Chronic Diseases in Canada Inside this issue

Chronic Diseases in Canada Inside this issue
Chronic Diseases in Canada
Volume 29 · Number 2 · 2009
Inside this issue
47
48
49
56
70
80
89
93
The role of public health in the health
of Canada’s children
Guest editorial by D Butler-Jones, MD
From rags to genes: mixing
research paradigms
Guest editorial by C Infante-Rivard, MD
Health outcomes in low-income children
with current asthma in Canada
T To, PhD; S Dell, MD; M Tassoudji, MSc; C Wang, MD
Deprivation and unintentional injury
hospitalization in Quebec children
M Gagné, MA; D Hamel, MSc
Injury data in British Columbia: policy
maker perspectives on knowledge transfer
C Mitton, PhD; YC MacNab, PhD; N Smith, MA;
L Foster, PhD
Complementary therapies for cancer patients:
assessing information use and needs
MJ Verhoef, PhD; L Trojan, BSc; GD Armitage,
MA; L Carlson, PhD; RJ Hilsden, PhD
The national lung health framework:
an opportunity for gender analysis
N Hemsing, MA; L Greaves, PhD
Announcements
Chronic Diseases in Canada
a publication of the Public Health Agency
of Canada
Howard Morrison
Principal Scientific Editor
(613) 941-1286
Robert A Spasoff
Associate Scientific Editor
Claire Infante-Rivard
Associate Scientific Editor
Elizabeth Kristjansson
Associate Scientific Editor
Michelle Tracy
Managing Editor
(613) 954-0697
CDIC Editorial Board
Jacques Brisson
Laval University
Neil E Collishaw
Physicians for a Smoke-Free Canada
James A Hanley
McGill University
Clyde Hertzman
University of British Columbia
C Ineke Neutel
University of Ottawa Institute on
Care of the Elderly
Kathryn Wilkins
Health Statistics Division
Statistics Canada
Chronic Diseases in Canada (CDIC) is a quar­
terly scientific journal focussing on cur­rent
evidence relevant to the control and pre­
vention of chronic (i.e. non-communicable)
diseases and injuries in Canada. Since 1980
the journal has published a unique blend of
peer-reviewed feature articles by authors
from the public and private sectors and
which may include research from such fields
as epidemiology, public/community health,
bio­statistics, the behavioural sciences, and
health services or economics. Only feature
articles are peer reviewed. Authors retain
responsibility for the content of their arti­
cles; the opinions expressed are not neces­
sarily those of the CDIC editorial committee
nor of the Public Health Agency of Canada.
Chronic Diseases in Canada
Public Health Agency of Canada
130 Colonnade Road
Address Locator 6501A
Ottawa, Ontario K1A 0K9
Fax: (613) 941-3605
E-mail: [email protected]
Indexed in Index Medicus/MEDLINE
To promote and protect the health of Canadians through leadership, partnership, innovation and action in public health.
— Public Health Agency of Canada
Published by authority of the Minister of Health.
© Her Majesty the Queen in Right of Canada, represented by the Minister of Health, 2009
ISSN 0228-8699
This publication is also available online at www.publichealth.gc.ca/cdic
Également disponible en français sous le titre : Maladies chroniques au Canada
Guest editorial
The role of public health in the health of Canada’s children
Dr. David Butler-Jones, Chief Public Health Officer
Although Canadians are among the health­
iest people in the world, some groups of
Canadians are more likely to experience
poorer health and premature death than
others.1 In general, health status follows
a step-wise gradient where people in less
advantageous socio-economic circumstances
are not as healthy as those at each
subsequently higher socio-economic level.2
Those with the lowest incomes and educa­
tion, in combination with other related
factors – inadequate housing, poor working
conditions, detrimental health behaviours,
limited access to health care, and who
lack early childhood support and/or social
supports – are more likely to develop poorer
physical and mental health outcomes than
those living in better circumstances.
Canada has strong social policy foundations
that have helped to make it more egalitarian,
and thereby, healthier. Programs like the
Canada and Quebec Pension Plans, Old Age
Security, Employment Insurance, publicly
funded health care and universal primary
and secondary education have all helped
to establish a minimum standard of living.
However, after 20 years of declines, income
inequality has increased in Canada over the
last decade.3
Two papers in the current issue of
Chronic Diseases in Canada highlight the
importance of social and economic factors
as determinants of children’s health. The
paper by Gagné and Hamel4 reported that
children in the most materially deprived
areas of Quebec had significantly higher
risk of hospitalization from transportationrelated injuries and from poisonings and
fires. Risks for sports injuries, however,
were lower for the most disadvantaged. A
similar, but more modest risk gradient was
observed according to measures of social
deprivation.
Determinants such as the natural and built
environments create the context for other
determinants of health such as income,
Chronic Diseases in Canada
employment, social networks and personal
behaviours. In terms of injury prevention,
community design features such as recre­
ational pathways and sidewalks, safe levels
of lighting, and compatible land uses can
ensure pleasant, safe spaces for both recre­
ational and transit activities. The absence
of safe streets and recreational areas influ­
ence the risk of injuries. Moreover, social
and economic factors in deprived neigh­
bourhoods, such as household structures,
can also play a role in the risk of childhood
injuries.
Dr. To and colleagues reported that low
income adequacy was associated with
higher rates of hospitalization, but lower
rates of doctor visits for asthma, among
children with asthma,5 suggesting poorer
outpatient management of asthma among
children with low income adequacy. While
Canadians take justifiable pride in their
universally insured health care, income still
plays a role in medical access to specialists6
such as respirologists. Socio-economic
differences in asthma medication use,
which contributes to appropriate asthma
management, may have also contributed
significantly to the differences in asthma
hospitalizations by income.
Understanding the causes of these inequa­
lities and developing interventions that
reach these groups are essential elements
of public health action.7 It is not enough to
focus solely on individual health choices
and behaviours, as peoples’ actions are
very much shaped by the social and
environmental conditions in which they
live and work. A balanced mix of targeted
interventions for high risk populations and
universal programs for all is more likely to
work in a country as vast and complex as
Canada.1, 8,9 This kind of balance ensures
that, regardless of personal circumstances,
Canadians experience those conditions
necessary for better health and for making
healthy choices the easier choices.
47
References
1.
Chief Public Health Officer. Report on the
state of public health in Canada. Ottawa:
Public Health Agency of Canada, 2008.
2.
Wilkins R, Tjepkema M, Mustard C,
Choinière R. The Canadian census mortality
follow-up study, 1991 through 2001. Health
Rep. 2008;19(3):25-43.
3.
OECD. Growing unequal?: income distri­
bution and poverty in OECD countries.
Country note: Canada [Internet]. Available
from: http://www.oecd.org/dataoecd/44/48/
41525292.pdf. Accessed February 09, 2009.
4.
Gagné M, Hamel D. Deprivation and unin­
tentional injury hospitalization in Quebec
children. Chron Dis Canada. 2009;29(2):56-69.
5.
To T, Dell S, Tassoudji M, Wang C. Health
outcomes in low-income children with
current asthma in Canada. Chron Dis
Canada. 2009;29(2):49-55.
6.
Asada Y, Kephart G. Equity in health
services use and intensity of use in Canada.
BMC Health Serv Res. 2007 March 11;7:41.
7.
Macintyre S. Inequalities in health in
Scotland: what are they and what can we
do about them? The Scottish Government,
Medical Research Council Social and Public
Health Sciences Unit, 2007 October.
8.
Health Disparities Task Group of the
Federal/Provincial/Territorial Advisory
Committee on Population Health and Health
Security. Reducing health disparities –
roles of the health sector: discussion paper.
Ottawa: Public Health Agency of Canada,
2004 December.
9.
World Health Organization. Closing the
gap in a generation: health equity through
action on the social determinants of health.
Commission on Social Determinants of
Health, 2008.
Vol 29, No 2, 2009
Guest editorial
From rags to genes: mixing research paradigms
Claire Infante-Rivard MD, PhD, James McGill Professor, McGill University,
Associate Scientific Editor, Chronic Diseases in Canada
The journal is publishing two papers1,2 in
this issue that underscore the contribution
of social and economic determinants to
outcomes such as hospitalization and the
use of health services among children with
injuries and asthma. Both papers show that
greater use of health services is generally
associated with lower economic status and,
though not as markedly, with social status
as well. These observations are taking
place in Canada where there is universal
access to health care. Other findings are
also of interest; for example, the To et al.
study1 reports that, whereas children from
low-income families were at higher risk
of hospitalization regardless of severity of
asthma status, they had lower utilization
of physicians’ services. On the other hand,
children with persistent asthma were less
likely to come from low-income families
or to be born to immigrant mothers. In
addition, children of immigrant mothers
were less likely to be hospitalized. The
Gagné and Hamel study2 may have used
a finer definition of social and economic
status, captured by two, distinct ecological
variables. This study shows that both
material and social status influence not only
hospitalizations in children with injuries,
but also the occurrence of injuries.
From a research perspective, a number
of issues lend themselves to discussion
based on these papers: (1) the sustained
importance of accounting for social and
economic status (hereafter referred to as
SES) when studying determinants of health
and the utilization of health services; (2)
the methods used to measure SES; (3) the
use of administrative or research databases
for research; and (4) the way that SES
influences health outcomes.
We are without a doubt in an era in which
“gene-talk” has powerfully come to dominate
biomedical research. The technology used
to study population genetics is evolving so
rapidly, is so complex, and the arguments
Vol 29, No 2, 2009
in favour of a major role for genetic risk
factors in complex outcomes are so
powerful that the role of SES may be seen
as remote and insignificant by comparison.
However, arguments could easily be
made that (a) once genetic variants are
found to be associated with diseases or
conditions, it does not necessarily mean
that we understand how they work; and
(b) the role of these genetic variants is
most likely influenced by and linked with
a number of environmental factors. Given
the picture obtained from the studies
discussed above, the same conclusions
about SES factors could be reached. Thus,
the complexity of SES as a determinant of
health and the utilization of health services
is such that, as with genetic research, SES
research deserves resources. It should
also involve outstanding methodological
advances and remain among public health
research priorities. However, prudence
is warranted when advocating more and
better research in this area based mainly on
the assumption that SES can be acted upon,
in comparison with other determinants.
Changing SES may be as difficult or even
more so than developing gene therapy!
What direction should research in this
area possibly take? Given the assumption
that genes and related physiological
mechanisms are turned on and off by
our environment, SES factors included,
and that researchers in one area may be
less than highly competent in the other,
initiation of a dialogue between the
biological and public health/sociological
research communities would be a very
useful approach to better understand both
types of determinants. It could potentially
lead to outstanding findings, such that the
results are also better understood. Whereas
it is probably relatively easy to agree that
the pathways to injuries and asthma are
both genetic and environmental, it may
be more difficult to become convinced
that similar pathways influence the use of
48
health services. However, it does remain
plausible, although possibly even more
complex, and with different weights for
different determinants.
The authors of both papers in question
efficiently used available databases
(administrative or research-oriented) to
pursue their analyses, as well as measures
of social and economic determinants that,
although useful and acceptable, were likely
limited, due in part to the constraints of
database research. As a result, the reported
findings may give an impression of “déjà
vu,” and definitely leave many questions
from apparently contradictory findings,
either within the studies themselves or
between these and previously published
ones. However, the authors had descriptive
goals, which they properly pursued. Such
results are of high interest to this journal
and they are welcome; it seems particularly
opportune for the public health community
to use available databases of good quality.
However, we would also like to encourage
other innovative and bold approaches, the
use of more refined methods to measure
SES, and alliances with the biology com­
munity to better understand the factors at
play in health and the utilization of health
services.
References
1.
To T, Dell S, Tassoudji M, Wang C. Health
outcomes in low-income children with
current asthma in Canada. Chronic Dis
Can. 2009;29(2):49-55.
2.
Gagné M, Hamel D. Deprivation and
unintentional injury hospitalization in
Quebec children. Chronic Dis Can. 2009;
29(2):56-69.
Chronic Diseases in Canada
Health outcomes in low-income children with current
asthma in Canada
T To, PhD (1, 2); S Dell, MD (1,3); M Tassoudji, MSc (1); C Wang, MD (1)
Abstract
ascertained longitudinally provides us
with a means to delineate the “persistent”
asthmatics from the “transients,” and to
evaluate prospectively the impact of fac­
tors present during the preschool years
on outcomes in school-aged children. The
three main purposes of this study are to
create a population-based longitudinal
cohort of Canadian children with current
asthma; to measure their HSU and health
outcomes compared to children without
asthma; and to assess the impact of asthma
on HSU and health outcomes in children,
adjusting for other risk factors.
Data collected from the Canadian National Longitudinal Survey of Children and Youth
(NLSCY) in 1994/95 and 1996/97 were used to measure longitudinal health outcomes
among children with asthma. Over 10 000 children aged 1 to 11 years with complete data
on asthma status in both years were included. Outcomes included hospitalizations and
health services use (HSU). Current asthma was defined as children diagnosed with asthma
by a physician and who took prescribed inhalants regularly, had wheezing or an attack
in the previous year, or had their activities limited by asthma. Children having asthma
significantly increased their odds of hospitalization (OR = 2.52; 95% CI: 1.71, 3.70) and
health services use (OR = 3.80; 95% CI: 2.69, 5.37). Low‑income adequacy (LIA) in 1994/
95 significantly predicts hospitalization and HSU in 1996/97 (OR = 2.68; 95% CI: 1.29,
5.59 and OR = 0.67; 95% CI: 0.45, 0.99, respectively). Our results confirmed that both
having current asthma and living in low-income families had a significant impact on the
health status of children in Canada. Programs seeking to decrease the economic burden
of pediatric hospitalizations need to focus on asthma and low-income populations.
Methods
Keywords: asthma, low-income, hospitalization, health services use
Data source
Introduction
Asthma has been determined to be the
most common reason for preventable hos­
pitalization in children.1-4 It is generally
agreed that asthma is an “ambulatory care
sensitive condition,” i.e. good outpatient
management should result in decreased
hospitalizations.5 Inadequate control of
asthma can be costly; it has physical
consequences and can lower a person’s
quality of life. In 1998, the major direct
health care costs (hospital care and drug
expenditures) for asthma totalled over
$402 million.6 According to the Institute
for Clinical Evaluative Sciences report by
To et al., the total annual indirect and
direct costs were reported to be $676 per
Canadian child with asthma in 1995.7 The
largest single component was hospital
admissions, accounting for 77% of the total
cost. The use of other health services (i.e.
visits to general practitioner, specialist and
emergency department) contributed 21%
to the total cost.8
Most information on HSU and health
outcomes among children with asthma is
ascertained from cross-sectional surveys or
administrative records.9,10 To our knowledge,
no systematic longitudinal survey data
based on a population have been collected
to ascertain the prevalence of asthma in
various age groups, their HSU and asthmaassociated morbidities. The NLSCY, a
longitudinal Canadian population-based
survey, provides a unique opportunity
to study childhood asthma prospectively
and obtain stable estimates of asthma
prevalence. Specifically, the asthma status
Longitudinal data from the NLSCY cycle 1
(C1) 1994/95 and cycle 2 (C2) 1996/97
were used in this study. The NLSCY is a
prospective longitudinal survey designed to
measure child development and health.11,12
A multi-stage clustered sampling scheme
was used in the survey for the study sam­
ple to be representative of the Canadian
population of children. The clusters were
designed to have sufficient sample sizes
within large geographic areas and within
seven key age groupings. Trained inter­
viewers from Statistics Canada went to
households and administered standardized
questionnaires to the person most knowl­
edgeable (PMK) about the child, i.e. the
biological mother in 89.9% of cases. The
overall response rate to the survey was
86.4% for C1 and 91.6% for C2. Informed
consent was obtained from the legal
guardians and/or the child, as appropriate.
A full description of the NLSCY is available
Author References
1 Child Health Evaluative Sciences, Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada.
2 Department of Public Health Sciences, University of Toronto, Toronto, Ontario, Canada.
3 The Division of Respiratory Medicine, The Hospital for Sick Children; University of Toronto, Faculty of Medicine, Toronto, Ontario, Canada.
Correspondence: Dr. Teresa To, Child Health Evaluative Sciences, Research Institute, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario M5G 1X8,
Tel.: 416-813-8498, Fax: 416-813-5979, Email: [email protected]
Chronic Diseases in Canada
49
Vol 29, No 2, 2009
at http://www.statcan.ca/bsolc/english/
bsolc?catno=89F0078XIE11
The NLSCY was comprised of all children
sampled for C1 in responding households.
Although some children who were partic­
ipants in C1 did not participate in C2 for a
variety of reasons (e.g. random deliberate
attrition to decrease sample size, loss to
follow-up, or the biological parent not
completing the survey), efforts were made
to keep the number of these children to a
minimum so that the longitudinal research
by age cohort at the national level was still
permitted.12 Included in the analyses for the
current study were a total of 10 148 children
(i.e. a weighted sample of 3 128 645),
aged 1 to 11 years at baseline (C1), whose
biological parent had responded to the sur­
vey and who had complete data on asthma
status in both cycles.
Definition of current asthma
Current asthma was ascertained if the PMK
reported an asthma diagnosis in his or
her child by a health professional and at
least one of the following: 1) the child took
prescribed inhalants on a regular basis; 2)
the child had wheezing or whistling in
the chest or had an attack of asthma
in the previous 12 months; or 3) the child’s
asthma prevented or limited participation
in school or other normal activities.13
The asthma status in cycles 1 and 2 was
further categorized into four groups: 1)
“no asthma” (i.e. no current asthma in
either cycle); 2) “new asthma” (i.e. current
asthma in C2 only); 3) “transient asthma”
(i.e. current asthma in C1 only); and 4)
“persistent asthma” (i.e. current asthma in
both cycles). The only children included
in our longitudinal analysis were those
categorized as having “persistent asthma”
(PA) or “no asthma” (NA) (based on a total
of 9 462 children or a weighted sample of
2 908 136). In the longitudinal multivariable
analysis, the health outcomes of each child
over time were examined using the General
Estimating Equation (GEE) Model14-17
incorporating current asthma status in both
cycles while considering the correlation
within individuals.
Vol 29, No 2, 2009
Outcome measures
Statistical analysis
Health outcomes included parental reports
of child hospitalization and their HSU at
both C1 and C2. Hospitalization was defined
as an overnight stay in the hospital in the
past 12 months. HSU was defined as help/
advice the parents sought from a physician
about a child’s health, and was determined
by the PMK’s response to the question: “In
the past year, how many times have you
seen or talked about your child’s health
with a physician?” Both hospitalization and
HSU were coded as dichotomous variables
(i.e. yes/no) to minimize the impact of
recall bias.
GEE14-17 for categorical longitudinal data
was used to incorporate “time-dependent”
covariates in modelling predictors of HSU or
hospitalization. GEE was used to account
for the longitudinal effects of the predictors
on HSU or hospitalization in C1 and/or C2.
Data on predictors collected in both cycles
were used to measure the overall adjusted
probability of HSU or hospitalization. The
carry-over effects (i.e. the effect of a risk
factor in C1 on HSU or hospitalization in
C2) were also considered.
Risk factors
Other independent variables examined for
potential confounding included sex, age,
maternal and child health status, child
chronic conditions, child history of wheeze,
parental smoking, maternal history of
asthma, maternal symptoms of depression
(not previously examined in relation to hos­
pitalization and HSU in asthmatics), mar­
kers of socio-economic status (SES) and
maternal immigration status.
Child health status was based on the
mother’s rating of the child’s health (i.e.
excellent/very good or good/fair/poor). The
child’s health conditions were recorded if
the PMK reported a diagnosis of allergies
or bronchitis by a physician. Wheezing or
whistling in the chest at any time in the
last 12 months characterized the child as
having wheeze.
Maternal health status was based on her
self-rating of health (i.e. excellent/very good
or good/fair/poor). Parental smoking was
based on whether cigarettes were smoked
daily, occasionally or not at all.
Income adequacy was classified into low-,
middle- and high-income groups based on
total household income and the number of
household members.11,18 A child’s mother
was considered an immigrant if she ever
reported having immigrated to Canada. The
child’s age was measured by year and kept
as a continuous variable.
50
The data publication guides by Statistics
Canada were followed.11,12 Longitudinal sam­
ple weights derived by Statistics Canada11,12
were applied in analyzing study population
characteristics so that the derived estimates
could be considered representative of the
total population of children aged 1 to
11 years at baseline. Coefficients of variation
derived by Statistics Canada11 were used
to determine the quality of the estimates.
Accordingly, estimates that did not meet
the Statistics Canada criteria were flagged.
For multiple variable analyses, standardized
sample weights were used to preserve the
original sample size, thereby avoiding an
overestimation of significance while main­
taining the same distributions as those
obtained when using population weights.19
Only statistically significant variables (p <
0.05) or known confounders such as sex,
age, LIA and maternal immigration status
were included in the final models.20 The
correlations between the covariates were
examined. The SAS statistical package (i.e.
SAS version 8.0, Cary, North Carolina, USA)
was used for all analyses.21
Results
Demographics/characteristics
Table 1 shows the overall prevalence of
asthma, hospitalization and HSU by the
four categories of asthma status. The overall
prevalence of reported asthma was 8.4% in
C1 and 9.4% in C2. Using our previously
defined asthma classification scheme,
4.0% had new asthma, 3.0% had transient
asthma and 5.3% had PA. Overall, 87.6%
of children had NA in both cycles. The
prevalence of hospitalization was the high­
est in children with PA in both cycles (i.e.
Chronic Diseases in Canada
16.8% and 8.3%) and lowest in children
with NA. Compared to the NA children, the
hospitalization rate was four times higher
in children with PA (i.e. 16.8% vs. 3.8%;
p < 0.0001); their HSU was also higher
(i.e. 95.3% vs. 80.3%; p < 0.0001).
Table 2 shows a comparison of baseline
characteristics between the NA children and
those with PA. Overall, in children with PA,
there was a higher percentage of boys aged
1 to 11 years compared to girls (i.e. 62.6%
vs. 37.4%; p < 0.001). Children aged 9 to
11 years had the highest prevalence of PA
compared with children aged 1 to 4 years
and 5 to 8 years (i.e. 6.8% vs. 4.2%; p <
0.001 and 6.8% vs. 6.6%; p < 0.001). At
baseline, the NA children reported better
health status than those with PA (i.e. 91.8%
vs. 61.0%; p < 0.001). More children with
PA were hospitalized (i.e. 16.8% vs. 3.8%;
p < 0.001) and used health services (i.e.
95.3% vs. 80.3%; p < 0.001) when com­
pared with the NA children. Furthermore,
children with PA had a higher percentage
of allergy (i.e. 45.0% vs. 11.8%; p < 0.001)
and a lower percentage of mothers who were
immigrants (i.e. 13.9% vs. 18.8%; p < 0.001).
Estimates of HSU and hospitalization
Our regression model in Table 3 showed that
being younger with current asthma (OR =
2.52, 95% CI: 1.71, 3.70) or not having
excellent or very good current health status
(OR = 3.10, 95% CI: 2.38, 4.03) increased
the risk of hospitalization; however, a child
whose mother is an immigrant (OR = 0.65,
95% CI: 0.43, 0.98) decreased the child’s risk
of hospitalization. However, a child’s being
younger, having asthma (OR = 3.80, 95%
CI: 2.69, 5.37), having allergy (OR = 1.61,
95% CI: 1.33, 1.94) and not having a very
good or excellent current health status (OR =
1.56, 95% CI: 1.27, 1.92) increased the risk
of HSU.
The carry-over effects of the covariates were
examined. Among all covariates studied,
LIA showed a statistically significant carryover effect on hospitalization and HSU
(OR = 2.68, 95% CI: 1.29, 5.59 and OR =
0.67, 95% CI: 0.45, 0.99). This indicates
that LIA, as measured at baseline, had
lasting effects on the hospitalization and
HSU in C2. Cross-sectional logistic regres­
sion analyses using the same risk factors
Chronic Diseases in Canada
at baseline were also explored and similar
results were obtained (results not shown).
Discussion
Results from this longitudinal populationbased study quantify the magnitude of
hospitalizations and HSU in children with
current asthma. Our findings support the
hypothesis that childhood asthma has a
significant impact on the risks of all-cause
hospitalization and HSU. Few previous
studies have examined this relationship in
children. Chen et al. reported that among
Canadians aged over 12 years, asthma was
a significant risk factor for overall hospi­
talization, and that the odds ratio for asthma
differ by socio-demographic variables, such
as age and household income.22 These fin­
dings were based on cross-sectional sur­
vey data and did not take into account the
potential changes in asthma status over
time. In the current study, asthma status
was assessed longitudinally, both at base­
line and follow-up. A major strength of this
study was the ability to examine the tem­
poral relationships between risk factors and
health outcomes, and assess children who
are persistently asthmatic or non-asthmatic
over time. Children with persistent asthma
had almost triple the risk of hospitalization
and quadruple the risk of HSU compared
to those with no asthma. The longitudinal
nature of this study makes the results more
powerful, as the analysis incorporated the
risk factors and outcomes over time while
taking into account the correlation within
individuals between baseline and follow-up.
