Volume 27, Number 1, 2006

Volume 27, Number 1, 2006
Volume 27, Number 1, 2006
Chronic Diseases in Canada
Table of Contents
1
Spatio-temporal distribution of hypothyroidism
in Quebec
Fabien Gagnon, Marie-France Langlois, Isabelle Michaud,
Suzanne Gingras, Jean-François Duchesne and Benoît Lévesque
9
A population-based analysis of health behaviours,
chronic diseases and associated costs
Arto Ohinmaa, Donald Schopflocher, Philip Jacobs,
Sandor Demeter, Anderson Chuck, Kamran Golmohammadi
and Scott W Klarenbach
25
Multiple exposures to smoking, alcohol, physical
inactivity and overweight: Prevalences according to the
Canadian Community Health Survey Cycle 1.1
Julia E Klein-Geltink, Bernard CK Choi and Richard N Fry
34
Letters
Chronic or non-communicable?
Status Reports
36
David Carle-Ellis
Acting Editor-in-Chief
(613) 952-3299
Robert A Spasoff
Associate Scientific
Editor
Sylvie Stachenko
Principal Scientific Editor
(613) 946-3537
Claire Infante-Rivard
Associate Scientific
Editor
Stephen B Hotz
Associate Scientific
Editor
Cathy Marleau
Desktop Publisher
Francine Boucher
Graphic Design
The epidemiology of self-reported fibromyalgia
in Canada
J Dayre McNally, Doug A Matheson and Volodko S Bakowsky
17
a publication of the Public Health
Agency of Canada
Two to three percent of infants are born with a
congenital anomaly, but who’s counting? A national
survey of congenital anomalies surveillance in Canada
Dana Paquette, R Brian Lowry and Reg Sauvé
39
Easy access to chronic disease surveillance information:
The NCD Surveillance Infobase
40
Calendar of Events
41
42
2005 Peer Reviewers
Indexes to Volume 26
Information for Authors
(on inside back cover)
Published by authority of the Minister of Health
©Minister of Public Works and Government Services Canada 2006
ISSN 0228-8699
CDIC Editorial Committee
Jacques Brisson
Université Laval
Neil E Collishaw
Physicians for a
Smoke-Free Canada
James A Hanley
McGill University
C Ineke Neutel
University of Ottawa
Institute on Care of
the Elderly
Kathryn Wilkins
Health Statistics Division
Statistics Canada
Clyde Hertzman
University of British
Columbia
Chronic Diseases in Canada (CDIC) is a quarterly scientific
journal focussing on current evidence relevant to the control and prevention 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 articles; the opinions expressed are not necessarily those of the CDIC editorial
committee nor of the Public Health Agency of Canada.
Subscription is free upon request.
When notifying us of a change of address,
please enclose your address label.
Chronic Diseases in Canada
Public Health Agency of Canada
130 Colonnade Road
Address Locator: 6501G
Ottawa, Ontario K1A 0K9
Fax: (613) 941-3605
E-mail: [email protected]
Indexed in Index Medicus/MEDLINE, PAIS (Public
Affairs Information Service) and EMBASE, the
Excerpta Medica database.
This publication is also available online at
http://www.phac-aspc.gc.ca/publicat/
cdic-mcc/index.html
Spatio-temporal distribution of hypothyroidism in
Quebec
Fabien Gagnon, Marie-France Langlois, Isabelle Michaud, Suzanne Gingras, Jean-François Duchesne and
Benoît Lévesque
Abstract
This study estimates the incidence and prevalence of hypothyroidism in Quebec, based on a
data bank produced by the Régie de l’assurance maladie du Québec (RAMQ) on the use of
thyroid hormones by persons insured under RAMQ’s public drug insurance plan between
1992 and 2001. In 2001, the prevalence of thyroid hormone use in women and men
respectively was 10.8 and 2.9 percent. Prevalence increases with age, reaching, among those
aged 65 and over, 21.9 percent in women and 8.0 percent in men in 2001. Incidence is highest
in women between the ages of 45 and 64 and in men aged 65 and over. Age-related incidence
is relatively stable in women but tends to increase in men. On a regional and local basis (by
Centre local de services communautaires [CLSC]), incidence rates up to 2.4 times higher than
anticipated on the basis of provincial incidence rates were observed.
Key words: distribution by age, distribution by sex, hypothyroidism, incidence, prevalence,
Quebec, spatial distribution, temporal trend
Introduction
Thyroid hormones, which act at the genome
level, perform many different functions
within many different systems. In addition to
being essential to neurological and intellectual development at the fetal stage and in
childhood, these hormones are also essential
to normal growth. They also have an effect
on the heart by accelerating cardiac rate and
contractility. Thyroid hormones influence
the respiratory centres, alter intestinal motility and increase bone remodelling as well as
protein uptake by the muscles. Finally, these
hormones influence the metabolism of
carbohydrates and lipids.1
Hypothyroidism is defined as a clinical
syndrome resulting from thyroid hormone
deficiency. However, because thyroid hormones are implicated in numerous functions, the signs and symptoms of hypo-
thyroidism can often be rather general and
difficult to characterize. The symptomatology of this endocrine disease is generally
subtle and insidious, at least in the early
stages. Consequently, hypothyroidism may
manifest as a variety of stigmas, including
hoarseness, psychomotor slowing, intolerance to cold, hair loss, skin coarsening and
dryness, weight gain, bradycardia and constipation. Certain signs, such as myxoedema
and slowing of the relaxation phase in
tendon reflexes, are more specific but not
always present.1 The condition can also
cause morbid complications in a wide variety of other conditions. For example, even
mild, sub-clinical hypothyroidism is associated with a partially reversible rise in
low-density lipoprotein (LDL) cholesterol.2
This can have clinically important consequences, as demonstrated in cohort studies
that point to a possible link between
sub-clinical hypothyroidism and cardiovascular disease.3,4 Hypothyroidism thus represents a far from negligible source of
morbidity, both from an individual and a
population standpoint.
Management of this chronic condition consists of lifelong treatment with levothyroxine,
as well as medical monitoring. Even when
the disease is stable, adjustments in therapy
may be required in a variety of situations
(pregnancy, aging, particularly in patients
suffering from coronary disease), or as a result of poor drug compliance.1 According to
the classification established by the Régie de
l’assurance maladie du Québec (RAMQ), sodium levothyroxine (Synthroid®) is the medication most frequently prescribed by Quebec
physicians after acetylsalicylic acid (Aspirin®).5
Recognized risk factors for hypothyroidism
include genetic predisposition, excessive
consumption of iodine or, conversely, iodine
deficiency, as well as certain iatrogenic
causes (radioiodide, surgery) and drugrelated causes (lithium, amiodarone, anticonvulsant drugs).1 A large number of
chemical products can interfere with thyroid
gland functioning and, possibly, the action
of thyroid hormones.6 However, there is still
considerable uncertainty regarding the clinical impact of such disturbances, given the
paucity of studies on human subjects.7 Accordingly, the purpose of this study is to provide guidance for etiological research in this
area.
Author References
Fabien Gagnon, CHUL-Centre de recherche du CHUQ, Unité de recherche en santé publique; Université de Sherbrooke, Faculté de médecine
Marie-France Langlois, Université de Sherbrooke, Faculté de médecine; Centre de recherche clinique du CHUS, Axe de physiopathologie endocrinienne
Isabelle Michaud, Université de Sherbrooke, Faculté de médecine
Suzanne Gingras, Benoît Lévesque, CHUL-Centre de recherche du CHUQ, Unité de recherche en santé publique; Institut national de santé publique du Québec
Jean-François Duchesne, CHUL-Centre de recherche du CHUQ, Unité de recherche en santé publique
Correspondence: Fabien Gagnon, 300 King Street East, Suite 300, Sherbrooke, Québec J1G 1B1; fax: (819) 566-2903; e-mail: [email protected]
Chronic Diseases in Canada
1
Vol 27, No 1, 2006
Method
Study population and period
The study population is that of Quebeckers
who were insured under the Régie de
l’assurance maladie du Québec’s (RAMQ)
public drug insurance plan during the period
of 1992 to 2001 inclusively. Prior to 1997,
only persons aged 65 and over, as well as
income security recipients and aboriginal
persons, were insured under this plan. In
1997, coverage under the public plan was
extended to all persons under the age of 65
who were not covered under a private drug
plan. These persons are referred to as
“participants.” Users of thyroid hormone
replacement products are defined as those
given a prescription for sodium levothyroxine (Synthroid® or Eltroxin®) or sodium
liothyronine (Cytomel®). Excluded from this
study are users of thyroid hormone replacement products who also take lithium, a drug
that can induce hypothyroidism.1
The first year of the study period, 1992,
encompasses a combination of new and old
cases. It was therefore selected as the base
level for identifying new users of thyroid
hormone replacement products as of 1993.
Moreover, since “participants” were added
to the population covered by the public
insurance plan in 1997, data from that year
can only be used to establish a new base
level for subsequent years and to identify
new users as of 1998. For persons under the
age of 65, the study period was divided into
two periods: 1993 to 1996 and 1998 to 2001.
However, for persons aged 65 and older who
were covered by the plan without interruption, the entire period of 1993 to 2001 was
used.
Variables
Since the data bank of RAMQ assigned an
anonymous identification number to every
user of thyroid hormone replacement products, we were able to gain access to the
following data: age or date of birth, sex,
Centre local de services communautaires
(CLSC) area and administrative region of
residence, and product name. Lithium use
status was also available for each subject
(except for the year 1992).
For each year from 1997 to 2001, three different records (one for income security recipients, one for persons aged 65 and over, and
another for public drug insurance plan
participants) provided the total number of
persons (in person-years) who were insured
under the plan, by age and sex, in the various health and social service regions of
Quebec. For years prior to 1997, population
data for income security recipients, as well
as for persons aged 65 and over, by region,
sex and age, were not available and were
therefore estimated based on the data relating to 1997 and 1998. This estimate was
based on the supposition that the change in
the size of the insured population that
occurred between 1997 and 1998 was comparable to the changes that occurred in earlier years, based on region, sex and age.
An additional record dealt with the total
number of persons insured, in 2002, under
the public drug insurance plan for each area
served by a CLSC, by age and sex. These
population data were applied to the 1998–
2001 period, according to CLSC area, age
and sex.
