus groups have been conducted in

Sydney. Many general practitioners were already practising aspects of EPC (e.g. care planning) prior to the government initiatives.

Staff such as dieticians and practise nurses were perceived as a useful adjunct to caring for chronic disease care although they were thought to be expensive in particular, by solo practitioners. The EPC activities were seen as helpful for organising care. Low numbers of general practitioners was seen as a barrier to effective chronic disease care.

Discussion and implications

The findings will be used to inform further focus group investigation. Exploratory practice visits and face to face interviews will be conducted to ascertain how chronic disease care is delivered and how general practitioners think it could be improved. The views of general practitioners provide unique information to augment division research on ways of strengthening structured care for chronic disease.

This research is being conducted in conjunction with The Department of General

Practice, University of Adelaide.

All in the mind: how clinicians interpret referral information concerning gamblers

J Patford, School of Sociology, Social Work and Tourism, University of Tasmania at

Launceston

The study explored the cognitive processes of counsellors, as manifested in verbal commentaries on written referrals concerning gamblers. Its primary aim was to conceptualise and classify the constituent elements of counsellors’ ‘process knowledge’.

The sample was purposive in order to ensure that participants differed in regard to professional qualifications, organisational setting and work experience. The referral information was presented in the form of two ambiguously written vignettes. Participants initially responded to these without interruption. Follow-up questions explored various aspects of case handling, in particular, the usefulness of other helpers and services. A categorisation system was developed by means of content analysis.

Findings from the present study build on previous work and have implications for professional training programs in social work, psychology and medicine.

2002 General Practice and Primary Health Care Research Conference — Proceedings

169

GP–hospital integration: what are we trying to achieve and how are we trying to do it?

PG Powell Davies, Centre for GP

Integration Studies, University of NSW,

F Reynolds, MF Harris

Objectives

To develop a framework to describe the main purposes for improving GP-hospital integration and show how main activities fit these purposes and which of the purposes are currently least strongly pursued.

Methods

A review was conducted of literature, reports and policy documents relating to GP-hospital integration. Interviews were conducted with staff from selected projects, and a Delphi process conducted with key informants and stakeholders. Preliminary conclusions were checked at a national stakeholder conference.

Principal findings

Five main purposes were identified:

X building better working relationships between GPs and hospitals

X preventing the need for acute care

X shifting care to more appropriate settings

X improving transitions of care

X improving access to care.

Nine major program types were identified: case management, shared care, hospital diversion, after hours care, GP-hospital communication, pre-admission planning, discharge planning, post acute care and hospital in the home. These each had the potential to contribute to more than one of the purposes, and program development typically proceeded from the basis of particular projects rather than through a systematic review of the potential under each of the purposes. The initiatives that were adopted tended to reflect the particular needs of the collaborating parties, and purposes which were not urgent for them — such as preventing the need for acute care and improving access to care — tended not to be pursued.

Discussion

The five main purposes make intuitive sense and fitted well with the initiatives that we studied. There appears to be a cycle where an investment in working relationships provides the basis for improving transitions of care and, to a much more limited extent, shifting care to more appropriate settings. If successful these reinforce each other, but do not seem to lead often to a focus on prevention or improving access to care.

This appears to reflect the fact that prevention and improving access to care do not immediately address the core concerns of

GPs (which relate largely to continuity of care) or hospitals (which relate largely to problems of budget and waiting lists).

Furthermore there is no integrated base in primary health care from which issues of prevention and improving access, which are fundamentally primary health care concerns, can be pursued.

It appears that a more balanced approach to

GP-hospital integration may require a more integrated policy framework which looks beyond the immediate interests of hospitals and general practice, and a more integrated primary health care to advocate for these interests.

This work was funded by the Department of

Health and Ageing.

More bridges: an evaluation of GP– hospital liaison officer positions in

Australia

F Reynolds, Centre for General Practice

Integration Studies, University of NSW,

PG Powell Davies, F Reynolds, J Oldroyd,

MF Harris

Objectives

X

To describe the use of GP hospital liaison positions, including:

X

– the GPLO workforce

– the location of the positions

– the way in which these positions operate.

To identify how this has developed since

1999.

X

To describe the perceived impact of GPLO positions on GP-hospital integration.

Methods

A survey of GP hospital liaison officers; a national survey of hospitals, stratified for state, location and public/private status; a census of Divisions of general practice; and four case studies.

170

2002 General Practice and Primary Health Care Research Conference — Proceedings

Principal findings

Since 1999 the number of GPLO positions had increased from 46 to 54, 76% in urban areas. Most GPLOs were doctors (64%) or nurses (24%), and mostly female (74%).

Most positions were part time (71%) and permanent (55%). Most were employed by a hospital (52%) and located there (64%).

45% were funded by a hospital and a further

19% were funded jointly by a division and a hospital. The greatest proportion of positions had existed for two years or more, but 28% were relatively new.

38% of divisions of general practice and 22% of hospitals reported having access to a

GPLO, some of whom worked across more than one hospital. The highest proportion of hospitals were urban (37%) then provincial/regional (28%) and rural and remote (14%). Of hospitals without access to a GPLO, 54% had a staff member nominated to oversee links with general practice.

Most GPLOs worked alone, but 43% through a GP liaison unit or similar. 79% of positions had a job description but only a minority a strategic plan or a written plan for their position. Most had some administrative support and reported to a committee, but only 36% of positions had been formally evaluated.

As in 1999, the main focus of the GPLO positions was on developing ongoing systems for linking GPs and hospitals rather than building relationships or solving individual problems. The case studies and surveys showed that they were seen as effective in this role.

Hospitals with a GPLO were significantly more likely to be involved in joint planning, projects or committee membership with

Divisions, and to have a formal agreement.

They were also more likely to be involved in shared care or hospital in the home with GP involvement, and to be aware of government policy as contributing to better links with general practice. Having an existing nominated staff member was also beneficial in most of these areas, but to a lesser extent.

Discussion

It appears that the use of GPLO positions has developed since 1999, with more positions and more GPLOs who are well established.

GPLOs are most used in hospitals with complex interfaces with general practices, and their use is associated with more sophisticated approaches to GP-hospital links and a wider range of activities. However having an existing staff member tasked with overseeing links with general practice also appears to be beneficial, and may be most appropriate for less complex circumstances.

It appears that a broad strategy for improving GP-hospital integration may need to focus on strengthening the use of identified staff members as well as liaison officers.

This work was funded by the Department of

Health and Ageing.

Is there anybody listening? The difficulties working as a female GP in remote and rural Western

Australia

S Roach, Western Australian Centre for

Remote and Rural Medicine, F Jefferies

Objectives

This research project aimed to:

X identify and examine barriers to female

GPs working and living in rural and remote locations at the practice, community, divisional, organisational, and State level in WA

X identify possible ways to overcome barriers that exist so as to attract and retain female GPs in rural and remote

WA.

Methods

The WA project utilised a survey that was developed by the NSW RDN and this was mailed to all 119 female GPs and registrars in

WA. Sixty GPs responded to the survey representing 50.4% of the population. The survey results were discussed at a focus group of representatives from diverse areas of WA, and then with a group of female GPs in the Pilbara as part of the Pilbara WACRRM weekend. Individual consultation also occurred with a number of GPs to further examine the issues. The Statewide rural postgraduate weekend in Fremantle in

October 2001 provided an opportunity to synthesise the information gained thus far and to workshop possible solutions.

Principal findings

The main issues that were raised in the survey were:

X limitations due to on-call requirements

X difficulty obtaining appropriate childcare

2002 General Practice and Primary Health Care Research Conference — Proceedings

171

X heavy work commitments (long consults/long hours)

X lack of part-time work

X professional isolation and lack of network support

X difficulties in accessing Continuing

Medical Education (CME) and adequate training

X social isolation

X spouse issues.

The issues were workshopped by a group of thirteen rural female GPs at the WACRRM sponsored post-graduate education weekend in Fremantle. Positive and practical solutions were sought to each of the issues and these formed the basis of an extensive list of recommendations. Some can be addressed immediately and others require system change that may be slow in coming.

Discussion

Despite the creation of a list of mutually agreed issues, it was clear throughout this project that individual GPs and specific communities placed varying degrees of importance on the issues. For this reason it is imperative that solutions are approached on a level that allows for individual input.

The results of this project acknowledge and specify the unique problems that exist in WA and offer ways to ensure that the health needs of rural West Australians are met.

Implications for policy, delivery or practice

It is of prime importance that all stakeholders invest time, effort and money into pursuing these recommendations towards positive conclusions. For the women who have taken the time to contribute to this study it is imperative that they begin to see action and practical outcomes that will result in real assistance to enable them to stay working in their communities. For too long, rural research has reported the same difficulties — it is now time for someone to listen and to take action.

Co-locating community nurses in general practice to improve care in disadvantaged communities: a comparison of models

V Rose, Centre for Health Equity Training

Research and Evaluation, E Harris, M Harris

Objectives

This project aimed to:

X identify and develop the role and responsibilities of GPs and community nurses in providing accessible, appropriate and quality primary health care to people living in disadvantaged communities.

X improve the linkage between general practices and community health services in disadvantaged areas by co-locating community nurses in general practices.

Methods

Community health nurses were co-located in two general practices located within disadvantaged communities. Two models of co-location were implemented and evaluated: a full-time model and a part-time model.

The evaluation comprised:

X record audits of a sample of GP records at six-month intervals.

X analysis of community health data to identify referral rates from GPs.

X semi-structured interviews with GPs, community nurses and patients.

Principal findings

Both co-location models had an impact upon

GP practice through increased referrals to community health and multi-disciplinary care. GPs in the full-time model reported an increase in the identification of psychosocial problems and referral of patients with these problems to the community nurse. The increase in referrals in both models impacted upon community nurse practice making this project time intensive for community health.

In the full-time model co-location resulted in increased recognition of community health services amongst the general community.

Teamwork was a feature of the full-time but not part-time model. The limited nature of contact between the GPs and community nurses in the part-time model made program development difficult and implementation slow.

172

2002 General Practice and Primary Health Care Research Conference — Proceedings

Discussion

The full-time model produced greater change. However, this was difficult to sustain both within the context of the demands on community health services and the commercial pressures operating on general practice. Both community health services and general practice are facing increasing pressure to change. This presents both a barrier and a potential opportunity for the development of innovative ways in which these different organisations and structures can work more effectively together.

Implications for policy, delivery or practice

This project has shown that it is possible to develop models of co-location between GPs and community nurses but that these need to be placed in a wider organisational culture and structure. There is an important role for community nurses that is different from that of a practice nurse especially in dealing with psychosocial issues, providing home-based care and in linking the practice to other community health services.

The development of an Aboriginal community controlled ethics committee in WA

J Rowbottom, Combined Universities Centre for Rural Health, T Wilkes

Objectives/methods

To describe the development of the Western

Australian Aboriginal Health Information and

Ethics Committee (WAAHIEC), it’s trial, tribulations and successes over a 5 year period of from 1996.

Principal findings

I don’t need to stress to this audience that

Aboriginal people have been well researched.

The problem has been that much past research has not been practically useful in terms of improving Aboriginal health.

To address this deficiency, the Office of

Aboriginal Health (OAH) of the Health

Department of Western Australia (HDWA) in partnership with the Western Australian

Aboriginal Community Controlled Health

Organisation (WAACCHO) established a

Western Australian Aboriginal Health

Information and Ethics Committee

(WAAHIEC) in August 1996.

The OAH of HDWA has invested resources into running the committee, with expertise from that office in terms of staffing, and in funding committee members’ expenses. The members have appointed by the minister if

2002 General Practice and Primary Health Care Research Conference — Proceedings health, having been nominated through

WAACCHO, the WA Community Controlled

Health Organisation. This organisation covers the entire state, and appears to be the most representative way of ensuring a fair process. I would like to share my experience of this from the perspective of a medical and scientific advisor to the committee over three years, during which time over 65 submissions were reviewed. The main problems we encountered were lack of consultation and community consent, in addition to exploitation of limited community resources.

Implications for policy, delivery or practice

To some researchers, Aboriginal control in the research process is a new concept. It shouldn’t be. It is important that Aboriginal people and researchers get together to work on the most important issues for research, and ensure these get priority.

The main message that we’d like to share with you is that Aboriginal research needs improved communication. We need to share a better understanding of the roles of the

Aboriginal community and the scientific researchers. Mutual respect is the key to this.

The success of the current committee stems from 3 important issues: getting the right partnerships, strong leadership and mutual respect between the scientific and the

Aboriginal community.

We’re not alone: family practice reform in Canada

G Russell, Saw Research Fellow, University of Western Australia

Objectives

To highlight the current challenges faced by family medicine in Canada.

To compare Canadian and Australian solutions to contemporary primary care problems.

Methods

The presentation will present a review of the current state of Canadian family practice. It will draw on the author’s experiences during a 2 year attachment to the Department of

Family Medicine at the University of Western

Ontario. Comparisons will be made between the Australian and Canadian approaches to family practice integration, workforce, and training. The academic implications of these comparisons will be highlighted.

173

Findings

The Canadian health care system has been widely admired for its quality, equity and accessibility. In recent years this publicly administered system has been beset by pressures associated with high technology, private medicine and workforce demographics. The effects have placed enormous pressures on health care delivery, particularly that practiced by the nations

30,000 family doctors (FDs).

Decreasing proportions of medical graduates are choosing a career in family medicine.

Their experienced FD colleagues are increasingly demoralised and have begun to limit the comprehensiveness of their care.