Our study showed that children from
low-income families were at higher risk
for hospitalization over time regardless
of asthma status, yet they had lower
HSU. Interestingly, other Canadian studies
have shown similar findings, especially
in hospitalization. A study conducted in
Manitoba showed that children aged 0 to
19 years in the lowest income quintile were
three times more likely to be hospitalized
than those living in the highest income
quintile in 1999.23 The Canadian Institute
of Child Health also reported a similar
trend in hos­pitalization by household
income among children.24 A recent study
in children born in a major Canadian
urban centre found that the relationship
51
between socio-economic disadvantage
and hospitalization for ambulatory caresensitive conditions (with asthma being
the most frequent diagnosis) and all-cause
hospitalization was large, consistent across
many conditions, remained stable over
time and persisted up to 9 years of age.25
Since Canada has a universal health care
system, access barrier to health care due
to affordability cannot account for these
differences. Factors leading to higher risk
of acute care use among children in the
lowest socio-economic stratum may include
higher disease prevalence, increased dis­
ease severity, multiple comorbidities,
poor health habits, crowded living condi­
tions, inconsistent patterns of preven­
tative care, nutritional problems and
poor physical fitness.25,26 Moreover,
socio-economic differences in the use of
inhaled corticosteroids have been reported
in children with asthma.27-30 Although
detailed information on asthma medication
use is not available in the current study,
it could have contributed significantly
to the differences in asthma and asthmarelated hospitalizations, which account for
a significant proportion of the overall mor­
bidity in the asthma population.31 Finally,
health system factors such as distribution
of specialist care may also contribute to the
differences in health outcomes. It has been
reported that in Canada, general practitioner
care is distributed fairly equally by income
according to needs; however, people with
higher incomes are significantly more likely
to seek specialist care than those with
lower incomes, making total doctor uti­
lization somewhat higher in the rich.32
Our finding on the distribution of asthma
by socio-economic status (SES) is also
interesting. We found that children with
persistent asthma were less likely to come
from low‑income families. This means that
the effect of asthma as a risk factor for health
care utilization could be confounded by the
effect of low income; however, this poten­
tial bias was minimized by adjustments in
multivariable analysis. Previous evidence
for the distribution of asthma by SES is
mixed, and findings varied depending on
the definition of asthma and the study
design. Cross-sectional survey studies
in the US found that SES was associated
with increased asthma prevalence in
Vol 29, No 2, 2009
Table 1
Prevalence of hospitalization and health service use by asthma status
Hospitalization
Asthma status
Health service use
(Cycle 1)
(n = 147 800)
(Cycle 2)
(n = 116 000)
(Cycle 1)
(n = 2 562 300)
(Cycle 2)
(n = 2 341 300)
%
%
%
%
n*
%
Persistent asthma
166 600
5.33
New asthma
126 000
4.03
6.07
5.84
89.84
92.94
94 600
3.02
9.41
4.75
93.78
76.72
2 741 500
87.62
3.77
3.29
80.32
73.03
Transient asthma
No asthma
16.77
8.29
95.27
91.45
* The number of children is weighted and values are rounded to the nearest 100; the unweighted n = 10 148.
Table 2
Characteristics of study population by asthma status
Total
Persistent asthma
No asthma
(n = 2 908 100)*
(n = 166 600)
(n = 2 741 500)
%
%
%
male
50.54
62.58
49.81
female
49.46
37.42
50.19
Baseline (cycle 1) characteristic
Sex
Age in years (row %)
1 to 4
37.87
27.62 (4.18)
38.50 (95.82)
5 to 8
35.14
40.46 (6.60)
34.81 (93.40)
9 to 11
26.69 (93.22)
26.99
31.92 (6.78)
Child with wheeze
13.69
85.22
9.34
Child with allergy
13.69
44.98
11.79
90.06
61.01
91.82
9.94
38.99
8.18
Child current health status
excellent or very good
good, fair, or bad
Low income adequacy
Immigrant mother
Biological mother with asthma
Health service use
Hospitalization
9.77
6.94 †
9.94
18.56
13.92 †
18.84
4.78
15.45
4.14
81.18
95.27
80.32
4.51
16.77 †
3.77
* The number of children is weighted and values are rounded to the nearest 100; the unweighted n = 9 462.
Percentages are adjusted for missing data and may not total to 100 due to rounding.
† The coefficient of variation is between 16.6% and 33.3%, which is considered marginal by Statistics Canada.
Vol 29, No 2, 2009
52
Chronic Diseases in Canada
Table 3
Adjusted odds ratios for hospitalization and health service use based on longitudinal logistic regression*
Hospitalisation
Health service use
p-value†
OR
1.57
NS
0.96
0.85
1.08
NS
0.94
< 0.001
0.84
0.82
0.85
< 0.001
1.71
3.70
< 0.001
3.80
2.69
5.37
< 0.001
1.30
0.96
1.78
NS
1.61
1.33
1.94
< 0.001
3.10
2.38
4.03
< 0.001
1.56
1.27
1.92
< 0.001
Low income adequacy
1.54
0.91
2.61
NS
0.58
0.44
0.76
< 0.001
Low income adequacy carry-over effect
2.68
1.29
5.59
0.008
0.67
0.45
0.99
0.046
Immigrant mother
0.65
0.43
0.98
0.039
1.06
0.88
1.28
NS
Cycle 2‡
1.01
0.79
1.29
NS
0.87
0.78
0.98
0.019
Risk factors
OR
95% CI
Male sex
1.25
0.99
Age (per 1 year increase)
0.91
0.88
Child current asthma
2.52
Child allergy
Child current health status (good, fair, or bad)
95% CI
p-value†
CI: confidence interval; OR: odds ratio; NS: not significant (at p < 0.05).
*
†
‡
Based on unweighted n = 9 462.
Based on longitudinal logistic regression using normalized weights.
Controlled for time in both cycles.
children.33,34 In Ontario, a longitudinal
population-based study using health admin­
istrative data showed that neighbourhood
income had no impact on the distribution of
asthma persistence in school-age children
after adjusting for other risk factors.35
The influence of immigration status is also
important to examine given that Canada
has a high immigration rate and the
immigrant population may be viewed
as vulnerable. Our results showed that
children with persistent asthma were less
likely to be born to immigrant mothers. This
agrees with studies from the US and other
countries36-39 and adds to the documenta­
tion of the importance of immigration and
acculturation in the development of asthma
and allergy. The lower reported asthma
prevalence in the immigrant population40-42
may be explained by immunoprotection as
a multifactorial phenomenon, the healthselection involved in the immigration
process and the potential reporting bias
due to language barriers and cultural
differences. Our study also found that
children with immigrant mothers are less
likely to be hospitalized. This is consistent
with a recent study conducted by Quah et
al. using the 2001 Canadian Community
Health Survey, which showed that hos­
pitalizations among visible minorities (81%
were immigrants) in Canada were lower
than among white Canadians.43 The lower
Chronic Diseases in Canada
rate of hospitalization also may be partly
explained by the good health status of
immigrants, rather than poor access, thus
highlighting the unique health patterns
among them.
Canadian studies using administrative data
have examined the seasonal patterns of
asthma hospitalizations in the province
of Ontario from 1988 to 2000.44, 45 Children
aged 0 to 4 and 5 to 9 years accounted for the
highest hospitalization rates. Furthermore,
young males were hospitalized at a rate of
two to three times that of females of the
same age,44, 45 a finding consistent with our
results. A downward trend in the total
number of hospitalizations in the general
population, most notably among young
males has been reported since the early
1990s.44,45 Our study also showed that
hospitalizations among children with
persistent asthma almost halved from
1994/95 to 1996/97 (i.e. 16.8% vs. 8.3%).
Among children participating in both
cycles, the percentage of children who
were hospitalized in 1996/97 decreased
from 1994/95 figures, while the number of
children diagnosed with asthma increased
during the same period. This trend may
be explained partially by the aging of
the cohort between the two time periods,
but may also reflect improvement in
asthma therapy and changes in health care
53
practices that have resulted in decreased
hospitalizations overall.
Some limitations were present in this study.
First, as in all studies based on survey data,
information on health care utilization was
reported by parents and is subject to recall
bias. There was a risk of undercounting
hospitalizations given that a child who was
hospitalized on numerous occasions
was counted only once in each year; in
addition, hospitalization was defined as
admission for any reason, and not just
for respiratory illnesses. However, the
consistency between the decreasing trend
in hospitalization observed in the current
study and that previously reported validates
our findings. HSU was defined based on the
question, “In the past year, how many times
have you seen or talked about your child’s
health with a physician?” This question
may bias the number of consultations
with physicians by double counting the
number of same‑day phone calls/visits and
excluding emergency visits to the hospital.
Therefore, we chose to dichotomize the HSU
outcome in order to reduce the potential
bias, and this may underestimate the
number of overall physician visits. In
general, these limitations on the reporting
of HSU variables may tend to decrease
the detectable differences in HSU in the
population; therefore, it is unlikely that
the burden attributable to asthma was
Vol 29, No 2, 2009
overestimated. Another limitation of this
study was the lack of information regarding
the severity of asthma. Some evidence
exists that the most severe asthmatics have
the highest morbidity and the most HSU;
therefore, it is possible that the most
severe asthmatics account for most of the
increased HSU.46-49
2.
Gadomski A, Jenkins P, Nichols M. Impact
of a medicaid primary care provider and
preventive care on pediatric hospitalization.
Pediatrics. 1998;101(3):1-10.
3.
Flores G, Abreu M, Chaisson CE, Sun D.
Keeping children out of hospitals: parents’
and physicians’ perspectives on how
pediatric hospitalizations for ambulatory
sensitive conditions can be avoided.
Pediatrics. 2003;112:1021-30.
Conclusions
This longitudinal population-based study
confirms that asthma is significantly asso­
ciated with higher hospitalization and HSU
in the pediatric population, while children in
low-income families are susceptible to
higher hospitalization and lower HSU.
Programs seeking to decrease the economic
burden of pediatric hospitalizations need to
focus on asthma and low-income populations.
4.
References
1.
Casanova C, Starfield BJ. Hospitalizations
of children and access to primary care: A
cross-sectional comparison. Int J Health
Serv. 1995;25:283-94.
Vol 29, No 2, 2009
14. Diggle PJ, Liang KY, Zeger SL. Analysis of
longitudinal data. Oxford: Claredon Press;
1994. Chapter 8, Marginal models; p. 146-68.
15. Carey V, Zeger SL, Diggle P. Modelling
multivariate binary data with alternating
logistic regressions. Biometrika. 1993;
80(3):517-26.
Brownson RC, Remington PL, Davis JR.
Chronic disease epidemiology and control.
2nd ed. Washington (D.C.): American Public
Health Association; 1998.
16. Lipsitz SR, Fitzmaurice GM, Orav EJ, Laird
NM. Performance of generalized estimating
equations in practical situations. Biometrics.
1994;50:270-8.
6.
Health Canada. Economic burden of illness
in Canada. Ottawa: Health Canada; 1998.
Report No.: ISBN 0-662-33144-3.
7.
To T, Ungar W. Outpatient health services
use by children in Ontario. Toronto: Institute
of Clinical Evaluative Studies; 2001. Report
No.: ISBN 0-9686318-6-X.”
17. Stokes ME, Davis CS, Koch GG. Categorical
Data Analysis: using the SAS system. First
ed. SAS Publishing; 1995. Advanced topic:
The generalized estimating equation (GEE)
method. p. 413-23.
Disclaimer
This analysis was based on the Statistics
Canada master file, National Longitudinal
Survey of Children and Youth, which
contains anonymised data collected in the
1994/95 and 1996/97 Special Survey. All
computations were prepared by the Child
Health Evaluative Sciences program at
The Hospital for Sick Children’s Research
Institute, and the responsibility for the use
and interpretation of these data is entirely
that of the authors. The opinions expressed
do not represent the views of Statistics
Canada.
13. To T, Vydykhan T, Dell S. Is obesity
associated with asthma in young children?
J Pediatr. 2004;144:162-8.
5.
Acknowledgements
Funding for this project was made available
through the Research Institute of The
Hospital for Sick Children in Toronto,
Ontario, Canada. Dr. Teresa To is supported
by the University of Toronto, Life Sciences
Committee, Dales Award in Medical
Research for 2007.
Flores G, Abreu M, Tomany-Korman S,
Meure J. Keeping children with asthma
out of hospitals: parents’ and physicians’
perspectives on how pediatric asthma
hospitalizations can be prevented.
Pediatrics. 2005;116:957-65.
12. Human Resources Development Canada,
Statistics Canada. National longitudinal
survey of children and youth. User guide.
Cycle 2, release 3. Ottawa: Minister of
Industry; 1999.
8.
9.
Ungar WJ, Coyte PC, Pharmacy Medication
Monitory Program Advisory Board.
Prospective study of the patient-level cost of
asthma care in children. Pediatr Pulmonol.
2001;32:101-8.
Health Canada. Respiratory disease in
Canada. Ottawa: Health Canada; 2001 Sep.
Report No.: ISBN 0-662-30968-5.
10. To T, Dell S, Dick P, et al. Burden of childhood
asthma [Internet]. Toronto: Institute for
Clinical Evaluative Sciences; 2004 May.
Available from: http://www.ices.on.ca/file/
ACF77.pdf. Accessed: March 17, 2008.
11. Human Resources Development Canada,
Statistics Canada. National longitudinal
survey of children and youth. User’s
handbook and microdata guide. Cycle 1,
release 2. Ottawa: Minister of Industry; 1997.
54
18. Statistics Canada. Low income cut-offs.
Ottawa: Statistics Canada; 1996.
19. Hosmer DW, Lemeshow S. Applied logistic
regression. New York: John Wiley and Sons,
Inc.; 1989.
20. To T, Cadarette SM, Liu Y. Biological, social,
and environmental correlates of preschool
development. Child Care Health Dev. 2001;
27(2):187-200.
21. SAS Institute Inc. SAS/STAT user’s guide,
version 8. STATS Publishing Inc.; 1999:
1375-428.
22. Chen Y, Dales R, Krewski D. Asthma and
the risk of hospitalization in Canada: the
role of socioeconomic and demographic
factors. Chest. 2001;119(3):708-13.
23. Brownell M, Martens P, Kozyrskyj A,
et al. Assessing the health of children
in Manitoba: a population-based study.
Winnipeg, Manitoba: Manitoba Centre for
Health Policy; 2001.
Chronic Diseases in Canada
24. Canadian Institute for Child Health. The
health of Canada’s children: a CICH profile.
Ottawa: Canadian Institute for Child Health;
1994.
34. Higgins PS, Wakefield D, Cloutier MM. Risk
factors for asthma and asthma severity in
nonurban children in Connecticut. Chest.
2005;128(6):3846-53.
25. Agha MM, Glazier RH, Guttmann A.
Relationship between social inequalities and
ambulatory care-sensitive hospitalizations
persists for up to 9 years among children
born in a major Canadian urban center.
Ambul Pediatr. 2007;7(3):258-62.
35. To T, Gershon A, Wang C, Dell S, Cicutto L.
Persistence and remission in childhood
asthma – a population-based asthma birth
cohort study. Arch Pediatr Adolesc Med.
2007;161(12):1197-204.
45. To T, Guttmann A, Dick P. Inpatient and
day surgery use by children in Ontario.
Toronto: Institute for Clinical Evaluative
Sciences; 2001. Report No.: 0-9699405-9-9.
36. Eldeirawi K, McConnell R, Freels S,
Persky VW. Associations of place of birth
with asthma and wheezing in Mexican
American children. J Allergy Clin Immunol.
2005;116(1):42-8.
46. Van Ganse E, Antonicelli L, Zhang Q,
et al. Asthma-related resource use and
cost by GINA classification of severity in
three European countries. Respir Med.
2005;100:140-47.
37. Hjern A, Haglund B, Bremberg S, RingbackWeitoft G. Social adversity, migration and
hospital admissions for childhood asthma
in Sweden. Acta Paediatr. 1999;88(10):
1107-12.
47. Bootman JL, Crown WH, Luskin AT. Clinical
and economic effects of suboptimally
controlled asthma. Manag Care Interface.
2004;17(1):31-6.
26. Sin DD, Svenson LW, Cowie RL, Man
SF. Can universal access to health care
eliminate health inequities between
children of poor and nonpoor families? A
case study of childhood asthma in Alberta.
Chest. 2003;124(1):51-6.
27. Kozyrskyj AL, Mustard CA, Simons FE.
Inhaled corticosteroids in childhood
asthma: income differences in use. Pediatr
Pulmonol. 2003;36(3):241-7.
28. Kozyrskyj AL, Mustard CA, Simons FE.
Socioeconomic status, drug insurance
benefits, and new prescriptions for
inhaled corticosteroids in schoolchildren
with asthma. Arch Pediatr Adolesc Med.
2001;155(11):1219-24.
29. Ungar WJ, Kozyrskyj A, Paterson M,
Ahmad F. Effect of cost-sharing on use
of asthma medication in children. Arch
Pediatr Adolesc Med. 2008;162(2):104-10.
30. Blais L, Beauchesne MF, Levesque S.
Socioeconomic status and medication
prescription patterns in pediatric asthma in
Canada. J Adolesc Health. 2006;38(5):607
e9-16.
31. To T, Gershon A, Tassoudji M, et al. The
burden of asthma in Ontario. Toronto,
Ontario: Institute for Clinical Evaluative
Sciences; 2006 Sep.
32. van Doorslaer E, Masseria C, Koolman X.
Inequalities in access to medical care by
income in developed countries. CMAJ.
2006;174(2):177-83.
33. Halfon N, Newacheck PW. Childhood
asthma and poverty: differential impacts
and utilization of health services. Pediatrics.
1993;91(1):56-61.
Chronic Diseases in Canada
38. Netuveli G, Hurwitz B, Sheikh A. Ethnic
variations in incidence of asthma episodes
in England & Wales: national study of
502,482 patients in primary care. Respir
Res. 2005;6:120.
39. Gibson PG, Henry RL, Shah S, Powell H,
Wang H. Migration to a western country
increases asthma symptoms but not
eosinophilic airway inflammation. Pediatr
Pulmonol. 2003;36(3):209-15.
44. Crighton EJ, Mamdani MM, Upshur RE.
A population based time series analysis
of asthma hospitalizations in Ontario,
Canada: 1998 to 2000. BMC Health Serv
Res. 2001:1-7.
48. Berggren F, Hjelmgren J, Lindgren B.
Differences in health care utilisation and
workdays lost between individuals with
and without asthma. Appl Health Econ
Health Policy. 2003;2(3):165-70.
49. Halterman JS, Yoos HL, Sidora K, Kitzman H,
McMullen A. Medication use and health care
contacts among symptomatic children with
asthma. Ambul Pediatr. 2001;1(5):275-9.
40. Gold DR, Acevedo-Garcia D. Immigration to
the United States and acculturation as risk
factors for asthma and allergy. J. Allergy
Clin. Immunol. 2005;116(1):38-41.
41. Kalyoncu AF. Symptoms of asthma,
bronchial responsiveness and atopy in
immigrants and emigrants in Europe. Eur
Respir J. 2002;19(5):980-81.
42. Choudhry S, Seibold MA, Borrell LN, et al.
Dissecting complex diseases in complex
populations: asthma in Latino Americans.
Proc Am Thorac Soc. 2007;4(3):226-33.
43. Quan H, Fong A, De Coster C, et al. Variation
in health services utilization among ethnic
populations. CMAJ. 2006:174:787-91.
55
Vol 29, No 2, 2009
Deprivation and unintentional injury hospitalization
in Quebec children
Mathieu Gagné, MA (1); Denis Hamel, MSc (1)
Abstract
Injuries disproportionately affect children from deprived areas. This study examines the
links between the material and social dimensions of deprivation and injury hospitalizations
in children aged 14 years or under from 2000 to 2004. Hospitalization data are from the
Quebec hospital administrative data system, whereas socio-economic characteristics of
individuals were estimated based on the smallest geographic areas for which Canadian
census data were disseminated. The Poisson regression model was used to calculate the
relative risks of hospitalization for seven categories of unintentional injury. A total of
24 540 injury hospitalizations were examined. Hospitalization in children is associated
with both dimensions of deprivation. Injuries to pedestrians and motor vehicle
occupants and injuries related to burns and poisonings are clearly associated with both
dimensions of deprivation. These inequalities should be considered in the development
of preventive measures.
Key words: Socio-economic factors, inequalities, children, hospitalizations,
Quebec, injuries, trauma, unintentional
Introduction
Unintentional injuries are the leading cause
of death and the third most common cause of
hospitalization in Quebec children aged
1 to 14 years. Although the majority of
children recover from an injury quickly
and completely, some are affected by
temporary, or even permanent, disabilities
that can significantly restrict their quality
of life.1
It is generally accepted that some children,
particularly those from deprived areas,
are at greater risk of sustaining an injury
than others.2 With respect to mortality, the
association between socio-economic factors
and risk of injury has been extensively
measured and illustrated.3-8 Some claim
that inequalities in mortality are increas­
ing,9 while others claim that gaps have
held constant despite the observed decline
in mortality rates.10
Research on the links between non-fatal
injuries and socio-economic factors has
yielded mixed results.11 A number of studies
carried out in Quebec in the early 1990s
show links between the deprived situation
of some populations in the Montreal area
and risk of injury in pedestrians and
cyclists aged 14 years or under.5,12-14,16
The relationship between socio-economic
characteristics and the risk of childhood
injury has also been observed elsewhere
in Canada.4,17,18 Some have observed
that unintentional injury hospitalizations
among children increased significantly
with deprivation,19-28 while others did not
find any association between the two
phenomena.29-31
Cubbin and Smith (2002) have identified
some reasons that may explain these
fluctuating results.11 First, the analyses do
not always account for the level of injury
severity nor define this level using a thresh­
old that reflects the use of health care
services. In the case of hospitalizations,
a number of extrinsic factors on injury
severity influence the likelihood of hospi­
talization, including bed availability,
distance between home and hospital,
concerns about whether the injury was
intentional or even patient prefences.11,33
These factors can affect case identifica­
tion in different ways. There is a risk of
obscuring or magnifying the relationship
between the injuries and deprivation
if the severity of injuries is considered
solely from the standpoint of services use.
Second, the injury mechanism (i.e. falls,
pedestrians, poisoning, etc.) is not always
examined in detail.17,18 Some authors
have nevertheless demonstrated that the
effect of socio-economic factors can go in
opposite directions when each mechanism
is studied independently.34 Last, Cubbin
and Smith (2002) emphasize that there are
many measures, sometimes inadequately
defined, by which socio-economic status
can be expressed, whereas the choice of
indicator is generally not justified by the
investigators.11
Finally, we emphasize that few studies
have focused on the social dimension of
deprivation, i.e. the effect of social cohe­
sion or isolation on injury risk, although
this dimension is equally identified as a
health determinant.55 Recently, a measure
of social fragmentation introduced to
explain the association between injuries
Author References
1 Institut national de santé publique du Québec
Correspondence: Mathieu Gagné, Institut national de santé publique du Québec, 945 Wolfe Avenue, 3rd Floor, Sainte-Foy, QC G1V 5B3,
Tel.: 418-650-5115, ext. 5702, Fax: 418-643-5099, Email: [email protected]
Chronic Diseases in Canada
56
Vol 29, No 2, 2009
and deprivation in Sweden yielded no
significant findings, after adjustment, for
economic deprivation.56
Table 1
Classification of unintentional injuries and main categories of external causes
Unintentional injuries
E800–E949
Study objective
Motor vehicle occupants
E810–E819; .0, .1 and .9
Bicyclists
E810–E819; .6 and E826.1
This study examines the links between
unintentional injuries and deprivation
in Quebec children. More specifically,
we intend to establish whether hospital
morbidity due to unintentional injury is
associated with the material and social
dimensions of deprivation in children aged
14 years or under in Quebec. We will then
verify whether this relationship varies
with the most important circumstances
surrounding the injury. Finally, we will
attempt to note whether the association
also applies to severe injuries to ensure
that the observed associations are not the
result of administrative variations or a
different use of health care services.