Data processing and statistical
analysis
In the course of this study, prevalence, as
well as crude and age-standardized incidence rates (direct standardization) were
TABLE 1
Incidence and prevalence of thyroid hormone use in Quebec from 1993 to 1996, in
seniors and social assistance recipients
Incidence
Eligible
a
population (PY)
Number of
new cases
Crude rate
(/100,000)
Adjusted rate
(/100,000)
Number of
cases
1993
962,170
12,302
1,278.57
1,384.44
65,722
6.83
1994
945,939
12,608
1,332.86
1,472.39
73,562
7.78
1995
929,708
13,498
1,451.85
1,550.84
81,726
8.79
1996
913,477
14,255
1,560.52
1,648.89
91,435
10.01
3,751,294
52,663
1,403.86
1,511.75
106,792
11.39
Sex
Women
Prevalence
Year
1993–1996
Prevalence (%)
b
p-trend value < 0.001
Men
1993
772,894
3,013
389.83
390.33
11,200
1.45
1994
756,046
3,331
440.58
420.70
13,194
1.75
1995
739,198
3,504
474.03
448.95
15,211
2.06
1996
722,350
4,135
572.44
515.00
17,921
2.48
2,990,488
13,983
467.59
443.49
22,285
2.98
1993–1996
b
p-trend value < 0.001
a
person-years
b
calculated for adjusted incidence rates
Vol 27, No 1, 2006
2
Chronic Diseases in Canada
FIGURE 1
Annual prevalence and age-adjusted incidence rates of thyroid hormone use in
Quebec from 1993 to 2001 in persons aged 65 and over
2000
25
1800
p-value = 0.712
20
1400
p-value < 0.001
1200
15
1000
800
10
Prevalence (%)
Adjusted incidence rate
(/100,000 person-years)
1600
600
400
Incidence in women
0
1993
Results
Prevalence in women
Prevalence in men
1994
1995
1996
used to describe the use of thyroid hormones. The weighting system employed was
based on the five-year age group structure of
the Quebec population insured by RAMQ,
via the summation of male and female population sizes during the period extending from
1998 to 2001.
The age-standardized rate ratio (SRR)—the
standardized incidence rate of a given area
over the provincial rate—was the measure
used to compare rates. The p-value associated with the SRR provided a means of determining whether differences were statistically
significant.8 Rate variation coefficients were
also presented in order to measure the rates.
In order for an SRR to be considered significantly different (by 1), both clinically and
statistically, three elements had to be present: the gap in relation to the province had to
be sufficiently large (a difference of at least
33 percent); rates had to be stable (a variance coefficient of no more than 16.5 percent); and, of course, the difference had to
be statistically significant (p-value ≤ 0.001).
The importance attributed to a gap is always
partly subjective. For this reason, it was
decided that the knowledge acquired concerning geographic variations in cardiovascular disease (CVD) would be used to
provide objective benchmarks. Like hypothyroidism, CVD is a multifactorial chronic
Chronic Diseases in Canada
1997
Year
1998
1999
2000
0
2001
disease. There are significant CVD mortality
gaps among industrialized countries. Since
we knew that the rate of CVD mortality in
Japan was 67 percent lower than in Canada,9
we felt that we were justified in assuming
that a gap should be at least equivalent to
half of this value (that is, 33 percent) before
variations in exposure to certain potential
risk factors should be considered. Moreover,
the statistical power of this study is unassailable, given the large population sizes that
were used in our calculations. It is important
to underscore, however, that this considerable statistical power may result in rejection
of the null hypothesis for very small differences. In order to offset this phenomenon,
the threshold of statistical significance was
set at a = 0.1% (p ≤ 0.001).
For each of the two study periods, provincial
incidence rates and prevalence were calculated by year and according to the following
age groups: < 15 years, 15 to 44 years,
45 to 64 years and ≥ 65 years. Since the population aged 65 and over was insured by
RAMQ during the period from 1992 to 2001
inclusively, annual rates for this age group
were calculated for each study year, beginning with 1993. Regional incidence rates and
prevalence were calculated for the period of
1998 to 2001, by year and by age group.
Rates for the entire period were also calculated for each of the 167 CLSC areas in
3
In order to determine the presence of a temporal trend, in the form of an increase or
decrease in annual provincial or regional incidence rates during the period under study,
linear rate modelling was employed. The
threshold of statistical significance for these
temporal trends was also set at 0.1%
(p ≤ 0.001).
5
Incidence in men
200
Quebec. However, rates by region and by
CLSC area were not calculated for the
1993–1996 period since there remained considerable uncertainty regarding insured population estimates for this period. All results
are stratified by sex.
Table 1 presents incidence and prevalence of
thyroid hormone use by seniors and social
assistance recipients for the period of 1993 to
1996, while Table 2 presents incidence and
prevalence of use for all persons covered under the public drug insurance plan during
the 1998–2001 period. In 2001, 1,705,570
women and 1,454,208 men (in person-years) were covered by the public drug
insurance plan. For the 1998–2001 period as
a whole, the adjusted incidence rate was
1,192/100,000 person-years (PY) for women
and 541/100,000 PY for men. Among
women, the incidence rate tends to increase
over time during the first study period of
1993 to 1996 (p < 0.001), but shows a
downward trend during the second period of
1998 to 2001 (p < 0.001). In the case of
men, the incidence rate tends to increase
during both of the periods under study
(p < 0.001). The overall incidence rate in
women was 3.4 times higher than that of
men for the 1993–1996 period, and 2.2 times
higher in 1998–2001. In 2001, 10.8 percent
of women insured under the public drug
insurance plan were taking thyroid hormone
replacement products, a proportion that falls
to 2.9% for men.
Figure 1 presents annual rates of thyroid
hormone use in persons aged 65 and over
from 1993 to 2001. In women, the incidence
rate is fairly stable over time (p = 0.712),
whereas in men the rate increases
(p < 0.001). In 2001, the prevalence of
Vol 27, No 1, 2006
FIGURE 2
Adjusted incidence rates of thyroid hormone use in Quebec, by age and sex,
from 1998 to 2001, in persons registered with the public drug insurance plan
2500
Women
Men
Adjusted incidence rate
(/100,000 person-years)
2000
1500
1000
500
0
< 15 years
15-44 years
45-64 years
65 + years
Age groups
thyroid hormone use in women and men in
this age group was 21.9 percent and 8.0
percent, respectively.
Figure 2 illustrates variations in the incidence of thyroid hormone use for different
age groups during the 1998–2001 period.
The highest incidence in women is found in
the 45-to-64 age group (1,998/100,000 PY).
In men, the highest incidence rate is found in
the 65-and-over age group (1,239/100,000
PY).
Incidence and prevalence of thyroid hormone
use, by administrative region during the
1998–2001 period, are presented in Table 3.
For women, Chaudière-Appalaches (age-standardized rate ratio [SRR] = 1.426) is the only
region in which the incidence rate is significantly higher than the provincial rate, based
on the criteria selected. For men, two regions
show an incidence rate that is significantly
higher than the provincial rate, namely
Chaudière-Appalaches (SRR = 1.778) and
Lower St. Lawrence (SRR = 1.491). Based on
the selected criteria, no region shows an incidence rate for either sex that is significantly
lower than the provincial rate. The highest
prevalence obtained by age group were in persons aged 65 and over residing in the
Chaudière-Appalaches region, where 30.6 percent of women and 16.4 percent of men in this
group were thyroid hormone users (data not
presented).
Vol 27, No 1, 2006
During the 1998–2001 period, annual
incidence rates for women showed a downward trend in Saguenay-Lac-Saint-Jean,
Montreal-Centre, Outaouais, ChaudièreAppalaches and Lanaudière, and an upward
trend in Montérégie (p < 0.001). Among
men, a downward variation was observed
only in the LanaudiPre region, while rates increased in Mauricie-Centre-du-Québec,
Estrie, Abitibi-Témiscamingue, GaspésieÎles-de-la-Madeleine and Montérégie.
Two additional figures (not published here
for technical reasons but available by request) show the geographical distribution,
for men and women respectively, of the
SRRs calculated for the 1998-2001 period, by
Centre local de services communautaires
(CLSC) area. Based on the selected criteria,
women in 17 CLSC areas show an incidence
rate of thyroid hormone use that is significantly higher than the provincial rate; incidence rates for men were significantly higher
than the provincial rate in 22 CLSC areas.
CLSC areas with incidence rates exceeding
the provincial rate for both women and men
were observed in the following regions:
Chaudière-Appalaches (6 in 10 CLSC areas
for women and 9 in 10 areas for men);
Lower St. Lawrence (3 in 9 CLSC areas for
women and 6 in 9 areas for men) and North
Shore (3 in 8 CLSC areas for women and 1 in
8 areas for men). CLSC rates up to 2.0 and
4
2.4 times the expected rate were observed in
women and men respectively (in ChaudièreAppalaches).
Moreover, for women and men respectively,
16 and 19 CLSC areas had incidence rates of
thyroid hormone use that were significantly
lower than the provincial rate, based on the
selected criteria. The regions with the largest
proportion of CLSCs with incidence rates
lower than the provincial rate were
Saguenay-Lac-Saint-Jean (3 in 7 CLSC areas
for women and 2 in 7 areas for men),
Montreal-Centre (2 in 35 CLSC areas for
women and 6 in 35 areas for men), the
Laurentians (2 in 7 CLSC areas for women
and 3 in 7 areas for men), and Montérégie
(4 in 19 CLSC areas for women and 5 in 19
areas for men). CLSC rates as low as 0.62
and 0.54 times the expected rate were
observed for women and men respectively
(in Montérégie).
Discussion
In locales where iodine intake levels are adequate, the usual prevalence of hypothyroidism is roughly one percent.10 For example, a survey conducted with 2,779 adults
in the municipality of Whickham, England,
showed that the prevalence of hypothyroidism was 1.4 percent in women and
less than 0.1 percent in men.11 The prevalence of thyroid hormone use measured in
2001 for the purposes of this study—namely
10.8 percent for women and 2.9 percent for
men—suggest that hypothyroidism is far
more common than first suspected. In fact,
these prevalences would appear to be more
reflective of the prevalence of sub-clinical
hypothyroidism, a condition defined as the
presence of a high concentration of thyreostimulin (or thyroid-stimulating hormone
[TSH], which is produced by the pituitary
gland to stimulate the thyroid), and normal
concentrations of thyroid hormones. Indeed,
investigations comprising biological measurements have demonstrated prevalences
of sub-clinical hypothyroidism of eight percent in women and three percent in men.12
Our own results suggest that the majority of
sub-clinical hypothyroidism cases are probably being treated in Quebec (although this
practice is not unanimously supported by
Chronic Diseases in Canada
TABLE 2
Incidence and prevalence of thyroid hormone use in Quebec from 1998 to 2001, in
persons registered with the public drug insurance plan
Incidence
Sex
Women
Year
Eligible
a
population (PY)
Prevalence
Number of
new cases
Crude rate
(/100,000)
Adjusted rate
(/100,000)
Number of
cases
Prevalence (%)
1998
1,672,977
21,073
1,259.61
1,223.87
145,987
8.73
1999
1,695,650
20,888
1,231.86
1,192.64
158,908
9.37
2000
1,693,898
20,993
1,239.33
1,195.26
171,587
10.13
2001
1,705,570
20,661
1,211.38
1,159.28
183,429
10.75
6,768,095
83,615
1,235.43
1,191.83
211,956
12.53
1998
1,406,844
6,299
447.74
492.25
29,003
2.06
1999
1,429,296
7,002
489.89
538.71
33,533
2.35
2000
1,440,160
7,397
513.62
566.41
38,069
2.64
2001
1,454,208
7,547
518.98
565.69
42,677
2.93
5,730,508
28,245
492.89
540.96
51,680
3.61
1998-2001
b
p-trend value < 0.001
Men
1998–2001
b
p-trend value < 0.001
a
person-years
b
calculated for adjusted incidence rates
the medical community13,14), or perhaps that
the prevalence of clinical hypothyroidism is
actually greater here than elsewhere.