Many FDs have left practice altogether in favour of working in heavily subsidised emergency and after hours services.

Family practice reform has been contemplated for over a decade — recent reform proposals, while courageous, place new and important demands on family doctors and their patients. These proposals, many of them generated by FD academics, specify patient rostering, major investments in information technology and the implementation of primary care networks.

Unlike Australian Divisions, these ‘real or virtual’ FD groups have a clinical focus — being designed to deliver 24 hour primary care to enrolled patients. The reform process has drawn heavily on the personnel and research agendae of academic Departments of Family Medicine.

Discussion

An understanding of the current situation requires an appreciation of the cultural, academic and philosophic foundations of

Canadian family practice. The presentation will encourage participants to reflect upon these issues as they apply to the Australian context — one providing similar challenges in terms of distance, isolation and urban consolidation.

Implications

Canada’s health care challenges are of particular relevance to Australian Primary

Care in 2002 The road to a solution, while uniquely Canadian, has important implications and lessons for Australia in general and its academics in particular.

New partnerships are required to address data needs and issues for community health planning

D Ruth, Department of Community Health,

The Royal Melbourne Hospital, Melbourne

Health, R Hurworth, N Sulaiman

Objective

This qualitative study brought together the views of data users and producers at primary health care, hospital, regional and state levels to identify data needs and issues for community health planning.

Methods

Eight focus groups and the Nominal Group

Technique were used to discuss local service providers’ and planners’ perceived current practice, issues and needs related to obtaining and using data for planning services in the Hume and Moreland local government areas of Melbourne. Groups were conducted with nominees from 2 municipal councils, 4 divisions of general practice, 3 community health services, 3 hospital networks, 8 community organisations, and the Department of Human

Services, Victoria.

Principal findings

The 66 participants had a broad range of experience in using data for needs assessment and service evaluation.

Participants reported that issues limiting the use of data related to:

X access to data (lack of awareness, gaps in data, contacting the right person, poor communication between data providers and users, resource constraints, lack of central access)

X quality of data (inconsistent definitions and collection, currency, ties to funding)

X applicability of data (unfriendly format, problems with aggregated versus small area data, non-matching data sets, lack of contextual information)

X local service providers’ roles and skills for population-based health planning.

Hence National Health Priority Areas, rather than local needs assessment, often were used to inform service planning. Participants suggested data, presentations and field support strategies that would help them to better access and use data to plan local needs-based health services.

174

2002 General Practice and Primary Health Care Research Conference — Proceedings

Discussion

In Victoria, Primary Care Partnerships (PCP) are required to develop and implement

Community Health Plans involving service planning, coordination and partnerships. To do this, PCPs need to identify and monitor key health issues and determinants in their catchment area and find appropriate data to support needs-based program design, implementation and evaluation. However, currently available information has limited usefulness for informing local community needs assessment. Our findings show that service providers need access to high quality data, presented in formats that are applicable to their local communities. They also need support in using data for needs assessment and service planning.

Examples are given of initiatives that address some of the data issues, including the production of a guide in partnership with data users and producers ‘Selecting and accessing population data: An information resource,

Department of Human Services, Victoria,

2001’. Additional strategies are suggested that could be implemented to improve the quality of data and its accessibility and applicability to community health planning, and to build the capacity of managers, practitioners and planners to use data.

Implications

To address the need for data for communitybased health planning requires a policy and funding commitment from government for data producers to work in partnership with local service planners to address data collection and usage issues.

Docs 4 docs: a survey of doctors’ attitudes to seeking medical attention

P Schattner, Department of General

Practice, Monash University, S Davidson

Objectives

The aims of this study were to measure the attitudes to the boundary between self and external care among doctors and to identify the barriers and facilitators doctors face in moving appropriately from self- to external care.

Methods

Phase one of the study consisted of in-depth interviews with GPs and medical specialists.

(The results of phase one were presented at the 2001 conference). The second phase of the study, on which the current presentation is based, involved a questionnaire survey of almost 900 doctors (GPs and medical specialists) randomly selected from stratifications in the Health Insurance

Commission database. The 32-item questionnaire comprised 4 sections: 1) beliefs about appropriate self-care; 2) personal experience of a consultation with another doctor; 3) the management of case vignettes; and 4) demographic data.

Principal findings

The was a 40% response rate, with 65% of the respondents being GPs. Three-quarters agreed that doctors present late with their medical problems, with 71 per cent being embarrassed to see another doctor for a diagnosis. GPs were more likely to be embarrassed than specialists (32% versus

19%). They were particularly reluctant to seek professional help for psychological problems (88% agreed). Three-quarters believed that doctors were more likely to make errors if they treated themselves, and almost the same proportion thought that doctors tend to downplay the significance of their own symptoms. A quarter thought it reasonable to treat chronic medical conditions, and 50% thought it acceptable to order blood tests on themselves. Nine per cent felt it was acceptable to self-prescribe psychotropic medication. Analysis of participants responses to a series of case vignettes indicated that specialists were more likely than GPs to seek formal health care, and doctors who had their own GP were more likely to choose appropriate treatment for the three conditions presented. Variables such as workload, self-efficacy in own ability and embarrassment in attending another practitioner were not significant predictors or whether doctors would seek formal care.

Discussion

It is difficult to define a boundary for appropriate self versus external care which is consistent across a wide variety of circumstances, such as screening, monitoring, and acute versus chronic, selflimiting versus serious, and physical versus psychological illnesses. However, the patterns which emerged in this study are generally consistent with those found in the literature. Neither demographic nor belief variables (concerns around confidentiality, workload, embarrassment and self-efficacy) consistently predict whether doctors are likely to manage conditions in themselves

‘appropriately’.

Implications for policy, delivery or practice

There remain cultural barriers to doctors formally seeking the advice of colleagues for

2002 General Practice and Primary Health Care Research Conference — Proceedings

175

their medical problems. These might not be easily changed once such behaviour becomes entrenched. ‘Docs 4 docs’ programs probably need to commence at an early stage or professional development, especially at the undergraduate and early postgraduate level.

Access to private psychiatrists: what are the trends? New story — old dilemmas

R Scheurer, Mental Health Policy and

Services Research Group, Queensland Centre for Schizophrenia Research, Wolston Park

Hospital, H Whiteford, D Chant, P White

Objective

This paper provides a descriptive analysis of the Medicare Benefit Schedule funding to private psychiatrists in Australia and illustrates the trends in billing by item number, gender and geographic area.

Method

Aggregate figures on Medicare Benefits

Schedule (MBS) items by private psychiatrists were extracted from the Health

Insurance Commission records for the 1993 to 2001 financial years.

Principal findings

Consultant psychiatrist services billed to the

HIC (n= 27, items 300–352) showed a

56.6% reduction in 15 MBS items. Item 319 displayed the greatest up ward trend and only marginally impacted on the overall private psychiatrists’ expenditure. Victoria and South Australia remain the States with comparatively high numbers of psychiatrist services billed per 100,000 population, with

Northern Territory the lowest. Private psychiatrists in Australia continue to be concentrated in the capital cities and tend to attract the majority of their patients from metropolitan areas.

Discussion

Relevance: An analysis of funding provided to private psychiatrists in Australia through the

HIC should assist in designing future strategies to improve this resource allocation.

Rigor: Medicare data on private psychiatrists consultations has been available since 1984–

85. The utilisation of Health Insurance

Commission (HIC) payments over time can be used as a proxy to measure the number and time spent by private psychiatrists in consultations.

Partnerships: Essentially private psychiatrists and private and public psychiatric services

176

need to form further collaborative links with general practitioners and allied health providers, particular in rural areas of

Australia.

Capacity: This research paper has the capacity to explore further the psychiatric specialty in relation to workforce resource management, analyse different models of service delivery or contribute to a future template for mental health care in a primary care setting.

Implication for policy, delivery or practice

Private psychiatrists constitute an important component of the mental health workforce.

The overall decline of MBS billing by psychiatrists and access to services is a major concern.

Teaching clinical skills in the community to Year 2 medical students

T Sen Gupta, School of Medicine, James

Cook University, J Vanlint, J Veitch,

R Hays

Introduction

The James Cook University (JCU) School of

Medicine is an innovative, community-based six-year undergraduate course. The course emphasises integration of basic and clinical sciences in a systems-based approach, with students exposed to clinical settings from early in the course. As students study each organ system in turn, they look at normal structure and function, and are introduced to some disease processes and relevant clinical skills. GPs were used to teach clinical skills to small groups of year two students in their practices, using general practice patients.

Three organ systems were studied in each semester of year two, so students had six half-day teaching sessions in general practice over the year. GPs had previously been involved in a number of facets of the course including teaching communication skills and practice visits.

Methods

Learning objectives for teaching clinical skills in each system were identified by groups of clinicians and basic scientists. GPs were recruited and provided with information and training. A variety of measures were used to evaluate this program including focus groups with students and GPs, direct observation of teaching sessions, questionnaires completed by participating GPs, and student performance on assessment items.

2002 General Practice and Primary Health Care Research Conference — Proceedings

Principal findings

This approach was popular with students and

GPs, with the general objectives of teaching clinical skills early in the course being met.

Students enjoyed the early clinical contact, the small group format, and the opportunity to spend time in a general practice. They felt they had a greater appreciation of the role and clinical skills of the GP and how general practices were organised. GPs were underconfident about their ability to teach this material, but were satisfied with the delivery of the sessions, and felt their own clinical skills had been enhanced in the process.

Discussion

Clinical skills can be taught in the community to ‘pre-clinical’ students in an integrated course. A number of issues emerged including the importance of using an integrated approach to define the key objectives of each teaching session, and limiting the amount of material taught; a systematic approach to teacher training and support; administration of the program; and ways in which to assess this material. Further evaluations are ongoing, but initial data suggest this approach is successful in teaching clinical skills.

Implications for policy, delivery or practice

As hospitals become more focused on shortstay and high turnover of patients, there will be an increased emphasis on communitybased education. This approach demonstrates that clinical skills can be taught in the community setting. Benefits to students apart from the acquisition of clinical skills include a deeper understanding and appreciation of general practice.

A primary health care strategy for

NSW

S Simpson, Centre for Health Equity

Training Research and Evaluation, E Harris

Objectives

A consultation process was undertaken for the NSW Department of Health in developing a statewide primary health care strategy and served the following purposes:

X engagement of all stakeholders (including

GPs and Divisions of General Practice) with an explicit chance to influence the direction of the project;

X information sharing with the opportunity for stakeholders to provide local examples of effective primary health care interventions and available evidence of the intervention’s effectiveness; and

X a chance to gauge the level of commitment to primary health care and some of the challenges to the strategy process.

Methods

The project is currently under way and will be completed by the end of April 2002. A draft outline of the proposed primary health care strategy is circulated to participants in advance, providing a basis for consultation using the following methods:

X individual interviews with the twenty

CEOs of Health Services in NSW — in person and/or by teleconference — using structured interview questions;

X stakeholder consultations with senior

Health Service staff in each Area Health

Service and other stakeholders including

GPs and Divisions within the Area. Using a premortem approach of three years hence, participants are being asked to identify three factors that would contribute to the strategy failing and three factors that it would contribute to its success; and

X consultation at key relevant forums within NSW including the Directors of

Community Health and the Directors of

Public Health.

Principal findings

The principal findings of the consultation process will not be known until late April

2002. However, the following outcomes are expected from the consultation process:

X identification of primary health care champions, examples of effective primary health care interventions and better practice, and supporting evidence;

X identification of barriers to the development, implementation and success of the strategy;

X a report on the consultation process with recommendations for addressing identified barriers.

Discussion

General practitioners are the main providers of primary health care in NSW. Together with community health services, community pharmacists, Emergency Departments and some community and non-government organisations they provide the bulk of out of

2002 General Practice and Primary Health Care Research Conference — Proceedings

177

home health care to people in the community. Currently the primary health care system in NSW is highly fragmented.

This has resulted in a lack of integration between the various primary health care providers (such as general practitioners and community health staff) but also a lack of integration with other community-based health services and hospitals. NSW Health has recognised the importance of building a strong primary health care system through the development of a statewide primary health care strategy. It is important that this strategy is developed in consultation and collaboration with GPs as the main providers of primary health care.

Implications for policy, delivery or practice

Engagement of GPs and Divisions as the main providers of primary health care, using an action research approach to the consultation process, should contribute to the development of effective health care policy for providers, contributing to improved service provision and practice.

The Victorian rollout of the EPC

Education, Support and Community

Linkages Program: evaluation reflections

J Sims, Department of General Practice,

University of Melbourne, L Naccarella,

M Potiriadis, T Tacticos, R Fenech,

W Newton, L Lippman

Objectives

The GP Education, Support and Community

Linkages Program (EPCESCL Program) has supported implementation and appropriate use of the MBS items since early 2000. The provision of training and resources to assist

GPs and other providers was intended to impact upon the usage and uptake levels of the Enhanced Primary Care (EPC) MBS items.

The aim of this evaluation was to assess the implementation of education, support and community linkages via Victorian Divisions of

General Practice.

Methods

A public health promotion evaluation model

(RE-AIM) and a capacity building framework were used to evaluate the program. The evaluators worked with General Practice

Divisions – Victoria (GPD-V) to collate secondary data concerning the program.