Pedestrians
E810–E819; .7
Poisonings
E850–E869
Falls
E880–E888
Fires and burns
E890–E899 and E924
Methods
Data sources
Data used for this study are drawn from
the records of the Quebec hospital client
information system, called MED-ÉCHO. All
Quebec children aged 14 years or under
were selected if they were admitted to
hospitals providing general and specialty
care from January 1, 2000, to December 31,
2004, for short-term physical care of unin­
tentional injuries. Case identification was
based on the external cause of trauma
and coded according to the rules of the
International Classification of Diseases,
Ninth Revision (E800–E949). Hospitalizations
due to medical or surgical complications
(E870–E879), adverse effects from the
therapeutic use of medications (E930–E949)
and after effects of injury (E905–E909)
were excluded (i.e. 2358 cases). In
addition, readmissions and transfers were
excluded to limit the effects of variations
related to service use and obtain a more
robust morbidity indicator.32 Based on these
criteria, we identified 24 540 unintentional
trauma-related events resulting in the
hospitalization of children 14 years or under
in Quebec during the period in question.
Our measure of severity is based on the
definition of the eligibility criteria used by
the Registre des traumatismes du Québec
(RTQ) [Québec Trauma Registry], which
gathers information on victims of severe
injuries. To be identified as severe, cases
had to meet one of the following criteria:
hospitalization for three days or more,
admission to the intensive care unit (ICU),
or death during hospitalization. This defi­
nition of a severe case was met by 24.8%
of the 24 540 cases.
The deprivation index
The administrative health databases in
Quebec do not contain socio-economic
information. For this reason, we used the
deprivation index developed by Pampalon
and Raymond (2000), which estimates an
individual’s socio-economic status using
an ecological approach, i.e. by attributing
to each individual admitted to hospital
the socio-economic level of his or her
neighbourhood. The index reflects relative
disadvantage in relation to total population,
which is expressed through a material and
a social dimension. The material dimension
reflects available economic resources or
poverty, whereas the social dimension
expresses the level of social cohesion or
isolation, i.e. the quality or fragility of the
social network.35
The index is obtained through the appli­
cation of principal component analysis of
six indicators taken from Canada’s 2001
census, chosen for their known association
with health inequalities.36 The material
dimension of the index primarily consists
of the following indicators: the proportion
of persons with no high school diploma,
the employment to population ratio and
the average personal income. The social
dimension of the index primarily consists
of the following indicators: the proportion
of persons who are separated, divorced or
widowed, the proportion of persons living
alone and the proportion of single-parent
families. For both dimensions, a value is
calculated for each dissemination area
(DA), which is defined as a small, relatively
stable geographic unit composed of one or
more adjacent dissemination blocks, with
400 to 700 inhabitants, for which all census
data are disseminated.35 The values are
then grouped into quintiles (i.e. groups of
20%) to create the index, ranging from the
most privileged (i.e. quintile 1) to the least
privileged (i.e. quintile 5).36 The key point
is that each DA can be linked to a postal
code, which appears in all administrative
health records in Quebec. This strategy can
be used to estimate the level of deprivation
of the individuals for whom information is
collected in our hospitalization records.
Analysis
Relative risks (RRs) of unintentional injury
hospitalization were calculated along
with confidence intervals of 95% (CI) by
Poisson regression modelling using the
GENMOD procedure (i.e. SAS, version 9.1)
for each material and social deprivation
quintile. The RRs were adjusted for dif­
ferences in age, sex and residence locationi
between quintiles, but also for the presence
of the other dimension of the index. For
each analysis, the RR is interpreted based
on the reference category, i.e. the most
i Census Metropolitan Area (CMA) of Montreal, other CMAs, other census agglomerations, small towns and rural areas.
Vol 29, No 2, 2009
57
Chronic Diseases in Canada
privileged quintile, for which the RR is set
at 1.00. The value associated with the other
quintiles expresses the RR of hospitalization
between the most privileged quintile and
the other quintiles.
Results
From 2000 to 2004, the annual average
number of unintentional injuries resulting
in hospitalization in children aged 0 to
14 in Quebec was 4908 (Table 2). This
amounts to an annual rate of 384 injury
hospitalizations per 100 000 children. This
rate is substantially higher in boys than
in girls (i.e. 480 per 100 000 and 285 per
100 000, respectively). It peaks in children
aged 0 to 4 years (i.e. 421 per 100 000),
declines in children aged 5 to 9 years (i.e.
336 per 100 000) and then rises again in
children aged 10 to 14 years (i.e. 403 per
100 000).
An examination of hospitalizations from a
socio-economic standpoint indicates that
they appear to be strongly associated with
the material dimension of the depriva­
tion index, with the rate increasing from
357 hospitalizations per 100 000 in children
in the most privileged quintile (i.e. Q1) to
426 in those from deprived areas (i.e. Q5).
The association with the social dimension
of deprivation is less pronounced. Children
in the most privileged category have a
slightly lower hospitalization rate than chil­
dren in the least privileged category (i.e.
367 vs. 401 per 100 000).
Injury mechanisms and location
There are varying degrees of differences in
most injury mechanisms between socioeconomic groups, which are generally
Table 2
Annual average numbers and injury hospitalization rates by age, sex, area of residence and the two dimensions of the deprivation
index in children aged 14 years or under, based on severity, for all of Quebec, 2000 to 2004
Injuries resulting in hospitalization
All
Severe
Number
Rate*
Number
Rate*
0 to 4 years
1561
420.6
365
98.2
5 to 9 years
1519
335.6
342
75.5
10 to 14 years
1829
403.1
511
112.6
Boys
3127
480.4
811
124.5
Girls
1781
284.5
406
64.9
1304
446.0
368
121.3
720
494.1
153
101.7
Age
Sex
Area
Small towns and rural areas (< 10 000 inhabitants)
Census agglomerations (10 000 to 100 000 inhabitants)
Other metropolitan areas (>100 000 inhabitants)
905
394.1
196
87.1
1978
327.4
500
83.7
Q1-Privileged quintile
841
356.6
182
78.8
Q2
907
356.0
213
85.6
Q3
975
380.7
230
90.6
Q4
1032
397.3
267
101.4
Q5-Deprived quintile
1152
426.0
325
116.2
Q1-Privileged quintile
1092
366.9
254
83.8
Q2
1088
374.1
265
90.5
Q3
1038
390.3
266
100.8
Q4
918
398.4
234
102.9
Census Metropolitan Area of Montreal
Material deprivation
Social deprivation
Q5-Deprived quintile
Total
773
401.3
197
104.2
4908
384.3
1217
95.3
* Rate adjusted for the other dimension of the index, age, sex and area of residence
Sources: INSPQ, MED-ÉCHO hospitalization records, 2000 to 2004
MSSS, demographic outlook based on the 2001 census
Chronic Diseases in Canada
58
Vol 29, No 2, 2009
more pronounced for the material
dimension of the index. The gradients are
particularly obvious for pedestrians, motor
vehicle occupants, bicyclists, poisonings
and fire and burn injuries (Table 3).
Transportation-related injury categories are
all strongly associated with the material
dimension of deprivation, i.e. children from
the least privileged areas have significantly
higher RRs than their peers from privileged
areas (i.e. motor vehicle occupants: RR=
1.69; pedestrians: RR=3.62; bicyclists:
RR=1.31). RRs for bicyclists are sig­
nificantly higher when analyses focus
exclusively on accidents involving motor
vehicles, with children from materially
deprived areas still at a disadvantage (i.e.
data not shown; RR=1.75). With respect
to the social dimension of the index,
children in the most deprived quintile
have higher risks of hospitalization due
to injuries suffered as a motor vehicle
occupant or pedestrian (i.e. RR=1.32 and
2.36, respectively). This is also found for
the categories of poisoning (i.e. material
dimension: RR=1.68; social dimension:
RR=1.66) and fires and burns (i.e. material
dimension: RR=2.05; social dimension:
RR=1.50).
In contrast, the falls category as a whole
shows no marked difference based on socioeconomic level (i.e. material dimension:
RR=1.01; social dimension: RR=1.00).
Hospitalizations for falls account for more
than half (i.e. 51.3%) of all unintentional
injury hospitalizations in the study pop­
ulaion, with this category covering a
wide variety of circumstances. A different
picture emerges (Table 4) when the
main circumstances surrounding falls
are identified. Children from materially
deprived areas have a higher RR than
their peers in privileged areas for falls on
stairs (i.e. RR=1.36), falls from the top of
a building (i.e. RR=2.39) and, to a lesser
extent, falls from one level to another (i.e.
RR=1.09), including falls from a bed or chair
(i.e. data not shown; RR=2.30). A similar
trend is observed for the social dimension
of the index only for falls from the top of
a building (i.e. RR=1.96). Conversely, the
RR for falls on same level from slipping,
tripping or stumbling is substantially lower
in children from deprived areas.
Moreover, between the injury location and
both dimensions of the index, there also
exists an association that changes, depend­
ing on the characteristics of the location in
questionii (Table 5). For example, children
from deprived areas have a higher RR of
home injuries than their peers in privileged
areas (i.e. material dimension: RR=1.50;
social dimension: RR=1.18). Conversely,
materially deprived children have a sig­
nificantly lower RR of recreational or sportrelated injuries than children from privileged
areas (i.e. RR=0.66).
Nature of main traumatic injury
and severity of injuries sustained
To ensure that the observed associations
were not caused by administrative vari­
ations or a differential use of health care
services, our analyses were considered from
the standpoint of injury characteristics.
First, we examined the nature of the main
traumatic injury by isolating a group of
similar mechanisms, i.e. transportation
accidents involving a pedestrian, bicyclist
or motor vehicle occupant (Table 6). With
respect to the material dimension, the RR
of a skull fracture and intracranial injury
appears to be significantly higher in chil­
dren from deprived areas compared with
children in privileged categories (i.e. mate­
rial dimension: RR=1.67; social dimension:
RR=1.52). A similar finding, but of greater
severity, emerges for lower limb injuries
(i.e. material dimension: RR=2.58; social
dimension: RR=1.70) in contrast to upper
limb fractures. Next, we repeated each
analysis carried out to this point by selecting
cases of severe injury only (Tables 7 to 9).
Overall, the examination of severe injuries
indicates that the measured associations
persist and are nearly always more pro­
nounced for the material dimension than
for the social dimension, for which there is
no clear trend.
Discussion
The results of the study show clearly that
in Quebec, children from deprived areas
are at greater risk of injury hospitalization
than children from privileged areas. Recent
data have confirmed the relationship
between socio-economic characteristics and
risk of childhood injury for all of Quebec,
as observed in the early 1990s for the
Montreal region.5,12-14,16 Our results suggest
that these socio-economic differences exist
not only in injury categories related to road
accidents, but also, to varying degrees, in
other categories such as injuries related
to fires and burns, poisonings and certain
circumstances surrounding falls, including
falls on stairs. Unintentional injury in Quebec
children are generally influenced by the
two dimen­sions of deprivation. Most studies
have underscored the association between
risk of injury and the material dimension of
deprivation (i.e. level of education, unem­
ployment, income, father’s occupation,
access to a vehicle, housing tenure, financial
difficulties, etc.),5,13-18,20,22,28,34,37 whereas the
social dimension was viewed only partially
through the lens of single-parent families,
usually used as an indicator of poverty.37,38
However, our results suggest that the two
dimensions of deprivation are independently
associated with risk of hospitalization fol­
lowing an injury and that their effects may
be cumulative. These results lend greater
insight to the observations made for the
entire Quebec population from 1997 to
2000,53 when no significant trend between
economic deprivation and unintentional
traumas had previously been observed.
Similarly, our results contrast with those
recently obtained in the Swedish context,
where no significant correlation between
social isolation and injuries persisted after
adjustment for economic deprivation.56
In order to minimize the effects of extrinsic
factors on the severity of injuries on the
probability of hospitalization, a severity
measure was used to limit the analysis of
injuries associated with a higher probability
of hospitalization. Thus, as Hippisley-Cox
et al. (2002) had observed for the Trent
ii Accident location only available for codes E850–E869 and E880–E928.
Vol 29, No 2, 2009
59
Chronic Diseases in Canada
Table 3
Average annual number, adjusted* hospitalization rate and RR† for main categories of unintentional injuries by the two dimensions
of the deprivation index in children 14 years or under, for all of Quebec, 2000 to 2004
Deprivation index
Trauma categories
Material dimension
Number
Social dimension
Rate
RR
95% CI
Number
Rate
RR
95% CI
841
907
975
356.6
356.0
380.7
1
1.00
1.07
0.96-1.04
1.02-1.11
1092
1088
1038
366.9
374.1
390.3
1
1.02
1.06
0.98-1.06
1.02-1.11
1032
1152
397.3
426.0
1.11
1.19
1.07-1.16
1.14-1.25
918
773
398.4
401.3
1.09
1.09
1.04-1.13
1.05-1.14
Motor vehicle occupants
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
16
24
30
36
42
8.0
10.4
11.5
12.9
13.6
1
1.30
1.43
1.61
1.69
0.97-1.73
1.08-1.89
1.22-2.12
1.28-2.24
32
31
36
25
23
10.5
10.1
13.1
11.5
13.9
1
0.96
1.25
1.10
1.32
0.77-1.20
1.01-1.55
0.87-1.39
1.03-1.69
Pedestrians
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
11
18
21
25
39
4.2
7.0
8.5
9.7
15.1
1
1.69
2.03
2.34
3.62
1.21-2.38
1.46-2.84
1.69-3.24
2.65-4.95
18
17
22
23
33
6.3
6.4
8.7
9.6
14.8
1
1.01
1.39
1.53
2.36
0.76-1.36
1.05-1.84
1.16-2.03
1.81-3.08
Bicyclists
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
63
72
79
90
96
27.1
28.0
30.5
34.7
35.4
1
1.03
1.13
1.28
1.31
0.89-1.20
0.97-1.31
1.10-1.49
1.12-1.52
90
86
85
76
63
30.5
29.3
32.0
32.7
32.7
1
0.96
1.05
1.07
1.07
0.84-1.10
0.92-1.20
0.93-1.23
0.92-1.25
Poisonings
Q1-Privileged quintile
Unintentional injuries
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
34
16.3
1
50
16.7
1
Q2
43
17.5
1.07
0.88-1.31
57
19.0
1.14
0.96-1.35
Q3
50
18.9
1.16
0.95-1.42
52
19.3
1.15
0.97-1.37
Q4
62
22.8
1.40
1.15-1.70
57
25.0
1.49
1.26-1.78
Q5-Deprived quintile
78
27.4
1.68
1.39-2.04
51
27.9
1.66
1.39-2.00
Falls
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
493
486
511
502
528
200.0
186.4
199.9
197.0
202.9
1
0.93
1.00
0.98
1.01
0.88-0.99
0.94-1.06
0.93-1.04
0.96-1.08
578
557
530
469
384
195.2
194.3
200.2
202.1
194.7
1
1.00
1.03
1.04
1.00
0.94-1.05
0.97-1.08
0.98-1.09
0.94-1.06
Fires and burns
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
15
17
20
23
33
6.1
6.6
7.7
8.9
12.4
1
1.09
1.27
1.46
2.05
0.80-1.50
0.93-1.73
1.08-1.98
1.52-2.74
19
20
24
24
20
6.5
7.1
9.2
10.3
9.7
1
1.09
1.42
1.59
1.50
0.83-1.45
1.08-1.86
1.21-2.08
1.12-2.01
4908
384.3
4908
384,3
Total
* Rate adjusted for the other dimension of the index, age, sex and area of residence
† Relative risk
Sources: MSSS, Med-Écho hospitalization records, 2000 to 2004
MSSS, demographic outlook
Chronic Diseases in Canada
60
Vol 29, No 2, 2009
Table 4
Average annual number, adjusted* hospitalization rate and RR† for the main categories of falls by the two dimensions
of the deprivation index in children 14 years or under, for all of Quebec, 2000 to 2004
Deprivation Index
Circumstances of falls
Material dimension
RR
Social dimension
Number
Rate
95% CI
Number
Rate
Q1-Privileged quintile
29
11.9
1
Q2
33
13.0
1.09
0.87-1.37
Q3
38
14.8
1.24
Q4
37
14.5
1.21
Q5-Deprived quintile
44
16.3
1.36
6
2.6
1
Q2
7
2.8
1.08
Q3
11
4.3
1.66
Q4
10
3.7
Q5-Deprived quintile
17
6.2
Q1-Privileged quintile
172
68.7
1
Q2
178
68.1
0.99
Q3
179
70.4
1.02
Q4
184
72.6
Q5-Deprived quintile
192
74.6
RR
95% CI
44
14.9
1
45
15.5
1.05
0.87-1.26
0.99-1.55
34
12.7
0.86
0.70-1.05
0.97-1.52
32
13.8
0.93
0.76-1.14
1.08-1.71
26
13.3
0.90
0.72-1.12
8
2.8
1
0.66-1.76
11
4.0
1.43
0.96-2.14
1.06-2.60
8
3.1
1.13
0.73-1.75
1.42
0.89-2.26
12
5.1
1.84
1.23-2.75
2.39
1.54-3.69
11
5.4
1.96
1.29-2.99
202
68.7
1
0.90-1.09
197
69.3
1.01
0.92-1.10
0.93-1.13
186
70.8
1.03
0.94-1.13
1.06
0.96-1.16
171
73.3
1.07
0.97-1.17
1.09
0.98-1.20
148
73.6
1.07
0.97-1.18
Falls on or from stairs (E880)
Falls from the top of building (E882)
Q1-Privileged quintile
Other falls from one level to another (E884)
Falls on same level from slipping, tripping or stumbling (E885)
Q1-Privileged quintile
163
64.3
1
162
54.6
1
Q2
142
53.6
0.83
0.75-0.92
153
53.4
0.98
0.89-1.08
Q3
135
53.0
0.82
0.74-0.92
152
57.1
1.05
0.95-1.16
Q4
134
53.2
0.83
0.74-0.92
123
53.0
0.97
0.87-1.08
Q5-Deprived quintile
116
45.7
0.71
0.63-0.80
100
50.8
0.93
0.83-1.04
Q1-Privileged quintile
38
15.3
1
19
17.9
1
Q2
36
13.7
0.89
0.73-1.10
20
15.9
0.89
0.74-1.06
Q3
47
18.4
1.20
0.98-1.46
24
14.7
0.82
0.68-0.99
Q4
37
14.8
0.96
0.78-1.19
24
15.7
0.87
0.72-1.06
Q5-Deprived quintile
39
15.5
1.01
0.81-1.25
20
12.4
0.69
0.56-0.86
85
36.7
1
109
36.3
1
Q2
89
35.0
0.95
0.83-1.09
106
36.2
1.00
0.88-1.12
Q3
101
39.1
1.06
0.93-1.22
111
41.6
1.15
1.02-1.29
Q4
100
38.2
1.04
0.91-1.19
94
41.2
1.13
1.00-1.29
Q5-Deprived quintile
121
44.1
1.20
1.05-1.38
74
39.3
1.08
0.94-1.24
Falls on same level from collision, pushing or shoving,
by or with another person (E886)
Other falls (E881, E883, E887 and E888)
Q1-Privileged quintile
* Rate adjusted for the other dimension of the index, age, sex and area of residence
† Relative risk
Sources: MSSS, Med-Écho hospitalization records, 2000 to 2004
MSSS, demographic outlook
Vol 29, No 2, 2009
61
Chronic Diseases in Canada
Table 5
Annual average number, adjusted* hospitalization rate and RR† for the place of injury by the two dimensions of the deprivation
index in children aged 14 years or under, for all of Quebec, 2000 to 2004
Deprivation Index
Injury location
Material dimension
RR
Social dimension
Number
Rate
95% CI
Number
Rate
Q1-Privileged quintile
308
128.0
1
Q2
344
138.3
1.08
0.91-1.29
Q3
373
154.4
1.21
Q4
406
167.4
1.31
Q5-Deprived quintile
467
191.6
1.50
Q1-Privileged quintile
140
49.7
1
Q2
118
43.4
0.87
Q3
115
42.9
0.86
Q4
103
40.6
84
32.6
Q1-Privileged quintile
56
21.9
1
Q2
45
16.8
0.77
Q3
48
18.5
0.85
Q4
46
18.1
Q5-Deprived quintile
49
19.3
RR
95% CI
399
146.8
1
409
154.6
1.05
0.89-1.24
1.01-1.44
364
145.0
0.99
0.84-1.17
1.10-1.56
382
167.9
1.14
0.97-1.35
1.26-1.79
345
173.9
1.18
1.00-1.41
138
41.6
1
0.68-1.12
127
41.1
0.99
0.78-1.25
0.68-1.10
124
45.9
1.10
0.87-1.40
0.82
0.64-1.05
97
41.4
1.00
0.78-1.26
0.66
0.51-0.85
72
38.5
0.93
0.72-1.19
60
19.8
1
0.62-0.95
49
17.2
0.87
0.71-1.07
0.69-1.05
55
20.5
1.04
0.85-1.26
0.83
0.67-1.03
45
18.8
0.95
0.77-1.17
0.88
0.71-1.10
36
18.2
0.92
0.73-1.15
Home
Recreational or sports area
Q5-Deprived quintile
Public building
Other specified location
Q1-Privileged quintile
21
9.5
1
28
8.9
1
Q2
25
10.2
1.08
0.80-1.45
29
9.4
1.06
0.80-1.39
Q3
26
10.2
1.08
0.80-1.45
25
9.2
1.03
0.78-1.37
Q4
21
7.9
0.83
0.61-1.14
20
9.1
1.02
0.76-1.37
Q5-Deprived quintile
31
10.2
1.08
0.79-1.48
21
12.6
1.43
1.06-1.92
Q1-Privileged quintile
199
90.0
1
273
91.6
1
Q2
229
91.2
1.01
0.90-1.14
281
93.2
1.02
0.91-1.13
Q3
243
93.2
1.03
0.92-1.16
275
99.8
1.09
0.98-1.21
Q4
258
97.8
1.09
0.97-1.22
215
96.1
1.05
0.94-1.17
Q5-Deprived quintile
277
100.1
1.11
0.99-1.25
163
93.0
1.02
0.90-1.14
Unspecified location
* Rate adjusted for the other dimension of the index, age, sex and area of residence.