It is important to note that among the thyroid
hormones considered in this study, sodium
liothyronine (Cytomel®) is not specific to the
treatment of hypothyroidism: This hormone
can also be used to treat refractory depression or to prepare patients for certain nuclear
medicine tests.1 However, our data bank
shows that this drug was taken by only 0.38
percent of those who used thyroid hormones
for the first time during the 1998–2001
period. Moreover, the large proportion of
seniors in our study population necessarily
results in an overestimation of prevalence,
since age adjustments are made only for
incidence. Still, the prevalences calculated
with respect to persons aged 65 and over are
comparable to those estimated during the
same period in seniors for The Canadian
Study of Health and Aging. In that study,
which was carried out between February
1991 and May 1992, the proportion of persons aged 65 and over who were using thyroid hormone replacement products was 8.8
percent for men and women combined
(compared to 11.2 percent for women and
Chronic Diseases in Canada
2.9 percent for men in the same group in our
own 1993 study).15
As for incidence, the rates of thyroid hormone use observed here (1,192/100,000 PY
in women and 541/100,000 PY in men for
the 1998–2001 period) are distinctly higher
than those measured for hypothyroidism in
the Whickham cohort follow-up (410/
100,000 PY in women and 60/100,000 PY in
men, at the end of a follow-up period of
twenty years).16 Also noteworthy is the fact
that case definitions are comparable, since
the identification of new cases in the
Whickham cohort follow-up was based on
the treatment decisions of physicians. It may
be that the treatment of sub-clinical hypothyroidism is more selective in England.
What is more, the two populations are quite
different: The first comprises persons aged
65 and over, as well as all social assistance
recipients, while the second is derived from
a random sample.
Age-related increases in incidence rates and
prevalence of hypothyroidism are a wellknown phenomenon.17 However, it is impossible to determine whether the temporal
variations observed in the rates measured
5
here (particularly the marked increase observed in men aged 65 and over between
1993 and 2001) reflect a real or apparent increase in disease. Such temporal trends may
reflect changes in the population’s consultation habits as much as changes in medical
practices (level of medical assessment, thyroid hormone assay methods, interpretation
of laboratory results, etc.). One thing is certain: The same phenomenon has been observed elsewhere in the world. According to
a general population study conducted in
Spain, the prevalence of thyroid hormone
use in that country increased by 164 percent
between 1992 and 2000.18
One of the primary problems encountered in
epidemiological studies of thyroid disease relates to the definitions used.10 The diagnosis
of hypothyroidism is based on the measurement of TSH. The secretion of thyroid hormones by the thyroid gland is in fact a
response to a negative feedback mechanism:
If there is thyroid insufficiency, the level of
TSH increases.1 TSH is therefore a marker of
thyroid activity. Different generations of
tests have been used to measure TSH. The
detection limit of first-generation tests was
somewhere between 5 and 10 BIU/L. Most
Vol 27, No 1, 2006
TABLE 3
Incidence and prevalence of thyroid hormone use, by Quebec region, from 1998 to 2001, in
persons registered with the public drug insurance plan
Incidence
Men
Prevalence
SRR
P-value
CV (%)
Number of
cases
1,495.18
1.255
0.0000
1.71
8,362
15.02
1,246.55
1.046
0.0112
1.74
8,619
13.43
1,139.78
0.956
0.0003
1.19
19,845
13.81
04 Mauricie et Centre-du-Québec
1,011.72
0.849
0.0000
1.38
14,849
12.14
05 Estrie
1,273.41
1.068
0.0001
1.67
8,999
13.31
06 Montreal-Centre
1,054.83
0.885
0.0000
0.70
54,285
11.28
07 Outaouais
1,281.48
1.075
0.0001
1.82
7,163
11.43
08 Abitibi-Témiscamingue
1,018.39
0.854
0.0000
2.69
4,427
12.74
09 North Shore
1,491.30
1.251
0.0000
3.02
2,573
13.53
11 Gaspésie – Îles-de-la-Madeleine
1,097.58
0.921
0.0027
2.72
3,416
10.90
12 Chaudière-Appalaches
1,699.91
1.426
0.0000
1.27
14,985
17.38
13 Laval
1,312.69
1.101
0.0000
1.57
9,750
13.29
14 Lanaudière
1,437.23
1.206
0.0000
1.47
10,359
12.64
15 Laurentides
1,050.59
0.881
0.0000
1.55
11,302
11.20
16 Montérégie
1,227.85
1.030
0.0014
0.87
32,625
12.41
10-17-18 Nord-du-Québec, Nunavik,
Terres-Cries-de-la-Baie-James
1,286.58
1.079
0.3042
7.44
397
9.49
Province
1,191.83
1
–
0.35
211,956
12.53
Sex
Women
Region
01 Lower St. Lawrence
02 Saguenay – Lac-Saint-Jean
03 Quebec City
Adjusted rate
(/100,000)
Prevalence (%)
01 Lower St. Lawrence
806.42
1.491
0.0000
2.61
2,545
5.19
02 Saguenay – Lac-Saint-Jean
607.30
1.123
0.0001
2.95
2,125
3.86
03 Quebec City
540.43
0.999
0.9641
2.07
4,793
4.14
04 Mauricie et Centre-du-Québec
484.42
0.895
0.0000
2.31
3,776
3.68
05 Estrie
586.07
1.083
0.0054
2.81
2,342
4.12
06 Montreal-Centre
441.69
0.816
0.0000
1.27
11,782
2.96
07 Outaouais
552.49
1.021
0.5388
3.38
1,678
3.06
08 Abitibi-Témiscamingue
583.49
1.079
0.0682
4.11
1,282
4.08
09 North Shore
652.87
1.207
0.0003
5.19
720
4.19
11 Gaspésie – Îles-de-la-Madeleine
472.60
0.874
0.0036
4.59
883
3.07
12 Chaudière-Appalaches
961.82
1.778
0.0000
1.92
4,702
6.33
13 Laval
546.30
1.010
0.7410
2.91
2,233
3.63
14 Lanaudière
613.55
1.134
0.0000
2.67
2,576
3.54
15 Laurentides
458.44
0.847
0.0000
2.76
2,770
3.12
16 Montérégie
507.78
0.939
0.0002
1.59
7,374
3.32
10-17-18 Nord-du-Québec, Nunavik,
Terres-Cries-de-la-Baie-James
530.72
0.981
0.8834
13.01
99
2.53
Province
540.96
1
51,680
3.61
Vol 27, No 1, 2006
6
–
0.60
Chronic Diseases in Canada
laboratories in Quebec now use secondgeneration tests, which have a detection
limit of approximately 0.1 BIU/L.19 The
changeover to more sensitive tests occurred
in the mid-1980s.19 Therefore, it is unlikely
that this transition accounts for the observed
trend, as it occurred well before the period
covered by the present study.
Moreover, the upper limit of the reference
interval for TSH has regularly declined over
the past two decades. Long set at 10
BIU/L,19,20 this limit had fallen to anywhere
between 0.4 BIU/L and 6.0 BIU/L by
1992.19,21 Although it occurred just before
the study period, this change in the interpretation of laboratory results may have been
introduced gradually and may therefore
explain part of the rate increases observed,
at least during the first study period (1993–
1996). Following the introduction of new
guidelines in 2002 by the National Academy
of Clinical Biochemistry in the U.S., the
upper limit of the reference interval was
again reduced.19 However, that change took
place after the period covered by the present
study.
This study does not provide an explanation
for the disparities and geographic clustering
observed in regional and local rates of thyroid hormone use. Given that auto-immune
forms represent the primary cause of hypothyroidism in regions where iodine intake
levels are adequate,1,10 increased susceptibility determined by genetic factors inevitably
presents itself as a possible explanation.
Exposure to certain environmental factors
may also play an etiological role, even in
auto-immune forms.22 Pesticides, halocarbons, phenolic compounds and phthalates
are the synthetic compounds most frequently studied for their toxic effects on thyroid function.7 It has not been possible,
within the framework of this study, to identify the proportion of users who began taking
thyroid hormones following treatment for
hyperthyroidism (iodine 131, subtotal thyrodectomy, antithyroid drugs). As many as
one third of all cases of hypothyroidism may
in fact have an iatrogenic origin.10 Thus, it is
possible that part of the geographic variations observed in rates of hypothyroidism
may be due to variations in the occurrence of
hyperthyroidism, or to variations in the
Chronic Diseases in Canada
modalities used to treat this disease. In order
to avoid confounding effects, users of thyroid hormone replacement products who
were also taking lithium for the treatment of
manic-depressive psychosis were excluded
from the study, since lithium can induce
hypothyroidism.1 Family history is considered to be the most powerful risk factor for
mood disorders and bipolar disorder in
particular.23
trends that were measured here reflect an
actual or merely apparent increase in disease. It will also be necessary to determine
whether regional and local disparities can be
explained by variations in thyroid patient
management practices. In the event that geographic variations in medical practices are
not present, the investigation of regional excesses should be pushed further, in the form
of etiological studies.
It is important to interpret results prudently.
In certain situations, the use of extrapolations to estimate eligible populations may
have resulted in inaccurate rate calculations.
What is more, even though the public drug
insurance plan now covers close to half of
the Quebec population (46 percent of
women and 41 percent of men in 200124),
differences exist between the population
insured under this plan and that covered by
private insurers. Finally, it is important to
understand that two participants covered for
only six months under the public plan, as a
result of having access to private coverage
during the rest of the year, were not counted
in this study as two persons covered under
the plan in that year, but as a single personyear. A variation of this phenomenon on a
regional basis could represent a source of
bias in terms of geographic analysis, particularly in situations where differences in types
of employment or in employment stability
result in more frequent movement between
the public plan and private insurance plans.
Still, such bias would be limited to workingage population groups, since all seniors are
covered under the public plan. However,
incidence rates for seniors could also be over
estimated in regions that have a large proportion of persons covered by a private plan
prior to age 65. Among the latter, all thyroid
hormone users who acquire coverage under
the public plan upon reaching age 65 will incorrectly be identified as new cases. Nonetheless, it is interesting to note that agestandardized rate ratio (SRR) calculations by
age group reveal that in regions where
excesses were observed, these excesses
manifested themselves in all age groups,
beginning at age 15.
Acknowledgements
The authors wish to thank the Human
Health intervention component of the
federal-provincial St. Lawrence Action Plan
Vision 2000 for financially supporting the
production of the data bank required to complete this study. We also wish to thank
Danielle Labrie Pelletier and France Bourque
of the information production and dissemination department of the Régie de l’assurance
maladie du Québec (RAMQ) for the establishment of the data bank, as well as Health
Canada’s Safe Environments Programme for
its financial support of the mapping of results, and Anne-Marie Lalonde for her contribution to the mapping. The thyroid
research being carried out by Dr. MarieFrance Langlois is funded in part by a Junior
2 Clinical Research Scholar Award provided
by the Fonds de la recherche en santé du
Québec.
References
1.
Greenspan FS. The thyroid gland. In:
Greenspan FS, Gardner D, editors. Basic &
clinical endocrinology. McGraw-Hill; 2004;
215–94.
2.
Danese MD, Ladenson PW, Meinert CL,
Powe NR. Effect of thyroxine therapy on serum lipoproteins in patients with mild
thyroid failure: A quantitative review of the
literature. J Clin Endocrinol Metab. 2000;85:
2993–3001.
3.