Semi-structured interviews were conducted with the CEO, EPC Coordinator and GP

Trainer from each Victorian division of general practice. These personnel provided feedback on how the EPCESCL program had

178

2002 General Practice and Primary Health Care Research Conference — Proceedings operated at state and divisional level. Further interviews with GPs and other health care staff trained during the program are currently being conducted. This information will be triangulated with that obtained from the program ‘providers’.

Principal findings

Interviewees noted achievements in the uptake of the MBS items during the course of the program’s operation. This trend had occurred in spite of some concurrent changes within the divisions, notably staff turnover, which could have influenced optimal program implementation.

Interviewees acknowledged GPD-V’s support in assisting program implementation. The

‘train the trainer’ model was seen as a valuable approach to adopt, although there were some reservations about its actual operation. Multiple strategies were used in the program’s education component. In general, CME events were used for awareness raising, for trainees to understand the EPC concept and broad parameters of the items’ use. In contrast, practice visits and academic detailing were used to impart greater detail and to provide technical resource to assist item usage. Overall, the practical systems support, rather than the awareness raising, was perceived to positively influence the items’ uptake.

Discussion

This work represents a range of partnerships: between researchers and stakeholders, namely Divisions of General Practice, their staff, members and associated community health providers. The overarching goal of these partnerships has been to conduct research and to ensure that systems are in place to enhance the dissemination and uptake of research findings. Optimism regarding the program’s sustainability was tempered by comments regarding the funded program’s short timeframe, and uncertainty about future funding, given finite divisional funds.

Implications for policy, delivery or practice

Capacity building and item usage may have been influenced by factors beyond the program’s direct control, including: the rollout of the Primary Care Partnerships and the introduction of other government initiatives e.g. ‘More Allied Health Services’.

These need consideration in interpreting the findings. However, divisions with clear objectives and who articulated planned strategies were more likely to report positive achievements. Where divisions integrated the program with others, respondents tended to

be more confident that the programs’ activities could be sustained in the future.

Falls prevention — developing a decision-support tool using the

Australian GP data model and core data set

J Sims, University of Melbourne, Department of General Practice, T Liaw, N Sulaiman,

K Hill, C Ng, C Pearce, M Rawlin

Objectives

To develop an evidence-based model for falls risk assessment and management within the framework of the Australian GP Data Model and Core Data Set.

Methods

A comprehensive range of vignettes was created to guide the design and development as well as verification and validation by multi-disciplinary clinician and consumer focus groups in an iterative cycle. The need for a controlled vocabulary, to incorporate user and work issues early in system design and for a standards-based approach to development in a distributed environment guided the literature review and project methodology. A rule-based logic engine was implemented based on the results of an extensive literature review and input from experts, clinicians and consumers in falls risk assessment and management. The Unified

Modeling Language (UML), Arden knowledge representation syntax and Extended Markup

Language (XML) messaging format was used to develop the falls knowledge base for delivery over the Internet. The prototype was tested in an Integrated Health Care Centre and a number of GPs.

Principal findings

The data model and development cycle addressed user and work issues and compliance to standards in knowledge acquisition and representation, vocabulary and messaging. The comprehensive input by clinicians and consumers significantly enhanced the term set and data model. The

GP Data Model varied in granularity across all the domains, being most superficial in the patient and clinical knowledge domains, particularly with regards to prevention and prescribing risk management. Further work is required in these areas.

Discussion

The study selected rigorous methodologies that assisted the development of the decision-support system and will enhance its dissemination. The use of the Arden Syntax to represent the knowledge and XML to implement the generic HL7 messaging format allowed the independent falls prevention knowledge base to be accessed via an

Internet Browser and Application Program

Interface (API) to desktop applications. The easy-to-comprehend Arden Syntax enhanced the understanding of the formalisation of the

‘rules’ and ‘knowledge’ among the multidisciplinary team members with varying technical expertise. This is probably the most flexible and constructive way to communicate and share knowledge in the distributed

‘Internet-based knowledge webs’ and

Electronic Health Record (EHR) environments. Proposed enhancements to the Arden Syntax to accommodate functional requirements and current limitations is potentially the most fruitful way forward.

Implications for policy, delivery or practice

Greater collaboration is needed between classification systems, terminologies, health information models and standards to achieve an ‘all of health’ reference terminology and information model for decision support.

Convergence of the international work on knowledge-based decision support in both the EHR and Internet (TCP/IP) environments will benefit Australia’s National Health

Information Plan and Strategy for EHR, knowledge bases and clinical decision support.

Supporting Divisions of General

Practice to enhance GP integrative activities: a role for the GP

Integration Index

D Southern, Department of General

Practice, The University of Melbourne,

D Young, D Dunt, C Harris, G Elsworth

Objectives

To assess the acceptability and feasibility of widespread uptake of the GP Integration

Index by Divisions of General Practice as part of their ongoing quality enhancement programs, and to obtain a nation-wide baseline of GP integration activities.

Methods

25 randomly selected Divisions of General

Practice, stratified by rural/urban and each state and territory, were invited to survey their GP membership using the GP

Integration Index with a comprehensive

Divisional Demographics Survey. Data was analysed using STATA statistical software package.

2002 General Practice and Primary Health Care Research Conference — Proceedings

179

Principal findings

The response rate across the 25 Divisions

(12 urban and 13 rural) was 54% of 3924 surveys distributed. 80% of Divisions achieved response rates greater than 50% and 48% had better than 70% response rates. Most survey respondents were

Australian graduates (75%), had Vocational

Registration (87%) and were in private general practice (88%). Respondents compared similarly with Health Insurance

Commission data, with the exception of gender and rurality. Compared with the national rate, the study population had fewer males (59% versus 66.8%) and fewer urban

GPs (70% versus 78%).

Findings demonstrate that the GP Integration

Index and its constituent factors can detect differences in the extent of integration in GP members of Divisions in the pilot sample. The effects of demographic characteristics, practice characteristics, GP engagement in relevant programs, and ‘Enabling Factors’ on the two main higher order factors of the GP index (Patient Care Management and

Community Health Role) were investigated using multiple regression analyses. The enabling factor ‘Knowledge of Local Services’ had the strongest observed effect on any individual variable, explaining 59% of the variance for the ‘Patient Care Management’ factor. GP engagement in relevant programs explained 36.1% of the variance for the

‘Community Health Role’ factor. Demographic characteristics demonstrated the least impact, explaining 1.4% of the total variance for the PCM factor and 2.5% of the total variance for the CHR factor.

Discussion

The external validity of the GP Integration

Index is well demonstrated by multiple regression analysis. Further analysis will show whether findings are confounding by GP characteristics and/or their patients within the divisional area. Results suggest that this is unlikely because the proportion of variance due to demographic and practice characteristics is relatively small. This will not influence the use of the instrument when comparing within divisions over time. The interpretation of the significance of findings for each division will need to be explored case by case, to clarify whether statistically significant differences are significant in policy or program terms.

Implications for policy, delivery or practice

Currently there is little baseline data on the level of integration of general practitioners with other health service providers at a national level. This current survey serves to

180

2002 General Practice and Primary Health Care Research Conference — Proceedings provide Divisions of General Practice with baseline information which will assist with informing their quality enhancement programs.

Evidence in action: FRAMS, a Falls

Risk Assessment and Management

System

N Sulaiman, University of Melbourne,

Department of General Practice, T Liaw,

C Ng, C Pearce, J Sims, K Hill

Objectives

The main objective is to demonstrate the implementation of computer-based guidelines for the early identification and management of falls risk factors in people over the age of

65 years by members if the health care team. Falling is a major public health problem and falls prevention one of the

National Health Priority Areas. The Falls Risk

Assessment and Management System

(FRAMS) will be presented to demonstrate its main features, main uses important issues encountered during the development and field-testing.

Methods

A formal modelling process was adopted to the clinical and data model. A generic standards-based method was used to translate a clinical and data model into a clinical application and knowledge base — the

FRAMS. The model is based on standardisation of health information, information management and software design. Technically, the FRAMS consists of an input/output device, a knowledge base, and an inference engine used for building the knowledge base.

A computerised knowledge base was developed to generate patient-specific assessments or recommendations. Clinicians may choose to accept or reject the recommendations. Development of the

FRAMS followed a standard computer-based implementation of guidelines and protocols, involving the following steps:

X modelling — deciding which knowledge representation to use.

X authoring — putting the guidelines into the knowledge representation

X dissemination — putting the guidelines into practice

X implementation — putting the guidelines into a computer executable form

X update — changing the guidelines as necessary.

In addition to the system, the team has also trialed and established methods to enhance the underlying terminology with relevant and user-friendly clinical terms, leading to the establishment of a well-grounded interface terminology and extended the GP Data Model and Data Set in the area of falls prevention in the elderly.

Results

The FRAMS software could be accessed through the web or through a desktop application such as Medicus

TM

software. The web-based model is available for testing.

GPs, allied health professional and consumers have been testing the functionality and usability of the model. The short version of the model is popular among testers and the feedback is encouraging. The more detailed version was time consuming, hence less favourable to GPs. However it may be suitable for use by nurses, physiotherapists and other primary health care professionals where consultations are less time constrained.

Discussion

Doctors have generally been positive about the use of the software packages and their decision making (accuracy, appropriateness) has been enhanced by in-built evidencebased information. The output of FRAMS is found to be particularly useful in mild to moderate risk assessment. The next phase is to develop and test a consumer version of

FRAMS to enhance empowerment, selfassessment of risk of falls and self care in the management of their individual risk factors, in partnership with health care providers.

After hours primary health care for the disadvantaged: a qualitative investigation

J Taylor, WA Research Unit, The Royal

Australian College of General Practitioners,

S Roach, T Robbins

Objectives

The project aims to:

X determine the after hours health care needs of the disadvantaged within the population served by Fremantle Hospital and Fremantle Hospital General Practice

X propose alternative models of after hours health care that meet the needs of the disadvantaged within the geographical area

X identify the determinants of coordinated interprofessional communication between local primary and tertiary health care services

X identify the process through which patients are currently triaged to

Fremantle Hospital General Practice.

Methods

In depth, semi-structured interviews were conducted and recorded with consumers, local GPs, local health services and hospital staff. Interviews were transcribed verbatim.

Transcriptions were analysed using the constant comparative method of grounded theory to develop a conceptual model of existing services and barriers to improved efficiency and access to after hours medical care by the population under study.

A workshop with stakeholders utilising the developed conceptual model to explore potential models of after hours medical care will also be conducted.

Principal findings

This project is currently a work in progress and analysis is expected to be completed by the end of May, 2002. Interim analysis indicates that many patients are using the

Fremantle Hospital Emergency Department after hours for non-emergency presentations due to a lack of other appropriate options.

The departments within the local hospital have a good rapport with the Hospital

General Practice and referrals between tertiary and primary care are well coordinated.

Discussion

This project aims to establish and promote integration between primary and tertiary health care professionals. It also seeks to encourage multi-disciplinary primary health care teams in an after hours setting. It involves a focussed investigation of consumer and health-provider issues and of the health and service satisfaction of disadvantaged persons.

Implications for policy, delivery or practice

The project will result in a model of after hours care that will be of benefit to the local disadvantaged population. It will enable continuity of care and therefore is expected to improve patient satisfaction and GP job satisfaction. A model of after hours care should also move the workload of nonemergency presentations from the

2002 General Practice and Primary Health Care Research Conference — Proceedings

181

emergency department to the more appropriate after hours primary health care.

Rural career choice — what influences medical students? The first stage of a longitudinal study

H Tolhurst, University of Newcastle

Objectives

The objectives of the project are:

X

To investigate the influences on the career intentions and choices of medical students, with a focus on social and cultural issues including gender issues.

X

To explore the attitudes of medical students to various models of rural practice including primary health care teams and the influence that the availability of different models of practice have on the career intentions and choices of medical students.

X

To explore the attitudes of students to working in various practice structures for example traditional partnerships, corporatised practices and salaried health services, and the influence that the availability of different practice structures have on the career intentions and choices of medical students.

Methods

Stage one of this project is a descriptive pilot study using qualitative data, collected using focus groups with medical students, and allied health students. This data will be used to develop an understanding of a number of issues, which relate to the participants’ career intentions. A purposive sample will be recruited for the focus groups.

The study will use maximum variation sampling, and will include focus groups, which are homogeneous in relation to one of these factors

X positive rural intention (rural club members and holders of rural scholarships selected on the basis of rural interest)

X no commitment to a rural careerundergraduate course

X no commitment to a rural careergraduate course

X male gender

X female gender.

182

Two focus groups will be conducted with each of these groups, one with students who are early in their undergraduate training and the other with students who are late in their undergraduate training. The focus groups will consist of 6 to 10 participants in each group.

This sampling method will provide sufficient data to identify the issues of importance in the phenomenon being studied. The focus groups will be audiotaped, and transcribed.

The text of the transcriptions will be analysed for content and themes. The content of the transcriptions will be coded using specific codes developed in relation to the research questions. Data will be interpreted to identify common themes and patterns.

Principal findings

The themes identified in the data analysis will be presented, and if there are differences between the student groups the differences will be discussed

Discussion

This project will present data on which further research can be developed to inform strategies for the recruitment of medical graduates into rural practice.

Implications for policy, delivery or practice

This research has important implications for policy in relation to strategies to recruit recent medical and health professional graduates into rural practice.

National Female Rural General

Practice Project

H Tolhurst, University of Newcastle,

N Lippert, J Talbot

Objectives

The aims of the project were to:

X identify and describe mechanisms/structures currently in place to support female GPs in rural and remote locations at the practice, community, divisional, organisation,

State/NT and Commonwealth levels and assess their effectiveness;

X identify additional support mechanisms that could be introduced to attract and retain more female GPs in rural and remote practice;

X identify the types of practice structures and working arrangements in which female GPs in rural and remote communities prefer to practice.