† Relative risk
Sources: MSSS, Med-Écho hospitalization records, 2000 to 2004
MSSS, demographic outlook
Chronic Diseases in Canada
62
Vol 29, No 2, 2009
Table 6
Average annual number, adjusted* hospitalization rate and RR† for the main traumatic injury categories by the two dimensions
of the deprivation index in children of 14 years or under, for all of Quebec, 2000 to 2004
Deprivation Index
Traumatic injury categories
Material dimension
Number
Rate
RR
Social dimension
95% CI
Number
Rate
RR
95% CI
Skull fractures and head traumas
Q1-Privileged quintile
26
11.6
1
41
13.9
1
Q2
31
12.6
1.08
0.86-1.37
39
13.3
0.96
0.79-1.17
Q3
43
16.7
1.44
1.15-1.80
47
17.7
1.27
1.06-1.54
Q4
55
20.8
1.79
1.44-2.22
43
18.4
1.33
1.09-1.61
Q5-Deprived quintile
54
19.4
1.67
1.34-2.09
40
21.1
1.52
1.25-1.87
Q1-Privileged quintile
28
11.0
1
34
11.4
1
Q2
30
11.2
1.01
0.80-1.28
33
11.4
1.00
0.80-1.24
Q3
24
9.5
0.86
0.67-1.10
31
11.6
1.02
0.81-1.26
Q4
33
13.0
1.18
0.93-1.49
26
11.2
0.98
0.78-1.24
Q5-Deprived quintile
30
11.9
1.08
0.84-1.38
22
10.8
0.95
0.74-1.22
Q1-Privileged quintile
10
4.3
1
21
7.0
1
Q2
17
7.0
1.62
1.15-2.28
16
5.5
0.79
0.59-1.05
Q3
20
7.9
1.83
1.30-2.57
18
7.1
1.02
0.77-1.35
Q4
19
7.3
1.69
1.20-2.40
19
8.2
1.18
0.89-1.56
Q5-Deprived quintile
31
11.2
2.58
1.85-3.60
24
11.8
1.70
1.29-2.24
Upper limb fractures
Lower limb fractures
* Rate adjusted for the other dimension of the index, age, sex and area of residence
† Relative risk
Sources: MSSS, Med-Écho hospitalization records, 2000 to 2004
MSSS, demographic outlook
Vol 29, No 2, 2009
63
Chronic Diseases in Canada
Table 7
Average annual number, adjusted* hospitalization rate and RR† for the main categories of severe unintentional injuries by the two
dimensions of the deprivation index in children 14 years or under, for all of Quebec, 2000 to 2004
Deprivation Index
Trauma categories
Material dimension
Social dimension
Number
Rate
RR
95% CI
Unintentional injuries
Q1-Privileged quintile
Q2
Q3
182
213
230
78.8
85.6
90.6
1
1.09
1.15
Q4
Q5-Deprived quintile
267
325
101.4
116.2
Motor vehicle occupants
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
6
12
11
16
17
Pedestrians
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
Bicyclists
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
Poisonings
Q1-Privileged quintile
Number
Rate
RR
95% CI
0.99-1.19
1.05-1.26
254
265
266
83.8
90.5
100.8
1
1.08
1.20
1.00-1.17
1.11-1.30
1.29
1.48
1.18-1.41
1.35-1.61
234
197
102.9
104.2
1.23
1.24
1.13-1.33
1.14-1.36
2.9
5.0
4.1
5.7
5.4
1
1.70
1.39
1.94
1.85
1.08-2.68
0.87-2.23
1.24-3.06
1.16-2.93
15
12
18
8
9
4.8
3.7
6.3
3.8
5.4
1
0.77
1.32
0.79
1.13
0.55-1.09
0.96-1.80
0.54-1.16
0.76-1.68
5
8
11
15
22
1.8
3.2
4.6
5.9
8.5
1
1.79
2.52
3.26
4.67
1.08-2.96
1.56-4.09
2.04-5.21
2.96-7.38
11
8
12
14
16
3.7
3.1
5.0
5.7
7.0
1
0.83
1.34
1.53
1.89
0.55-1.24
0.93-1.94
1.07-2.21
1.32-2.72
16
16
23
23
30
6.8
6.2
9.1
8.9
11.4
1
0.91
1.33
1.31
1.67
0.67-1.24
0.99-1.78
0.97-1.76
1.24-2.23
24
23
26
19
16
8.1
8.1
9.8
8.3
8.3
1
0.99
1.21
1.02
1.02
0.67-1.24
0.99-1.78
0.97-1.76
0.76-1.36
7
2.8
1
7
2.2
1
Q2
9
3.4
1.18
0.76-1.85
9
3.2
1.47
0.94-2.30
Q3
9
3.4
1.21
0.77-1.91
11
4.4
2.03
1.32-3.12
Q4
13
5.0
1.74
1.14-2.67
11
4.8
2.19
1.42-3.39
Q5-Deprived quintile
16
5.8
2.05
1.35-3.13
14
7.1
3.22
2.10-4.94
Falls
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
77
85
83
87
101
31.5
33.0
32.6
33.9
38.1
1
1.05
1.03
1.07
1.21
0.91-1.20
0.90-1.19
0.93-1.24
1.05-1.40
91
94
92
87
68
30.4
32.9
35.1
37.8
34.3
1
1.08
1.16
1.25
1.13
0.95-1.23
1.01-1.32
1.09-1.42
0.98-1.31
Fires and burns
Q1-Privileged quintile
Q2
Q3
Q4
Q5-Deprived quintile
10
10
12
16
22
4.3
4.1
4.6
6.2
8.2
1
0.97
1.09
1.45
1.93
0.66-1.44
0.74-1.60
1.01-2.08
1.36-2.76
12
14
15
17
13
3.9
5.1
5.7
7.4
6.3
1
1.31
1.49
1.92
1.64
0.93-1.85
1.05-2.10
1.37-2.69
1.13-2.37
* Rate adjusted for the other dimension of the index, age, sex and area of residence
† Relative risk
Sources: MSSS, Med-Écho hospitalization records, 2000 to 2004
MSSS, demographic outlook
Chronic Diseases in Canada
64
Vol 29, No 2, 2009
Table 8
Average annual number, adjusted* hospitalization rate and RR† for the main circumstances of falls resulting in
severe injuries by the two dimensions of the deprivation index in children 14 years or under, for all of Quebec, 2000 to 2004
Deprivation Index
Circumstances of falls
Material dimension
Number
Rate
RR
Social dimension
95% CI
Number
Rate
RR
95% CI
Falls on or from stairs or steps (E880)
Q1-Privileged quintile
3
1.2
1
5
1.8
1
Q2
6
2.4
2.00
1.05-3.78
7
2.4
1.34
0.81-2.21
Q3
6
2.4
1.97
1.03-3.76
5
2.0
1.10
0.64-1.89
Q4
5
1.8
1.47
0.74-2.92
6
2.4
1.35
0.79-2.31
Q5-Deprived quintile
8
3.0
2.51
1.32-4.79
4
2.3
1.25
0.70-2.24
Q1-Privileged quintile
2
0.7
1
3
0.8
1
Q2
2
0.8
1.04
0.40-2.72
3
0.9
1.13
0.53-2.42
Q3
3
1.4
1.86
0.78-4.40
2
0.6
0.74
0.30-1.79
Q4
2
0.6
0.81
0.29-2.22
3
1.3
1.61
0.76-3.43
Q5-Deprived quintile
5
1.6
2.16
0.92-5.08
3
1.8
2.12
0.98-4.61
Q1-Privileged quintile
21
8.3
1
26
8.6
1
Q2
27
10.5
1.26
0.98-1.63
28
9.8
1.14
0.90-1.45
Q3
26
10.4
1.25
0.96-1.63
29
11.1
1.29
1.02-1.64
Q4
29
11.4
1.37
1.06-1.78
26
11.4
1.32
1.03-1.69
Q5-Deprived quintile
32
12.0
1.44
1.10-1.87
26
12.9
1.50
1.17-1.93
Falls from or out of building or other structure (E882)
Other falls from one level to another (E884)
Falls on same level from tripping, slipping or stumbling (E885)
Q1-Privileged quintile
31
12.5
1
32
10.9
1
Q2
32
11.9
0.95
0.76-1.19
30
10.4
0.96
0.77-1.20
Q3
23
8.8
0.70
0.55-0.90
32
12.2
1.13
0.90-1.40
Q4
29
11.6
0.92
0.73-1.17
29
12.7
1.17
0.93-1.46
Q5-Deprived quintile
26
10.2
0.82
0.63-1.05
17
8.5
0.78
0.60-1.02
Falls on same level from collision, pushing or shoving by or with another person (E886)
Q1-Privileged quintile
9
3.4
1
12
3.0
1
Q2
7
2.8
0.81
0.52-1.25
14
3.7
1.25
0.84-1.87
Q3
10
3.7
1.09
0.71-1.66
15
2.8
0.94
0.60-1.45
Q4
7
2.8
0.82
0.51-1.30
17
3.3
1.11
0.72-1.71
Q5-Deprived quintile
7
2.9
0.84
0.52-1.36
13
2.7
0.90
0.55-1.47
Q1-Privileged quintile
12
5.2
1
16
5.2
1
Q2
11
4.6
0.89
0.61-1.28
16
5.5
1.06
0.77-1.44
Q3
15
5.9
1.15
0.81-1.63
17
6.4
1.21
0.89-1.65
Q4
15
5.7
1.11
0.78-1.59
15
6.8
1.29
0.94-1.78
Q5-Deprived quintile
23
8.2
1.58
1.12-2.23
12
6.4
1.22
0.86-1.73
Other falls (E881, E883, E887 and E888)
* Rate adjusted for the other dimension of the index, age, sex and area of residence
† Relative risk
Sources: MSSS, Med-Écho hospitalization records, 2000 to 2004
MSSS, demographic outlook
Vol 29, No 2, 2009
65
Chronic Diseases in Canada
Table 9
Average annual number, adjusted* hospitalization rate and RR† for the incident location of severe injuries,
by the two dimensions of the deprivation index in children 14 years or under, for all of Quebec, 2000 to 2004
Deprivation Index
Incident location
Material dimension
Number
Rate
RR
Social dimension
95% CI
Number
Rate
RR
95% CI
Home
Q1-Privileged quintile
80
32.5
1
Q2
91
36.5
1.12
0.92-1.38
96
32.4
1
109
40.3
1.24
1.03-1.50
Q3
91
36.7
1.13
Q4
112
45.9
1.41
0.92-1.38
95
38.2
1.18
0.97-1.43
1.15-1.73
106
45.4
1.40
1.16-1.69
Q5-Deprived quintile
125
48.1
1.48
1.21-1.82
94
47.0
1.45
1.19-1.77
Q1-Privileged quintile
27
9.9
1
24
7.2
1
Q2
25
9.5
0.97
Q3
19
7.2
0.73
0.67-1.39
25
8.5
1.18
0.82-1.69
0.50-1.06
23
8.2
1.14
0.79-1.64
Q4
18
6.8
Q5-Deprived quintile
18
6.8
0.69
0.47-1.02
20
8.6
1.18
0.82-1.72
0.69
0.46-1.04
15
7.7
1.07
0.71-1.61
Q1-Privileged quintile
6
2.7
1
8
2.8
1
Q2
8
3.3
1.22
Q3
8
3.0
1.12
0.75-1.97
8
2.7
0.96
0.61-1.51
0.68-1.85
9
3.2
1.17
0.75-1.82
Q4
7
2.7
Q5-Deprived quintile
8
2.8
1.01
0.60-1.69
8
3.5
1.25
0.80-1.97
1.03
0.61-1.74
4
2.4
0.87
0.50-1.49
Q1-Privileged quintile
6
5.5
1
Q2
6
4.1
8
3.9
1
0.74
0.52-1.07
11
4.5
1.17
0.84-1.62
Q3
8
Q4
7
4.2
0.76
0.53-1.10
7
3.9
1.01
0.74-1.37
3.7
0.68
0.47-0.99
7
4.6
1.18
0.84-1.66
11
4.8
0.88
0.61-1.26
5
5.8
1.48
1.04-2.12
Q1-Privileged quintile
26
13.1
1
45
14.2
1
Q2
35
14.5
1.11
0.87-1.42
48
15.7
1.11
0.90-1.37
Q3
45
17.2
1.32
1.04-1.67
54
20.0
1.41
1.15-1.73
Q4
49
18.1
1.38
1.09-1.76
38
17.2
1.21
0.98-1.51
Q5-Deprived quintile
63
20.7
1.58
1.24-2.01
32
19.8
1.39
1.11-1.76
Recreational or sports area
Public building
Other specified location
Q5-Deprived quintile
Unspecified location
* Rate adjusted for the other dimension of the index, age, sex and area of residence
† Relative risk
Sources: MSSS, Med-Écho hospitalization records, 2000 to 2004
MSSS, demographic outlook
Chronic Diseases in Canada
66
Vol 29, No 2, 2009
Region in the United Kingdom, associations
persist upon examination of severe injuries
even when injury mechanisms are studied
independently.22 These findings, however,
should be interpreted with caution, given
the weak frequency upon which they are
based. Moreover, the nature of the injuries
sustained also appears to be associated
with socio-economic disparities, skull frac­
tures and traumas, and lower limb
fractures occurring proportionally more
often in children from deprived areas. These
results are particularly significant, as they
suggest that severe injury hospitalizations
(i.e. skull fractures, intracranial injuries
and lower limb fractures) are strongly
associated with deprivation, whereas no
association was found for somewhat less
severe injuries (i.e. upper limb fractures)
involving mechanisms known for their
strong association with deprivation.
Whereas there are numerous associations
between socio-economic characteristics
and injury risk, how the causal link oper­
ates has not been clearly demonstrated.
These associations could be attributable
to multiple individual or environmental
factors that directly or indirectly influence
injury risk.41 A significant amount of
the research conducted to date has con­
centrated on individual characteristics to
explain the differences observed in the
risk of young children sustaining an injury,
including family characteristics such as
early motherhood,26,42,43 single-parent
families,3,38 or even the number of children
in the household.14 Other authors have also
pointed to the effect of parents with a low
level of education.8,14 These factors, usually
associated with poverty, would particularly
influence the immediate level and nature
of child supervision. Pless et al. (1989)
emphasize that the links between child
behavioural characteristics and injury risk
in pedestrians and bicyclists were more
tenuous than the links between family or
neighbourhood characteristics and such
injuries. It is obvious here that children
from socially isolated environments are at
greater risk of hospitalization following an
injury, perhaps in part, because the quality
of the social network affects the parent’s
mental health,52 a risk factor in child
injuries. Still another possibility is that, in a
single-parent situation, the parent assumes
Vol 29, No 2, 2009
several responsibilities alone, which may
limit the nature of supervision provided to
the child.54
Other approaches have shed new light on
the contribution of environmental factors,
including the home environment, which
can directly or indirectly affect children.
With respect to road injuries, for example,
a number of authors have suggested that
children from deprived areas live in neigh­
bourhoods that have particular char­
acteristics. The arrangement of roads and
buildings can influence risk exposure due
to the volume and speed of motor vehicle
traffic, on-street parking and the lack of
safe play areas, causing children to use the
roads for “recreation”.24,41,44 In addition,
children from deprived areas are more likely
to walk to school compared with their peers
in privileged areas, and be accompanied
by an adult less often.45,46 It is difficult
to separate the variations attributable to
individual or environmental characteristics.
However, recent observations through
multilevel analysis have demonstrated
an effect on injury risk independent of
home environment—regardless of the
effects of individuals’ socio-economic
characteristics.26,37,47
With regard to the other mechanisms, other
home environment characteristics have
been identified as influencing the risk of
injury. Housing conditions can affect injury
risk48 in that poor housing is less likely to
meet existing safety standards, especially
for electrical and heating systems and
stairs. Overcrowded, dilapidated housing
generally tends to be occupied by materially
deprived families. Moreover, it appears
that safe practices and the ownership of
safe equipment is less common in families
from deprived areas than in those from
privileged areas.49 From this viewpoint,
studies carried out in the United Kingdom
suggest that hospitalizations for poisoning,
particularly from the use of benzodi­
azepine, antidepressants, and cough and
cold medications, were significantly higher
in children from deprived areas.20 The
authors explain that the association is due
to a greater exposure to these substances in
deprived areas, owing to the quantity and
availability of potentially toxic products,
the locations where they are stored and the
67
containers used to store them. Finally,
poverty plays an important role in the lack
of ownership of safe equipment and the
adoption of safe practices.49 In other words,
the observed differences could be bridged
more easily by eliminating economic bar­
riers and distributing safe, affordable or free
equipment, for example.
Limitations of this study
The data forming the basis of our study
include all hospitalizations in Quebec.
However, this knowledge base does not
include a known, standardized indicator
that would establish a severity level of
injuries sustained. Consequently, we used
an approximate measure based on the
eligibility criteria used by the RTQ data­
base. This could give rise to criticism
that the measure’s capacity to eliminate
administrative variations has not been
proven. Furthermore, the MED-ECHO data­
base contains no socio-economic infor­
ma­tion. To circumvent this problem, we
used the deprivation index developed by
Pampalon and Raymond (2000). However,
an ecological index opens the door to
similar errors, i.e. the socio-economic
characteristics measured for a DA do not
correspond to those of the families within
the area. Given the number of studies in
which this relationship has been observed
through individual and ecological data,
we believe that this limitation is of little
relevance in this case.
Conclusion
The results of this study suggest that dep­
rivation in Quebec children is associated
with hospitalization risks for a great range
of unintentional injuries for both dimen­
sions of deprivation. In addition, exami­
nation of the injuries sustained tells us that
the associations noted are not the result of
differential health care services use or even
administrative variations, given severe
injuries are equally related to deprivation.
These findings should be taken into account
when developing preventive strategies.
The use of socio-demographic characteristics
to identify children at risk of injury in
order to develop targeted interventions
has recently come under criticism. These
Chronic Diseases in Canada
critics have instead recommended using a
population-based approach adapted to each
population sub-group,43 because restricting
preventive interventions to children from
deprived areas means a significant number
of injury victims would not benefit from
the interventions. On the other hand, it
appears that a physical modification of
the environment is more successful in
preventing injuries than most educational
programs.50 From this perspective, measures
aimed at modifying infrastructure to reduce
traffic, while taking into account socioeconomic inequalities related to injuries in
young pedestrians, have produced positive
results in injury rates, in absolute terms,
and in reducing relative inequalities.51
Acknowledgements:
The authors wish to thank Yvonne
Robitaille, of the National Public Health
Institute of Quebec, for her sound advice
and thoughtful suggestions. We would also
like to thank Robert Pampalon, also of the
National Public Health Institute of Quebec,
for his recommendations on the use of the
deprivation index.
References
1.
2.
Polinder S, Meerding WJ, Toet H et al.
Prevalence and prognostic factors of disability
after childhood injury. Pediatrics. 2005;116:
810-17.
Dowswell T, Towner E. Social deprivation
and the prevention of unintentional injury
in childhood: a systematic review. Health
Educ Res. 2002;17:221-37.
5. Dougherty G, Pless IB, Wilkins R. Social
class and the occurrence of traffic injuries
and deaths in urban children. Can J Public
Health. 1990;81:204-9.
6. Edwards P, Roberts I, Green J et al. Deaths
from injury in children and employment
status in family: analysis of trends in
class specific death rates. Clinical research
edition. BMJ. 2006;333:119.
7. Laflamme L, Diderichsen F. Social differences
in traffic injury risks in childhood and
youth–a literature review and a research
agenda. Inj Prev. 2000;6:293-98.
17. Faelker T, Pickett W, Brison RJ.
Socioeconomic differences in childhood
injury: a population based epidemiologic
study in Ontario, Canada. Inj Prev.
2000;6:203-8.
18. Gilbride SJ, Wild C, Wilson DR, et al. Socioeconomic status and types of childhood
injury in Alberta: a population based study.
BMC Pediatr. 2006;6:30.
8. Scholer SJ, Hickson GB, Ray WA.
Sociodemographic factors identify US
infants at high risk of injury mortality.
Pediatrics. 1999;103:1183-188.
19. Durkin MS, Davidson LL, Kuhn L, et al.
Low-income neighbourhoods and the
risk of severe pediatric injury: a smallarea analysis in northern Manhattan. Am
J Public Health. 1994;84:587-92.
9. Roberts I, Power C. Does the decline in
child injury mortality vary by social class?
A comparison of class specific mortality
in 1981 and 1991. BMJ. 1996;313:784-86.
20. Groom L, Kendrick D, Coupland C et al.
Inequalities in hospital admission rates for
unintentional poisoning in young children.
Inj Prev. 2006;12:166-70.
10. Birken CS, Parkin PC, To T, et al. Trends
in rates of death from unintentional injury
among Canadian children in urban areas:
influence of socioeconomic status. CMAJ.
2006;175:867.
21. Hasselberg M, Laflamme L, Weitoft GR.
Socioeconomic differences in road traffic
injuries during childhood and youth: a
closer look at different kinds of road users.
J Epidemiol Community Health. 2001;55:
858-62.
11. Cubbin C, Smith GS. Socioeconomic
inequalities in injury: critical issues in
design and analysis. Annu Rev Public
Health. 2002;23:349-75.
12. Joly MF, Foggin PM, Zvagulis I, et al.
Bicycle accidents among children in the
urban environment. Can J Public Health.
1989;80:351-54.
3. Blakely T, Atkinson J, Kiro C, et al. Child
mortality, socioeconomic position, and oneparent families: independent associations
and variation by age and cause of death.
Int J Epidemiol. 2003;32:410-18.
13. Joly MF, Foggin PM, Pless IB. Geographical
and socio-ecological variations of traffic
accidents among children. Soc Sci Med.
1991;33:765-69.
4. Brownell M, Friesen D, Mayer T. Childhood
injury rates in Manitoba: socioeconomic
influences. Can J Public Health. 2002;
Suppl 2:93:S50-S56.
14. Pless IB, Verreault R, Tenina S. A casecontrol study of pedestrian and bicyclist
injuries in childhood. Am J Public Health.
1989;79:995-98.
15. Pless IB, Peckham CS, Power C. Predicting
traffic injuries in childhood: a cohort
analysis. J Pediatr. 1989;115:932-38.
Chronic Diseases in Canada
16. Pless IB, Verreault R, Arsenault L, et al. The
epidemiology of road accidents in child­
hood. Am J Public Health. 1987;77:358-60.
68
22. Hippisley-Cox J, Groom L, Kendrick D et
al. Cross sectional survey of socioeconomic
variations in severity and mechanism of
childhood injuries in Trent 1992-7. Clinical
research edition. BMJ. 2002;324:1132.
23. Laing GJ, Logan S. Patterns of unintentional
injury in childhood and their relation
to socio-economic factors. Public Health.
1999;113:291-94.
24. Mueller BA, Rivara FP, Lii SM, et al.
Environmental factors and the risk for
childhood pedestrian-motor vehicle col­
lision occurrence. Am J Epidemiol. 1990;
132:550-60.
25. Pomerantz WJ, Dowd MD, Buncher CR.
Relationship
between
socioeconomic
factors and severe childhood injuries.
J Urban Health. 2001;78:141-51.
Vol 29, No 2, 2009
26. Reading R, Langford IH, Haynes R, et al.
Accidents to preschool children: comparing
family and neighbourhood risk factors.
Soc Sci Med. 1999;48:321-30.
27. Rivara FP, Barber M. Demographic analysis
of childhood pedestrian injuries. Pediatrics.
1985;76:375-381.
28. Silversides JA, Gibson A, Glasgow JF, et al.
Social deprivation and childhood injuries
in North and West Belfast. Ulster Med J.
2005;74:22-28.
29. Kelly S, Miles-Doan R. Social inequality and
injury: do morbidity patterns differ from
mortality? Soc Sci Med. 1997;44:63-70.
30. Lyons RA, Lo SV, Heaven M, et al. Injury
surveillance in children–usefulness of
a centralised database of accident and
emergency attendances. Inj Prev. 1995;
1:173-76.
31. Lyons RA, Delahunty AM, Heaven M, et al.
Incidence of childhood fractures in affluent
and deprived areas: population based study.
BMJ (Clinical research ed). 2000;320:149.
37. Kendrick D, Mulvaney C, Burton P, et al.
Relationships between child, family and
neighbourhood characteristics and child­
hood injury: a cohort study. Soc Sci Med.
2005;61:1905-915.
38. Roberts I, Pless B. Social policy as a cause
of childhood accidents: the children of lone
mothers. BMJ. 1995;311:925-28.
39. Alwash R, McCarthy M. Measuring severity
of injuries to children from home accidents.
Arch Dis Child. 1988;63:635-38.
40. Walsh SS, Jarvis SN, Towner EM, et al.
Annual incidence of unintentional injury
among 54,000 children. Inj Prev. 1996;
2:16-20.
41. Towner E, Dowswell T, Errington G, et
al. Injuries in children aged 0-14 years
and inequalities. Newcastle: Health
Development Agency, Community Child
Health, 2005.
42. Braun PA, Beaty BL, DiGuiseppi C, et al.
Recurrent early childhood injuries among
deprived children in primary care settings.
Inj Prev. 2005;11:251-55.
32. Robitaille Y, Gratton J. Les chutes chez les
adultes âgés : vers une surveillance plus
fine des données d’hospitalisation. Québec:
Institut national de santé publique du
Québec, 2005. 19 p.
43. Kendrick D, Marsh P. How useful are
sociodemographic characteristics in iden­
tifying children at risk of unintentional
injury? Public Health. 2001;115:103-7.
33. Lyons RA, Brophy S, Pockett R, et al.
Purpose, development and use of injury
indicators. Int J Inj Contr Saf Promot.
2005;12:207-11.
44. Roberts I, Norton R, Jackson R, et al. Effect
of environmental factors on risk of injury of
child pedestrians by motor vehicles: a casecontrol study. BMJ. 1995;310:91-94.
34. Reimers A, Laflamme L. Neighbourhood
social and socio-economic composition and
injury risks. Acta Paediatr. 2005;94:1488-494.
45. Towner EM, Jarvis SN, Walsh SS, et al.
Measuring exposure to injury risk in
school children aged 11-14. BMJ. 1994;
308:449-52.
35. Pampalon R, Raymond G. Indice de
défavorisation matérielle et sociale : son
application au secteur de la santé et du
bien-être. Santé, société et solidarité 2003;
1:191-208.
36. Pampalon R, Raymond G. A deprivation
index for health and welfare planning in
Quebec. Chronic Dis Can. 2000;21:104-13.
Vol 29, No 2, 2009
48. Shenassa ED, Stubbendick A, Brown MJ.
Social disparities in housing and related
pediatric injury: a multilevel study. Am J
Public Health. 2004;94:633-39.
49. Hapgood R, Kendrick D, Marsh P. How
well do socio-demographic characteristics
explain variation in childhood safety prac­
tices? J Public Health Med. 2000;22:307-11.
50. Peek-Asa C, Zwerling C. Role of
environmental interventions in injury
control and prevention. Epidemiol Rev.
2003;25:77-89.
51. Jones SJ, Lyons RA, John A, et al. Traffic
calming policy can reduce inequalities in
child pedestrian injuries: database study.
Inj Prev. 2005;11:152-56.
52. Laflamme L. Explaining socio-economic
differences in injury risks. Inj Control Saf
Promot. 2001;8:149-53.
53. Hamel D, Pampalon R. Trauma and depri­
vation in Quebec. Québec: Institut national
de santé publique du Québec, 2002. 8 p.
46. Roberts I, Norton R. Auckland children’s
exposure to risk as pedestrians. N Z Med J.
1994;107:331-33.