Hak AE, Pols APH, Visser TJ, Drexhage HA,
Hofman A, Witteman JCM. Subclinical
hypothyroidism is an independent risk factor
for atherosclerosis and myocardial infarction
in elderly women: The Rotterdam study.
Ann Intern Med. 2000;132:270–8.
Only population surveys that comprise biological measurements would provide a
means of determining whether the temporal
7
Vol 27, No 1, 2006
4.
Imaizumi M, Akahoshi M, Ichimaru S,
Nakashima E, Hida A, Soda M, Usa T,
Ashizawa K, Yokoyama N, Maeda R,
Nagataki S, Egushi K. Risk for ischemic heart
disease and all-cause mortality in subclinical
hypothyroidism. J Clin Endocrinol Metab.
2004;89:3365–70.
5.
Les dix médicaments les plus cofteux et les
plus souvent prescrits en 2003. L’actualité
médicale. June 9, 2004;10.
6.
Zoeller RT, Dowling ALS, Herzig CTA,
Iannacone EA, Gauger KJ, Bansal R. Thyroid
hormone, brain development, and the
environment. Environ Health Perspect. 2002;
110:355–61.
7.
Brucker-Davis F. Effects of environmental
synthetic chemicals on thyroid function.
Thyroid. 1998;8:827–56.
8.
Bouyer J, Hémon D, Cordier S, Derriennic F,
Stücker I, Stengel B, Clavel J. Épidémiologie :
Principes et méthodes quantitatives. Paris:
Les Éditions INSERM;1993.
9.
Levi F, Lucchini F, Negri E, La Vecchia C.
Trends in mortality from cardiovascular and
cerebrovascular diseases in Europe and
other areas of the world. Heart. 2002;88:
119–24.
10. Vanderpump MPJ, Turnbridge WM. The
epifemiology of thyroid diseases. In:
Braverman LE, Utiger RD, editors. The
thyroid: A fundamental and clinical text. 8th
ed. Philadelphia: Lippincott Williams &
Wilkins; 2000;467–73.
Vol 27, No 1, 2006
11. Tunbridge WMG, Evered DC, Hall R,
Appleton D, Brewis M, Clark F, et al. The
spectrum of thyroid disease in a community:
The Whickham Survey. Clin Endocrinol.
1977;7:481–93.
12. Vanderpump MP, Tunbridge WM. Epidemiology and prevention of clinical and
subclinical hypothyroidism. Thyroid. 2002;
12:839–47.
13. Vanderpump M. Subclinical hypothyroidism: The case against treatment. Trends
Endocrinol Metab. 2003;14:262–66.
14. Surks MI, Ortiz E, Daniels GH, Sawin CT, Col
NF, Cobin RH, Franklyn JA, Hershman JM,
Burman KD, Denke MA, Gorman C, Cooper
RS, Weissman NJ. Subclinical thyroid
disease: Scientific review and guidelines for
diagnosis and management. JAMA. 2004;
291(2):228–38.
15. Krueger PD, Raina P, Braun EA, Patterson C,
Chambers LW. Prevalence and risk factors of
hypothyroidism: Findings from the Canadian
study of health and aging. Canadian Journal
on Aging. 2001;20:127–35.
16. Vanderpump MPJ, Tunbridge WMG, French
JM, Appleton D, Bates F, Clark F, et al. The
incidence of thyroid disorders in the community: A twenty-year follow-up of the
Whickham Survey. 1995;43:55–68.
17. USPSTF. Screening for thyroid disease,
Chapter 20. Guide to clinical preventive
services: Report of the U.S. Preventive
Services Task Force. 2nd ed. Williams and
Wilkins. 1996;209–18.
8
18. Diaz Madero A, Lopez Ferreras A. Estimate
of the prevalence of hypothyroidism in
Castilla y Leon and its progress from 1992 to
2000 through the consumption of thyroid
hormones [in Spanish]. Rev Esp Salud
Publica. 2001;75:345–52.
19. Baloch Z, Carayon P, Conte-Devolx B,
Demers LM, Feldt-Rasmussen U, Henry JF,
LiVosli VA, Niccoli-Sire P, John R, Ruf J,
Smyth PP, Spencer CA, Stockigt JR. Guidelines Committee, National Academy of Clinical Biochemistry. Laboratory medicine
practice guidelines. Laboratory support for
the diagnosis and monitoring of thyroid disease. Thyroid. 2003;13:3–126.
20. Sox C H, Jr, editor. Thyroid function. Common diagnostic tests: Use and interpretation.
Philadelphia, PA: American College of Physicians; 1990;151.
21. Wallach, J. Serum thyroid-stimulating hormone (TSH;thyrotropin). Interpretation of
diagnostic tests: A synopsis of laboratory
medicine. Boston: Little, Brown; 1992;448.
22. Brix TH, Kyvik KO, Hegedus L. A population-based study of chronic autoimmune
hypothyroidism in Danish twins. J Clin
Endocrinol Metab. 2000;85:536–9.
23. Merikangas KR, Low NC. The epidemiology
of mood disorders [Abstract]. Curr Psychiatry Rep. 2004;6:411–21.
24. Statistics Canada. 2001 Census of Canada:
Profile – Quebec. URL: http://www12.
statcan.ca/english/census01/products/
standard/prprofile/prprofile.cfm?G=24.
Accessed: October 2004.
Chronic Diseases in Canada
The epidemiology of self-reported fibromyalgia in
Canada
J Dayre McNally, Doug A Matheson and Volodko S Bakowsky
Abstract
Fibromyalgia (FM) is a poorly understood condition characterized by chronic diffuse
musculoskeletal pain. This study describes the self-reported epidemiology of FM in Canada
using data collected from the Canadian Community Health Survey, Cycle 1.1 (2000). FM
prevalence rates with corresponding 95 percent confidence intervals were calculated. The
Canadian prevalence rate was 1.1 percent with a female-to-male ratio of six to one. In women,
rates increased with age up to 65 years, declining thereafter. Data collected on-ageat-diagnosis is presented and demonstrates a surprising number of newly diagnosed FM cases
among people in their 20s and 30s, signifying that FM is a problem for people of all ages. The
association with FM and a number of sub-populations was also investigated. With respect to
geography and environment, the FM prevalence rate in women was shown to be
approximately two percent in all Canadian regions except Quebec, where it was 1.1 percent.
Further analysis by language suggested that geographical and cultural differences might best
explain this observation. Finally, an association with a number of behavioral and
socioeconomic determinants of health, including weight, is presented.
Key words: Canada, epidemiology, fibromyalgia, prevalence
Introduction
Fibromyalgia (FM) is a controversial
rheumatologic disorder of uncertain etiology
and pathogenesis, characterized by chronic
widespread non-articular musculoskeletal
pain. The classification criterion most commonly used to define cases, both clinically
and in research, is the 1990 American College of Rheumatology (ACR) definition.1
This definition requires the presence of
chronic widespread pain of at least three
months duration and the presence of at least
11 of 18 possible tender points on clinical
exam. In addition to pain, FM patients often
report disturbing physical and psychological
symptoms including altered sleep patterns,
fatigue, cognitive problems and mood disturbances. Some have argued that these latter features should also be included in the
diagnostic criteria.2–4
Although many aspects of FM, such as
pathophysiology and treatment, are controversial, the substantial impact on patient
quality of life and the socioeconomic costs of
this disorder are without debate. Numerous
studies have shown that FM affects not only
physical health, but also emotional and
mental health, leading to restrictions in daily
living and leisure activities.3,5,6 FM is often
accompanied by a considerable degree of
work disability, an increased likelihood of
receiving financial support and consistently
higher health resource utilization.7–10 If previously reported values for FM prevalence
are correct, (one to two percent of the general population) approximately 500,000
Canadians suffer from FM, with an estimated cost of 350 million dollars to the
Canadian health care system.11 Given the
large impact that FM has at both the
individual and population levels, further
descriptive epidemiology of the disorder
would be helpful.
The widespread acceptance of FM as a diagnostic entity over the past decade has
created a scenario where large-scale epidemiological studies using self-reporting are
now possible. For example, two Europeanbased epidemiological studies focusing on
rheumatology, and including results on FM,
have recently been published, using selfreporting data.12,13 Our study used data collected by Statistics Canada in a national
health survey, the 2000/2001 Canadian
Community Health Survey (CCHS), Cycle
1.1, to carry out the first Canadian-based
large-scale descriptive epidemiological study
on FM. More specifically, the prevalence of
FM and its association with a number of
socioeconomic, demographic and behavioural determinants of health were
investigated.
In addition to providing more current data, a
FM study utilizing the CCHS data provides
numerous advantages over existing
European and North American studies. To
date, only small-scale studies have been carried out in North America, and, although of
undisputed value, the results obtained are
difficult to extrapolate to the national level.
These types of studies are invariably carried
out on relatively homogenous populations
and can be influenced by the health care dynamics within the area.14,15 In particular,
small-scale studies may suffer from a referral
bias as they are generally carried out in tertiary care centers where patients are not typical of the general community. The CCHS
Author References
J Dayre McNally, Department of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada
Doug A Matheson, Emergence Consulting, Manotick, Ontario, Canada
Volodko S Bakowsky, Department of Medicine, QEII Health Sciences Centre and Dalhousie University, Halifax, Nova Scotia, Canada
Correspondence: Volodko S Bakowsky, QEII Health Sciences Centre, Nova Scotia Rehabilitation Centre, Room 245, 1341 Summer Street, Halifax, Nova Scotia,
Canada B3H 4K4; fax: (902) 473-7019; e-mail: [email protected]
Chronic Diseases in Canada
9
Vol 27, No 1, 2006
survey design and sample size reduces these
biases. As well, the large size has allowed for
the estimation of prevalence in a variety of
subgroups—calculations not possible with
the smaller North American studies and not
performed with the larger European studies.
Methods
CCHS overview
The present analysis is based on the
cross-sectional data from Cycle 1.1 of the
CCHS, conducted by Statistics Canada and
carried out over a 12-to-14-month period
beginning in 2000. The survey’s design and
execution have been detailed elsewhere.16
Briefly, the CCHS uses the area frame
designed for the Canadian Labour Force
Survey as its primary sampling frame. A
multistage stratified cluster design was used
to draw a representative sample of dwellings, totaling 131,535 individual Canadians.
The target population included household
residents in all ten provinces, excluding
households on Indian reserves or Crown
land, Canadian military bases and some
remote areas. Selection of individual respondents was designed to target individuals
aged 12 or older with an over-representation
of those between 12 and 19 and those over
the age of 65. In 82 percent of the households one person was randomly selected for
an in-depth interview, and in the remaining
18 percent two persons were selected. If the
selected individual was unavailable after
repeated attempts, another member of the
household was asked to provide a proxy
interview.
Study population
As part of the interview, it was explained to
the participants that the survey was focused
on certain chronic health conditions. These
were described as “‘long-term conditions”
that had lasted or were expected to last six
months or longer and that had been diagnosed by a health professional. Those
respondents who answered affirmatively to
the survey question CC_Q041 (“Remember,
we’re interested in conditions diagnosed by
a health professional. Do you have
fibromyalgia?”) were recorded as having
FM. Those individuals who self-reported
Vol 27, No 1, 2006
having FM were also asked to provide the
age at which they were diagnosed (survey
question CC_Q042).