2002 General Practice and Primary Health Care Research Conference — Proceedings

Methods

Research commenced on the project in

September 2000 and was completed in June

2001. During the course of the research, indepth interviews were conducted with approximately 120 female GPs in selected rural and remote Divisions of General

Practice in every State and the Northern

Territory. Ninety per cent of the interviews were face-to-face interviews with the female

GP at her practice, home or other convenient location, e.g. Division Office. The remainder of the interviews were by telephone.

Transcripts of the interviews were analysed for content and a number of consistent themes were identified.

Principal findings

Despite demographic, geographical and practice differences, three consistent themes emerged from the interviews with the female rural GPs. These themes were the women’s commitment to rural life and work, and the need for the principles of flexibility and fairness to form the basis for the development of rural health policies and programs to support female GPs in rural and remote practice, and in rural medical training.

The women interviewed wanted policies and programs developed, which reflected the need for spouse employment, maternity leave, child care, retraining on re-entry to the workforce, social support, flexible work and training arrangements.

The issue of fairness was also identified as being a crucial factor underlying an environment in which the women interviewed would feel valued, welcomed and supported.

The women interviewed expressed a desire that the contribution that they make to rural practice be recognised and valued by their colleagues as well as their communities. In particular, they wanted fair and flexible practice structures; clarity and fairness in employment conditions; improved remuneration for some types of work such as long consultations, and provision of mental health services; availability of child care for after-hours work; and the development of after-hours safety provisions.

Discussion

The priorities of study participants in relation to professional and social support needs differ from those identified in research on rural doctors as an aggregate group where women are in the minority. There is the need for the development of strategies directed at specifically recruiting and retaining women in rural practice.

Implications for policy, delivery or practice

The results of this study have been used in the development of recommendations from

GPPAC to the DHAC in relation to the development of policy on rural recruitment and retention. The work of registered nurses in the context of rural and urban general medical practice settings in Australia: toward a model of effective practice.

Building research capacity together: a description of the NSW Primary

Health Care Research Capacity

Building Program

V Traynor on behalf of the NSW Primary

Health Care Research Capacity Building

Program (NSW PHC)

Objectives

NSW PHC aims to increase the capacity for primary health care (PHC) research and to provide leadership in developing a research culture in PHC (including general practice).

The objectives of the program are to:

X develop the research skills of primary health care workers

X establish structures to support research in primary health care

X encourage and support the transfer of research findings into practice

X develop organisational capacity and culture in primary health care to support research in an ongoing way

X link the development of research capacity with the broader primary health care and research agenda

X explore and develop opportunities for additional funding for research and capacity building in primary health care.

Methods

NSW PHC is a collaborative endeavour between the four University departments of general practice and rural health at the

Universities of Newcastle, NSW and Sydney.

A range of stakeholder groups including consumers, allied health, Divisions of General

Practice, the Royal Australian College of

General Practitioners and the Commonwealth

Department of Health and Ageing (CDHA) are also represented. The Program has funding from the CDHA until 2004 and operates

2002 General Practice and Primary Health Care Research Conference — Proceedings

183

under a state wide 3 year strategic plan which provides the overarching framework within which partners implement locally relevant capacity building activities.

Principal findings

Since its inception in 2001, NSW PHC has a number of achievements, including: the establishment of pilot PHC research networks across NSW; development of education modules on conducting and using the results of PHC research; a program of evaluation workshops for GPs and Divisions; development of PHC research resources; implementation of regional meetings and conferences; input into state and national policy developments; and the establishment of multi-disciplinary collaborations around

PHC research and evaluation.

Discussion and implications for policy, delivery or practice

Building research capacity in PHC is both challenging and complex. Engaging such a broad target audience is difficult and attempts to effect real change must operate across local, state and national systems.

NSW PHC presents one approach for dealing with such complexity, through a state wide collaborative program of local capacity building activity. This approach has allowed us to more effectively utilise the diversity of experience and expertise across the program partners, thus building our capacity to deliver both a comprehensive and consistent program to our stakeholders.

Building the capacity of primary care practitioners to use research findings: an evidence-based approach to developing a curriculum in critical thinking

L Trevena, Department of General Practice,

The University of Sydney, M Kidd

Objectives

This paper outlines an evidence-based process for developing a curriculum in ‘Basic

Critical Thinking Skills’ for primary care providers.

Methods

A search of major databases and key journals was conducted to identify the literature on the effectiveness of strategies to increase research capacity. Effective strategies for increasing the uptake of research were mapped against the level of evidence for their effect, and the results of a fourteenregion consultation/audit process of the primary care sector in NSW. Gaps were identified and a list of desirable skills for acquisition via a CME curriculum was drafted.

Finally, the available evidence of the effectiveness of different educational modalities, their cost and accessibility was tabulated.

Principal findings

Many strategies have been identified as being effective in increasing research uptake in practice. Recall/reminder systems, and academic detailing can improve evidencebased prescribing, preventive health practices and disease management. The consultation/audit process conducted within this study showed that these were currently being implemented by other groups in

Australia. Effective education strategies for increasing research uptake included: basic training in searching Medline; learning critical appraisal skills; clinical audit and feedback; and use of tailored risk communication to facilitate guideline uptake with patients as partners. General practitioners prefer evidence-based summaries for use in clinical practice, a substantial proportion do not know the meaning of basic terminology used in research publications, and general practice publication rates fall well-short of other disciplines despite its pivotal role in health care delivery in most countries. Mapping this against our audit/consultation process showed a striking absence of CME activity in all of these areas with most CME activity on clinical topics or practice-management IT.

Although there is good evidence that educational initiatives must be interactive, relevant and relate closely to daily practice, our audit/consultation process highlighted significant access difficulties for many rural practitioners in achieving this. We therefore note the emerging appreciation of web-based interactive learning environments as a possible alternative for some practitioners.

Discussion

The process described above therefore combines current evidence on the effectiveness of known educational strategies with the development of a local consultation and audit-based implementation plan. The resultant curriculum in Basic Critical Thinking

Skills will be available nationally as a set of clinically relevant units available free of charge via ‘Medeserv’ as an on-line interactive learning experience or as a smallgroup resource package for face-to-face delivery. One topic per fortnight is planned for release between May and October 2002.

Aspects of the course will be shown during the presentation of this paper to illustrate the results.

184

2002 General Practice and Primary Health Care Research Conference — Proceedings

Implications for policy, delivery or practice

It is anticipated that this course will provide an accessible and relevant resource to fill an identified gap in CME that will result in an increase in the skills and confidence of

Australian primary care providers to find and apply research in their daily practice.

Introduction to the National

Divisions Youth Alliance

R Trigger, Management Committee, National

Divisions Youth Alliance

Despite the deteriorating psychosocial health of youth people in Australia … The strengthening of relationships between young people and GPs has the potential to significantly improve current and future population health outcomes.

This requires research into more innovative and effective ways of delivering health services to young people from national, state/territory and local perspectives, as well as making current services more appealing to youth consumers.

A number of Divisions of General Practice,

State-based branches of the Australian

Divisions of General Practice (SBOs) and individual general practitioners have been working to improve general practice service delivery to young people. Centres for

Adolescent health Many more recognise that there is a significant need now for a more concerted effort to address youth health issues.

The National Divisions Youth Alliance (NDYA), an initiative driven by general practice, provides a central body for the development of partnerships with stakeholder bodies, coordination of youth health initiatives, examination of best practice models and collation of information on the existing skills and innovations already evident across many

Divisions of General Practice.

The NDYA program aims to increase the profile of youth health within general practice and provide the Divisions and their members with information, strategies and support for best practice in delivering youth health services.

In 2002, NDYA will undertake a national mapping exercise to gather information on youth health focused programs conducted within the 123 Divisions of General Practice.

NDYA is currently collating and analysing divisional responses to a scoping questionnaire designed to collect preliminary data on divisional programs to help inform the development of a more sophisticated and evidence-based consultative methodology.

NDYA will be seeking nominations to form a reference group with academic and consumer representation and other key stakeholders later in the year to act as an advisory body to the Alliance.

Length of consultation and GP characteristics in Australian general practice

H Britt, L Valenti, AIHW General Practice

Statistics and Classification Unit, University of Sydney, GC Miller

Objectives

To describe the length of consultations in

Australian general practice and relationships between consultation length and selected GP characteristics.

Method

BEACH is a continuous national study of general practice activity enrolling about

1,000 GPs per year. Information is collected about GP and patient characteristics, and morbidity and treatment for 100 consecutive encounters per GP. Additional studies on specific topics are conducted with subsamples.

The GP characteristics of interest in this study were GP sex, GP age group (<45, 45–

54 and 55+ years), and practice location

(urban vs rural; where urban = RRMA groups

1 and 2 and rural = RRMA groups 3, 4, 5, 6 and 7).

This study employs a cluster sample survey design, with the GP as the cluster. The mean consultation length and 95% confidence intervals were calculated in SAS, using the method described by Kish

1

, to calculate robust standard error to adjust for clustering.

Principal findings

This sub-study involved 929 GPs and included 30,776 encounters designated as claimable from Medicare as A1 items of service and for which start and finish times were recorded between April 2000 and March

2001.

The mean consultation length for these encounters was 14.8 minutes (95%CI:

14.52–15.07) and median consultation length was 13 minutes, with a range of 1 to 106 minutes.

2002 General Practice and Primary Health Care Research Conference — Proceedings

185

Female GPs conducted consultations which were significantly longer than their male counterparts. At the 9,839 encounters with female GPs mean consultation length was

15.9 minutes (95%CI: 15.38–16.32) while the 20,937 encounters with male GPs had a mean length of 14.3 minutes (95%CI:

13.97–14.64).

Urban or rural practice setting was not associated with consultation length, mean consultation length being 14.8 minutes in urban areas and 14.7 minutes in rural areas.

The age group of the GPs was not associated with consultation length. The average length of consultations with GPs aged <45 years was 14.6 minutes, compared to 14.7 minutes for GPs aged 45–54 years and 15.1 minutes for GPs aged 55 or more.

Combining GP sex, age group and practice location, the group with shortest mean consultation time (13.3 minutes, 95%CI:

12.7–13.9) was young (<45 years) male GPs practicing in urban settings. The group with the longest (mean) consultations (16.7

minutes, 95%CI: 14.3–19.1) was female GPs aged 55 or older, practicing in rural locations.

Discussion

There is very little current and reliable data available on consultation length in Australian general practice. This paper goes some way toward addressing this lack of published data by describing consultation length in relation to several GP characteristics. Further research into relationships between GP characteristics, patient characteristics, morbidity managed and consultation length is required.

Implications for policy, delivery or practice

These results have significant implications for future workforce planning in general practice.

The longer consultation length of female GPs combined with increasing feminisation of the

GP workforce could result in a decrease in number of consultations provided.

Reference

1. Kish L 1965. Survey Sampling. New

York: John Wiley & Sons.

186

Improving the exchange of existing electronic clinical information between acute and primary care: a re-engineering of communication systems

A Wallace, School of Rural Health,

University of Melbourne, K Kranz, G Pullen,

B Hsu-Hage

Objectives

Information management is increasingly important in clinical practice, yet continues to exist in a variety of paper and digital formats. This project seeks to develop a bridge between these formats allowing effective, timely, and secure transfer of data between acute and primary care providers while being compatible across both modern hardware and legacy systems.

Methods

The approach that will be deployed and studied is based on a single secure server that operates as an electronic mailbox not connected to the Internet or any other wide area communications network. Only authenticated users (recognised data sources and recipients) will be granted dial-up access to this stand-alone computer. Initially the project scope will be limited to physicians’ letters and can be readily expanded to include the transfer of any other existing electronic information from GV Health, subject to the availability of appropriate information and infrastructure and patient consent.

To retrieve relevant patient data from an acute care setting, the GP Practice will dial the dedicated mailbox, go through the authentication process and then collect their files whereupon they will be imported into the practice software for integration with the clinical records. Initially, PIT files are used for their flexibility in associated data tags and inter-operability with current medical practice software. It is proposed that individual practice locations will be able to specify their preferred data format affording the system more flexibility. Additional file formats may then be added as they evolve. It is planned that supported formats at present would include JPEG, DOC, TXT, and PDF files.

Discussion

Consistently the public hospital system’s uptake of electronic records has lagged behind that of the private general practice sector following the Commonwealth’s Practice

Incentive Program. This project re-engineers the electronic communication systems aiming to foster the primary and acute health

2002 General Practice and Primary Health Care Research Conference — Proceedings

ongoing partnership and provide enhanced quality of patient care.

Implications

Potentially this project stands to deliver multiple benefits to the organisation and running of both public hospitals and general practice. Public hospitals will be able to electronically send information in any desired digital format to multiple clients in a secure and cost-effective manner, both in absolute financial terms and in term of manpower.

General practices will be spared from lost or incomplete medical records and thus will be enabled to make more accurate clinical decisions. This is long-needed system that will benefit public hospitals, general practice, and consumers through an effective information communication and management strategy.

A shared care model for cancer patients

AM Ward, Department of General Practice,

University of Western Australia, D Joske,

D Joseph

Aim

The primary aim of the project was to develop a specifically designed shared care delivery partnership between GPs and specialists for haematology cancer patients.