47. Haynes R, Reading R, Gale S. Household
and neighbourhood risks for injury to 514 year old children. Soc Sci Med. 2003;
57:625-636.
69
Chronic Diseases in Canada
Injury data in British Columbia: policy maker perspectives
on knowledge transfer
C Mitton, PhD (1, 2); YC MacNab, PhD (2,3); N Smith, MA (1); L Foster, PhD (4)
Abstract
Provincial and regional decision makers in the injury prevention field were interviewed
in British Columbia (B.C.) to obtain their views about best processes for the transfer
or dissemination of relevant data. These decision makers (n = 13) indicated that data
should provide them with a holistic and comprehensive picture to support their decision
processes. In addition, they felt information about injury types and rates should be
linked backward to determinants or causes and forward to consequences or outcomes.
This complete chain of data is needed for planning and evaluating health promotion
interventions. It was also felt that data providers needed to devote more effort to fostering
effective receptor capacity, so that injury prevention professionals will be better able
to understand, interpret and apply the data. These findings can likely be generalized to
other jurisdictions and policy areas, and offer additional insight into the practicalities of
knowledge transfer and exchange in researcher/decision maker partnerships.
Keywords: burden of injury, knowledge transfer, knowledge utilization,
policy, injury prevention
Introduction
Much effort can and has been invested in
generating data about the impact and bur­
den of chronic diseases in Canada, but do
we know how best to utilize these data in
policy and practice? Knowledge transferi
(KT) refers to a process whereby infor­
mation is made available to decision makers
through interactive engagement. Over the
past decade, researchers and policy makers
have described what inhibits or encourages
KT.1-5 Based on these findings, guidance
about mechanisms or strategies for effec­
tive implementation of KT has been
published.6-7 Common recommendations
include establishing ongoing collaborative
relationships between the researcher and
decision maker;6,8-9 fostering appropriate
attitudes, values, culture and capacity
within health care organizations;3 and
offering clear and timely communication
in a shared language appropriate to the
target audience.2,6,10 However, it remains
important to pursue a more substantial
evidence base around KT practices to
ensure research and data-collection efforts
are directed appropriately. Empirical case
studies with actual datasets in particular
contexts, such as the one reported in this
paper, should advance our understanding
and may offer potential for immediate
improvements to practice. While the find­
ings here relate most directly to those
interested in injury prevention and health
promotion, the issues generated from this
study should also be applicable to KT
for chronic diseases as well as in health
care contexts.
Methods
Data collection involved semi-structured
interviews with 13 key decision makers
with direct knowledge of injury prevention
policy in B.C. The interview schedule
is appended. The following organizations
were represented in the sample: the B.C.
Ministry of Health (recently divided into
the Ministry of Health Services and a sep­
arate new Ministry of Healthy Living and
Sport), the Provincial Health Services
Authority (PHSA) and two of the five regional
health authorities (RHAs). Interviewees
were primarily senior to middle managers,
with job titles such as Executive Director,
Director, Manager or Project Lead. These
individuals would typically be responsible
for broad planning, priority setting and/or
evaluation functions. The interviews were
conducted in June and July 2005.
The interviewer provided each inform­
ant with sample data to look over (see
Tables 1 and 2, and Figure 1, as examples).
Informants were then asked general
questions about what sources of data they
currently accessed and to provide feedback
about the sample data sets. Each interview
was audio recorded (with permission)
and transcribed. Analysis proceeded using
the constant comparison method, i.e.
themes and sub-themes were developed
i For details see: http://www.researchtopolicy.ca/whatwehavelearned/develop_approach.asp. Related terms include knowledge translation,
knowledge exchange, knowledge utilization, and research dissemination.
Author References
1 Faculty of Health and Social Development, University of British Columbia Okanagan
2 Child and Family Research Institute of British Columbia
3 Dept. of Health Care and Epidemiology, University of British Columbia
4 Faculty of Human and Social Development and Department of Geography, University of Victoria
Correspondence: Craig Mitton, PhD, Assistant Professor, Faculty of Health and Social Development, University of British Columbia Okanagan, 3333 University Way, Kelowna, BC V1V 1V7,
Tel: 250-807-8704, Fax: 250-807-8085, Email: [email protected]
Chronic Diseases in Canada
70
Vol 29, No 2, 2009
inductively.11 Written notes prepared by the
interviewer provided an additional source
of data. Ethics approval was granted by the
University of British Columbia Behavioural
Research Ethics Board.
The policy environment for
injury prevention in B.C.
The five RHAs have been given the respon­
sibility to develop and implement plans
and programs for injury prevention. They
must determine how much of a priority to
give to injury prevention compared to other
possible initiatives, assess the suitability
and effectiveness of program options,
engage other sectors where appropriate and
evaluate the success and relative costs
and benefits of their efforts. The RHAs are
at varying stages of development in injury
prevention policy. The PHSA provides
support with data gathering, analysis and
knowledge translation activities and is
an important source of data on patient
safety, particularly with respect to adverse
drug events, nosocomial infections and
radiation-related injury.
The role of the B.C. provincial government is
primarily one of stewardship. The province
also leads surveillance and monitoring
efforts. The Health Authority Division of
the Ministry of Health Services negotiates
performance agreements with the RHAs.
These presently include benchmarks for
falls and, generally speaking, are meant to
hold the RHAs to public account for their
outcomes, based on the resources allocated
to them. The Healthy Children, Women and
Seniors Branch (recently transferred to help
create the new Ministry of Healthy Living
and Sport) is the primary policy making and
advisory centre for the provincial gov­
ernment on injury matters, along with rec­
ommendations from the Provincial Health
Officer.
The Provincial Health Officer, who now has
functions in both the Ministry of Health
Services and the new Ministry of Healthy
Living and Sport, has provided assistance to
RHAs in setting their benchmarks related
to falls and has produced a special report on
falls among the elderly that: 1) outlines the
magnitude of this issue; 2) measures
the impact on the health care system; and
Vol 29, No 2, 2009
3) provides recommendations for evidencebased prevention strategies.12 Each RHA
must apply these in its own context, of
course. Further research and data-collection
efforts, in support of developing policy
recommendations, have been undertaken
since 1997 by the British Columbia Injury
Research and Prevention Unit (BCIRPU),
located at the Children’s and Women’s
Health Centre of British Columbia. These
roles were outsourced following the dis­
mantling of the Ministry’s own internal
Office of Injury Prevention in 2001. The
BCIRPU is expanding its data sources by
developing agreements with the Workers’
Compensation Board (i.e. WorkSafeBC),
Statistics Canada, the Canadian Institute for
Health Information (CIHI), the Insurance
Corporation of British Columbia (ICBC),
the RCMP and BC Ambulance Services.
Other key stakeholders (e.g. Red Cross,
RCMP) are involved through such
mechanisms as the BC Injury Prevention
Leadership Network, a provincial steering
committee meant to provide guidance and
advice on broad policy directions. Still other
entities, such as the BC Aboriginal Health
Network, the BC Sport & Recreation Injury
Free Advisory Committee and the BC Falls
Prevention Coalition, also exist. The BC
Healthy Living Alliance, a consortium of
primarily chronic disease organizations also
plays a supporting role in issues related to
healthy living, including injury prevention.
The province, PHSA and RHAs are aware
of the interest of these groups, but are not
always fully aware of their role or how they
might link to other stakeholders. In short,
the policy environment pertaining to injury
prevention in B.C. is somewhat fragmented,
consisting of multiple departments, agen­
cies and organizations.
Injury data used in the research
Within this policy environment, researchers
at the University of British Columbia have
been working on a set of statistical meth­
ods to further delineate injury data in this
province. This case study is part of an
ongoing research program, entitled Burden
of Injury in BC and Its Local Communities:
Information and Evidence for Communitybased Prevention Strategy, Health Policy
and Service Provision (short-titled Burden
71
of Injury in BC), currently funded by the
Canadian Institutes of Health Research.
The main objectives of the Burden of Injury
in BC project are: 1) to develop a synthesis
of analytic methods for a systematic
burden-of-injury research framework that
encompasses space-time surveillance moni­
toring, burden assessment, risk assessment,
research dissemination and knowledge
translation; and 2) to apply these methods
to examine burden-of-injury mortality
and disability in B.C. and its local
communities.13-14
During the policy maker interviews, inform­
ants were presented with sample tables
of burden-of-injury profiles for each of the
five RHAs and for B.C. as a whole. This
included mortality and hospitalization
counts and rates for major causes of injuries,
stratified by gender and based on 1991
to 2000 population-based administrative
data.13-14 Tables 1 and 2 present the
corresponding cause-specific statistics on
burden-of-injury, measured by years of life
lost to premature death (YLLs), years of
life lived with disability (YLDs), and dis­
ability adjusted life years (DALYs).15 These
burden-of-injury measures were derived
based on observed injury mortality and
hospitalization counts for the five health
authorities and for B.C. as a whole for the
calendar years of 1991-2000.14 The YLL,
YLD and DALY estimates in Tables 1 and
2 were derived using disability weight and
duration estimates from the 1990 Global
Burden of Disease study, specified with
zero age weighting and an annual 3%
discount rate.14-15 Note that the DALYs
are “health gap” measures that allow
the combined impact of mortality and
morbidity to be incorporated and assessed
simultaneously. The DALY measures were
developed under the Global Burden of
Disease project,15 as population health
indicators for public health assessment
and as a “currency” for cost-effectiveness
analysis with respect to priority setting
and evaluation of health interventions.14-15
Informants were also asked to view
two sample maps that depicted annual
iatrogenic injury risk estimates for chil­
dren and youth aged 1 to 19 years for
16 geographic subdivisions of the health
authorities, i.e. health service delivery
Chronic Diseases in Canada
areas (HSDAs). One of the maps, presented
in Figure 1, highlights the HSDAs with
high/low iatrogenic injury risks for the
male population. Bayesian estimates of
annual relative risks, quantified by the
ratios of HSDA rates over a B.C. average
were derived in order to prepare this
map.13,16 An HSDA was identified as having
a high (or low) iatrogenic injury risk if the
95% interval estimates (i.e. the upper and
lower limits) of the relative risk were above
(or below) one.13,16
much factual and descriptive data are avail­
able about injury incidence and prevalence,
i.e. what types of injury occur and the
demographics of those injured. This is use­
ful, but these policy makers also wanted
to know about determinants and contexts
of injury, i.e. they sought information to
help them understand why circumstances
in their own communities are the way they
are, and whether there are any unique local
concerns, needs or circumstances that ought
to be acknowledged and addressed.
Results
Decision makers also wanted to be able
to look at data from many possible angles
and to consider injury findings in light of
a range of variables. Demographics of the
people involved, times and locations of
incidents, and conditions in the social or
physical environment are among the factors
that might be relevant in understanding
and explaining the local injury picture.
This kind of information would likely come
from a retrospective investigation and a
description of each incident.
Findings from the interviews relate to both
the content and the processes that should
lead to effective dissemination and uptake
of injury data by researchers and decision
makers in B.C. The content of data sets
needs to provide a rounded perspective
that addresses both injury causes and con­
sequences. Dissemination processes should
be targeted to specific audiences and recog­
nize that individuals and organizations
may have differing degrees of capacity to
use information.
Data content
The interviewees were familiar with and
have used DALY measures for health plan­
ning purposes, though one suggests that
“[we] probably don’t use them as much as
we should, but we definitely use them” [#3].
However, they suspect that other audiences
are less familiar with this measure and
would be better versed with indicators
like morbidity and hospitalization rates.
One interviewee suggested that DALYs
might not have as much traction as other
measures, because they are not linked or
aligned to the “strategic goals of the health
system” [#1]. Health system managers
in B.C. are not held accountable by their
performance agreements with the province
for changes in this measure as they are for
some other targets. This suggests, as do
findings reported below, that institutional or
systemic barriers are important influences
on the KT process.
These decision makers wanted rich data so
that they could view policy issues in their
full complexity. They wanted a data “chain” so
that they could look both forward and back­
ward from specific injury events. Currently,
Chronic Diseases in Canada
I think what I would like to be able to
see is, okay, what are the various types
of falls? What are the causes and the
impacts and who is most at risk? And
so, we need to know first of all, who’s
falling, when are they likely to fall and
what are they doing when they are
falling? [#1]
We know motor vehicle crashes [are]
number one, but we need to break it
down. What does that actually mean?
Is it commercial drivers, is it alcohol,
is it due to lack of seatbelt use? [#3]
In addition, determinant and context data
are necessary in order for planners to
decide how best to intervene to solve the
problem and carry out evaluative efforts.
The informants had a particular interest
in these functions and desired data that
would assist them in effective planning and
evaluation efforts.
I think it’s a matter of articulating,
understanding the issues and the deter­
minants, and how the solutions come
in to make a difference, because the
solution is obviously predicated on an
understanding of the determinants. [#1]
72
When I think about data, I’m not
thinking about rate of; I’m constantly
linking it back to… to evidence for
interventions and the data related
to that, and linking it back then to
my population and the data of my
population so that I can develop an
appropriate intervention. It’s so impor­
tant to link health data with evidence
for effective interventions. [#9]
Policy makers also wanted data about the
aftermath of injury. This would require
following injury cases over time and
gathering information about medium- to
long-term consequences at the individual
and population levels (including full
costing). The “business case” for investing
in injury prevention depends upon good
knowledge of outcomes.
The outcome—figuring out, do they
recover; do they go back to work; are
they able to walk; are they able to
resume previous activities or does it
precipitate a downward spiral in their
health? And, as so often with the
elderly, that’s what happens. You know,
we can talk about ActNow [a provincial
government health promotion ini­
tiative], we can talk about healthy
living and exercise and so forth [but]
often, a setback like that for an elderly
person is extremely detrimental to
their health. [#1]
For policy decisions, we’ve really got
to turn it [injury data] into a bit of a
business case and a business case
model [for investment in prevention
interventions]. … We’re trying to con­
vince government that this is impor­
tant to pay attention to, because this
burden of illness on the system is
really a surrogate for the costs we’re
spending in the healthcare system. …
So, my advice would be to take
this information and convert it into
something that the decision makers
on the resource allocation ledger can
understand, interpret and begin to
believe that, if I invest in this other
end of the continuum, I’m going to
take some pressure off the far end—a
very compelling argument. … For them
to know what percentage of spending
Vol 29, No 2, 2009
contributes from the overall injuries on
sports injuries, drowning and burns,
they can then start targeting their
response. [#2]
However, according to these informants, the
data that would flesh out this data chain are
unavailable or difficult to obtain in B.C.
Processes for data dissemination
Several comments directly address dissemi­
nation processes. A range of ideas was raised
by the interviewees (e.g. using “knowledge
brokers”) about how the information
embedded in indicator data might best be
transferred to these decision makers, but
all emphasized that the credibility of who
transfers the knowledge was important
and that the knowledge be tailored to each
specific target audience. Those who set
priorities, for instance, may need different
data about the costs and outcomes of injury
than those who are responsible for detailed
program evaluation.
It’s really dependent on what you’re
using it for and who your audience is.
From a decision making point of view,
if you understand what they say, then
it doesn’t matter. You can look at raw
numbers or raw tables and you can
gather the information that you need
to make the decision. But it’s more in
putting a rationale, or putting a busi­
ness case forward or that kind of stuff
where you need to have the whole
array of tools. [#5]
The visual depiction of data in map format
was seen as having appeal for non-experts
(e.g. board or senior-management decision
makers). “If you want to disseminate this
information widely, to people in policy
or practitioners, or whoever you want to
digest this information, I think the maps
are more effective.” [#1] Maps were seen
as a quick way to transmit information to
audiences that are pressed for time and
that are unable or unwilling to read lengthy
reports or academic journal articles. In this
way, they are comparable to the briefing
notes or fact sheets touted by organizations
like the Canadian Health Services Research
Foundation (www.chsrf.ca) as an effective
means of reaching senior decision makers
with research-based evidence.
other people. … It’s nice to see where
you are pictorially in relation to the rest
of the province.” [#7] Interestingly, one
respondent from the provincial government
framed the issue of comparison in the
context of the province’s specific intention
to encourage competition among the health
authorities:
I like the whole idea of spatial mapping
around injuries; I think that’s the way
we should go … we’re sort of setting
up the health authorities to be quite
competitive, so we need to be able to
map which health authorities have got
the highest injury rates and relative
risks of certain injuries across the
province. [#3]
Maps were identified as a valuable commu­
nication tool and a good way to make a
point: “If I’m going to be using [data] for
the purposes of communication to others,
ok, I’ll go for the spatial mapping.” [#6]
Of course, analysis and interpretation are
embedded in the argumentative use of data
in any form:
Maps were seen as particularly valuable for
presenting comparisons. “You can look at
it on the page and compare [yourself with]
Table 1
Years of life lost to premature death (YLL), years of life lived with disability (YLD) and disability adjusted life years (DALYs)
in British Columbia males, by health authority and cause of injury, 1991-2000 cumulative total
Males
Injury
Health authorities
Interior
Fraser
Vancouver
coast
Vancouver
Island
Northern
B.C.
B.C.
Years of life lived with disability (YLD)
Road traffic injuries
8500
10 096
5489
5951
4656
Other transport injuries
1691
1424
607
1039
1186
5946
258
372
312
241
175
1357
Falls
3848
5526
3630
3564
2037
18 604
Burns/fires/scalds
1121
1370
1119
982
1033
5624
Drowning
174
336
181
141
120
952
Sports injuries
562
766
467
470
275
2540
Natural and environmental factors
244
180
125
135
155
838
4219
4810
1872
2943
2738
16 583
94
152
97
70
52
464
Other unintentional injuries
8622
9826
5921
7118
6088
37 576
Suicide and self-inflicted injuries
1008
1925
1183
1104
602
5822
Other intentional injuries
1219
2126
1687
1290
1116
7438
Poisoning
Machinery injuries
Suffocation and foreign bodies
Vol 29, No 2, 2009
73
34 692
Chronic Diseases in Canada
Table 1 (continued)
Years of life lost to premature death (YLL), years of life lived with disability (YLD) and disability adjusted life years (DALYs)
in British Columbia males, by health authority and cause of injury, 1991-2000 cumulative total
Males
Injury
Health authorities
Interior
Fraser
Vancouver
coast
Vancouver
Island
Northern
B.C.
B.C.
11 081
68 973
Years of life lost to premature death (YLL)
Road traffic injuries
19 008
19 357
9565
9962
Other transport injuries
2781
2794
1904
2350
2914
12 743
Poisoning
5736
17 897
26 815
10 009
3115
63 571
Falls
3295
4284
5125
3702
1042
17 448
Burns/fires/scalds
1112
968
912
720
830
4542
Drowning
2024
2130
1781
1905
1337
9177
138
78
233
28
64
541
Natural and environmental factors
1269
373
508
374
678
3202
Machinery injuries
1089
845
243
681
897
3755
871
1433
1248
938
409
4899
Sports injuries
Suffocation and foreign bodies
Other unintentional injuries
Suicide and self-inflicted injuries
Other intentional injuries
2009
1550
1189
1595
1255
7598
16 335
21 106
19 110
15 252
7861
79 664
2420
4764
4720
2116
2021
16 040
Disability adjusted life years (DALYs)
Road traffic injuries
27 508
29 452
15 054
15 913
15 737
103 664
Other transport injuries
4473
4218
2510
3389
4099
18 689
Poisoning
5994
18 269
27 127
10 249
3289
64 928
Falls
7143
9810
8755
7266
3079
36 053
Burns/fires/scalds
2233
2338
2031
1701
1863
10 166
Drowning
2197
2466
1962
2046
1457
10 129
700
844
701
498
339
3082
Sports injuries
Natural and environmental factors
1513
553
633
509
833
4040
Machinery injuries
5307
5655
2115
3625
3635
20 338
Suffocation and foreign bodies
965
1585
1344
1008
461
5363
Other unintentional injuries
10 631
11 376
7110
8713
7343
45 173
Suicide and self-inflicted injuries
17 343
23 031
20 293
16 356
8463
85 486
3638
6890
6407
3406
3137
23 478
Other intentional injuries
Chronic Diseases in Canada
74
Vol 29, No 2, 2009
Table 2
Years of life lost to premature death (YLL), years of life lived with disability (YLD) and disability adjusted life years (DALYs)
in British Columbia females, by health authority and cause of injury, 1991-2000 cumulative total
Females
Injury
Health authorities
Interior
Fraser
Vancouver
coast
Vancouver
Island
Northern
B.C.
B.C.
Years of life lived with disability (YLD)
Road traffic injuries
3971
4501
2495
2690
2222
15 880
Other transport injuries
350
290
149
226
246
1260
Poisoning
136
224
143
141
112
755
2692
3707
2607
2668
1137
12 811
535
771
524
516
373
2720
26
34
8
25
7
101
Sports injuries
146
149
108
116
65
584
Natural and environmental factors
179
154
80
109
83
605
Machinery injuries
226
336
183
252
253
1250
2
3
2
2
1
9
2646
3341
2214
2333
1833
12 366
816
1579
908
880
555
4738
302
426
331
228
314
1600
4423
4098
29 465
Falls
Burns/fires/scalds
Drowning
Suffocation and foreign bodies
Other unintentional injuries
Suicide and self-inflicted injuries
Other intentional injuries
Years of life lost to premature death (YLL)
Road traffic injuries
Other transport injuries
8242
7846
4857
532
368
467
251
388
2005
Poisoning
2224
4428
8197
3173
1304
19 327
Falls
2841
3108
3411
3441
658
13 458
Burns/fires/scalds
532
776
139
526
454
2428
Drowning
594
422
503
684
168
2372
27
4
55
23
0
110
Natural and environmental factors
311
150
97
149
185
892
Machinery injuries
106
26
0
10
50
193
Suffocation and foreign bodies
324
600
591
639
35
2188
Other unintentional injuries
271
175
305
193
150
1093
Suicide and self-inflicted injuries
4008
6815
6364
5168
1424
23 780
Other intentional injuries
1873
2260
1878
1425
644
8080
7113
6320
45 345
Sports injuries
Disability adjusted life years (DALYs)
Road traffic injuries
Other transport injuries
12 213
12 348
7352
881
658
616
477
634
3266
Poisoning
2360
4652
8340
3314
1416
20 082
Falls
5533
6814
6018
6109
1795
26 270
Burns/fires/scalds
1068
1547
663
1042
827
5148
Drowning
620
456
512
709
176
2472
Sports injuries
173
154
163
139
65
694
Natural and environmental factors
490
303
178
257
268
1497
Machinery injuries
333
362
183
262
303
1443
Suffocation and foreign bodies
325
603
593
640
36
2197
Other unintentional injuries
2917
3516
2518
2526
1983
13 460
Suicide and self-inflicted injuries
4824
8395
7272
6048
1979
28 518
Other intentional injuries
2174
2686
2209
1653
957
9680
Vol 29, No 2, 2009
75
Chronic Diseases in Canada
Figure 1
High and low iatrogenic injury risk estimates for male British Columbia children and youth aged 1 to 19 years,
by health service delivery area, 1991-2000
1991
1992
1993
1994
1995
1996
1997
1998
1999
High relative risk
Low relative risk
2000
Chronic Diseases in Canada
Relative Risk
76
Vol 29, No 2, 2009
Figure 2
The injuries “data chain”
Information about
determinants and
contexts of injury
Information
about injuries
Information about
consequences and
outcomes of injury
Business case for
specific priorities
and actions
Information for
planning and
implementing
interventions
Something that’s spatial like this is
really effective if you’re trying to
make a point. It’s data that’s [sic] being
presented in the context of an issue
and there’s an argument to go along
with it and there’s, you know, if it’s part
of a whole package trying to illustrate
something, then this kind of spatial
representation can be really helpful,
because it puts it in context. [#9]
Finally, informants noted that effective
uptake of injury data, however presented,
depends on the capacity of organizations
like the RHAs. This includes both individual
knowledge and skills held by data managers,
policy analysts and program developers,
and the systems needed to allow these
individuals to employ their knowledge to
affect practice.
Discussion
Effective KT requires that research and
data be framed to fit the information needs
of decision makers. “Researchers tend to
describe the past and present with a focus
on the ‘what.’ Decision makers want and
need explanation and prediction. They
need to know the ‘why’ and the ‘what if’.”17
Thus, it is perhaps understandable why
Vol 29, No 2, 2009
Information from
the evaluation
of interventions
policy makers in the B.C. injury prevention
field stress the importance of knowledge
about the determinants of injury and the
potential and expected outcomes of RHA
interventions.
Based on this research, we suggest the
following approach: In order for KT to
occur in a given context, researchers should
be prepared to collect and present data in
multiple formats, reflecting the range of
decision makers’ needs and capacities.
Our respondents in this particular context
endorsed the value of visual depictions for
senior executive members. Data should also
be organized in a way that feeds the stages in
the policy development and planning cycle,
from agenda setting, policy formulation and
decision making to policy implementation
and evaluation.18-19 In particular, priority
setting and evaluation are central to the
work of health system decision makers.