CCHS variables
The CCHS included numerous questions
related to health status, various determinants of health, and health service utilization. Following the identification of the study
population, the associations between FM
and a variety of additional characteristics
were investigated. The socio-demographic
variables included in the analysis were
gender; geographic location (Atlantic
Canada, Quebec, Ontario, the Prairie provinces, and British Columbia); urban (census
districts with a population density greater
than 400 inhabitants per square kilometer)
versus rural; immigrant status; Francophone
status; and age.
In addition, various determinants of health,
including education, income, weight, smoking and alcohol consumption, were investigated and the variables were grouped as
follows: To control for age and sex, men and
women were analyzed separately, and only
those between the ages of 35 and 65 were
used for the analysis. The association
between weight and FM was assessed using
Body Mass Index (BMI). Smoking history
was categorized as daily; occasional but
former daily; occasional; nonsmoker but
former daily; nonsmoker but former occasional; or never smoked. Alcohol consumption was categorized as regular, occasional
or former drinkers, as well as a group who
had never consumed alcohol. Subjects were
also grouped into four categories based on
the level of education attained: postsecondary education; some level of postsecondary education; completion of secondary education; or completion of less than
secondary school graduation. Income was
assessed using data provided by CCHS that
grouped the sample into quartiles (lowest,
lower middle, upper middle and highest
income) based on absolute income levels.
Statistical analysis
Prevalence rates were calculated among the
various subgroups described above. The
Rao-Wu bootstrap re-sampling technique
10
was used to calculate the corresponding 95
percent confidence interval for the point estimates. This technique corrects for the sampling error built into the CCHS complex
survey design caused by stratification, multiple selection stages and the unequalled probabilities of respondent selection.17,18 More
specifically, the Rao-Wu bootstrap method
estimates the sample variance by resampling from within the sample frame. The
statistical analyses were conducted using
SPSS software (release 11.1) and SPSS macros available through the Statistics Canada
remote access service. Statistics Canada
protects the validity of the data provided and
in certain instances, when the number of
observations is small and the output returns
a high coefficient of variation (CV), results
are withheld. More specifically, when the CV
is greater than 33.3, the estimate of variance
is considered meaningless and the point estimate ignored as it is deemed too unreliable
to publish. Additionally, when the CV was
calculated between 16 and 33.3, the point
estimate and confidence intervals are
retained, but the results should be interpreted cautiously as the estimated variance
used for deriving the confidence interval
may not be reliable.
Results
Demographic studies
Based on the CCHS data, 1.1 percent (95%
CI: 1.0–1.2) of the Canadian population
self-reported having health professionally diagnosed FM. Analysis by gender shows that
FM is a disorder predominantly affecting
women (1.8%; 95% CI: 1.7–2.0) with a
prevalence six times higher than that
observed in men (0.3%; 95% CI: 0.2–0.4).
The FM prevalence, and corresponding confidence intervals for a number of population
characteristics including age, immigrant
status and geographical location, are shown
in Table 1.
The prevalence of FM in women is initially
low in those younger than 25 years of age
(0.2%; 95% CI: 0.1–0.4) and then increases
until reaching a maximum in the 55 to 64
age grouping (4.2%; 95% CI: 3.6–4.8),
before declining in the elderly. The prevalence was constant in men over the age of
Chronic Diseases in Canada
TABLE 1
Prevalence and 95% confidence intervals of self-reported, health professionally
diagnosed FM in men and women according to individual characteristics.
The Canadian Community Health Survey, Cycle 1.1 (2000)
Men
Percentage
Women
CI
Percentage
CI
0.23
0.10–0.36
Age group
< 25
a
25–34
a
0.79
0.58–1.00
35–44
0.46
0.27–0.65
1.79
1.47–2.11
45–54
0.58
0.35–0.80
3.26
2.78–3.74
55–64
0.47
0.22–0.72
4.21
3.58–4.84
> 65
0.42
0.22–0.63
1.75
1.49–2.06
Rural
0.31
0.20–0.41b
2.03
1.75–2.31
Urban
0.33
0.25–0.42
1.79
1.63–1.94
0.28
0.22–0.35
Area
Immigrant status
Born in Canada
1.93
1.78–2.08
0.49
b
0.24–0.74
1.46
1.17–1.75
Atlantic provinces
0.27
0.15–0.39b
2.11
1.78–2.44
Quebec
0.21
0.11–0.30b
1.12
0.82–1.40
0.39
b
1.94
1.70–2.18
b
2.13
1.82–2.44
b
Immigrant
Region
Ontario
Prairie provinces
0.43
0.24–0.53
0.21–0.66
British Columbia
0.27
0.14–0.41
2.29
1.92–2.66
Overall prevalence
0.33
0.26–0.40
1.83
1.69–1.96
a
Insufficient observations to calculate the point prevalence; coefficient of variation is greater than 33.
b
Coefficient of variation is between 16 and 33.
35, but was too low to be estimated accurately in men under the age of 35. The combination of a gender preference and a high
prevalence in the age group making up the
largest portion of the population result in
almost 65 percent of all reported FM cases
being in women between the ages of 35 and
65.
Age-at-diagnosis observations
A number of interesting observations were
made using the data collected on age at diagnosis (data on men were excluded due to the
small number of cases). First, a comparison
of the respondent’s current age with age at
diagnosis showed a disproportionately high
number of women diagnosed in the years
corresponding to the introduction of the ACR
FM definition. Grouping the data into five-
Chronic Diseases in Canada
year intervals shows that there were an estimated 47,000 new FM cases in the last five
years, 88,000 cases five to ten years ago (i.e.
immediately following the introduction of
the ACR definition) and 56,000 cases ten to
fifteen years ago. A transient rise in the
number of new cases per annum is common
following both the general acceptance of a
new disease entity and the introduction of
either a new (more sensitive) diagnostic test
or set of diagnostic criteria. An assessment of
those cases diagnosed in the five years preceding data collection showed that only six
percent of the newly diagnosed FM cases
occurred in the group over the age of 60,
while 27 percent occurred in those under the
age of 35.
Using the Statistics Canada definition of
urban areas, no difference in the self-
11
reported prevalence was evident between
rural and urban sub-groups for either men or
women (Table 1). When disease prevalence
was calculated (among women between 35
and 65) for different geographical regions,
Quebec and Ontario were the only areas
with point estimates below two percent. The
Quebec value (1.1%; 95% CI: 0.8–1.4) was
nearly half the value of all other Canadian
regions—a statistically significant result.
Considering the relatively low prevalence of
FM in Quebec, the leading French-speaking
province in Canada, a more in-depth analysis according to language and province of
residence was performed. Figure 1 shows
the prevalence for francophone women
between the ages of 35 and 65, based on
whether they reside in Quebec or elsewhere.
The graphs show that francophone women
living outside of Quebec have a FM prevalence similar to the rest of the country
(Figure 1A), while those who live within
Quebec have a significantly decreased likelihood of reporting FM. Figure 1C shows that
there is no overall difference in FM prevalence between the English and French
speaking populations living outside of
Quebec.
Table 1 also lists the self-reported prevalence
for native-born Canadians and for immigrants. While the estimated prevalence for
men was similar between the two groups,
immigrant women appear to have a significantly lower prevalence (1.5%; 95% CI:
1.2–1.8) than native-born women (1.9%;
95% CI: 1.8–2.1). The self-reported prevalence of FM among immigrant women and
Canadian-born women was compared for
four different groups of women over the age
of 35 (Table 2). The data shows that the condition is less prevalent among immigrant
women in all three age groups under 65,
reaching statistical significance in the 45-to54 and 55-to-64 age groups. No difference in
prevalence rates was observed for the age
group representing those females over the
age of 65.
Socioeconomic results
To evaluate the link between socioeconomic
status and FM, we determined the prevalence of FM according to education and
Vol 27, No 1, 2006
TABLE 2
Prevalence and 95% confidence intervals of self-reported, health professionally
diagnosed FM in immigrant and Canadian-born women by age.
The Canadian Community Health Survey, Cycle 1.1 (2000)
income level in women between 35 and 65
(Table 3). Women in the lowest income
quartile were more likely (3.4%; 95% CI:
2.8–4.1) than women from the highest
(2.2%; 95% CI: 1.8–2.6) to report a diagnosis of FM. Similarly, men (no age restrictions) from the poorest households were
more likely (1.2%; 95% CI: 0.5–1.8) to report FM compared to the wealthiest group
(0.3%; 95% CI: 0.2–0.4). For education, the
only statistically significant difference
among the four groups was between women
who had completed a post-secondary education (2.0%; 95% CI: 1.8–2.3) and those who
had not completed their secondary school
studies (1.5%; 95% CI: 1.3–1.8). No such
trend was evident for men.
Age
group
< 25
5–34
35–44
Percentage
CI
2.03
Percentage
0.10–0.35b
0.18
0.87
Immigrant women
0.63–1.11
1.67–2.40
FM prevalence proportion
All languages
B
1.39–2.01
2.57
1.61–3.54 b
0.53
1.86
b
1.09
1.71
1.34–2.07
1.23–2.50
a
Insufficient observations to calculate the point prevalence; coefficient of variation is greater than 33.
b
Coefficient of variation is between 16 and 33.
C
Non-Quebec residents
2
1
0
Vol 27, No 1, 2006
0.61
2.20
4.04–5.58
3
Place of residence
0.46
b
3.02–4.18
Francophones
Quebec
0.45–1.91
4.81
4
Canada
0.93
N/A
b
3.60
5
Quebec
N/A
a
55–64
FIGURE 1
Comparison of FM prevlaence and 95% confidence intervals by language and
province of residence for women between the ages of 30 and 65.
The Canadian Community Health Survey, Cycle 1.1 (2000)
A
a
To consider the effect of smoking, respondents between the ages of 35 and 65 were
grouped into one of six categories, ranging
from daily smoker to having never smoked.
No difference in self-reporting was apparent
between any of the groups for men. However, the results suggested that women who
had never smoked were less likely (2.0%;
95% CI: 1.7-2.4) to report having FM when
compared to women with any level of smoking history (data not shown). While these
data were inconclusive for women with
moderate smoking histories, those who were
daily (3.3%; 95% CI: 2.8–3.8) or formerly
daily smokers (3.5%; 95% CI: 2.9–4.1) were
To examine the association between weight
and FM, the respondents were divided into
four groups according to body mass index
(BMI). The results for men show a similar
prevalence in all 4 BMI categories (Table 4).
In women, there was a clear trend towards
higher self-reported FM with rising levels of
BMI. Those with a BMI of greater than 30
were almost twice as likely to self-report
when compared to the group with a BMI less
than 24.
Ratio of immigrant to
Canadian born
CI
45–54
> 65
Behavioural determinants of
health
Canadian-born women
Canada
Place of residence
English
French
Primary language
12
clearly more likely to report suffering from
FM.
The relationship between frequency of alcohol consumption and FM prevalence was
analyzed by grouping respondents according
to drinking frequency. In women, the results
indicate that prevalence is lower among
those having never consumed alcohol
(1.6%; 95% CI: 1.0–2.0) when compared
with any of the other groupings (data not
shown). In addition, the data also suggest
that the prevalence of FM among both
women and men who currently consume
alcohol is lower when there is more frequent
consumption (regular consumption: 2.4%;
95% CI: 2.09–2.80 versus occasional consumption: 3.3%; 95% CI: 2.8–3.7).