The proposed outcome was improved communication and coordination of care between the hospital and community services and increased involvement of general practitioners in the overall delivery of cancer patient care.

Objectives

X

To develop the communication strategies for the shared care process including:

X

– a patient-held record (PHR) to supplement existing record systems and

GP guidelines for the management of common side-effects of chemotherapy.

To develop strategies to improve the rapport between GPs and cancer specialists.

Methods

In order for shared care to address the major problems and concerns of the key stakeholders, i.e. general practitioners, specialists, patients and nurses, these groups were interviewed.

The shared care model was then developed based on the results of these interviews. The model comprised four components:

X timely specialist — GP communication

X the patient-held record (PHR)

X a nurse coordinator — senior oncology/haematology nurse

X

GP upskilling package.

The shared care model was implemented for a period of six months for the pilot project.

Twelve haematology cancer patients took part in the pilot. An evaluation was conducted in which all participants including

GPs, patients, nursing staff and the haematologist were interviewed and a clinical audit of the accuracy and content of the PHR was undertaken.

Principle findings

X

Improved communication (both quality and amount) between cancer specialist and GPs.

X

Increased communication between patients and their health care practitioners.

X

Increased GP confidence in taking a greater role in the management of patients with cancer.

X

Improved timeliness of communication due to faxing letters and the PHR.

X

Reduction in the number of visits to hospital, particularly for rural cancer patients.

X

Patients felt empowered and more secure with the process and welcomed GP involvement.

Discussion

This project has been innovative in addressing stakeholder issues across the hospital/community interface. GPs are now an integral part of patient care during their cancer treatments, where before this was impossible due to poor communication and poor recognition of the GP role. The model is also highly transferable and a RCT is planned with other cancer patient groups.

Implications for policy, delivery or practice

The issues addressed by the model are universal in the Australian health care system. The number of patients diagnosed with cancer is increasing and they require ongoing treatment and support. Currently

2002 General Practice and Primary Health Care Research Conference — Proceedings

187

care for cancer patients is divided between specialists in hospitals, GPs and other community services. This division is the cause of many problems in communication and coordination of care between GPs and the hospital sector This model has been designed to address this division of care.

SARNet, The South Australian

Research Network for Primary

Health Care — a practical strategy to meet the challenges of capacity building

E Farmer, K Weston, PHC RED Program,

Department of General Practice, Flinders

University

This presentation describes a practical strategy to develop the research capacity of primary care practitioners through the creation of a research network. Such a network has the potential to help overcome some of the many system challenges to capacity building and change, such as limited funding, and lack of communication and support.

At Flinders PHCRED we have developed a model for research and evaluation capacity building aimed at all Primary Health Care

(PHC) practitioners. We suggest practitioners fall into four natural groups according to their level of research and evaluation experience and expertise. The four groups encompass practitioners with potential to develop research skills, those already participating to a small degree in research, those managing their own projects(s) and/or undertaking a higher degree, through to those academic practitioners with or undertaking a doctoral qualification. A practical strategy for involving practitioners at all levels and building capacity is to create a research network.

A number of practice-based research networks are in place in countries such as the

United Kingdom and are regarded as

‘research laboratories as essential to advancing the scientific understanding of medical care as bench laboratories are to advancing knowledge in the basic sciences’

(Green and Dovey, 2001). Various models have been created and various strategies employed in the development of these networks. The approaches which best suit the local Australian context will be discussed and form the basis for our whole-system approach to creating a research network, namely SARNet.

SARNet (The South Australian Research

Network for Primary Health Care) is open to all PHC professionals, has no joining fee, and will aim to meet the needs of urban and rural practitioners, practices, groups, Divisional and Department of General Practice staff, general practice registrars and students.

Some practical strategies to be adopted through the network include provision of bursaries (to start or continue a research project), writing grants of $500 (for grant application or manuscript preparation), and assistance to attend an annual meeting, workshop or course.

In addition, the appointment of a joint

Research Fellow between a local Division of

General Practice and Flinders Department of

General Practice will promote collaborative research between these organisations and enhance networking within the primary health care community. Mentoring through the research network is essential for developing and sustaining a new culture of research-aware and research-oriented practitioners.

The combination of these strategies also expands the pool of research-aware and research experienced mentors and role models in Divisions and in primary care practice, which will have a significant additional impact on the vocational training of general practice registrars. Arguably, earlier and more effective involvement of registrars in research and evaluation training is the most powerful way of creating a longterm and sustainable culture change.

The management of depression in general practice

I Wilson, Department of General Practice,

The University of Adelaide, K Duszynski,

A Mant

Objective

To explore the management and progression of depression in general practice.

Methods

The Medic-GP database of medical records from 59,000 patients, from 6 practices around Australia, covering the period July

1994 to July 2000, was interrogated to determine patients diagnosed with depression during the first year of the database (July 1994 to June 1995).

Computerised records for a random sample of these patients was examined in full to extract data on managements used, outcomes and recurrences.

188

2002 General Practice and Primary Health Care Research Conference — Proceedings

Results

A total of 5,889 patients identified from original search, 600 were randomly selected and examined in detail, with 382 (63.7%) found to be diagnosed with depression.

Females represented almost 60% of the overall sample of 382 patients and 57.3% of identified patients were diagnosed during the

12-month period.

The modal age group for women was 25–34 years and for men 35–44 years.

Antidepressants were prescribed for 95.3% of diagnosed patients. Of those patients whose therapy was initiated during the year

52.2% received an SSRI, while 38.1% received a TCA. The most common SSRI was fluoxetine and the most common TCA, dothiepin. Almost one-quarter of patients started on an SSRI were changed to another drug, most commonly a TCA (12.2%).

Almost one-half of patients started on a TCA required a change of therapy, most commonly to an SSRI (35.9%). SSRIs were most commonly initiated in those aged < 55 years, while TCAs were most common in those aged > 55 years. Most prescriptions

(54%) were taken for less than 9 months.

Anxiety (8.9%) and drug abuse (5.2%) were the most common co-morbid conditions. Of the 382 patients followed for the 5 years of this study, 155 (40.6%) were noted to have recovered and 14 (3.7%) attempted suicide.

There were no completed suicides. Referrals to other therapists were made for 21.6% of the patients and hospital admission occurred for 7.3%. Of the 155 patients who were noted to have recovered, 50.3% had a recurrence. The mean disease free interval was 1.4 years. Almost 30% were said to have recovered with antidepressant prescription for less than 3 months.

Discussion

This study reflects the diagnosis and treatment of depression of a cohort of patients in 1994–95. It highlights the chronicity of depression. Of those diagnosed as being depressed, 40.6% of patients were noted to have recovered and 50.3% of these were subject to a recurrence. In many ways depression behaves ‘more like asthma than appendicitis’.

Almost all patients in the cohort were treated with medication. The preponderance of short courses and the high recovery rates suggests that the antidepressants may not have been warranted for what may be minor depression.

Implications

This study confirms the chronicity of depression and raises questions concerning the treatment of depression in general practice. In particular it indicates the almost universal use of antidepressants and the tendency to short duration usage, which is not in line with current thinking.

Integrated primary mental health care — in practice

H Winefield, Department of Psychology,

University of Adelaide, D Turnbull,

J Marley, J Taplin, J Clifford

Objectives

A research team of general practitioners and psychologists explored the feasibility for both

GPs and patients, of a model of integrated primary mental health care. Major issues were the acceptability to both GPs and their patients of referral to an onsite clinical psychologist, the use made of the psychologist by staff and clients of the practice, the outcomes for both during the first year of this arrangement, and the sustainability of the arrangements in financial terms.

Methods

A clinical psychologist works three days per week at a large general practice, with full integration of casenotes and accounting systems. A representative sample of patients

(N = 466, a 78% consent rate) provided data on levels of psychological distress and attitudes to help-seeking at the practice.

Participants who accepted the possibility of help for psychological distress via the practice were followed up after an average of

22 weeks and re-assessed. Information is being collected from the 12 GPs working at the practice, as a one-year follow-up to the data presented at this conference in 2001, which will show their experiences of having the psychologist onsite.

Principal findings

Patients completing the Depression Anxiety and Stress Scale showed 16–17% prevalence of moderate to extremely severe distress.

Eighty participants (17.2%) expressed a wish to discuss the feelings described in the questionnaire with somebody from the practice staff. When interviewed 5 months later, about half had in fact done so, mostly with their GP and having raised the issues themselves. Most (90%) found these discussions quite or very helpful and average

2002 General Practice and Primary Health Care Research Conference — Proceedings

189

psychological distress had significantly reduced over the follow-up period.

In addition to seeing clients for evidencebased treatment of psychological problems upon referral from the GPs, the psychologist helps GPs with assessment and medico-legal reports, helps GPs learn what psychology can offer and which patients are likely to respond well, is helping the local Division of General

Practice to select routine screening measures for patient psychological state, and testing the suitability of the practice as a training placement for (5th and 6th year) clinical psychology Masters students.

Discussion

Our preliminary study has established the effectiveness and feasibility of collaborative

GP-psychologist care for primary health care patients with high-prevalence mental disorders. In addition to providing direct treatment the psychologist engages in transfer of knowledge with the GPs in ways relevant to the Better Outcomes in Mental

Health Care initiative. This early trial of the model needs to be extended in order to test its generalisability. In so doing the knowledge base and research skills of both professions will be enhanced.

Implications for policy, delivery or practice

The researchers seek opportunities to extend the model in a quasi-experimental design of greater scientific rigour. Through such efforts the natural alliance of physical and mental health primary care providers may improve the well-being and productivity of

Australians.

Proactive asthma care revisited: updated results from an Australian pilot study

N Glasgow, R Yates, Academic Unit of

General Practice and Community Care,

Canberra Clinical School, P Dugdale,

J Beilby, C Holton

Objective

To evaluate a system of proactive asthma care on asthma health outcomes in 5–12 year old children in the Australian Capital

Territory (ACT) and to report updated results of this pilot study.

Design

Pragmatic, twelve-month, randomisedcontrolled trial commenced in May 2000 involving 24 GPs and 174 children with carerdefined asthma, cluster-randomised into two groups by medical practice.

190

Intervention

Proactive asthma care implemented as the

3+ Visit Plan plus prompted reminders.

Control

Usual asthma care.

Participants

All ACT primary school children who returned a year 2000 school health survey and who additionally: consented to involvement in an asthma study; had a survey response suggestive of moderate-to-severe asthma; reported having a regular GP who also consented to participate in the Trial.

Method

All patients received asthma care from their regular general practitioner (GP) but control patients (n=73) received usual care from 12 control GPs whilst intervention patients

(n=101) received a structured system of proactive asthma care from 12 intervention

GPs. At trial entry, both groups were prompted to attend their GP for an initial asthma review. After this, intervention patients only were reminded to attend by the researchers prompting Practices to recall patients according to the 3+ Visit Plan schedule.

Main outcome measures

Process measures: number of proactive GP asthma review visits post Trial entry; number of written asthma action plans administered; number of children completing the 3+ visit plan.

Clinical measures: number of emergency department (ED) attendances for asthma; number of days absent from school for asthma; number of symptom free days.

Biological measures: Lung function changes

(spirometry).

Principal findings (awaiting verification)

At baseline, groups were equivalent on key markers. At Trial end, the following results were found for the intervention group compared to the control group:

Process measures: Significantly more proactive GP asthma review visits (rate ratio for proactive visits in last 12 month: 3.75,

95% CI: 2.37, 5.91, p <0.001); significantly more written asthma action plans (OR: 2.36,

95% CI: 1.06, 5.28, p = 0.04); significantly more children completing the 3+ Visit Plan

(OR 13.60, 95% CI: 3.13, 59.15 p <0.001).

2002 General Practice and Primary Health Care Research Conference — Proceedings

Clinical measures: Trends towards improvements were seen in ED attendance for asthma, school absence for asthma and number of symptom free days, but these did not reach statistical significance.

Biological measures: No consistent differences were found between groups on measures of lung function.

Discussion

Prompting works. Recall for review leads to greater GP attendance and more proactive care. The system of proactive care implemented in this Trial is viable for both patients and practitioners and shows trends towards improved asthma health outcomes, such as decreased emergency department attendance for asthma, less days absence from school and increased number of symptom-free days. However, the uptake of such care is not total. This may be due to both patient and practitioner factors. As this has implications for the delivery and impact of proactive care, these factors need elucidation in order to develop optimal asthma management systems which can be more widely embraced. Focus groups (not reported here) have been conducted to this end.

Low survey response rates — Can they be increased? Can they be used?

R Yates, Academic Unit of General Practice and Community Care, Canberra Clinical

School, N Glasgow

Objective

To examine the effect of different distribution methods and incentives on response rates for three health-related questionnaires.

Method

Cross-sectional surveys conducted between

1998 and 2000 using three carer-completed survey instruments (SI): SI1 — an 8-item general health and development questionnaire given annually to all ACT primary school new entrants (approximately

4500 4–6 year old kindergarten children) as part of a routine health screen and including an optional Maternal and Child Health

(MACH) nurse assessment; SI2 — a validated

28-item respiratory health questionnaire given to a subset of kindergarten children

(approximately 1000 per year) according to their responses to SI1; SI3 — a 16-item respiratory health and chronic disease questionnaire administered initially to all year

1–6 ACT primary school children

(approximately 28,000 7–12 year old children) in April 2000 and subsequently given as a resurvey to a sample of approximately 1000 of these same children, randomised by school, in September 2000.