Data sets should be built by researchers
in a way that better accommodates and
supports these endeavours. Of course,
researchers themselves are also limited by
time, funding and data availability in what
they are able to provide.
77
Access to the chain of information will allow
key stakeholders to follow forward and
backward linkages between determinants,
situations, interventions and outcomes.
This more complex and holistic view runs
against the tendency of researchers and
other experts to provide extensive and
detailed information about a narrow or
circumscribed subject. However, such data
may be more costly to collect and more
complex to interpret, so there are distinct
tradeoffs to be made.
To engage decision makers early in the
development of research and determination
of data needs would be of key importance,
as the literature has argued.4,9,20 Published
data reports are more likely to be seen as
relevant, understood and, perhaps, uti­
lized when there is early and up-front
participation of this nature from decision
makers.21-22
Finally, these informants made the point
quite clearly that information in itself is
insufficient to have an impact in the pol­
icy context. Information needs to fit into a
structure that can accommodate and cat­
alyze it. Much literature to date has focused
on the interchange between knowledge
producers and users at the individual
level.23 While this is no doubt important,
our research also suggests, in line with
other recent work on KT, that receptor
organizations’ systems and structures may
determine how well evidence and data can
be employed in support of health promotion
policy objectives.23-25
Our findings are summarized diagrammat­
ically in Figure 2. The injury data that
would be valued by these B.C. policy
makers make up a chain of information;
we have highlighted the interplay between
the links in the chain. In our view, this
way of looking at KT brings new insight to
this field. The nature of decision making
is not as linear as this simplified diagram
appears, of course. The determinants of
injury also likely help shape outcomes or
consequences—elderly people who are more
prone to injury in certain contexts, such as
in long term care facilities, may also have
different and less successful outcomes as a
result of their age and frailty. Appropriate
decisions would also have to account for
Chronic Diseases in Canada
and model such inter-relationships. This
suggests a need for multi-level analysis,
though again, there are trade-offs between
this better representation of the real context
in which injuries occur (so important for
designing appropriate policy interventions)
and the greater demands that such complex
analyses place on decision makers’ infor­
mation processing abilities. One of the
ongoing objectives of the Burden of Injury
in BC project is to explore and develop KT
methods that facilitate effective commu­
ni­cation of complex analytic results and
uptake of relevant information for policy
and priority considerations with respect
to injury monitoring, prevention and
intervention.
The primary limitation of the current
study was the small sample size. It may
be that more extensive sampling would
have produced other views and further
insight into the critical issues around KT
and burden-of-injury data. That said, our
sampling strategy purposefully included
a range of decision makers in different
organizations and with different roles and
levels of responsibility in injury preven­
tion policy and program development.
Convergence of perspectives was observed.
However, our informants did not include
those identified as data managers, i.e. the
persons most likely to handle and interpret
detailed data sets on behalf of the RHAs.
Further study of the actual practice of
injury prevention policy undertaken by
RHAs might help indicate how information
is actually employed, and whether the
detailed data sets asked for here will, in
fact, be used or if perhaps decision makers
are simply responding to uncertainty or
anxiety, or postponing hard choices, by
asking for more information.
In looking ahead, further research is
required to help elaborate and explain why
the data chain that is needed for decision
making purposes may be inadequate in
the current B.C. policy environment. Is
fragmentation the problem or rather the
data collection systems? Would progress
be advanced through the establishment
of a system-wide electronic health record?
Working backward from identified needs
(as gathered here) to guide systemic reform
efforts would be a more grounded approach
Chronic Diseases in Canada
than current practice, and likely would
lead to improved uptake and, ultimately,
desired actions at both the individual and
system levels.
6.
Lavis J, Robertson D, Woodside JM, et al. How
can research organizations more effectively
transfer research knowledge to decision
makers? Milbank Q. 2003;81:221-48.
Conclusion
7.
Jacobson N, Butterill D, Goering P.
Development of a framework for knowledge
translation: understanding user context.
J Health Serv Res Policy. 2003;8:94-99.
8.
Elliott H, Popay J. How are policy makers
using evidence? Models of research util­
isation and local NHS policy making.
J Epidemiol Commun Health. 2000;54:461-68.
9.
Lomas J. Using linkage and exchange to
move research into policy at a Canadian
foundation. Health Affair. 2000;19:236-40.
In this case study we consulted with
decision makers around KT approaches
related to one particular set of information:
injury data. A number of points raised
by the respondents are relevant to doing
effective KT. We presume that decision
makers would have a similar perspective
around other policy areas for which they
might be mandated to act, and thus believe
these insights are transferable beyond
injury data. Nonetheless, testing the infor­
mation found here, e.g. with other chronic
diseases, would be useful. In our view,
these findings speak to the relevance of
the policy environment and the stages in
decision makers’ policy cycle that must
be understood more fully to ensure ade­
quate uptake and utilization of research
knowledge.
References
1.
Landry R, Amara N, Pablos-Mendes A,
Shademani R, Gold I. The knowledgevalue chain: a conceptual framework for
knowledge translation in health. Bull World
Health Organ. 2006;84:597-602.
2.
Innvaer S, Vist G, Trommald M, Oxman A.
Health policy-makers’ perceptions of their
use of evidence: a systematic review.
J Health Serv Res Policy. 2002;7:239-44.
3.
Dobbins M, Cockerill R, Barnsley J, Ciliska,
D. Factors of the innovation, organization,
environment, and individual that predict the
influence five systematic reviews had on
public health decisions. Int J Technol Asses.
2001;17:467-78.
4.
Bowen S, Martens P, the Need to Know
Team. Demystifying knowledge translation:
learning from the community. J Health Serv
Res Policy. 2005;10:203-11.
5.
Lavis J, Ross S, Hurley J, et al. Examining
the role of health services research in public
policymaking. Milbank Q. 2002;80:125-54.
78
10. Roos NP, Shapiro E. From research to
policy: what have we learned? Med Care.
1999;27:JS291-JS305.
11. Parry KW. Constant comparison. The Sage
encyclopedia of social science research meth­
ods. Lewis-Beck MS, Bryman A, Liao TF,
editors. Thousand Oaks: Sage, 2004:180-81.
12. British Columbia, Ministry of Health
Planning, Office of the Provincial Health
Officer. Prevention of falls and injuries
among the elderly. Victoria (Canada): Office
of the Provincial Health Officer, 2004.
Available from: http://www.health.gov.bc.ca/
pho/pdf/falls.pdf.
13. MacNab YC, Kmetic A, Gustafson P, Shaps S.
An innovative application of Bayesian
disease mapping methods to patient safety
research: the Canadian iatrogenic injury
study. Stat Med. 2006;25:3960-980.
14. MacNab YC. Mapping disability-adjusted
life years: a Bayesian hierarchical model
framework for burden of disease and injury
assessment. Stat Med. 2007;26:4746-769.
15. Murray CJL, Lopez AD, editors. The global
burden of disease. Cambridge (MA): Harvard
University Press, 1996.
16. MacNab YC. Spline smoothing in Bayesian
disease mapping. Environmetrics. 2007;
18:727-44.
Vol 29, No 2, 2009
17. Fraser I. Organizational research with
impact: working backwards. Worldviews
Evid Based Nurs. 2004;1:S52-S59.
Appendix
Interview schedule
1
Please describe your role in relation to
injury prevention and control, and/or
planning, delivery and policy making
pertaining to injury in B.C.
2
Please briefly describe the data and
information that you currently access
and use in your above-defined role in
relation to burden-of-injury in B.C.
18. Howlett M. Policy development. In The
handbook of public administration in
Canada. Dunn C, ed. Toronto: Oxford
University Press, 2003:173-91.
19. Howlett M, Ramesh M. Studying public
policy: policy cycles and policy subsystems.
Toronto: Oxford University Press, 1995.
20. Vingilis E, Hartford K, Schrecker T, Mitchell B,
Lent B, Bishop J. Integrating knowledge
generation with knowledge diffusion and
utilization. Can J Public Health. 2003;
94:468-71.
21. Wong K, Gardner S, Bainbridge DB,
Feightner K, Offord DR, Chambers LW.
Tracking the use and impact of a community
social report: where does the information
go? Can J Public Health. 2000;91:41-45.
9
3
Are the current data and information
that you use adequate?
4
What are the strengths and limitations
of the currently available data and
information?
5
[Show participant data tables on
laptop.] Please refer to the on-screen
tables providing detailed information
on regional variations in injury-specific
mortality, morbidity and burden.
[Pause, perhaps 10-15 minutes, to
review data.] What are the most
relevant data to you in your current
role? Why is this the case?
23. Hanney S, Gonzalez-Block M, Buxton M,
Kogan M. The utilization of health research
in policy-making: concepts, examples and
methods of assessment. Health Res Policy
Syst. 2003 Jan 13;1:2.
6
Again referring to the data in the tables,
what would be the most useful means
of presenting this information to you
and your colleagues to help ensure
that the data are actually used?
24. Reimer B, Sawka E, James D. Improving
research transfer in the addictions field: a
perspective from Canada. Subst Use Misuse.
2005;40:1707-720.
7
If these data were made available
to you in the manner you have just
described, how would you see yourself
using this information?
25. Belkhodja O, Amara N, Landry R,
Ouimet M. The extent and organizational
determinants of research utilization in
Canadian health services organizations. Sci
Commun. 2007;28:377-417.
8
Would you see this information
contributing directly to priority setting
and resource allocation activity in [the
Ministry/your organization]? Please
describe how.
22. Kothari A, Birch S, Charles C. “Interaction”
and research utilisation in health policies
and programs: does it work? Health Policy.
2005;71:117-25.
Vol 29, No 2, 2009
79
[Show participant geographical analy­
sis on laptop.] Please refer to the
information previously depicted in
tabular form now presented in a spatial
map on the screen. [Pause, perhaps
10-15 minutes, to review map(s).] Is
this a useful depiction of these data, or
would you prefer the data in tabular
form? Why?
10 Noting that we are developing a
knowledge-transfer strategy to assess
how this information could be dis­
seminated to policy makers, is there
anything else that you could tell us to
help us in this process?
Thank you very much for participating in
this survey.
Chronic Diseases in Canada
Complementary therapies for cancer patients:
assessing information use and needs
MJ Verhoef, PhD (1); L Trojan, BSc (2); GD Armitage, MA (2); L Carlson, PhD (3); RJ Hilsden, PhD (1)
Abstract
Many cancer patients seek complementary therapies (CTs) for cancer management;
however, relatively little is known about patients’ CT information seeking behaviour.
Therefore, we assessed: 1) cancer patients’ use of the types and sources of CT information;
2) their information preferences; and 3) their understanding of the phrase “scientific
evidence or proof that a therapy works.” We collected data from 404 patients attending
the Tom Baker Cancer Centre (TBCC) in Calgary and 303 patients calling the Cancer
Information Service (CIS) helpline. In most cases, patients wanted information on the
safety of CTs, how CTs work and their potential side effects. Physicians and conventional
cancer centres were the most desired sources of CT information, but relatively few patients
obtained information via these sources. Although patients were aware of the meaning
of scientific evidence, they often used information based on non‑scientific evidence, such
as patient testimonials. The creation of a supportive care environment in conven­
tional cancer treatment centres, by providing CT information, may help address cancer
patients’ concerns and alleviate some of the stress that may have been caused by the
cancer diagnosis.
Keywords: CAM, cancer, alternative therapies, complementary therapies,
information use and needs, helpline
Introduction
Complementary therapies (CTs), some­
times referred to as complementary and
alternative medicine (CAM), consist of a
“group of diverse medical and health care
systems, practices and products that are not
presently considered part of conventional
medicine.”1 Studies assessing the extent of
CT use among cancer patients estimate that
7% to 91% of cancer patients report using
some form of complementary therapy.2-8
Although there has been an increase
in research to ascertain the safety and
effectiveness of CTs, mostly with respect
to symptom control,9,10 the use of CTs still
outpaces the evidence.
To date, most research has focused on
information needs regarding conventional
cancer treatments. A recent, systematic
review of cancer patients’ information
needs and sources11 identified that patients
most frequently seek treatment-related
information, such as treatment options
and side effects, and that patients tend to
consult a wide range of sources, including
health care providers, other cancer patients,
friends and family members, print material,
telephone helplines and the Internet.12-14 It
appears that cancer patients seek sensitive
information from telephone helplines
and choose Web sites for basic and less
sensitive information.15 Although many
patients access the Internet,16 dependence
on Web sites for health information can
be problematic, particularly as it relates to
CTs. Furthermore, even though the quality
of some Web sites is excellent, others lack
information regarding the safety or efficacy
of CTs,17 and may report misinformation, or
conflicting or inconsistent information.17-19
Consequently, there is the potential for
harm if such advice is followed,17,19 thus
placing users at risk.
Cancer patients seek CT information in
part because they are interested in an
alternative to conventional medicine.20 CTs
provide patients with a holistic treatment
approach and give them a sense of hope20-21
or control;22 however, they often become
frustrated with the overwhelming amount
of CT information and are unsure of what
information is credible.20 Despite the impor­
tant role that information appears to play
in cancer management,11 relatively little
is known about CT information seeking
behaviour, such as what information
patients consider credible and which
information sources they trust. Therefore,
we conducted a study to assess: 1) the
types and sources of CT information that
cancer patients use; 2) the information
they prefer; and 3) the meaning of the
phrase, “scientific evidence or proof that
a therapy works” to them. The results of
this study will be of interest to both cancer
patients who desire information on CTs
and those who treat or provide care to
patients interested in CT use.
Author References
1 Departments of Community Health Sciences and Medicine, University of Calgary, Calgary, Alberta, Canada
2 Health Systems and Workforce Research Unit, Calgary Health Region, Calgary, Alberta, Canada
3 Departments of Oncology and Psychology, University of Calgary, Calgary, Alberta, Canada
Correspondence: Dr. Marja J. Verhoef, Department of Community Health Sciences, University of Calgary, 3330 Hospital Drive NW, Calgary, AB T2N 4N1,
Tel.:403-220-7813, Fax: 403-270-7307
Chronic Diseases in Canada
80
Vol 29, No 2, 2009
Methods
Sampling
Participants were recruited in 2004 from
two settings, the TBCC in Calgary, Alberta,
Canada; and the Canadian Cancer Society’s
Canada-wide telephone helpline, the CIS.
This sampling approach23 was chosen to
focus on a range of individuals diagnosed
with cancer. It allows a comparison of CT
use and information seeking between people
in two very different contexts. The first was
a consecutive sample of new and follow-up
adult cancer patients attending the TBCC
outpatient clinic who were approached
by a research assistant in the clinic wait­
ing areas. The second was a consecutive
sample of cancer patients calling the CIS
telephone helpline who were recruited by
telephone information specialists.
Data collection
Data were collected by means of a pretested, structured questionnaire developed
specifically for this study. The questionnaire
included basic demographic information,
information about the respondent’s type
of cancer, diagnosis and treatment history,
and questions regarding their use of and
their search for information about CTs.
Most questions had a yes/no, multiple
choice or Likert scale response format.
Respondents were also encouraged to
provide additional comments about CT
use for cancer. The questionnaire was pretested in a sample of 20 cancer patients at
the TBCC to assess clarity and face-validity
of the questions, as well as the recruitment
strategy. Interested participants at the TBCC
provided consent and completed the ques­
tionnaire on-site. Completion of the survey
took approximately 10 minutes and most
were completed before attending the
scheduled clinic appointment. When
requested by the participant, assistance
was provided in reading and completing
the questionnaire. CIS helpline callers
who agreed to participate were mailed a
questionnaire with a postage-paid return
envelope. Follow-up telephone calls were
made to those who had not returned the
questionnaire within three weeks. A second
questionnaire was mailed if the telephone
follow-up was unsuccessful.
Vol 29, No 2, 2009
Data analysis
Descriptive data analysis was performed (i.e.
frequency tables, percentages and means)
and contingency analyses (i.e. chi-square
or t-tests) were used to assess associations
between socio-demographic variables, CT
use and CT information seeking charac­
teristics using SPSS data analysis software.
Respondents’ written comments and sug­
gestions were analyzed using descriptive
qualitative analysis.24
The study was approved by the Conjoint
Health Research Ethics Board at the
University of Calgary.
Results
Although some demographic and disease
characteristics of the two samples (i.e.
TBCC and CIS) were different, data on
information needs and information seeking
were fairly similar. When TBCC and CIS
data are significantly different, separate
analyses will be presented.
TBCC participants
Four hundred and eighty-four cancer
patients in the TBCC Outpatient Clinic
were invited to participate in the survey
and 404 patients (i.e. 84%) completed
it. Most (i.e. 76.6%) survey participants
were visiting the TBCC for a follow-up
appointment, while others were visiting
for a first-time appointment (i.e. 14.1%)
or for treatment (i.e. 5.8%). Reasons for
not participating included “no time,” “too
stressed,” “anxious” or “overwhelmed.”
CIS participants
There were 572 cancer patients who called
the CIS telephone helpline that were
invited to participate; of these, 394 (i.e.
68.9%) agreed to receive a questionnaire
by mail. Of the 388 patients who received
the questionnaire, 303 (i.e. 78.0%)
returned a completed questionnaire. This
included 12 (i.e. 4%) participants who
completed the survey during the telephone
follow‑up. Reasons for not participating
were very similar to those provided by
TBCC participants, except for “recently
diagnosed,” which was more common in
the CIS sample. Six questionnaires (i.e.
81
1.5%) were returned with an incorrect
mailing address. Overall, the CIS response
rate was 54%.
Qualitative analysis
Respondents who voluntarily provided
written comments and feedback included
137 from the CIS (i.e. 45.2%) and 24 (i.e.
5.9%) from the TBCC. The majority of
the comments expanded on the response
choices provided in the survey questions
and centred around three areas, i.e.
barriers to finding CT information, the
need for specific “evidence” and specific
CT information needs. Respondents’ com­
ments will be used to illustrate the results
of the quantitative analysis.
Demographic and disease characteristics
Table 1 shows demographic and disease
treatment characteristics of the TBCC and
CIS study participants. Gender, age distri­
bution, years since first diagnosis, types
of cancer (i.e. breast and genitourinary)
and cancer treatment history (i.e. surgery,
chemotherapy and hormone therapy) were
significantly different between the groups
(p < 0.05); however, in both the TBCC and
CIS samples, the three most common cancers
were breast, colorectal and genitourinary
(mostly consisting of prostate cancer). The
high proportion of females among CIS callers
has been previously reported25 and may
explain the higher percentage of respon­
dents with breast cancer in the CIS sample.
About 11% of TBCC and CIS respondents
(i.e. 11.4% and 10.9%, respectively)
reported multiple cancers. This appears
rather high; however, in this category,
respondents had also included metastases.
The data show that CIS participants were
diagnosed with cancer more recently than
TBCC participants.
CT use
In the questionnaire, CTs were described
as “herbs, mental imagery, meditation,
yoga, naturopathy, chiropractic and many
others. These therapies are different than
conventional cancer treatments such as
surgery, radiation, chemotherapy and/
or hormone therapy and are usually not
prescribed by physicians.” Table 2 reports
data on CT use before and after diagnosis
Chronic Diseases in Canada
by each group (i.e. TBCC and CIS). Whereas
CT use among CIS patients increased little
after the cancer diagnosis, it increased
substantially among TBCC patients. In
both samples, however, previous CT use
appeared to be a strong predictor of CT
use after diagnosis (i.e. p < 0.001 for both
samples). In both samples, females were
significantly more likely to use CT than
males (i.e. p = 0.006 in the TBCC sample
and p = 0.012 in the CIS sample). Age
was only related to CT use in the TBCC
group: patients aged 50 or younger were
more likely to use CTs than those over
age 50 (i.e. p = 0.008). Types of CTs were
grouped into nine categories and are listed
in Table 3. Significant differences in CT use
between the two groups were only found
for herbs and supplements and special
diets. Respondents were asked to check
all reasons for using CT treatments that
applied to them. Although in a different
order, the three most important reasons
for CT use in both groups were to improve
health, strengthen the immune system,
and enhance well-being and quality of life
(Table 4). CIS respondents mentioned “to
enhance well-being” and “to cure cancer”
significantly more than TBCC respondents.
CT information seeking
Among the proportion of the 256 (i.e.
36.8%) respondents who had sought CT
information, there was a marginal difference
between the two groups, i.e. 35.9% for
TBCC respondents and 38.0% for CIS
respondents). Table 5 shows the variation
in information seeking in the two samples
by sex, age and previous CT use. Although
the proportion seeking information differed
significantly by all three variables in the CIS
sample, it was only significantly different
for previous CT use in the TBCC sample.
Respondents who found the information
they needed differed significantly between
the two samples (p = 0.027). In the TBCC
sample, 44.6 % of respondents found most
of the information they needed compared
to 31.5% in the CIS group; 38.8% in the
TBCC group versus 55.9% found some of
the information they needed; and 16.5%
in the TBCC group compared to 12.6% in the
CIS group did not find what they wanted.
Of the information seekers who did not
find all the information they needed (n =
118), 43.2% indicated that the available
Chronic Diseases in Canada
information was too general. It was too
limited for 39.0%, confusing for 28.8%,
overwhelming for 24.4% and contradictory
for 23.7%. The only category in which the
two groups differed significantly, i.e. 33.3%
in the TBCC group and 14.5% in the CIS
sample, was the information made false
promises. Respondents checked all that
applied for this question.
needed to receive the therapy (21.3%).
Differences between TBCC and CIS
responses were insignificant. In addition,
information seekers wanted information
that is current (66.9%), explains how
the therapy works (63.5%), is easy to
understand (58.9%), is scientifically tested
(55.8%), is specific (41.5%) and provides a
range of information sources (47.1%).
In the written comments, many respondents
talked about barriers to finding information.
For example, “Cancer patients can’t give
up [looking for information]. Sometimes
doctors wait too long to do treatments;
patients need to find other options. It’s too
bad that many patients don’t know about
other options (treatments or resources)
available to them … you can’t stop.” For
those who found information, they also had
difficulty trusting the information: “When
you are diagnosed, people literally come
out of the woodwork with statements like,
‘I know a rural GP or an alternative centre
who are regularly curing cancer.’ How on
earth is a lay person supposed to be able
to determine the credibility of such claims?”
Another asked, “Is there a unified source of
credible information available that consults
scientists, oncologists, GPs and ‘alterna­
tive’ practitioners and provides intelligent
conclusions?”
In written comments, respondents
expressed interest in CT information on
a variety of other topics. One respondent
wanted information about a specific cancer
type: “There is so much cancer research, but
no one seems to know much about bladder
cancer; no support groups, etc. You’re on
your own. The Centre for Integrated Healing
[a centre in Vancouver, British Columbia
that takes a holistic, healing approach to
cancer care] is all the support we get.” Or
about preventing recurrence: “I would be
interested in knowing if there is more therapy
I should take even though I have been clean
for 5 ½ years. ... Is there a way to help
re-occurrence [sic] through complementary
therapy?” The alleviation of the side effects
of conventional treatments interested
several respondents: “If complementary
therapies help to alleviate side effects and
protect healthy cells and tissues, it would
be so beneficial to us undergoing chemo
treatment.”
As identified in the survey, conflicting
information
and
the
sometimesoverwhelming volume of information was
a challenge. “It seems to me one group says
one thing and another comes along and
says just the opposite. I think I would like
some straight answers I could really trust.”
Preferred types of CT information
Respondents were asked what three types
of CT information (i.e. out of a list of 8)
were most important to them; several
respondents identified more than three.
Among those who sought information (n =
256), 71.5% considered the most important
information to be the safety of the therapy;
67.2%, an explanation of how the therapy
works; 62.8%, the potential side effects
of the therapy; and 62.4%, proof that the
therapy improves well-being. This was
followed by cost (48.6%), information from
other patients (48.2%), proof that the ther­
apy could cure cancer (39.5%) and time
82
Used versus preferred sources
for CT information
Of those who sought CT information, there
were marked differences between the people
from whom the respondents actually
obtained information and their preferred
informants (Table 6). Although health
professionals such as physicians were often
the preferred information source, they
were seldom used, and whereas friends
or relatives were the most frequently used
sources of information, they were not the
preferred sources. This was also the case
for respondents who were asked where they
got the CT information they sought. Even
though the Internet, health newsletters and
books were used often, the respondents
would have preferred to go to conventional
cancer centres.