Discussion
Employing self-reporting of major illness and
health events is the most practical method of
assessing disease status in large population
studies. Self-reporting of diagnosis has been
criticized by some because of misclassification concerns, resulting in potential under or overestimation of disease prevalence
and societal burden. However, numerous
studies assessing the agreement between
self- and physician-reported diagnoses have
demonstrated a satisfactory accuracy with
respect to both sensitivity and specificity for
the majority of disease states, including the
rheumatic conditions rheumatoid arthritis
and osteoarthritis.19,20 Currently, the only
Chronic Diseases in Canada
data that call into question the validity of
self-reporting for FM were published in a
study that used a combination of telephone
interviewing and physical exam screening to
estimate prevalence.8 The study’s authors
state that because only 30 percent of those
they ultimately classified as having FM were
previously aware of their condition, the true
FM prevalence is three times higher than
what is commonly estimated using selfreporting. Curiously, however, the study
fails to advocate for or even discuss the
importance of identifying this group of previously unrecognized FM sufferers. What
makes the absence of this discussion interesting is that the authors proceed to address
and answer this very question within the
same article. For example, when the health
status of the previously diagnosed group is
compared to that of the undiagnosed group,
the undiagnosed are shown to have significantly better self-perceived health and less
work disability than do the diagnosed. Furthermore, the authors found no deterioration
in perceived health over time in either
group.8 Considering that numerous other
studies have demonstrated FM to be a
chronic, non-degenerative, non-progressive
disorder lacking adequate treatment (and
preventative measures), the corollary would
be a low likelihood of those previously
undiagnosed to consult a medical practitioner about their condition and, ultimately,
receive an FM diagnosis. We would therefore be comfortable in stating that this evidently healthier group of individuals might
be missing from our estimate of prevalence.21–23 Moreover, if the prevalence is indeed higher than that predicted by the present work, as well as by other studies, further
research on FM and its effects becomes even
more important.
A number of the findings in this study confirm the results of work in other countries,
suggesting that the identified Canadian FM
population may be similar to those described
in other, often ACR-criteria-based, previous
studies. First, the estimated Canadian FM
prevalence rates of 0.3 percent for men and
1.8 percent for women are consistent with
the results from epidemiological studies conducted in the last several years.12–15 These
findings suggest that prevalence appears to
be similar across developed countries.
Second, the present work shows an increasing prevalence of FM in women up to until
late middle age, followed by a marked
decrease in the elderly population.13,15,24
The CCHS age-at-diagnosis data allowed for
some previously unrecognized and interesting observations to be made regarding the
natural history of the disease. Presently,
most of the FM literature describes FM as a
disorder predominantly affecting late-
TABLE 3
Prevalence and 95% confidence intervals of self-reported, health professionally
diagnosed FM in men and women according to income and education.
The Canadian Community Health Survey, Cycle 1.1 (2000)
Men
Percentage
Women
CI
Percentage
CI
Income
Lowest quartile income
1.15
0.51–1.80 a
3.43
2.75–4.11
Second quartile income
0.61
0.32–0.90 a
2.48
2.11–2.85
0.46
0.27–0.65
a
2.67
2.27–3.07
0.29
0.15–0.43
a
2.21
1.79–2.63
Less than secondary
0.34
0.21–0.48
1.53
1.30–1.77
Secondary graduate
0.27
0.12–0.41
1.93
1.59–2.27
Some post secondary
0.34
0.19–0.49
1.59
1.21–1.97
Post-secondary graduate
0.34
0.23–0.46
2.04
1.81–2.26
Third quartile income
Highest quartile income
Education
a
Coefficient of variation (CV) between 16 and 33
Chronic Diseases in Canada
13
middle-aged women. This conclusion is not
surprising given the almost five-fold difference in prevalence between the 30-to-34 and
the 55-to-59 age groups. However, CCHS
age-at-diagnosis data shows that this does
not mean that women under 35 are not
afflicted with the disease, as demonstrated
by the fact that almost 30 percent of cases
diagnosed in the five years preceding data
collection occurred in those under the age of
35. Unexpectedly, the same analysis showed
an almost negligible number of new FM
cases (< 6%) in those over the age of 60.
These findings show that FM is not just a
problem for those in their late middle and
senior years, but can afflict women of all
ages. From these data, it could be suggested
that it is the chronic, unremitting nature of
the disorder that leads to the high prevalence
in late middle age. More age-at-diagnosis
data should become available following the
completion of further CCHS cycles.
The present study describes significantly
lower self-reporting among the Quebec
sub-population. Previous studies of other
medical conditions have demonstrated that
prevalence can vary by Canadian region for
other conditions. For example, research on
both sinusitis and chronic pain revealed that
Quebec has the lowest Canadian rate for
each.25,26 The authors of these two studies
conclude that the observed differences are
best explained by environmental, rather
than cultural or genetic phenomena, due to
the finding that rates of chronic pain among
non-francophones in Quebec were the same
as those among Quebec francophones, while
rates for francophones outside Quebec
tended to be the same as those for nonfrancophones in the same province of residence.26 Given the parallel between FM and
chronic widespread pain, and the comparable findings that were noted for FM in this
study, we could draw similar conclusions.
However, it is less than clear what the previous studies’ authors imply by “environment”. Environmental phenomena, in this
research context, could correspond to either
geography or local culture. For example, it is
possible that francophones residing in other
areas of Canada, as well as anglophones
residing in Quebec, have been to some
degree assimilated into the local culture. For
this reason, although the results suggest
Vol 27, No 1, 2006
TABLE 4
Prevalence and 95% confidence intervals of self-reported, health professionally
diagnosed FM in men and women according to Body Mass Index (BMI)a.
The Canadian Community Health Survey, Cycle 1.1 (2000)
Men
BMI Grouping
Underweight BMI < 24
Average 24 < BMI < 27
Percentage
0.39
0.58
Women
CI
Percentage
CI
0.23–0.56
b
2.16
1.81–2.51
0.33–0.83
b
2.78
2.29–3.28
b
3.65
2.84–4.46
4.10
3.42–4.75
Overweight 27 < BMI < 30
0.47
0.17–0.76
Obese BMI > 30
0.55
0.35–0.76 b
a
Body Mass Index of an individual is calculated as the weight (kg) divided by the square of the
height (meters).
b
Coefficient of variation (CV) between 16 and 33.
geography as an important factor, the role of
cultural influences cannot be excluded.
Sub-group analyses showed that immigrant
women are less likely to report having FM.
Again, multiple potential explanations exist,
including decreased genetic susceptibility,
different geographical or cultural exposures,
and even the landing of relatively healthier
women screened by the immigrant health
examination. Interestingly, analysis of immigrants and Canadian-born women by age
demonstrated a potential convergence of FM
prevalence later in life, possibly following
years of exposure to the same and as yet
unidentified conditions as Canadian-born
women. These conditions might be geographical in nature, though one cannot disregard the gradual assimilation into local
cultures, as mentioned above, as an alternative explanation for the convergence.
Further complicating the question of the
roles of environment and geography in the
etiology of FM are our results showing no
difference in prevalence between urban and
rural respondents. One previous study, carried out in Pakistan, demonstrated higher
prevalence in rural areas for numerous rheumatic diseases, including FM.27 Here, the
observation was attributed to a socioeconomic effect, since more affluent urban areas
demonstrated prevalence rates lower than
those from underprivileged rural regions. In
Canada, it is possible that the lack of a difference between urban and rural area prevalence is due to comparable standards of
living between these two settings. Nonetheless, this finding of similar prevalence rates
Vol 27, No 1, 2006
in Canada is somewhat problematic as it
calls into question the often suggested role of
exposure to environmental pollutants, usually associated with urban living, in the etiology of FM.
Regarding our analyses of the socioeconomic factors of education and income,
the findings not surprisingly indicate that the
prevalence of FM declines with increasing
income, consistent with what has been
observed in other studies.27–29 It is interesting
to note that the prevalence of FM does not
appear to be inversely related to education,
despite the fact that education is usually
strongly correlated with increased income.
An attractive, but yet unproven explanation
could be that lower income is not a predisposing condition for FM, but rather a result
of developing the disorder. An additional,
less straightforward explanation for these
associations would be that high education
and low income represent markers for other
co-existing or correlating population characteristics, including emotional processes,
which could be more common among individuals with FM.
The results of the BMI, alcohol and smoking
investigation raise both some interesting
issues and present some unclear findings
(Table 4). To our knowledge, this study
demonstrates the first clear association
between BMI and FM. A number of potential
explanations for this association exist. First,
increasing weight could predispose an individual to developing FM. For example, obesity may lead to a relative hormonal
imbalance, similar to what occurs with
14
central obesity and glucose intolerance, predisposing to disease.30 Alternatively, reduced physical activity, not uncommon
among FM sufferers, may result in weight
gain. Alcohol and smoking have been linked
to the development of numerous disease entities.31,32 Despite the lack of a clear doseresponse relationship, the results of this
study suggest that those who abstain from
smoking and drinking are less likely to
report having FM. Moreover, the observed
paradoxical decrease in FM prevalence
among the regular alcohol consumption
group compared to those with more occasional consumption might be explained by
an aversion or low tolerance to alcohol. With
the exception of studies reporting more
musculoskeletal and chronic pain among
smokers, and more pain and functional disability in FM patients who smoke, our study
provides some of the first evidence suggesting an association between tobacco use and
FM.33–35 Finally, considering the potentially
higher stress and anxiety levels in individuals with FM, there is a possibility that the
observed relationships between FM and
drinking, smoking and overeating represent
coping mechanisms.
Conclusion
Large scale population studies on selfreported diseases can be used to answer
public health questions. In this study, we use
data from a large national health survey to
carry out a large-scale, Canadian-based,
descriptive epidemiological study on FM.
The CCHS’s large sample size and broad collection of descriptive variables allowed for
the analysis of a variety of sub-groups,
which was not possible in previous and
smaller North-American-based studies. The
heterogeneity of the respondents should
reduce biases intrinsic to studies carried out
on smaller homogenous populations, which
use diagnoses made by a discrete and often
limited number of researchers. Despite these
advantages, it must be recognized that prevalence values and associations based on
self-reported cross-sectional data show correlations without evidence of cause and
effect. For this reason, some of the findings
presented here require verification and further investigation. For example, our results
Chronic Diseases in Canada
note an association between various determinants of health, including smoking, body
mass index and FM. It is not known whether
these variables are risk factors, a result of the
condition, or are merely correlated with
other factors such as socioeconomic status.
In addition, further exploration of the difference between FM prevalence in Quebec and
that in the rest of Canada, and whether this
simply represents differences in diagnosis or
reporting would be important. If the associations identified in this study are determined
to represent true risk factors, it would open
the way for the development of preventative
health measures.
6.
Kaplan RM, Schmidt SM, Cronan TA. Quality of well being in patients with
fibromyalgia. J Rheumatol. 2000;27(3):
785–9.
7.
Henriksson C, Liedberg G. Factors of importance for work disability in women with
fibromyalgia. J Rheumatol. 2000;27(5):
1271–6.
8.