SI1 and SI2 constitute the kindergarten screen whilst SI3 constitutes the child health survey.

Distribution methods: For all questionnaires, distribution was by school. Children were urged to return questionnaires within a week of delivery but with actual collection time two weeks post distribution. For the kindergarten screen, distribution methods varied by year.

In 1998 and 1999, SI1 and SI2 were both distributed separately but lag time in delivery of the two questionnaires differed from between one to six months in 1998 to between four to six weeks in 1999. In 2000, both questionnaires were given together (in one package) with carers requested to selfselect eligibility for completion of SI2 according to written instructions. For the child health survey (SI3), the use of incentive prizes rather than distribution method differed between the two surveys. In the initial survey (April 2000), no incentive prizes were used, whereas in the resurvey

(September 2000), children returning questionnaires within one week received a small prize. (Surveys returned within the second week were still collected but no incentive prize was given.)

Results

Kindergarten screen response rates remained at approximately 80% for SI1 over all three years. However, response rates for SI2 rose from 52% when the two questionnaires were given separately (1998 and 1999) to 80% when the questionnaires were given together

(2000). Child health survey (SI3) response rates were 52% when no incentives were given and rose to 92% when incentive prizes were used.

Discussion

Delivery methods and incentives can raise survey response rates. However, incentive prizes add to survey costs and are ethically difficult. In our experience, acceptable response rates are also attainable by varying the method of delivery of two questionnaires.

When carers are asked to self-select completion of a second questionnaire (SI2) included with a first questionnaire (SI1), response rates for SI2 increased from 52% to 80%. There may be several reasons for this. Firstly, by including both questionnaires together, lag time between questionnaire completion is eliminated. This may increase the relevance of the second questionnaire to

2002 General Practice and Primary Health Care Research Conference — Proceedings

191

the first and so increase response rate. A second possibility is colour preference. (In

2000, for ease of identification, SI2 was printed on blue paper). It is possible that people prefer to complete blue rather than white questionnaires. A third possibility in our example is that a subtle incentive is present.

Specifically, SI1 includes the offer of a free

MACH nurse assessment. When SI2 is presented together with SI1 it is possible that the second (unrelated) questionnaire may be linked to this subtle incentive and account for the increase in response rate from 52% to 80%. Linking surveys to legitimate incentives is therefore desirable but may be difficult to achieve in practice.

Given this, understanding the viability of low response rates is important. Ways in which we have addressed this will be discussed.

Patient perceptions of worthwhile treatment for chronic low back pain

M Yelland, Centre for General Practice,

University of Queensland, P Schluter,

M McKernon

Objectives

The literature provides no data on the definition of ‘worthwhile treatment’ for chronic low back pain, an area replete with treatments with little sound evidence for effectiveness. This study aims to describe patients’ perceptions of minimum and desired reductions in pain and disability in a treatment program.

Methods

Data were collected from all 110 patients who enrolled in a randomised controlled trial on the treatment of chronic low back pain.

Prior to treatment, they were asked prior to treatment to state their minimum acceptable percentage reductions in pain and disability and then what percentage reduction they desired with treatment.

Principal findings

The median (interquartile range) minimum acceptable reduction was 25% (20%, 50%) for pain and 35% (20%, 50%) for disability.

This compared with desired reductions of

80% (60%, 100%) for pain and 80% (50%,

100%) for disability. The internal consistency of pain and disability responses was high.

There was a significant association between minimum acceptable reductions and desired reductions, but no association between these two factors and patient age, gender, severity or duration of pain or ratings for disability, anxiety or depression. The minimum acceptable reductions for pain and disability

192

12 months after enrolment in the trial were poorly correlated with those at enrolment.

Conclusions and implications for practice

There is a wide disparity between the minimum acceptable reduction and the desired reduction in pain and disability in chronic low back pain sufferers facing a regime of injections and exercises. Inquiring directly about patients’ perceptions in these areas is important in realistic goal setting as other patient characteristics do not predict their responses. The high internal consistency for pain and disability responses suggest that asking about pain reduction alone is sufficient.

Randomised controlled trial of prolotherapy injections, saline

injections and exercises in the

treatment of chronic low back pain

M Yelland, Centre for General Practice,

University of Queensland, P Glasziou,

N Bogduk, P Schluter, M McKernon

Objectives

This trial aimed to assess the effectiveness and durability of a prolotherapy and exercise protocol in the treatment of chronic low back pain. Prolotherapy is a treatment that aims to strengthen weakened ligaments, which are believed to be a common cause of chronic musculoskeletal pain. The prolotherapy protocol used for chronic low back pain in general practice in Australia involves the repeated injection of tender ligaments with a solution of 20% glucose and 0.2% lignocaine to stimulate inflammation and subsequent strengthening of ligaments. This leads to a reduction in pain and disability. This process is aided by flexion/extension exercises and vitamin and mineral supplements.

Methods

The study was open to adults with nonspecific low back pain on most days in the preceding six months and who had failed other treatments. Using a 2x2 factorial design, participants were randomised primarily into a prolotherapy injections group or a saline injections group and secondarily to an exercises group or normal activity group. Trial staff and participants were blinded to the type of injections given.

Outcomes included levels of pain, disability, physical (PCS-12) and mental (MCS-12) component summary scales and medication usage. They were administered by blinded, independent observers before treatment and two and a half, four, six and 12 months after the commencement of treatment.

2002 General Practice and Primary Health Care Research Conference — Proceedings

Principal findings

One hundred and ten eligible, consenting participants with a mean duration of pain of

14 years, were enrolled between April and

November, 2000. Follow-up rates were 99% at six months and 96% at 12 months.

In all groups, mean pain and disability levels showed significant reductions from baseline at all follow-up assessments. These ranged from 26% to 44% for pain and from 30% to

44% for disability, with no significant differences occurring between groups. The proportion in each group at six and 12 months achieving at least 50% reduction in pain ranged from 33% to 50% and at least

50% reduction in disability ranged from 34% to 49%, with no significant differences occurring between groups. PCS-12 results improved significantly in all groups, but there were no changes in MCS-12 results and medication usage.

Discussion and implications for practice

In chronic non-specific low back pain, ligament injections, with or without exercises, result in significant and sustained reductions in pain and disability, but no attributable effect has been shown for glucose lignocaine over saline or to exercises.

Possible explanations for these responses include the counterirritation effect of the needles, the distension of ligaments with solution, the vitamin/mineral supplements, the placebo response and regression to the mean over time. The results cast doubt on the purported mechanisms of action of prolotherapy. Its role in practice is only where less invasive therapies have failed and the patient has been fully informed of the potential benefits and risks of prolotherapy.

Evaluating general practitioners’ views on the enhanced primary care items for care planning and case conferencing: a one-year follow-up

TM Blakeman, N Zwar, Department of

General Practice, South Western Sydney Area

Health Service, MF Harris

Objectives

Previous interviews with general practitioners in South West Sydney found a need for organisational and educational support to enable adoption of the Enhanced Primary

Care Medicare items for care planning and case conferencing. A one-year follow-up study was conducted with the same GPs to evaluate use of these items and identify strategies that influence their uptake.

Methods

A qualitative study of GPs’ responses to a semistructured face-to-face interview, conducted between September and

November 2001. GPs were asked about use of the items, perceived value, strategies facilitating uptake, difficulties encountered and further support required. Thematic analysis of the GPs responses was done from field notes and tape recordings of the interviews.

Principal findings

Twenty-five of the 30 GPs in the original study were available and consented to the follow-up. Of these 17 had performed care plans and 11 case conferences in the previous 12 months. Benefits of using the items were improved patient understanding and satisfaction, improved communication with other health professionals. The GPs predominantly identified practice and divisional strategies as facilitating uptake of the items. Barriers to uptake were time, organisation problems such as the complexity of arranging case conferences and problems with care plan forms and lack of knowledge of the items by other health professionals.

Over half the GPs who had undertaken case conferences felt that the time taken to set up the conference limited their use.

Recommendations included the need for practice, divisional and health service personnel to be involved in identifying suitable patients and actively supporting care planning and case conferencing.

Discussion

GPs were positive about the value of the items but continue to face a number of obstacles to incorporating care planning and case conferencing into their practices. One year on from the first set of interviews time, organisation, education and resources remain key issues raised by GPs.

Implications

Further support for care planning and case conferencing is needed and involves organisational and service developments and reorientation in general practices, GP divisions and health services. This project is an example of researchers trying to respond to a major change in how GPs work and provide evaluation in time frames that can inform policy and practice at a number of levels — GP practice, Divisions, state funded health services and the Commonwealth.

2002 General Practice and Primary Health Care Research Conference — Proceedings

193

Chronic pain among general practice patients: a study of prevalence and management

E Comino, Department of General Practice,

Fairfield Hospital, O Hermiz, N Zwar

Chronic pain is an important reason for consulting general practitioners (GPs) and is responsible for significant morbidity. Yet few studies have investigated the impact of chronic pain among general practice patients.

The aim of this study was to assess patient report of chronic pain and treatment used.

Method

A cross-sectional study of consecutive patients aged more than 20 years was conducted at 34 general practices in South

Western Sydney. Patients completed a questionnaire that asked about the presence of chronic pain described as ‘pain that had lasted for more than three of the last six months’, it’s severity and what treatment they had received. Patients with dementia or pain due to cancer were excluded.

Results

1,614 patients attending 33 practices participated. Applying the study criteria, 677 patients (41.9%) patients reported chronic pain. Compared to patients without chronic pain, Patients with chronic pain reported a significant reduction in their health status and quality of life. When categorised according to severity, 22.6% had mild,

51.0% moderate and 26.4% severe pain.

77.0% (n= 521) of patients had taken medication for chronic pain during the previous month with the majority of these

(66.5%, n=346) taking it at least once a day and most (91.9% n= 479) reporting some relief from pain. However there were many patients using medication for chronic pain

(33.4% n= 174) who agreed with the statement that they needed more or stronger medication. Overall, most patients with chronic pain (66.1%) believed that they had received as much information as they needed for their pain. During the last month, 28.8%

(187) of patients with chronic pain reported that they had seen a specialist, 9.0% had seen another GP and 16.4% (111) another health professional.

Conclusion

These results suggest that, for adults attending general practice in South Western

Sydney, pain is an important issue. Not only is it responsible for significant morbidity, among those with chronic pain it results in significant interference with daily activity, consumption of medication and use of

194

specialist services. These results suggest that there are opportunities to develop interventions to address these issues in the general practice setting that will help patients to achieve better pain relief. Further research is now under way to develop exercise and lifestyle interventions suitable for application in the general practice setting to further assist patients who experience chronic health problems such as chronic pain that may benefit. These will involve development of partnerships with other health and community-based organisations.

2002 General Practice and Primary Health Care Research Conference — Proceedings

Breakfast sessions

Joining the IMET (n=1 trials) program: a new program of research needs collaboration

Professor Paul Glasziou, Professor Chris Del Mar

Centre for General Practice, The University of Queensland Medical School

This workshop will cover the theory and practice of using Individual Medication Effectiveness Tests

(also known as n-of-1 trials) in routine general practice settings. Such ‘trials’ provide practical clinical tools that improve the precision of pharmacological management, where individual response to treatment is variable. Examples of their useful application include stimulants for

ADHD, and NSAIDs for osteoarthritis or chronic pain.

The Centre for General Practice at The University of Queensland is offering an Individualised

Medication Effectiveness Testing service to determine whether a patient is a responder to a particular medication in certain conditions. Doctors throughout Australia, and their patients, are able to take advantage of this free, ongoing service, which is accessible by post and telephone methods, making it suitable for rural and remote patients as well as those in metropolitan areas.

This introductory session will:

X discuss the purpose and methods of IMETs

X when to use them

X how to request an IMET

X how to interpret the results

X pitfalls in setting up your own IMETs for new clinical topics.

The workshop will cover some example cases to show how IMETs have improved patients’ clinical management.

Note: This is an introductory session. A series of follow-up workshops will be offered at various venues around Australia during the next 12 months to assist others in locally developing the use of

IMETs.

2002 General Practice and Primary Health Care Research Conference — Proceedings

195

Are health inequalities widening or narrowing across the SES gradient? Current issues in health inequalities and implications for a national research agenda

Convenors

HIRC- PHC Network Liz Harris, CHETRE; John Furler, Department of General Practice, University of

Melbourne; Network Coordinator: Julie McDonald ([email protected])

Professor John Catford

Dean of the Faculty of Health and Behavioural Sciences at Deakin University; Board member of the Health Inequalities Research Collaboration

Dr John Furler chaired the meeting, which was attended by 30 delegates. John explained the background of the HIRC PHC network.

HIRC was established in 1999 to develop a health inequalities research agenda within Australia and aims to be intersectoral and multidisciplinary. Its strategic objectives are:

X

X

X

X increase national focus on reducing health inequalities build national capacity and support for research and development in health inequalities establish close collaboration among researchers, practitioners and policy developers promote the uptake of research findings in policy, practice and evaluation.

As one initiative to meet its goal and strategic objectives, HIRC has funded the development of three research and development networks. The networks are funded for 2 years. These networks are primary health care; child, youth and family; sustainable communities.

Each network is to emphasise Indigenous, rural and remote populations. HIRC will not run a separate Indigenous research agenda, but will collaborate with organisations engaged in

Indigenous health research. There are regular teleconferences between the networks, as a network panel, to share information about activities and members and to strengthen coordination. The commonality between the PHC and Child, Youth and Families networks particularly requires a coordinated approach. Each network reports to HIRC annually through the network panel.