Vol 29, No 2, 2009
Table 1
Demographic and disease characteristics of TBCC and CIS respondents
Characteristic
TBCC
n = 404
Female (%)
Age range (mean)
CIS
n = 303
p-value
39.0
77.9
< 0.001
18 to 91 (60.0)
25 to 88 (57.3)
0.007
37.6
39.9
0.561
Education (%)
High school or less
More than high school
Years since diagnosis (range, mean)
Years (%)
62.4
60.1
0.491
< 1 to 23 (3.5)
< 1 to 29 (2.1)
< 0.001
2003 to 2004 71.8
2003 to 2004 38.6
2001 to 2002
9.8
2001 to 2002 25.7
1999 to 2000
5.8
1999 to 2000 16.7
< 1999 12.5
< 1999 18.9
17.1
42.9
< 0.001
7.9
5.9
0.309
24.5
10.6
< 0.001
Surgery
42.8
61.7
< 0.001
Radiation therapy
38.6
38.0
0.858
Chemotherapy
29.5
44.2
< 0.001
Hormone therapy
21.5
16.2
< 0.001
Types of cancer % (top three)
Breast
Colorectal
Genitourinary
Treatment history (%) (previously or currently received)
Table 2
Percentage of TBCC and CIS respondents using complementary therapies (CTs)
CT use
TBCC
n = 404
CIS
n = 303
p-value
Prior to cancer diagnosis
21.5
34.7
< 0.001
Since cancer diagnosis
30.4
36.3
0.101
Table 3
Categories of complementary therapies currently used by TBCC and CIS respondents (% of respondents)
CT categories
Examples
TBCC
n = 94
CIS
n = 103
p-value
Herbs and supplements
Aloe vera, essiac, saw palmetto, flax
44.7
28.2
0.016
Mind-body therapies
Meditation, hypnosis, support groups, relaxation,
visualization/imagery
31.9
40.8
0.129
Energy therapies
Acupuncture, homeopathy, Chinese and ayurvedic medicine,
exercise, Tai Chi, therapeutic touch, yoga
31.9
28.2
0.678
Vitamins and minerals
Amino acids, iron, vitamins A, B, C
16.0
23.3
0.184
Physical therapies
Chiropractic, massage
14.9
11.7
0.524
Special diets
Naturopathy, juicing diets, Gerson therapy
11.7
22.3
0.045
Extracts and concentrates
Hydrogen peroxide, lycopene, laetrile
7.4
8.7
0.722
Spiritual therapies
Prayer, faith healing and other spiritual rituals
2.1
5.8
0.184
Vol 29, No 2, 2009
83
Chronic Diseases in Canada
Table 4
Reasons for using complementary therapies (% of respondents)
Reasons for CT use
TBCC
n = 123
CIS
n = 110
p-value
To improve health
82.1
74.5
0.160
To enhance well-being and quality of life
65.5
82.7
0.003
To strengthen immune system
73.1
68.2
0.403
To give hope
42.0
45.5
0.539
Because CTs are less toxic/invasive than conventional therapies
40.3
38.2
0.796
To feel in control of cancer treatments
44.5
32.7
0.046
To supplement cancer treatments provided by doctor
33.3
40.9
0.232
To ease side effects of therapy
29.4
39.1
0.148
To cure cancer
23.5
43.6
0.002
To relieve symptoms
25.2
37.3
0.064
Table 5
Percentage of TBCC and CIS respondents who had sought CT information
by gender, age and previous CT use
Characteristics
Gender
Age
TBCC
n = 404
Male
32.1
25.4
Female
41.7
41.1
p-value
0.052
0.022
43.8
51.8
≤ 50 years
> 50 years
32.9
31.9
0.053
0.001
Yes
69.4
61.5
No
26.8
25.1
< 0.001
< 0.001
p-value
Previous CT use
p-value
Use of evidence
For the majority of information seekers
(i.e. 60.5%), evidence or proof that a
complementary therapy works meant that
there was scientific research to prove its
effectiveness; however, if a scientific report
stated that the CT which a respondent was
using was ineffective, 47.9% in the TBCC
sample and 33.6% in the CIS sample (p =
0.007) would continue to use it. Evidence
that a CT works could also mean that the
information came from a trusted source
(51.6%); the CT worked for others who
used it (50.4%); it was “my doctor’s
recommendation” (27.0%); and “my gut
feeling” (17.7%). The two groups differed
significantly with respect to “personal
Chronic Diseases in Canada
CIS
n = 303
experience has proven that it works” (TBCC
sample, 47% versus CIS sample, 26.4%).
Some respondents strongly endorsed
the need for scientific evidence in their
comments. As one person said, “I would like
to see independent bodies such as university
studies giving credibility to alternative treat­
ment.” Others expressed concern with the
lack of evidence-based information and
the possible consequences: “I have been
to a naturopath—not specifically for cancer
treatment—but am always sceptical, as
there is not much proven about these herbs
and maybe they could do more harm than
good.” Several respondents were interested
in patient testimonials or “softer” evidence,
for example, “I would like to discuss the
84
possibility of using complementary therapy
with someone who has had prostate cancer
and who used that therapy” and, “I hope
to read many testimonials on how such
treatment has improved the life and health
of cancer patients, and even cured them.
Surely all these people cannot be biased.”
However, others are interested in both
types of evidence and take responsibility
for their treatment decision, for example,
“Mostly, I want to hear from the scientific
community if there are any dangers in using
a particular therapy. Then I want to hear
from people who have tried it—patients
and practitioners. What is their personal
experience? Then I’d still weigh the cost to
me (i.e. time and money) and make my
own decision.”
Communication with physicians
about CT use
A large percentage of CT users (i.e. 60.7%
in the TBCC group and 67% in the CIS
group; p = 0.609) reported having told
their doctors about their CT use. Of the
combined samples, 30.3% of users did not
inform their doctors, and the remaining
6.0% indicated that they would like to tell,
but felt they could not do so. Of the CT
users who told their doctors (n = 147),
18.2% of users reported that their doctors
were very supportive and 43.2% of them
had somewhat supportive doctors. Almost
nine per cent of respondents (i.e. 8.8%)
reported having both doctors who were
Vol 29, No 2, 2009
Table 6
CT information sources identified, used and preferred by TBCC and CIS respondents (% of respondents)
TBCC
n = 144
CT information source
CIS
n = 112
Used
Preferred
Used
Preferred
Physician
13.6
76.5
13.4
75.7
Patient/survivor
22.1
50.7
40.2
58.6
Complementary practitioner
32.9
50
36.6
47.7
9.3
28.7
10.7
35.5
Nurse
10.7
23.5
15.2
20.7
Friend/relative
61.4
17.6
58.9
16.2
From whom
Pharmacist
Counsellor/psychologist
Health food store employee
3.6
6.6
9.8
18
17.9
5.9
24.1
8.1
From where
Conventional cancer centres
7.9
57.4
20.2
57.1
Internet
57.9
50.7
57.8
35.7
Health newsletters
40.7
41.9
50.5
38.4
Health organizations
12.9
33.1
28.4
38.4
Books/library
39.3
29.4
52.3
30.4
Scientific journals
14.3
31.6
18.3
24.1
Telephone cancer information services†
41.1
24.1
49.7
Magazines/newspapers
25
17.6
30.3
13.4
Television
6.4
5.9
12.8
7.1
† “Telephone cancer information services” this response option was included only on the CIS questionnaire
supportive as well as doctors who were
not, and 10.8% of users had unsupportive
doctors. The remaining 18.9% of CT users
did not know whether their doctors were
supportive.
In their comments on the questionnaires,
several respondents discussed the need
for physicians to have more knowledge
about CTs and to be willing to discuss
CT with patients early in the treatment.
As one respondent said, “It is useless to
tell your oncologist about herbal remedies.
He has no knowledge of them, nor time to
research them, so he also [sic] dismisses
them as inconsequential or harmful, with
no evidence to support his view.” Another
user stated, “Cancer patients need more
information about complementary therapy
as soon as they are diagnosed with cancer,
and my doctor would be the best person to
do this.”
Vol 29, No 2, 2009
Discussion
For this study, we recruited two consecutive
samples of cancer patients from two
different populations. The TBCC and CIS
samples differed in terms of the percentage
of females (which impacts on type of
cancer and treatment statistics), time since
diagnosis and previous CT use. With a
few exceptions, we found that the data are
remarkably similar regarding information
seeking, and used and preferred information
sources, even though we were dealing with
a very unique sample of CIS individuals who
actively seek cancer-related information
and who may be different from other
information seekers, in terms of beliefs and
attitudes.
The decision to use CTs is highly personal
and complex.22 The reasons participants
in the current study stated for using CTs
reiterate those reported in the literature.3, 26-28
In most cases, respondents looked to CTs
85
to improve their health, strengthen their
immune system or enhance their wellbeing and quality of life. Physicians, other
patients, scientific research or personal
intuition may, independently or jointly,
influence a patient’s decision to use CTs.
Information seeking may bring several
benefits, such as increased involvement
in making treatment decisions, improved
ability to cope after diagnosis and treatment,
reduced anxiety and mood disturbances,
and improved communication with family
members.11 It is therefore important that
patients are supported in their search for
information and that they have access to
accurate, comprehensive information.
Perhaps the most intriguing finding is the
difference between used and preferred
sources of CT information. In both sam­
ples, patients preferred information from
conventional health care providers (i.e.
physicians, pharmacists and nurses) and
conventional cancer centres; however,
Chronic Diseases in Canada
Table 7
Resource books and Web sites
Books
Ernst E, Pittler MH, Wider B, Boddy K. Oxford handbook of complementary medicine. Oxford: Oxford
University Press, 2008. Oncology: p. 342-8. Evidence for all treatments is rated.
Rakel D, Integrative medicine, 2nd ed. Philadelphia: Saunders, and imprint of Elsevier Inc. 2007. Section
13: Integrative oncology – an overview: p. 809-99. Evidence for all treatments is rated.
Ernst E, Pittler MH, Wider B, editors. The desktop guide to complementary and alternative medicine:
an evidence-based approach, 2nd ed. Philadelphia: Mosby, an imprint of Elsevier Limited, 2006.
Cancer: p. 80-8. Evidence for all treatments is rated.
Kligler B, Lee R. Integrative medicine: principles for practice. New York: McGraw-Hill, 2004.
Chapter 23: Integrative approach to oncology: p. 535-49.
Kohatsu W. Complementary and alternative medicine secrets: Q & As about integrating CAM therapies
into clinical practice. Philadelphia: Hanley & Belfus, Inc. 2002. Chapter 55: Approach to specific cancers:
p. 377-88.
Web sites
The University of Texas MD Anderson Cancer Center Complementary/Integrative Medicine
Education Resources: http://www.mdanderson.org/departments/cimer
Memorial Sloan Kettering Cancer Center: http://www.mskcc.org/mskcc/html/44.cfm
CAMline: http://www.camline.ca/
Center for Health and Healing, a service of Beth Israel Medical Center in New York:
http://www.healthandhealingny.org/
National Center for Complementary and Alternative Medicine (NCCAM): http://nccam.nih.gov/health/
Natural Medicines Comprehensive Database – Clinical Management Series:
http://www.naturaldatabase.com/(S(st2arzb2hbi2v355rtipno2p))/nd/ClinicalMngt.aspx?cs=&s=ND
Natural Medicines Comprehensive Database: http://www.naturaldatabase.com
Natural Standard Database: http://www.naturalstandard.com/
there was a large gap between preferred
and actual sources used. The preference
for information from their doctors has
been reported in earlier studies,11 yet
relatively few patients asked their doctors
for information. This may stem from a
reluctance to use valuable resources (e.g.
doctor’s time) when patients perceive
others need these resources more, or they
may question a physician’s willingness to
talk about CTs or his or her knowledge of
this subject. Consequently, these patients
may find it easier to obtain information
from family or friends,21 the Internet,14
health newsletters, books and the library.
Unfortunately, health information from
Web sites often contains conflicting, wrong
or incomplete information18,19 regarding
the safety or efficacy of CTs,17 and family
or friends may not have the necessary
knowledge to provide evidence-based
information.
Given study participants’ use of CTs and
their preferences for current, scientifically
Chronic Diseases in Canada
based and easily understood information
about this subject, health care providers
and conventional cancer centres have an
important role to play in disseminating
information.29 It is especially important
that health care providers bridge the gap
between preferred information sources and
those used and open up discussion about
this matter with their patients, because so
many patients do not report CT use to their
physicians.18 Compared to the literature,30-31
a fairly large number of participants in
this study reported their CT use to their
physicians; however, over a third did not.
Uncertainty regarding their physician’s
support of their CT use or, as reported in
previous research, a feeling of discomfort
in discussing CT use with their conven­
tional health care providers20 may result in
patients seeking opportunities to discuss
and gain support for CT use elsewhere.
Since patients may have already collected
CT information prior to talking with a
health care provider (e.g. obtaining basic
86
information from the Internet),14 it has
been suggested that as cancer information
seekers become more skilled at finding
information, their needs may shift from
seeking information to requiring assis­
tance with interpreting information.32,33
Although the need for health care
providers to be cognizant of available CT
information sources has been previously
identified,29,30,32,34 health care providers
are also challenged to find valid and com­
prehensive CT information that they can
discuss and share with their patients.
Since changing or conflicting information
regarding CTs can be confusing for both
health care providers and patients seeking
information,35 current, comprehensive infor­
mation sources would be helpful. CIS
telephone helplines may be able to assist
health care providers in developing the
skills to communicate about CTs, as well
as provide them with resources they can
use to assist patients in decision making.
Information that would be available to both
patients and health care providers could
include how to choose a complementary
practitioner or natural health products. A
guide such as “Complementary Healthcare:
A Guide for Patients” may be a helpful
start.36 It includes helpful information on
where to find CT practitioners and what
questions to ask of them. High-quality,
evidence-based CT information is increas­
ingly available; Table 7 lists some available
resources. Furthermore, the development of
the relatively new discipline of integrative
oncology reflects a shift in focus from
biomedical cancer treatment to the more
comprehensive concept of cancer care.
Integrative oncology has been defined
as “the ability to integrate the best of
complementary and mainstream care using
a multidisciplinary approach, combining
the best of mainstream cancer care and
rational, data-based, adjunctive CTs.”37 This
development will most likely lead to further
development of and guidelines regarding CTs.
The results also showed that patients rely
on or trust information sources of nonscientific, research-based evidence. Although
the importance of scientific evidence is
without question, the literature increasingly
points out that non-scientific evidence
factors are important to consider as well.
Sackett et al. have defined evidence-based
Vol 29, No 2, 2009
medicine as “the integration of the best
research evidence with clinical expertise and
patient values.”38 This definition highlights
that clinical expertise, partially based on
empirical observation, may provide impor­
tant information above and beyond what
can be learned from clinical trials. In
addition, it highlights that the patient has
important knowledge which is unavailable
to the health care provider. Whereas clinical
expertise and patient values are limited by
their subjective nature, scientific evidence
is limited in its bias towards “objectivity,”
attempt to control and discounting of
important subjective factors. Scientific evi­
dence, clinical expertise and patient values
combined will greatly contribute to optimal
evidence-based patient care.
This study raises important issues regard­
ing patients’ need for and use of CT infor­
mation, despite being limited due to its
cross-sectional nature, the general nature
of the questions in the questionnaire and
two very specific samples that do not allow
generalizations to the larger population of
CT information seekers. Such information
may be helpful for patients, in further
clarifying their questions, and for health
care providers, in understanding patients’
inquiries and learning how to address them.
Acknowledgements
Support for this project was provided
by the Alberta Heritage Foundation for
Medical Research. We want to thank the
Outpatient Clinic of the Tom Baker Cancer
Centre and the Cancer Information Service
of the Canadian Cancer Society for the
cooperation of its staff and its support of
this project.
References
1.
NCCAM:
National
Center
for
Complementary and Alternative Medicine.
Expanding Horizons of Health Care. What
is CAM? [Internet]. National Institutes of
Health, Bethesda (MD) [cited 2008, Nov
25]. URL: http://nccam.nih.gov/health/
whatiscam/.
2.
Ashikaga T, Bosompra K, O’Brien P,
Nelson L. Use of complementary and alter­
native medicine by breast cancer patients:
prevalence, patterns and communication
with physicians. Support Care Cancer.
2002;10(7):542-48.
3.
Conclusion
Cancer patients would prefer to receive
CT information from conventional health
professionals and cancer centres. They also
want this information to be cancer-specific
and comprehensive. Cancer information
services can play a valuable role in the
provision of CT information to both patients
and conventional health care professionals.
Patients seek information from a number
of sources and evaluate the trustworthiness
and validity of that information in different,
sometimes conflicting, ways. Therefore, it
is important to acknowledge that patients
may use information based on scientific
research as well as “softer” evidence, such
as patient testimonials. The creation of a
supportive care environment through CT
information provision may help address
some of the concerns of cancer patients
and alleviate some of the stress that may
have been caused by the cancer diagnosis.
Vol 29, No 2, 2009
4.
5.
6.
7.
Eng J, Ramsum D, Verhoef M, Guns E,
Davison J, Gallagher R. A population-based
survey of complementary and alternative
medicine use in men recently diagnosed
with prostate cancer. Integr Cancer Ther.
2003;2(3):212-16.
Ernst E, Cassileth BR. The prevalence of
complementary/alternative medicine in
cancer: a systematic review. Cancer. 1998
Aug 15;83(4):777-82.
McKay DJ, Bentley JR, Grimshaw RN.
Complementary and alternative medicine in
gynaecologic oncology. J Obstet Gynaecol
Can. 2005;27(6):562-68.
8.
Yates JS, Mustian KM, Morrow GR, et al.
Prevalence of complementary and alter­
native medicine use in cancer patients
during treatment. Support Care Cancer.
2005;13(10):806-11.
9.
Mitchell SA, Beck SL, Hood LE, Moore K,
Tanner ER. Putting evidence into prac­
tice: evidence-based interventions for
fatigue during and following cancer and its
treatment. Clin J Oncol Nurs. 2007 Feb;
11(1):99-113.
10. Tipton JM, McDaniel RW, Barbour L, et al.
Putting evidence into practice: evidencebased interventions to prevent, manage,
and treat chemotherapy-induced nausea
and vomiting. Clin J Oncol Nurs. 2007
Feb;11(1):69-78.
11. Rutten LJ, Arora NK, Bakos AD, Aziz N,
Rowland J. Information needs and sources
of information among cancer patients: a
systematic review of research (1980-2003).
Patient Educ Couns. 2005 Jun;57(3):250-61.
12. Basch EM, Thaler HT, Shi W, Yakren S,
Schrag D. Use of information resources by
patients with cancer and their companions.
Cancer. 2004 Jun1;100(11):2476-83.
13. Carlsson M. Cancer patients seeking infor­
mation from sources outside the health care
system. Support Care Cancer. 2000 Nov;
8(6):453-57.
14. Hardyman R, Hardy P, Brodie J, Stephens R.
It’s good to talk: comparison of a telephone
helpline and website for cancer information.
Patient Educ Couns. 2005 Jun;57(3):315-20.
15. Bott J. An analysis of paper-based
sources of information on complementary
therapies. Complement Ther Clin Pract.
2007 Feb;13(1):53-62.
Richardson MA, Sanders T, Palmer JL,
Greisinger A, Singletary SE. Complementary/
alternative medicine use in a comprehen­
sive cancer center and the implications for
oncology. J Clin Oncol. 2000;18:2505-14.
16. Chen X, Siu L. Impact of the media and
the Internet on oncology: Survey of cancer
patients and oncologists in Canada. J Clin
Oncol. 2001 Dec 1;19(23):4291-97.
Tough SC, Johnston DW, Verhoef MJ,
Arthur K, Bryant H. Complementary and
alternative medicine use among colorectal
cancer patients in Alberta, Canada. Altern
Ther Health Med. 2002;8(2):54-60,62.
17. Walji M, Sagaram S, Sagaram D, MericBernstam F, Johnson C, Mirza NQ,
Bernstam EV. Efficacy of quality criteria to
identify potentially harmful information: a
cross-sectional survey of complementary
87
Chronic Diseases in Canada
and alternative medicine web sites. J Med
Internet Res. 2004 Jun 29;6(2):e21.
18. Molassiotis A, Xu M.
issues of web-based
herbal medicines in
cancer. Complement
Dec;12(4):217-27.
Quality and safety
information about
the treatment of
Ther Med. 2004
28. Correa-Velez I, Clavarino A, Eastwood H.
Surviving, relieving, repairing, and boosting
up: reasons for using complementary/
alternative medicine among patients with
advanced cancer: a thematic analysis.
J Palliat Med. 2005 Oct;8(5):953-61.
19. Schmidt K, Ernst E. Assessing websites on
complementary and alternative medicine
for cancer. Ann Oncol 2004;15(5):733-42.
29. Mackenzie G, Parkinson M, Lakhani A,
Pannekoek H. Issues that influence patient/
physician discussion of complementary
therapies. Patient Educ Couns. 1999 Oct;
38(2):155-59.
20. Washburn A, Thomsen C. Cancer patients
and CAM: exploring information needs.
Abstract. Focus Altern Complement Ther.
2003;8(4):552.
30. Diaz JA, Griffith RA, Ng JJ, Reinert SE,
Friedmann PD, Moulton AW. Patients’ use
of the Internet for medical information.
J Gen Intern Med. 2002 Mar;17(3):180-85.
21. Leydon GM, Boulton M, Moynihan C, et
al. Cancer patients’ information needs
and information seeking behaviour: in
depth interview study. BMJ. 2000 Apr 1;
320(7239):909­-13.
31. Cohen L, Cohen MH, Kirkwood C, Russell
NC. Discussing complementary therapies
in an oncology setting. J Soc Integr Oncol.
2007 Winter;5(1):18-24.
22. Balneaves LG, Bottorff JL, Truant T,
Davison BJ. Information needs of breast
and prostate cancer patients using comple­
mentary therapies: a pilot study. Abstract.
Can Oncol Nurs J. 2005;15(1):42.
23. Maxwell JA. Qualitative research design: an
interactive approach. Thousand Oaks (CA):
Sage Publications; 1996:96-98.
24. Morse JM, Richards L. Readme first for
a user’s guide to qualitative methods.
Thousand Oaks (CA): Sage Publications;
2002:149-150.
25. Eng JL, Monkman DA, Verhoef MJ,
Ramsum DL, Bradbury J. Canadian Cancer
Society Information Services: lessons
learned about complementary medicine
information needs. Chronic Dis Can. 2001;
22(3-4):102-7.
26. Molassiotis A, Fernadez-Ortega P, Pud D, et
al. Use of complementary and alternative
medicine in cancer patients: a European
survey. Ann Oncol. 2005 Apr;16(4):655-63.
27. Vapiwala N, Mick R, Hampshire MK,
Metz JM, DeNittis AS. Patient initiation of
complementary and alternative medical
therapies (CAM) following cancer diagnosis.
Cancer J. 2006 Nov–Dec;12:467-74.
Chronic Diseases in Canada
37. Deng GE, Cassileth BR, Cohen L, et al.
Integrative oncology practice guidelines.
J Soc Integr Oncol. 2007 Spring; 5(2):65-84.
38. Sackett DL, Straus SE, Richardson WS, et al.
Evidence-based medicine: how to practice
and teach EBM. 2nd ed. London: Churchill
Livingstone; 2000:1.
32. Walji M, Sagaram S, Meric-Bernstam F,
Johnson CW, Bernstam EV. Searching for
cancer-related information online: unin­
tended retrieval of complementary and
alternative medicine information. Int J Med
Inform. 2005 Aug;74(7-8):685-93.
33. Bright MA, Fleisher L, Thomsen C, Morra
ME, Marcus A, Gehring W. Exploring
e-health usage and interest among cancer
information service users: the need for
personalized interactions and multiple
channels remains. J Health Commun.
10 Suppl. 2005;1:35-52.
34. Lee CO. Communicating facts and know­
ledge in cancer complementary and alter­
native medicine. Semin Oncol Nurs. 2005
Aug;21(3):201-14.
35. Ernst E, Pittler MH, Wider B, Boddy K.
Complementary/alternative medicine for
supportive cancer care: development of the
evidence-base. Support Care Cancer. 2007
May;15(5):565-68.
36. Pinder M, Pedro L, Theodorou G, Treacy
K. Complementary healthcare: a guide for
patients. [Internet]. London: The Prince’s
Foundation for Integrated Health; 2007
[cited 2008, Nov 25]. URL: http://www.fih.
org.uk/document.rm?id=19
88
Vol 29, No 2, 2009
The national lung health framework:
an opportunity for gender analysis
N Hemsing, MA (1); L Greaves, PhD (2)
Abstract
Smoking related respiratory diseases in Canada represent a huge social and economic
burden for both women and men. This article addresses the potential impact of the
National Lung Health Framework for reducing disparities between women and men in
respiratory health and between sub-populations of women and men. A preliminary analysis
of the existing framework documents indicates that sex and gender factors, differences
and influences have not yet been clearly or sufficiently identified. Yet, there are sex and
gender issues related to tobacco prevention and cessation, lung health and lung disease.