White KP, Nielson WR, Harth M, Ostbye T,
Speechley M. Does the label “fibromyalgia”
alter health status, function, and health service utilization? A prospective, within group
comparison in a community cohort of adults
with chronic widespread pain. Arthritis
Rheum. 2002;47(3):260–5.
9.
Acknowledgements
The authors wish to acknowledge the assistance of Statistics Canada, which not only
made available the data from the CCHS
(Cycle 1.1), but also provided the statistical
routines for the Rao-Wu bootstrap analysis.
The authors also wish to thank Loren Matheson for her thoughtful consideration of the
manuscript.
References
1.
Wolfe F, Smythe HA, Yunus MB, Bennett
RM, Bombardier C, Goldenberg DL, et al.
The American College of Rheumatology
1990 criteria for the classification of
fibromyalgia. Report of the multicenter
criteria committee. Arthritis Rheum.
1990;33(2):160–72.
2.
White KP, Harth M. Classification, epidemiology, and natural history of fibromyalgia.
Curr Pain Headache Rep. 2001;5(4):320–9.
3.
Tuzun EH, Albayrak G, Eker L, Sozay S,
Daskapan A. A comparison study of quality
of life in women with fibromyalgia and
myofascial pain syndrome. Disabil Rehabil.
2004;26(4):198–202.
4.
Noller V, Sprott H. Prospective epidemiological observations on the course of the disease
in fibromyalgia patients. J Negat Results
Biomed. 2003;2(1):4.
5.
Wolfe F, Hawley DJ. Measurement of the
quality of life in rheumatic disorders using
the EuroQol. Br J Rheumatol. 1997;36(7):
786–93.
Chronic Diseases in Canada
Wolfe F, Anderson J, Harkness D, Bennett
RM, Caro XJ, Goldenberg DL, et al. Work
and disability status of persons with
fibromyalgia. J Rheumatol. 1997;24(6):
1171–8.
10. Penrod JR, Bernatsky S, Adam V, Baron M,
Dayan N, Dobkin PL. Health services costs
and their determinants in women with
fibromyalgia. J Rheumatol. 2004;31(7):
1391–8.
11. White KP, Speechley M, Harth M, Ostbye T.
The London Fibromyalgia Epidemiology
Study: Direct health care costs of fibromyalgia syndrome in London, Canada. J
Rheumatol. 1999;26(4):885–9.
12. Picavet HS, Hazes JM. Prevalence of self
reported musculoskeletal diseases is high.
Ann Rheum Dis. 2003;62(7):644–50.
13. Carmona L, Ballina J, Gabriel R, Laffon A.
The burden of musculoskeletal diseases in
the general population of Spain: results from
a national survey. Ann Rheum Dis. 2001;
60(11):1040–5.
14. Wolfe F, Ross K, Anderson J, Russell IJ,
Hebert L. The prevalence and characteristics
of fibromyalgia in the general population.
Arthritis Rheum. 1995;38(1):19–28.
15. White KP, Speechley M, Harth M, Ostbye T.
The London Fibromyalgia Epidemiology
Study: the prevalence of fibromyalgia syndrome in London, Ontario. J Rheumatol.
1999;26(7):1570–6.
16. Beland Y. Statistics Canada health reports
2002;13(3):9–13.
17. Rao J, Wu C. Resampling interference with
complex survey data. J Am Stat Assoc.
1998;83:231–41.
15
18. Rao J, Wu C, Yue K. Some recent work on
resampling methods for complex surveys.
Surv Methodol. 1992;18:209–17.
19. Bergmann MM, Byers T, Freedman DS,
Mokdad A. Validity of self-reported diagnoses leading to hospitalization: a comparison
of self-reports with hospital records in a prospective study of American adults. Am J
Epidemiol. 1998;147(10):969–77.
20. Haapanen N, Miilunpalo S, Pasanen M, Oja
P, Vuori I. Agreement between questionnaire
data and medical records of chronic diseases
in middle aged and elderly Finnish men and
women. Am J Epidemiol. 1997;145(8):
762–9.
21. Baumgartner E, Finckh A, Cedraschi C,
Vischer TL. A six year prospective study of a
cohort of patients with fibromyalgia. Ann
Rheum Dis. 2002;61(7):644–5.
22. Mengshoel AM, Haugen M. Health status in
fibromyalgia: A follow-up study. J
Rheumatol. 2001;28(9):2085–9.
23. Wolfe F, Anderson J, Harkness D, Bennett
RM, Caro XJ, Goldenberg DL, et al. A prospective, longitudinal, multicenter study of
service utilization and costs in fibromyalgia.
Arthritis Rheum. 1997;40(9):1560–70.
24. Forseth KO, Gran JT. The prevalence of
fibromyalgia among women aged 20–49
years in Arendal, Norway. Scand J
Rheumatol. 1992;21(2):74–8.
25. Chen Y, Dales R, Lin M. The epidemiology of
chronic rhinosinusitis in Canadians. Laryngoscope. 2003;113(7):1199–205.
26. Wang PP, Badley EM. Consistent low prevalence of arthritis in Quebec: Findings from a
provincial variation study in Canada based
on several Canadian population health
surveys. J Rheumatol. 2003;30(1):126–31.
27. Farooqi A, Gibson T. Prevalence of the major
rheumatic disorders in the adult population
of north Pakistan. Br J Rheumatol. 1998;
37(5):491–5.
28. Reeder BA, Liu L, Horlick L. Sociodemographic variation in the prevalence of cardiovascular disease. Can J Cardiol. 1996;
12(3):271–7.
29. Valat JP, Goupille P, Vedere V. Low back
pain: Risk factors for chronicity. Rev Rhum
Engl Ed. 1997;64(3):189–94.
Vol 27, No 1, 2006
30. Vega GL. Obesity and the metabolic syndrome. Minerva Endocrinol. 2004;29(2):
47–54.
31. Bartecchi CE, MacKenzie TD, Schrier RW.
The human costs of tobacco use (1). N Engl J
Med. 1994;330(13):907–12.
32. Rehm J, Gmel G, Sempos CT, Trevisan M.
Alcohol related morbidity and mortality.
Alcohol Res Health. 2003;27(1):39–51.
Vol 27, No 1, 2006
33. Andersson H, Ejlertsson G, Leden I. Widespread musculoskeletal chronic pain associated with smoking. An epidemiological
study in a general rural population. Scand J
Rehabil Med. 1998;30(3):185–91.
34. Eriksen WB, Brage S, Bruusgaard D. Does
smoking aggravate musculoskeletal pain?
Scand J Rheumatol. 1997;26(1):49–54.
35. Yunus MB, Arslan S, Aldag JC. Relationship
between fibromyalgia features and smoking.
Scand J Rheumatol. 2002;31(5):301–5.
16
Chronic Diseases in Canada
A population-based analysis of health behaviours,
chronic diseases and associated costs
Arto Ohinmaa, Donald Schopflocher, Philip Jacobs, Sandor Demeter, Anderson Chuck, Kamran Golmohammadi
and Scott W Klarenbach
Abstract
Health behaviours influence the future incidence of certain common chronic diseases and thus
have an impact on health status and utilization of health care services and costs. We analyzed
person-level data of the Albertan adult population from the Canadian Community Health
Survey, Cycle 1.1 (2000) to determine health care costs associated with specific health
behaviours (smoking, sub-optimal diet, physical inactivity) and chronic disease states (heart
disease, diabetes, COPD). We found that 74.7 percent of the population exhibited one or more
risk behaviours, while 10.5 percent had one or more of the chronic diseases of interest. Greater
health care utilization and costs were noted in groups exhibiting risk behaviour and chronic
disease states. Approximately 31 percent of health care costs in Alberta were attributable to
people having one or more of the three chronic diseases. Our findings of higher health care
costs incurred by those exhibiting unhealthy behaviour prior to development of disease, as
well as by those with multiple co-existent diseases, are important indicators to guide future
prevention and treatment strategies of chronic illness.
Key words: Canada, chronic illnesses, health behaviour, health economics, health survey,
population surveillance, risk behaviour, WHO
Introduction
1
A recent World Health Organization (WHO)
document has called for a unified and global
strategy towards the prevention of specific
chronic diseases, namely chronic obstructive
pulmonary disease (COPD), diabetes
mellitus, heart disease, and lung and
colorectal cancer. The development of these
diseases has been linked to a common set of
risk behaviours (tobacco use, sub-optimal
nutrition and diet, and inadequate physical
inactivity), and they are therefore preventable to some degree. Their prevalence is rapidly increasing, and they have been
recognized as incurring a significant economic cost for society.2,3 While some investigators have conducted detailed costing
studies of specific chronic conditions4–6 and
risk factors,7,8 the WHO vision indicates that
we need a more comprehensive view of
“disease” costs. Such a vision would incorporate a wide spectrum of the population,
including not only those with the disorders
of interest, but also those at risk of future development of the disease. Further, it is increasingly recognized that diagnoses which
occur in combinations will have cumulative
impacts on costs,9–11 and thus it may not be
appropriate to focus only on one disease
entity.
Currently, only blunt conceptual tools are
available to deal with global or populationlevel resource issues. In Canada, as in many
other countries, a “top down” methodology
(or collective approach) to study health care
costs has been developed by Health Canada,2 which is based on service provider information, not on information obtained from
individuals comprising the population of interest. The Health Canada approach omits
several important but as yet undeveloped areas where global burden analysis needs to be
extended, including the measurement of
out-of-pocket costs and the analysis of risk
factors and disease co-morbidities. The relationship between health care costs and personal risk factors, in particular, cannot be
addressed at the population level using previously employed top-down methodologies.
Population health surveys are instruments
that can potentially be harnessed to explore
these important issues, though they have not
yet been exploited to conduct populationlevel economic analyses.
The purpose of this paper is to estimate the
cost of health services for adults in Alberta
from a population-based perspective using
individual-level data, with specific inquiry
into the burden attributable across a broad
spectrum of the population and using the
WHO framework. This spectrum ranges
from those with no high-risk behaviours to
those with one or more risk behaviours, and
then includes those with the chronic disease
of interest, including single and multiple
chronic illnesses.
Method
Our analysis entails the identification of individuals with risk behaviours and disease
Author References
Arto Ohinmaa, Philip Jacobs, Anderson Chuck, Department of Public Health Sciences, University of Alberta, Edmonton, Alberta, Canada
Donald Schopflocher, Health Surveillance, Alberta Health & Wellness, Government of Alberta, Canada
Sandor Demeter, Section of Nuclear Medicine, University of Manitoba, Winnipeg, Manitoba, Canada
Kamran Golmohammadi, Department of Radiology and Diagnostic Imaging, University of Alberta, Edmonton, Alberta, Canada
Scott W Klarenbach, Department of Medecine, University of Alberta, Edmonton, Alberta, Canada
Correspondence: Scott W Klarenbach, 11-107 Clinical Sciences Building, 8440-112 Street, University of Alberta, Edmonton, Alberta, Canada T6G 2G3;
fax: (780) 407-7878; e-mail: [email protected]
Chronic Diseases in Canada
17
Vol 27, No 1, 2006
states, quantifying per-person units of health
care resource utilization, and applying ageand-gender-specific costs to each unit of resources to enable determination of person-level costs. The population of interest is
the non-First Nation (including off-reserve
FN), non-institutionalized population of
Alberta aged 20 and over in budget year
2000/1. Person-level data from the 2000–
2001 Canadian Community Health Survey
(CCHS), Cycle 1.1, a household survey of
non-institutionalized persons in the general
population conducted by Statistics Canada,
was used for the analysis.12 Respondents to
this survey answered a variety of questions
dealing with personal and family characteristics, health status (including mental health
and presence of chronic conditions), and
health care service utilization. This survey
utilizes a multistage stratified cluster design
and provides cross-sectional data representative of 98 percent of the Canadian population over the age of 12. It attained an 80
percent overall response rate.13
The population was sub-divided into groups
based on the presence (or absence) of COPD,
heart disease and diabetes, either in isolation
or combination. Presence of disease states
were obtained from individual responses to
questions which asked whether the persons
had been diagnosed with COPD, heart disease or diabetes by a health professional.