PHC Network aims

X

To provide leadership and champion a focus on health inequality through PHC.

X

X

X

To develop a broad consensus of priority issues for the PHC Research Network and HIRC.

To build the capacity of PHC health equity research and development.

To provide a range of mechanisms for information exchange and support.

Speaker: Prof John Catford

Prof Catford presented data from the Victorian Burden of Disease Study that can be accessed through http://www.dhs.vic.gov.au/phd/topics/research.htm.

Analysis of relevant local data for trends over time demonstrates that changes are occurring quite rapidly, and that some gaps between advantaged/disadvantaged are narrowing.

Inequalities in gender can change rapidly over a decade. Eg lung cancer mortality following major reduction in tobacco smoking.

196

2002 General Practice and Primary Health Care Research Conference — Proceedings

Geographical differences in life expectancy are apparent in rural areas and urban fringes, especially in males. Why are some population groups changing faster than others? We don’t know the answer.

In terms of years of life lost, there is a trend to closing of the gap between SES groups from 1996 to 1999. This may be an artifact due to methods of measurement of groups, but it is worthy of investigation, to find out whether increased employment status, or increased quality of primary health care services has contributed. The answer would help in terms of public policy.

Northern Territory project: Strong Women, Strong Babies, Strong Culture. Focus on families and mothers, addresses issues of low birthweight, builds on cultural norms and seems to have some impact.

Change in risk factors. Lifestyle and health behaviours account for large part of burden of disease,

Many factors contribute to decreased prevalence of smoking, but improvement is observable. The smoking rate among unemployed has at last dipped down. It is exciting to suggest the gap may be closing in smoking prevalence.

Ambulatory Care Sensitive (ACS) admissions study. Copies of study are available on public health website of Victorian Dept of Human Services (www.dhs.vic.gov.au/phd/topics/research.htm). ACS examples include avoidable hospital admissions such as acute asthma, vaccine preventable admissions for pneumonia.

Factors affecting ACS: disease impact and prevalence, impact of prevention, propensity to seek

PHC, socio-economic barriers to care, quality and effectiveness of primary care, hospital utilisation patterns and barriers. No well developed published studies in Australia comparing ACS. Yet admission rates for asthma vary by primary care partnerships, region. These data have been driving the focus on primary care investment with the Department of Human Services, in the ongoing battle for resources.

Future steps

1 Health promotion and early intervention, new ways of working and thinking, collaboration across sectors at community level.

2 Research: We need

– intervention studies

– studies to explain observed change: Analysis and interpretation of trends in local level data can complement intervention research, and usefully inform policy.

– interdisciplinary, inter-sectoral, multilevel partnerships

3 Information systems: We are very data poor, but learning the power of information and the power of databases. We need to be helpful in planning future strategies.

– good local area information systems

– good data on access, and interpretation of these data.

The PowerPoint slides for this session are available at: www.phcris.org.au/events/conference_frameset.html

2002 General Practice and Primary Health Care Research Conference — Proceedings

197

Conference evaluation report

Belinda Lowcay, Research Officer, Primary Health Care Research and Information Service

Key findings

X

The conference attracted 274 delegates from every State and Territory in Australia, representing academic departments of general practice and rural health, Divisions of General

Practice, Government departments, other health organisations, research centres, and the

Royal College of General Practitioners.

X

The conference was rated as good or excellent by most delegates and the conference met their expectations.

X

The diversity of the insights the delegates listed they gained from attending the conference shows how the conference program provided many benefits for each of the people who attend.

X

The Keynote and Plenary speakers were most frequently listed as being the best part of the conference, followed by the opportunities to network.

X

Pre-conference workshops, included for the first time, complemented the conference program, with interactive sessions to explore specific issues in-depth.

X

The ‘New methods to discover best practices in health care’ workshop was particularly successful, with 85 participants attending.

X

There is strong delegate support for the GP & PHC Research Conference to continue.

I must say I found the conference really enjoyable and valuable. I was able to catch up with some former colleagues as well as meet a number of people who are working in similar research areas to me. I had a number of discussions which prompted further (useful!) thoughts about my own

PhD research, and I found a range of papers & posters in which I was interested. In short, I really felt like I was in the right place! (Academic Dept)

My third attendance — it is easily the best GP research conference and a MUST. (RACGP)

I got a lot out of the primary health care research conference last week, in particular Susan

Horn’s address on Friday morning which challenged some of the current wisdom about cost cutting interventions in primary care delivery. (Victorian State Office, DoHA)

About the conference

The 2002 GP & PHC Research Conference was held in Melbourne on 30 and 31 May, at the Carlton

Crest Hotel. The conference was funded by the Commonwealth Department of Health and Aged

Care and convened by the Primary Health Care Research & Information Service (PHC RIS). The conference theme was both timely and relevant to a wide audience of researchers and practitioners, with a high number of poster and paper abstracts submitted.

The theme ‘Research: Making a difference to health and health care’ was reflected in the keynote addresses and in presentations throughout the pre-conference workshops and the two-day conference.

The evaluation

Overall, the conference was well received, with favourable feedback, constructive suggestions and thoughtful comments. The conference met delegates’ needs to increase their awareness of current

GP&PHC research, present their own work, and establish or maintain relationships with others working in the GP&PHC field. A total of 128 conference evaluation forms were completed and returned (47%) (see Appendix I for the paper version of the questionnaire).

198

2002 General Practice and Primary Health Care Research Conference — Proceedings

Conference attendance

The conference was promoted in a number of ways, primarily using a targeted approach with personal invitations posted to past conference attendees, and by email to allied and public health organisations. Other PHC RIS services were used such as the Email Alert mailing list, the newsletter GPinfonet, and the website. Delegates heard about the conference in more than one way*, primarily by personal invitation (32%), Email Alert (32%), word of mouth (29%) and

GPinfonet (23%).

*multiple responses given

Table 1

Personal invitation

32%

How delegates heard about the conference (n=128)

Email alert

32%

Word of mouth

29%

Gpinfonet

23%

PHCRIS

Website

9%

* multiple responses given

Other

9%

Total

n=171

134%*

The conference attracted 274 delegates from every State and Territory in Australia who work in general practice and primary health care (GP&PHC). Eighty-one conference delegates also attended the pre-conference workshops (as well as a further 69 participants making 150 workshop participants in total).

The conference attendance list shows that delegates were primarily from academic departments of general practice and rural health (52%), Divisions of General Practice (15%), Government departments (11%), other health organisations (10%), research centres (5%) and the Royal

College of General Practitioners (3%) and held positions as outlined in Table 2.

Table 2

Research/

Project

Officer

38%

Positions held by delegates (n=127)

Academic

35%

GP

20%

Policy

Maker

11%

Other

6%

* multiple responses given

Consumer

Rep.

3%

Health

Practitioner

2%

Total%

N=148

115%*

Overall rating of the conference and attendee expectations

The majority of delegates (87%) gave the conference an overall rating of good or excellent.

Table 3 Overall ratings of the conference (n=125)

What is your overall rating of the conference?

Excellent

29%

Good

58%

Fair

13%

Poor

0

My third attendance — it is easily the best GP research conference and a MUST. (ID 16,

Research/Project Officer, RACGP)

Reasons for attending were*: to hear about current GP and PHC research projects, results and developments (41%); to meet with researchers and professionals working in the GP&PHC field

(34%); to present their own work (23%), and; to increase their knowledge and learn new methodologies (12%). Ninety per cent indicated that the conference had met their expectations adequately or more than adequately.

*multiple responses given

2002 General Practice and Primary Health Care Research Conference — Proceedings

199

Table 4 Ratings of conference meeting expectations (n=125)

Did the conference meet your expectations?

More than adequately

34%

Adequately

56%

Not adequately

10%

Conference aims

The Conference focused on five key areas:

X raising awareness of current GP&PHC research

X identifying research questions relevant to consumers, practitioners, managers and policy makers

X selecting rigorous qualitative and quantitative methodologies that are appropriate to the purpose and setting of the research

X developing viable and sustainable ongoing partnerships between researchers and stakeholders to design and conduct research, and to enhance the dissemination and uptake of research findings

X building the capacity of the research workforce to undertake research, and the capacity of managers, practitioners and policy makers to use research findings.

Half of the delegates indicated that the first aim had been well met, and the other aims had been somewhat met.

Table 5 Ratings of how well the conference aims were met

The conference increased my:

awareness of current research ability to identify relevant research questions ability to select rigorous methodologies ability to develop and maintain partnerships ability to disseminate and uptake capacity (as researcher) capacity (as policy maker/practitioner) total

A lot

50%

12%

7%

30%

19%

8%

10%

20%

Somewhat

46%

61%

62%

49%

59%

55%

49%

54%

Not at all

3%

25%

27%

18%

20%

26%

19%

20%

N/A

1%

2%

4%

3%

2%

11%

22%

6%

Keynote speakers, presentations and networking

There were two International and two Australian keynote speakers, who addressed the following questions in their presentations:

X

X

X

X

How do we identify the most important and relevant research questions?

What is the best way to conduct research for maximum impact?

How do we translate research findings into policy and practice?

How do we overcome system challenges?

General practice and primary health care research was presented over the two-day conference during concurrent paper and poster sessions. In these sessions, 54 papers and 90 posters were presented. There were opportunities for networking during breaks in the sessions.

200

2002 General Practice and Primary Health Care Research Conference — Proceedings

Most delegates rated the keynote speakers as good or excellent (82%), the concurrent sessions and the posters as good or excellent (84%) (Table 6); the two day format as an appropriate length of time (98%); and the opportunities for networking as being enough (84%) (Table 7).

Table 6 Ratings of the keynote speakers, concurrent sessions and posters (n=124)

How would you rate the:

keynote speakers?

concurrent sessions?

posters?

totals

Excellent

30%

25%

26%

27%

Good

52%

59%

55%

56%

Fair

14%

14%

18%

15%

Poor

3%

1%

0

1%

N/A

1%

1%

1%

1%

Table 7

n=124

Ratings of the two day format and opportunities for networking

What do you think of the length of the conference?

Too short

1%

Too many

Regarding opportunities to network, there were: n=126 3% totals 2%

Appropriate

98%

Enough

84%

91%

Too long

1%

Not enough

13%

7%

Insights gained from attending the conference

Delegates were asked to list up to 3 new insights they had learnt, or work practices they intended to change, as a result of attending the conference. The 209 insights gained were diverse and covered a broad range of issues, such as methodology (50%), the development of partnerships

(34%), avoiding duplication of research (20%), the research and policy relationship (15%), consumer involvement (13%), and specific ideas or skills. The diversity of insights gained shows how the conference program provided many benefits for each of the people who attend.

Be innovative in selecting research topics and really assessing the evidence before doing research. (ID 24, Honors student)

Randomised controlled trials are an appropriate methodology for some research but we need to acknowledge that they have their limitations, especially in general practice where the patients have complex problems and where all sorts of factors impact on the appropriateness of an intervention. (ID 5, Academic/GP, Uni/private practice)

The non-linear relationship between research questions and methodologies. (ID 82,

Academic/Research Officer, Uni)

Further consider ways of influencing policy, taking into account the policy making process...(ID

39, Research/Policy Officer, SBO)

Politicians, policy and other external factors can lead areas of research. (ID 86, Statistician, Uni)

...the idea that we fund ‘safe’ research in Australia, and sometimes we should take a risk with funding... (ID 44, Research/Project Officer, Uni)

Researchers still grapple with how to involve consumers in research. (ID 12, Research/Project

Officer, Uni)

The value of undertaking a literature review to determine what (if any) similar research is being undertaken across Australia and internationally, so as not to reinvent the wheel. (ID 18, Policy

Maker, Govt)

The need for stakeholders to work together to formalise a research agenda for Australia. (ID 18,

Policy Maker, Govt)

Getting involved in GP research and using available data. (ID 90, GP, private practice)

2002 General Practice and Primary Health Care Research Conference — Proceedings

201

More coordination of research activities is essential. (ID 101, Research/Project Officer, Division)

When, and when not to, collaborate — common goals, complementary skills. (ID 108, Academic,

Uni)

The huge gap between research and policy, and the perceived inability by researchers to cross it.

(ID 122, Policy Maker, Division)

The importance of reducing fragmentation and hence duplication of GP&PHC research (especially research programs being run by Divisions). There appear to be a lot of researchers doing similar work, or work that has been done previously. (ID 38, Research/Project Officer, Uni)

What was liked best

The Keynote and Plenary speakers (in particular Susan Horn and Simon Stewart) (33%) were most frequently listed as being the best part of the conference, followed by the opportunities to network

(28%), posters (11%), particular presentations (9%), the opportunity to hear about current research (7%), papers (6%), and, conference content (6%).

....in particular those [sessions] given by Prof. S. Horn and S. Stewart. Relevant and focused on what is achievable by proper and appropriate research. (ID 57, Research/Project Officer)

The opportunity to be challenged in our ways of thinking about research, such as Susan Horn’s and James Dunbar’s presentations and the extremely professional presentations of Simon Stewart and Susan Horn. (ID 106, Research/Project Officer)

Presentations by Susan Horn and Simon Stewart [were] illuminating and thought provoking. (ID

127, GP)

.... I like the ‘smorgasbord’ effect of the conference, of hearing what’s happening around

Australia... (ID 39, SBO)

The two paper sessions I attended....were refreshing as they were held in an environment of exchange and collaborative thinking. (ID 25, Policy Maker, Govt)

What could be improved

Delegates mostly did not like particular aspects related to the venue, such as room set-up, lighting, and seating, cigarette smoke, and temperature (18%). They also indicated that the keynote speakers were not challenging enough (14%), there was not enough discussion time or

Q&A time (12%), and that there were attendance problems associated with the scheduling of concurrent sessions (11%).