In particular, we consider the specific respiratory health needs and experiences of women
to demonstrate the need for sex and gender-based analysis within the framework. For
example, while there is inconsistent evidence regarding quit rates, women and men have
different cessation patterns and reasons for smoking. Although creating a Canada-specific
approach to lung health is an important initiative, the sex and gender issues associated
with respiratory disease and health need to be explicitly addressed in the planning and
development stages of the framework in order to have a beneficial and lasting impact on
both women and men.
Key words: smoking, respiratory diseases, National Lung Health Framework, NLHF,
chronic obstructive pulmonary disease, COPD, sex, gender, women
Introduction
Respiratory diseases in Canada represent
a huge social and economic burden. Lung
cancer, chronic obstructive pulmonary dis­
ease (COPD) and pneumonia, the three lead­
ing respiratory causes of death in Canada,
were responsible for 15.6% of deaths in men
and 13.5% of deaths in women in 2004.1
The current planning and development
of a National Lung Health Framework
(NLHF) is an integral step to improving
the respiratory health of Canadians. This
framework emerged in March 2006 from
a workshop entitled “Breathing Matters,”
which united stakeholders in the mandate
to develop a national action plan to
improve respiratory health in Canada.
Under the auspices of the Canadian Lung
Association, the process is being led by
an interim steering committee that has
coordinated subsequent workshops to
guide the framework-development process.2
Released in August 2008, the NLHF
document will be used to form an action
plan and guide decision makers and
stakeholders in strategic planning.3
The creation of a comprehensive framework
has the potential to improve the respiratory
health of Canadian women and men from
prevention to diagnosis, management and
treatment. The framework also seeks to
address some of the health challenges facing
diverse sub-populations of Canadians. For
example, the documents produced during
the framework development process spec­
ify the need to “[address] the needs of
vulnerable populations,” and consistently
identify First Nations persons, youth and
immigrants as important sub-populations
for respiratory health initiatives.4 The
four key strategies in the framework docu­
ment,i which deal with everything from
health promotion and disease detection to
policy and research, indicate that actions
must aim to not only improve overall health,
but also the disparities between Aboriginal
and non-Aboriginal populations.3 The
steering committee has identified many
research- and practice-based issues, such as
the importance of examining relationships
between respiratory health, vulnerable
populations and environmental factors,
and the need to improve provider-patient
support and increase the use of spirometry
as a diagnostic tool.3,4
Developing a national framework is a
significant challenge, given the wide range
of acute and chronic respiratory conditions
affecting Canadians. These include dis­
eases as varied as asthma, tuberculosis,
sleep apnea, pneumonia, influenza, COPD
and lung cancer, each of which have
i (1) health promotion, awareness and disease prevention; 2) disease detection and management; 3) policy, partnerships and community/
systems support; 4) research, surveillance and knowledge translation
Author References
1 British Columbia Centre of Excellence for Women’s Health
2 British Columbia Centre of Excellence for Women’s Health
Correspondence: Natalie Hemsing, E311–4500 Oak Street, P.O. Box 48, Vancouver, BC V6H 3N1, Tel.: 604-875-2424 ext. 3793, Fax: 604-875-3716, Email: [email protected]
Chronic Diseases in Canada
89
Vol 29, No 2, 2009
unique causal, diagnostic, management
and treatment issues. Furthermore, these
four issues vary between women and men,
and among sub-populations of women and
men. Hence, the sex and gender issues
associated with respiratory disease and
health need to be explicitly addressed in
the framework in order to have a beneficial
and equitable impact for both women and
men. However, a preliminary analysis of the
existing framework documents2-5 indicates
that sex and gender factors, and diversity
related differences and influences have not
been sufficiently identified, and that sex
and gender analysis is not identified as a
key analytical tool for strategic planning.
In response to this lack, this article exa­
mines some of the respiratory health needs
of women to highlight how these omissions
within the current framework may fail to
capture sex- and gender-based differences
between women and men.
Why integrate a sex
and gender lens?
If significant improvements in lung health
are to be made, sex and gender analysis
must be an integral part of planning and
program initiatives. Sex- and gender-based
analysis (SGBA) is a tool that promotes
consideration of a range of issues related
to both the research process and the appli­
cation of knowledge in program or policy
development activities such as the NLHF.
An SGBA is recommended by Health
Canada’s Women’s Health Strategy,6 and
is also integrated into the work of the
World Health Organization (WHO).7 Using
such an approach helps to improve our
understanding of how the influences of
sex (i.e. biological) and gender (i.e. social
and cultural aspects) determine health and
disease. The effectiveness of how we
design and implement sex- and gendersensitive policies and programs is partially
determined by such analyses.8 Utilizing
SGBA would allow the national strategy to
address the unequal distribution of disease
among women and men and among subpopulations of women and men, including
Aboriginal groups and those with low
incomes.
Vol 29, No 2, 2009
The reviewed documents underpinning
the NLHF2,4,5 fail to clearly or consistently
articulate a sex- or gender-based approach;
nor do they indicate whether sex or
gender has informed the development or
implementation processes. Occasionally, the
differences between women’s and men’s
respiratory health needs and issues are
discussed. Importantly, the increasing
smoking rates in women are identified as a
timely issue,3,5 as is the growing prevalence
among women of COPD and lung cancer,
in part due to the relative lag in women’s
smoking compared to men.3 In addition,
pregnant women are cited as an important
population when creating cessation pro­
grams.5 Yet the mention of women in the
NLHF documents is additive and sporadic
in comparison to other populations such
as youth, First Nations people and immi­
grants,3,4 and the need to examine the
specific health needs of women and men
and sub-populations of women and men is
not consistently identified. Overlooked in
the NHLF is the fact that all populations are
gendered, and their health concerns need
to be addressed accordingly.
Sex and gender influences
on lung health
Numerous sex- and gender-based influences
and factors must be considered to develop
a framework that addresses respiratory
health needs. To illustrate this point, we
consider some of these needs, focussing on
the context of women’s respiratory disease
prevention, diagnosis and treatment issues.
For example, many respiratory diseases
affect women and men disproportionately.
Women have higher rates of asthma, COPD
is increasingly becoming a woman’s dis­
ease, and mortality rates for lung cancer
have been increasing among women in
Canada since 1987, yet decreasing among
men.1 There are also more lung cancers
among women who have never smoked,
compared to men who have never
smoked.9,10 Furthermore, certain subpopulations of girls and women, including
low socio-economic groups and non-white
minorities, have disproportionately higher
rates of respiratory disease. Non-white and
low-income women tend to have less access
to health care resources and suffer more
often from disease and disabilities.11,12
90
Tobacco use is a key factor in the devel­
opment of respiratory disease. Gendered
patterns of smoking and exposure to smoke,
and biological, hormonal and genetic
factors overlap and influence women’s
susceptibility to respiratory diseases.13
Estrogen may influence the metabolism
of cigarette smoke, resulting in increased
damaging effects.10,12,14-16 Evidence shows
that women who smoke less than men
show similar levels of impaired lung
function, and smoking decreases women’s
lung function more than men’s.12,17,18
Meanwhile, tobacco marketing has been
gendered, effectively and increasingly
targeting women, portraying smoking as
glamorous and as a method to stay thin.19,20
The industry has also developed genderspecific tobacco products. Many women
have been marketed “light” cigarettes with
higher yields of N-nitrosamines, which may
be partially responsible for the increased
lung cancer rates in women.12 Second hand
smoke also impacts women differently than
men, given the lag in overall smoking trends
between men and women is resulting in
more non-smoking women living with men
who smoke.19,21
There are also sex and gender issues
associated with diagnosis. Women and
men report different symptoms and women
develop COPD at a younger age.12 Women
are also less likely to report sputum
production than men, due to gendered
norms and ideals.22 In addition, women are
often under-diagnosed or misdiagnosed for
certain diseases, due to these differences
in presentation as well as gender bias in
the health system. For example, women
with COPD are more often diagnosed with
asthma than men.13,23 Moreover, even when
women and men present the same symp­
toms, providers may not interpret symptoms
in the same way.24
In general, compared to men, women with
respiratory diseases tend to report more
hospitalizations, more limitations in activity
and higher rates of anxiety and depression
associated with respiratory diseases.1,23,25-29
Changes in physical appearance associated
with COPD and lung cancer may be
especially troubling for women, who
are encouraged to meet gendered social
standards of beauty.28 Finally, pulmonary
Chronic Diseases in Canada
rehabilitation has been shown to be more
effective for men over time.22 All of these
factors shape women’s and men’s lung
health from prevention to treatment and
need to be addressed and included within
the planning stages to produce a framework
that will significantly improve women’s
and men’s respiratory health.
Conclusion
The NLHF can significantly improve the
dissemination and uptake of knowledge
related to respiratory health by drawing
links between sex- and gender-specific
research in both tobacco use and exposure,
as well as respiratory health and disease.
There is a clear need for the integration of
spheres of knowledge on tobacco, gender,
and respiratory disease.12 The highest rates
of mortality are associated with diseases
that are primarily associated with smoking
or smoke exposure, such as COPD and
lung cancer. There is inconsistent evidence
regarding gender differences in cessation.ii
Regardless, women and men do smoke
for different reasons and have different
cessation patterns.20 One potential strategy
for addressing some of these issues is the
formation of a NLHF working group on sex,
gender and diversity issues, which could
include researchers and decision makers
in lung health, women’s health, men’s
health and tobacco use and prevention.
Connections must be strengthened
between research, programs and policy so
that emerging sex- and gender-specific
findings are effectively translated to
health care settings and decision makers.40
By addressing these and other sex- and
gender-related factors, the NLHF could
lead the way in effectively responding to
all “vulnerable” groups.
for respiratory disease, both compared to
men and among sub-populations of women.
We have discussed some of the unique
respiratory health concerns of women
in particular to make a case for a more
detailed, consistent and mainstreamed
need for a sex and gender lens to guide
the NLHF and action plan. The full and
comprehensive implementation of a sex
and gender analysis would also necessitate
an exploration of the unique respiratory
health needs of men, and contribute to
a systematic assessment of gendered
responses aimed at men. The NLHF can
seek to improve the respiratory health of all
groups, through the creation of initiatives
to support and encourage further sex and
gender-based research and interventions.
Although very little research has examined
specific respiratory health issues for subpopulations, such as people on low incomes
or of particular ethnic groups, the NLHF
can address and respond to these issues
by utilizing an SGBA to encourage thought
on women in Canada and steer Canada
toward future sex-, gender- and diversitybased research, programming, policy and
analysis.
1.
Public Health Agency of Canada. Life and
breath: respiratory disease in Canada. Ottawa:
Public Health Agency of Canada; 2007.
2.
Canadian Lung Association. Leading.
Acting. Together. Building Canada’s first
national lung health action plan: report
from the all-stakeholder workshop on lung
health. Paper presented at: All-stakeholder
workshop on lung health. 2007 April 26-27;
Ottawa, Canada.
3.
Canadian Lung Association. National lung
health framework. Ottawa: Canadian Lung
Association; 2008.
4.
Canadian Lung Association. Respiratory
health in Canada asset map: preliminary
report. Ottawa: Canadian Lung Association;
2007.
5.
de Lusignan S, Chan T, Cohen A, Thana L,
Hague N, van Vlymen J. Health education
and prevention for people with severe
mental illness: a cross-sectional study of
general practice computer records. Primary
care mental health. 2005;3(3):221-33.
6.
Health Canada. Women’s health strategy.
Ottawa: Health Canada; 1999.
7.
World Health Organization. The WHO
gender mainstreaming strategy. Geneva:
WHO. Retrieved August 18, 2008
from:
http://www.who.int/gender/
mainstreaming/strategy/en/index.html.
8.
Health Canada. Exploring concepts of gen­
der and health. Ottawa: Health Canada;
2003.
9.
Patel JD, Bach PB, Kris MG. Lung cancer
in US women: a contemporary epidemic.
JAMA. 2004;291:1763-768.
Acknowledgements
This commentary is based on an evidence
review on women’s respiratory health,
prepared for and funded by the Provincial
Health Services Authority (PHSA)
Population and Public Health Program. The
activities of the British Columbia Centre of
Excellence for Women’s Health are made
possible through a financial contribution
from Health Canada. However, the views
expressed are those of the authors and
not necessarily those of PHSA or Health
Canada.
The identification of sex- and genderbased analysis as a key analytic tool would
encourage researchers, decision makers
and other stakeholders to account for these
differences. In sum, there are different
reasons why women smoke, as well as
varying experiences of care and treatment
ii
References
10. Tang DL, Rundle A, Warburton D, et al.
Associations between both genetic and
environmental biomarkers and lung cancer:
Although evidence from CTUMS (2006) shows that in Canada, women (60.1%) have slightly higher quit rates than men (57.9%), studies examining gender
differences in quit rates often reveal that women have lower quit rates, more difficulty quitting smoking and greater relapse rates than men.30-36 Yet, other
evidence suggests that gender differences are minimal or non-existent,37, 38 or that it is, in fact, men who have lower quit rates and greater difficulty
quitting smoking.39
Chronic Diseases in Canada
91
Vol 29, No 2, 2009
evidence of a greater risk of lung cancer in
women smokers. Carcinogenesis. 1998;19:
1949-953.
20. Sarna L, Aguinaga Bialous S. Why tobacco
is a women’s health issue. Nurs Clin North
Am. 2004;39(1):165-80.
11. Steele CB, Richmond-Reese V, Lomax S.
Racial and ethnic disparities in HIV/AIDS,
sexually transmitted diseases, and tuber­
culosis among women. J Womens Health.
2006;15(2):116-22.
21. Siegfried JM. Women and lung cancer:
does oestrogen play a role? Lancet Oncol.
2001;2(8):506-513.
12. Greaves LJ, Richardson LA. Tobacco use,
women, gender, and chronic obstructive
pulmonary disease: are the connections
being adequately made? Proc Am Thorac
Soc. 2007; 4:675-79.
13. Caracta C. Gender differences in pul­
monary disease. Mt Sinai J Med. 2003;
70(4):215-24.
14. Zang EA, Wynder EL. Differences in lung
cancer risk between men and women:
examination of the evidence. J Natl Cancer
Inst. 1996;88(3/4):183-92.
15. Ben-Zaken Cohen S, Pare PD, Man SFP,
Sin DD. The growing burden of chronic
obstructive pulmonary disease and lung
cancer in women: examining sex differences
in cigarette smoke metabolism. Am J Respir
Crit Care Med. 2007;176:113-20.
16. Sin DD, Cohen SB–Z, Day A, Coxson H,
Pare PD. Understanding the biological
differences in susceptibility to chronic
obstructive pulmonary disease between
men and women. Proc Am Thorac Soc.
2007;4:671-74.
17. Prescott E, Bjerg AM, Andersen PK,
Lange P, Vestbo J. Gender difference in
smoking effects on lung function and risk
of hospitalization for COPD: results from
a Danish longitudinal population study.
Eur Respir J. 1997;10:822-27.
18. Canadian Lung Association. Women and
COPD: a national report. Ottawa: Canadian
Lung Association; 2006.
19. Ernster V, Kaufman N, Nichter M, Samet J,
Yoon SY. Women and tobacco: moving from
policy to action. Bull World Health Organ.
2000;78(7):891-901.
Vol 29, No 2, 2009
22. Camp PG, Goring SM. Gender and the
diagnosis, management, and surveillance
of chronic obstructive pulmonary disease.
Proc Am Thorac Soc. 2007;4:686-91.
23. French CT, Fletcher KE, Irwin RS. Gender
differences in health-related quality of life
in patients complaining of chronic cough.
Chest. 2004;125:482-88.
24. Chapman KR, Tashkin DP, Pye DJ. Gender
bias in the diagnosis of COPD. Chest. 2001;
119:1691-695.
25. Sinclair A, Tolsma D. Gender differences
in asthma experience and disease care in
a managed care organization. J Asthma.
2006;40(8):945-53.
26. Belloch A, Perpiñá M, Martínez-Moragón
E, de Diego A, Martínez-Francés M. Gender
differences in health-related quality of life
among patients with asthma. J Asthma.
2003;40(8):945-53.
27. Woods S, Sorscher J, King J, Hasselfeld K.
Young adults admitted for asthma: does
gender influence outcomes? J Womens
Health. 2003;12(5):481-85.
28. Johnson JL, Campbell AC, Bowers M, Nichol
A-M. Understanding the social consequences
of chronic obstructive pulmonary disease:
the effects of stigma and gender. Proc Am
Thorac Soc. 2007;4:680-82.
29. Sarna L, Brown JK, Cooley ME, et al.
Quality of life and meaning of illness of
women with lung cancer. Oncol Nurs
Forum. 2005;32(1):E9-19.
30. Scharf DS. Are there gender differences
in smoking cessation, with and without
bupropion? Pooled- and meta-analyses of
clinical trials of Bupropion SR. Addiction.
2004; 99:1462-469.
92
31. Bohadana A, Nilsson F, Rasmussen T,
Martinet Y. Gender differences in quit rates
following smoking cessation with combi­
nation nicotine therapy: influence of base­
line smoking behavior. Nicotine Tob Res.
2003;5(1):111-16.
32. Bjornson W, Rand C, Connett JE, et al.
Gender differences in smoking cessation
after 3 years in the lung health study. Am
J Public Health. 1995; 85:223-30.
33. Westmaas JL, Langsam K. Unaided smoking
cessation and predictors of failure to quit
in a community sample: effects of gender.
Addict Behav. 2005;30(7):1405-424.
34. Ward KD, Klesges RC, Zbikowski SM,
Bliss RE, Garvey AJ. Gender differences
in the outcome of an unaided smoking
cessation attempt. Addict Behav. 1997;
22(4):521-33.
35. Wetter D, Kenford S, Smith S, Fiore M,
Jorenby D, Baker T. Gender differences in
smoking cessation. J Consult Clin Psychol.
1999;67(4):555-62.
36. Osler M, Prescott E, Godtfredsen N, Hein H,
Schnohr P. Gender and determinants of
smoking cessation: a longitudinal study.
Prev Med. 2000;29(1):57-62.
37. Killen J, Fortmann S, Varady A, Kraemer H.
Do men outperform women in smoking
cessation trials? Maybe, but not by much.
Exp Clin Psychopharmacol. 2002;10(3):
295-301.
38. Etter J-F, Prokhorov A, Perneger TV.
Gender differences in the psychological
determinants of cigarette smoking.
Addiction. 2002;97(6):733-43.
39. Walsh PM, Carrillo P, Flores G, Masuet C,
Morchon S, Ramon JM. Effects of partner
smoking status and gender on long term
abstinence rates of patients receiving
smoking cessation treatment. Addict Behav.
2007;32(1):128-36.
40. Pederson AP, Hoyak KAK, Mills S, Camp PG.
Reflecting the changing face of chronic
obstructive pulmonary disease: sex and
gender in public education materials
on COPD. Proc Am Thorac Soc. 2007;
4:683-85.
Chronic Diseases in Canada
Announcements
Call for papers
Conferences
In honour of the Vancouver Olympics and
Paralympics, Chronic Diseases in Canada
wishes to devote its winter 2010 issue (to
be published in February 2010) to articles
on the relationships between health habits,
especially physical activity, and chronic
disease prevention. Ideally, these articles
will report the results of empirical research
using Canadian databases, or application
of other results to the Canadian population.
The deadline for submission of such articles
will be June 5, 2009. The top 3 papers –
gold, silver and bronze! – will be published.
2009 CDC Diabetes Conference
April 21-24, 2009
Long Beach, California
http://www.cdc.gov/diabetes/
conferences/index.htm
Chronic Diseases in Canada
7th International Conference on Diet and
Activity Methods
June 5-7, 2009
Washington, D.C.
http://icdam.org/
International Scientific Conference on
Nutraceuticals and Functional Foods
June 9-11, 2009
Zilina, Slovakia
http://www.foodandfunction.com/
Third International Chronic Disease
Conference
November 23-26, 2009
Calgary, Alberta
http://www.cdmcalgary.ca/index.
php?lang=english
Canadian Public Health Association
Conference
June 7-10, 2009
Winnipeg, Manitoba
http://www.cpha.ca/en/conferences/
conf2009.aspx
93
Vol 29, No 2, 2009
CDIC: Information for Authors
Chronic Diseases in Canada (CDIC) is a quarterly
Book/Software Review: Usually solicited by the
References: In AVancouver [email protected] (consult a
scientific journal focussing on the prevention and
editors (500B1,300 words), but requests to review
recent CDIC issue for examples); num­bered in
control of non‑ communicable diseases and injuries
are welcomed. Abstract not required.
superscript in the order cited in text, tables and
content of articles may include research from such
Submitting Manuscripts
[email protected] if more); without any automatic reference
fields as epidemiology, public/community health,
Submit manuscripts to the Editor‑in‑Chief, Chronic
numbering feature used in word processing; any
biostatistics, the behavioural sciences, and health
Diseases in Canada, Public Health Agency of Canada
unpublished
services or economics. CDIC endeav­ours to foster
130 Colonnade Road, Address Locator: 6501G, Ottawa,
communications used (discouraged) to be cited in
communication on chronic diseases and injuries
Ontario K1A 0K9, e‑mail: cdic‑[email protected]‑aspc.gc.ca.
the text in parentheses (authors responsible for
in Canada. Its feature articles are peer reviewed. The
figures; listing up to six authors (first three and Aet
among public health practitioners, epidemiolo­gists
obtaining
authors
are
on illustrations not applicable) to the “Uniform
Requirements for Manuscripts Submitted to
Tables and Figures: Send vector graphics only.
concise­ness and technical accuracy. Although CDIC
Biomedical
the
Each on a separate page and in electronic file(s)
is a publication of the Public Health Agency of
International Committee of Medical Journal Editors,
separate from the text (not imported into the text
Canada, contributions are welcomed from both
authors should refer to this document for complete
body); as self‑ explanatory and succinct as possible;
the public and pri­vate sectors. Authors retain
details before submitting a manuscript to CDIC (see
not too numerous; numbered in the order that they
responsibility for the contents of their papers, and
<www.icmje.org>.
are mentioned in the text; explanatory material for
Agency of Canada.
Article Types
approved
(section
per­mis­sion);
scientific quality, public health relevance, clarity,
as
general
written
educators. Submissions are selected based on
Journals”
in
personal
Since
CDIC editorial committee nor of the Public Health
adheres
or
and researchers, health policy plan­ners and health
opinions expressed are not necessarily those of the
CDIC
observations/data
by
responsible for verifying accuracy of references.
tables
Checklist for Submitting
Manuscripts
in
footnotes,
identified
by
lower‑case
superscript letters in alpha­betical order; figures
limited to graphs or flow charts/templates (no
Cover letter: Signed by all authors, stating that all
photographs), with software used specified and
have seen and approved the final manuscript and
titles/footnotes on a separate page.
have met the authorship including a full statement
regarding any prior or duplicate publication or
Number of copies: If submitting by mail, one
submission for publication.
complete copy, including tables and figures; one
figures, references) in the form of original research,
First title page: Concise title; full names of all
copy of the manuscript on diskette. If submitting by
surveillance reports, meta‑analyses or methodological
authors and institutional affiliations; name, postal
e‑mail to cdic‑[email protected]‑aspc.gc.ca, please fax or
papers.
and e‑mail addresses, tele­phone and fax numbers
mail the covering letter to the address on the inside
for corresponding author; separate word counts for
front cover.
Peer‑reviewed Feature Article: Maximum 4,000
words for main text body (excluding abstract, tables,
Status Report: Describe ongoing national programs,
copy of any related supple­men­tary material, and a
abstract and text.
studies or information systems bearing on Canadian
public health (maximum 3,000 words). Abstract not
Second title page: Title only; start page numbering
required.
here as page 1.
Workshop/Conference
Report:
Summa-rize
Abstract:
Unstructured
(one
paragraph,
no
significant, recently held events relat­ing to national
headings), maximum 175 words (100 for short
public health (maximum 1,200 words). Abstract not
reports); include 3B8 key words (preferably from
required.
the Medical Subject Headings (MeSH) of Index
Medicus).
Cross‑Canada Forum: For authors to present or
exchange information and opin­ions on regional or
Text: Double‑spaced, 1 inch (25 mm) margins, 12
national surveillance findings, programs under
point font size.
development or public health policy initiatives
(maximum 3,000 words). Abstract not required.
Acknowledgements: Include disclosure of financial
and material support in acknowledgements; if
Letter to the Editor: Comments on articles recently
anyone is credited in acknowledgements with
published in CDIC will be consid­ered for publication
substantive scientific contributions, authors should
(maximum 500 words). Abstract not required.
state in cover letter that they have obtained written
permission.
Vol 29, No 2, 2009
94
Chronic Diseases in Canada
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

Download PDF

advertisement