I refuse to attend any conference ever again at this hotel. PHC conferences should be held in smoke free venues. (ID 102, Research/Project Officer, Uni)

Some of the keynote speeches were less than inspiring... (ID 126, Academic, Uni)

There was an absence of discussion on some of the most pertinent issues effecting PHC at the moment. (ID 3, Research/Project Officer, Govt/Uni)

Limited time to discuss methodology issues (ID 4, Academic GP, private practice/Uni)

More interaction between audience and speaker. Less didactic oral presentations. (ID 50,

Academic, Uni)

Too many formal presentations. Could have had a couple of facilitated workshops. (ID 104,

Academic GP, private practice)

Conference organisation

Organising the conference involved a number of factors, some being: pre-conference preparation

(conducted by Conference Logistics); the development of a cross referenced handbook (which included maps, the program, abstracts of papers and posters, and an index of presenters) that was printed and distributed to all delegates; a web page of conference information was included on the PHC RIS website; The Carlton Crest Hotel was booked as the venue; conference meals and a

202

2002 General Practice and Primary Health Care Research Conference — Proceedings

conference dinner on the Thursday evening were catered for by the Carlton Crest Hotel; Coco’s

Lunch were booked to entertain during the dinner.

Almost all delegates rated the pre-conference organisation as excellent or good (96%); the handbook as excellent or good (98%); the venue, catering, dinner as good or excellent (81%), and; the web page as good (36%). Thirty-four per cent had not accessed the web page for conference information.

Table 8 Ratings of the organisational aspects

How would you rate the:

conference organisation? n=126 venue?

catering?

dinner?

handbook?

web page?

totals n=127 n=124 n=119 n=127 n=121

Excellent Good

51%

28%

35%

21%

49%

21%

35%

45%

43%

55%

24%

49%

36%

42%

Fair

3%

22%

6%

8%

2%

7%

8%

Poor

1%

7%

3%

2%

0%

2%

2%

N/A

Did not attend=54

Did not access=41

13%

Handbook projected a higher quality of preparation than the 2001 handbook. (ID 66, Research

Officer, Uni)

Further comments

Chairs of sessions need to ensure that particular individuals do not dominate question sessions — important to make an effort to give as many people as possible an opportunity to ask questions.

(ID 94, Academic, Uni)

... we should promote this conference overseas, so the audience (not just the speakers) take on an international flavour. (ID 13, Academic, Uni)

There should be fewer papers and more interactive sessions (such as workshops, forums etc). (ID

22, Academic, Uni)

People who are simply reporting on a completed project or one in progress should be confined to poster sessions unless they are prepared to generalise from their project. (ID 22, Academic, Uni)

Was pleased to be able to attend the conference. While there were a few of the usual playing safe saying the usual thing, there were also some great speakers, such as Horn, and I appreciate the chance to be able to listen to them. I’ve organised conferences before and know how much work goes into them and that you can’t please everyone — so well done! (ID 45, Policy Maker &

Research/Project Officer, Govt)

The Australian (academic) professors of medicine need to inject more thought into their presentations and we need positive people with workable solutions... Every year we hear the same negativity. Speakers like Dr Horn and Prof Wilkin are great because they are so positive and focus on how we can make a difference. (ID 61, Manager, Division)

Lack of State involvement. (ID 79, Policy Maker, Govt)

Why is this conference not attended by more Divisions? (ID 96, Research/Project Officer,

Division)

I was disappointed with the lack of discussion around the PHCRED program, with no provision in other programs made for this specifically. Given it is meant to be a major driver in the PHC research area, its progress and lessons learnt seemed to be sadly missing... (ID 106,

Research/Project Officer, Uni)

I would suggest specific advertised slots for special interest groups to meet. Greater audience participation — not just Q&A. (ID 108, Academic, Uni)

Seems unfair to restrict completed work to poster dissemination route, whilst giving more time to incomplete work. (ID 109, Academic, Uni)

2002 General Practice and Primary Health Care Research Conference — Proceedings

203

I liked the poster sessions and the way they were grouped. (ID 118, Academic, Uni)

I think it would have been far better if abstracts were vetted more strictly and so fewer papers and posters were accepted — it should be quality not quantity! (ID 114, Academic, Uni)

Sponsor more novice GP and PHC researchers. (ID 62, GP, private practice)

... I think the concurrent sessions and posters highlighted the disgraceful dearth of funding currently available for new PHC research. (ID 106, Research/Project Officer, Uni)

Other indicators of conference achievements

During the month after the conference (June 2002), the conference page on the PHCRIS website was accessed 183 times, with Susan Horn’s power point presentation accessed 44 times and

James Dunbar’s accessed 21 times.

Immediately after the conference various PHCRIS staff members received unsolicited feedback from a number of delegates, including keynote speakers. Comments included:

Delegate from: Institute for Clinical Outcomes Research, and the University of Utah, USA

Thank you for all your and your staff’s efforts to put together the pre-conference course and notebook, and for the whole conference. It was delightful working with your team, who were so competent. There was no stress because of all your wonderful planning (email, 4/6/02).

I thought the conference was terrific (email 8/6/02).

Delegate from: National Primary Care R&D Centre, Manchester, UK

I very much enjoyed the conference and particularly enjoyed meeting you and your colleagues again. Thank you very much for inviting me and for all your hospitality. I think you and your colleagues did a brilliant job of organising the event, and you should be proud of the achievement. I am sure you felt exhausted by Friday evening. Apart from the conference itself, I had many interesting discussions with people over the course of the conference (email, 5/6/02).

Delegate from: Greater Green Triangle University Department of Rural Health, Flinders and Deakin

Universities

I thoroughly enjoyed the conference and met lots of interesting people. Your choice of presentations and the way the meeting was arranged worked really well. Thanks for the chat on

Wednesday evening. I will follow up on the options (email, 3/6/02).

Delegate from: R&Q, GP Branch, DoHA

Thanks Belinda for the [email] reminder about the evaluation — a great idea. Also thanks for your hard work behind the scenes — it is much appreciated (email, 5/6/02).

Delegate from: GP Branch, DoHA

Excellent conference. Posters: about time to impose some standards, and to have posters reproducible electronically for the website. I personally don’t like the mini presentations — prefer to walk around and talk to the owners.

AAAGP: integrate their presentations with the conference (email, 8/6/02)

Delegate from: Victorian State Office, DoHA.

I got a lot out of the primary health care research conference last week, in particular Susan

Horn’s address on Friday morning which challenged some of the current wisdom about cost cutting interventions in primary care delivery (email, 4/6/02).

Delegate from: Policy Officer, General Practice Divisions, Victoria

I enjoyed the conference, although was unwell so unfortunately missed some of the sessions I’d been looking forward to. The organisation was impressive, and the range of presentations really interesting (email, 4/6/02).

Delegate from: Department of Epidemiology & Preventive Medicine, Monash University

I must say I found the conference really enjoyable and valuable. I was able to catch up with some former colleagues as well as meet a number of people who are working in similar research areas

204

2002 General Practice and Primary Health Care Research Conference — Proceedings

to me. I had a number of discussions which prompted further (useful!) thoughts about my own

PhD research, and I found a range of papers & posters in which I was interested. In short, I really felt like I was in the right place! There seemed to be a bit of defensiveness about perceived attacks on RCTs, which I found interesting (a bit misplaced do you think)? There seems to be a bit of resistance from some quarters perhaps, to the broadening of focus from ‘GP’ to ‘PHC’, with an associated broadening of the range of research questions, methods and approaches. As a nonmedical person, I certainly felt comfortable with the breadth of focus and range of approaches.

One of the best sessions I attended, in terms of promoting interesting discussion, was the one on multi-disciplinary health care. Hope you also found the conference enjoyable — well done to

PHCRIS on organisation (email, 3/6/02).

Delegate from: the University of Newcastle

Congratulations to everyone at PHCRIS on an excellent and very interesting conference — it was great to hear some new international speakers presenting new ideas and research which I had not previously heard about (email, 5/6/02).

Delegate from: Centre for General Practice Integration Studies, School of Public Health and

Community Medicine Uni of NSW

Excellent follow up! [re emailed conference evaluation] (email, 4/6/02)

Appendix I Evaluation questionnaire

2002 General Practice and Primary Health Care Research Conference — Proceedings

205

206

2002 General Practice and Primary Health Care Research Conference — Proceedings

Index of presenters

Albert, E .........................................128

Aldrich, R ......................................... 70

Allwell, L, ......................................... 94

Amir, L ...........................................128

Artridge, J .......................................129

Askew, D .................................. 88, 130

Atkinson, V....................................... 88

Bailey, L.................................... 89, 131

Barclay, L........................................132

Barton, C ........................................132

Bastian, H ........................................ 45

Beacham, B............................... 90, 133

Berrisford, K ....................................134

Bettess, A .......................................135

Binell, M........................................... 91

Blignault, I ....................................... 91

Boase-Jelinek, K...............................136

Boreland, F......................................137

Bridges-Webb, C ............................... 92

Brookman, D ...................................138

Bubner, T......................................... 93

Burns, JR .................................. 94, 139

Butler, R, ......................................... 94

Byron, J ..........................................140

Catford, J ........................................196

Cavenagh, L ....................................141

Chalmers, E...................................... 95

Connelly, LB ....................................141

Cook, C...........................................128

Couzos, S......................................... 96

Del Fante, P...................................... 65

Del Mar, C .................... 51, 97, 142, 195

Deveny, E .......................................143

Dick, M ................................... 143, 144

Dunbar, J ......................................... 28

Dunbar, N ........................................ 98

Duszynski, A....................................145

Farmer, EA ......................................146

Fitzgerald, M..................................... 98

Freebody, P .....................................147

Furler, J ..........................................147

Gentle, I .......................................... 91

Gialamas, A.....................................148

Gill, G .............................................149

Glasziou, P ................... 50, 99, 149, 195

Goodyear-Smith, F ...........................150

Gorman, D ....................................... 99

Hansen, E .......................................151

Harris, C .........................................100

Harris, M ................................... 81, 101

Hawkes, A .......................................102

Heal, C ...........................................102

Hicks, RE ........................................152

Holton, C.........................................152

Horn, S ............................................ 35

Horsburgh, T ...................................153

Jacobs, B ........................................153

Jolley, G..........................................103

Joyce, C ..........................................154

Kalucy, L ...............................1, 86, 155

Laven, G......................................... 104

Lewi, F ........................................... 156

Litt, JCB.......................................... 105

Lowcay, B ................................156, 198

Lyle, D ........................................... 157

Mackson, JM.................................... 158

Malone, D ....................................... 106

Marker, J ........................................ 159

Martin, C .................................159, 160

McAvoy, B....................................... 161

McCallum, Z .................................... 106

Mickan, S.................................162, 163

Millard, F ........................................ 163

Mitchell, G ....................... 107, 164, 165

Morrissey, M ................................... 108

Mors, L ........................................... 166

Mutch, A ......................................... 108

Nagle, C ......................................... 166

Nelson, M ....................................... 167

Neumayer, B, .................................. 109

Nguyen, HT..................................... 110

O’Dea, K...........................................72

Oldroyd, J ................................168, 169

Patford, J ........................................ 169

Patterson, K ......................................21

Peady, L, ........................................ 109

Pedlow, K ....................................... 110

Pilotto, L ............................................ 1

Pirotta, M........................................ 111

Pond, D .......................................... 112

Powell Davies, G .............................. 170

Richards, J ...................................... 113

Roach, S..................................113, 171

Roche, AM ...................................... 114

Rose, V....................................114, 172

Rowbottom, J .................................. 173

Rowley, K .........................................72

Russell, G ....................................... 173

Ruth, D ...................................115, 174

Schattner, P .................................... 175

Scheurer, R..................................... 176

Sen Gupta, T................................... 176

Sherwood, J .................................... 116

Simpson, S ..................................... 177

Sims, J ............................ 117, 178, 179

Skinner, K.........................................72

Skinner, M ........................................72

Southern, D .................................... 179

Stewart, S ........................................57

Stocks, N ........................................ 117

Sulaiman, N .............................118, 180

Taft, A............................................ 119

Taylor, J ................................... 91, 181

Temple-Smith, MJ ............................ 120

Tolhurst, H...............................121, 182

Tolhurst, H...................................... 182

2002 General Practice and Primary Health Care Research Conference — Proceedings

207

Trasancos, C....................................122

Traynor, V.......................................183

Trevena, L............................... 123, 184

Trigger, R........................................185

Valenti, L ........................................185

Veitch, C .........................................123

Vines, R ..........................................124

Walker, C ........................................124

Walker, R ........................................125

Wallace, A .......................................186

Ward, AM ........................................187

Weston, K .......................................188

White, G .......................................... 72

Wilkin, D .......................................... 24

Wilson, I .........................................188

Winefield, H.....................................189

Wood, L ..........................................125

Yates, R .................................. 190, 191

Yelland, M .......................................192

Yelland, M .......................................192

Ziaian, T .........................................126

Zwar, N .................................. 193, 194

208

2002 General Practice and Primary Health Care Research Conference — Proceedings

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project