The presence or absence of other diseases,
such as arthritis or cancer, may have
occurred but were not incorporated into the
analysis.
Persons without any of the three reported
chronic diseases were classified by risk
behaviour categories: no risk behaviour, or
one or any combination of smoking, inadequate nutrition and physical inactivity.
Smokers were defined as those who indicated that they smoked daily or occasionally
in response to the question “At the present
time, do you smoke cigarettes daily, occasionally or not at all?” “Adequate nutrition”
was defined as consumption of five or more
servings of fruit and vegetables daily (the
current Health Canada standard), which was
derived from a series of questions within the
survey. “Physical activity” was defined by
the Physical Activity Index,14,15 which is
derived from several questions in the survey
pertaining to physical activity. Those classified as “inactive” were defined as having
sub-optimal physical activity (as opposed to
those classified as “moderate” or “active”).
The frequencies of respondents for the risk
factors and chronic disease were adjusted
with population-based weights to obtain
population estimates. The distribution of
health states among valid cases was used to
redistribute a health state to those with missing information. Responses to service utilization questions provided subject-level data
on the number of days in hospital, and the
number of visits to family doctors and specialists over a one-year period. Per-person
units of health care resource utilization was
based on responses of valid cases in the
CCHS. The utilization rates of multiple
chronic diseases were estimated from the
Canadian sample, due to the small number
of people in these groups in Alberta.
Unit costs were developed for the aforementioned services using Alberta cost data. The
average cost of a typical hospital day was
TABLE 1
Distribution and prevalence of risk behaviours and chronic disease by age group in Alberta (CCHS 2000–01)
Age groups
20 – 45
a
Risk and disease groups
b
Population
45–64
b
65+
Percent
Population
Percent
b
Population
All ages
b
Percent
Population
Percent
No risk behaviors
183,001
15.30
93,542
14.33
40,898
14.34
315,036
14.77
One risk behavior
371,491
31.07
193,550
29.66
82,060
28.78
642,976
30.14
>1 risk behavior
609,657
50.99
296,029
45.36
73,595
25.81
951,454
44.60
12,113
1.01
24,847
3.81
40,416
14.17
91,666
4.30
Heart disease only
COPD only
1,433
0.12
4,970
0.76
8,255
2.90
17,764
0.83
17,249
1.44
32,201
4.93
27,742
9.73
89,268
4.18
Heart and diabetes
128
0.01
5,446
0.83
8,068
2.83
17,103
0.80
Heart and COPD
Diabetes only
651
0.05
1,208
0.19
2,355
0.83
4,991
0.23
COPD and diabetes
0
0.00
237
0.04
1,172
0.41
1,765
0.08
All 3 conditions
0
0.00
539
0.08
565
0.20
1,391
0.07
1,195,722
100.00
652,567
100.00
285,124
100.00
2,133,413
100.00
Total
a
Chronic health conditions other than those specified may also coexist, and include asthma, fibromyalgia, arthritis, back problems, high blood pressure,
migraine headaches, epilepsy, cancer, stomach or intestinal ulcers, urinary incontinence, inflammatory bowel disease, dementia, glaucoma, cataracts,
thyroid disease, neurologic disease including stroke, chronic fatigue syndrome, food allergies and multiple chemical sensitivities.
b
The final population was derived by using the distribution of health states among valid cases to redistribute the number of missing cases to a corresponding
health state.
Vol 27, No 1, 2006
18
Chronic Diseases in Canada
obtained from Alberta Health and Wellness
(AHW), measured as the total inpatient
facility) cost divided by total inpatient days,
as estimated from the provincial Management Information System (MIS) data base
for the budget year 2000–01.16–18 A
weighted-average, province-wide, costperdiem statistic ($780) was obtained from
all hospitals.
Since physicians bill the provincial payment
plan for services and procedures, the data
from AHW was used to calculate physicianassociated costs. We calculated an average
physician-billing per day of hospitalization
according to patient age (20–44, 45–64,
65+). This per diem fee was added to the
daily hospital facility cost for each recorded
day of care. We estimated total general practitioner (GP) office billings per visit for the
province by age group. The cost of diagnostic services attached to GP visits was added
to this statistic, so that the total GP cost included examination and diagnostic costs.18
The cost of a specialist visit was also calculated by patient age. Specialist visits were
divided into two groups: those which were
made in specialists’ offices and those which
were made in hospital outpatient clinics. For
the office visits, we calculated an average
cost and added the costs per visit for diagnostic services. For outpatient hospital
visits, we added a hospital outpatient facility
fee based on the province-wide Alberta
Ambulatory Care Classification System cost
per visit (adjusted for age) to the physician
fee to obtain a total cost per outpatient
visit.18
We added to the CCHS estimates the costs of
those who died during 2000, as these persons would not appear in the CCHS, yet
would have received services. We estimated
the age-specific health care cost of deaths for
all persons who died in 2000, including persons with one of the three chronic conditions
as the major diagnosis for death according to
Alberta mortality statistics,19 which we
valued using the last six months of life cost
(average lifetime during the budget year) in
Manitoba.20
As the CCHS is based on self-reported utilization data, it may be subject to errors of
recall. In order to determine the degree of
Chronic Diseases in Canada
error in our estimates and establish face
validity, we compared the population-level
costs for physician and hospital inpatient
and outpatient care in Alberta, as documented by AHW budget data, to the estimated results using our methodology, and
with respect to the population over 20 years
of age.
Our analysis had three components. In the
first component, we estimated the number
of persons in each health status group,
which include disease states and risk behaviour. In the second component, we estimated the hospital and physician utilization
and cost per person by age category and
health status. Finally, with the third component, we calculated global health care system-wide costs by health status, inclusive of
mortality cases.
Results
The total non-First Nations population in
Alberta aged 20 and over according to the
CCHS analysis was 2.13 million. The breakdown of persons by group is shown in Table
1. Approximately 15 percent of the population in each age group exhibited none of the
specific risk behaviours, while 75 percent of
the population had one or more of them
present. Among those without chronic disease, the proportion of subjects exhibiting
risk behaviours decreased with age from
82.1 percent (20–45 years) to 54.6 percent
(65+). Those with one or more of the three
chronic conditions comprised 10.5 percent
of the estimated Albertan population, with
disease prevalence rising with increased age.
The unit costs for physician visits and hospital days for the three age groups in Alberta
are shown in Table 2. With the exception of
family physician visits, fewer resources are
used per visit or hospitalization day with
increasing patient age, reflecting that fewer
investigations and procedures are performed
as age increases for any single day. (Note
that hospital stays typically are longer
among older persons).
The health care utilization statistics are
shown in Table 3. There is a trend for higher
health care utilization rates when moving
across the risk behaviour spectrum in every
age group. Furthermore, the number of
family physician visits is approximately double for those subjects identified with a
chronic condition of interest compared to
those not exhibiting risk behaviours. The
number of physician visits is substantially
higher when more than one chronic disease
is present, especially in the youngest age
group. Hospitalizations increase very rapidly
when moving from the no-risk behaviour
population to the multiple chronic diseases
category. The number of specialist visits
increases gradually by age; however, the
variation in the visits by health status is not
large within each age group. The increasing
trend of health care utilization by increased
FIGURE 1
Average annual total health care costs per person by age in different risk
behaviour and chronic disease groups in Alberta, 2000–01
$5,000
$4,000
20 - 44
$3,000
45 - 64
65 +
$2,000
$1,000
$0
No risk
More than Only heart
1 risk
risk
disease
behaviour
behaviour
19
Only
COPD
Only
diabetes
More than 1
chronic
condition
Vol 27, No 1, 2006
FIGURE 2
Total health care cost in different risk behaviour and combined disease groups
among three age groups of adult Albertans, 2000–01 (millions of dollars)
Millions
$400
$350
$300
$250
20 - 44
45 - 64
65 +
$200
$150
$50
$0
1 risk behaviour
risk behaviour and number of chronic diseases was statistically significant (95 percent
CI), with a few exceptions—mainly in the
specialist visit category (Table 3).
The average annual cost per person by
health status is shown in Figure 1. Similar to
utilization data, total annual costs increase
through the risk behaviour spectrum: from
no-risk behaviour to some risk behaviours,
and from one chronic disease to more than
one coexisting chronic illness. In those with
a single disease of interest, subjects with diabetes alone incurred the lowest costs. Costs
increased with increasing age for heart disease and COPD, but not diabetes mellitus
alone. Examples of the increment in costs for
groups with disease compared to the no-risk
groups include a six-fold increase for 45 to
64 years olds and a 7.5 fold increase for
those 65 years and over with heart disease.
For diabetes, this increase from no risk is
approximately four times for the two youngest age groups and three times for the
oldest.
The total annual cost in Alberta of the identified services on a population basis was $1.49
billion (excluding the cost of deaths during
the year) and was distributed by health
status and age as per Figure 2. Approximately 7.8 percent of health care costs in this
population were incurred by those without
the aforementioned risk behaviours or disease states (14.8 percent of population), with
Vol 27, No 1, 2006
More than 1 risk
behaviour
Mortality-related costs in Alberta were $187
million. Of these costs about 42 percent were
attributable to heart disease, two percent to
diabetes and five percent to COPD. The
projected cost of death in 2000–01 would
increase the individual-based health care
cost estimate by 12.6 percent to $1.68
billion.
In validating this costing method, we used as
the gold standard the total budgeted cost
estimates for hospital and physician services
in Alberta. The AHW health care budget
data for the adult population showed $2.06
billion, resulting in a difference of 18 percent between the two estimates.
$100
No risk
is responsible for a substantial burden of disease, while COPD has a smaller impact on
total health care costs, especially in younger
age groups.
1 or more chronic
conditions
relatively more costs incurred by younger
age groups. The population that exhibited
one or more risk behaviours were responsible for 61.1 percent of the health care costs,
although they comprise 74.7 percent of the
population. Persons with the three chronic
diseases alone or in any combination
accounted for 31.1 percent of total health
care costs, although they comprise 10.5 percent of the adult population.
Health care costs associated with the presence of chronic disease is presented in Figure
3. While costs associated with heart disease
alone represents about 14.2 percent of all
health care costs, the prevalence is 4.3 percent. Similarly, diabetes alone (4.2 percent
of the population) or in combination with
heart disease (0.8 percent of the population)
Discussion
Combining the person-level CCHS risk
behaviour and utilization data with unit cost
data from Alberta for health care services,
we estimated the Alberta adult population
health care costs, including costs for those
exhibiting specific risk behaviour characteristics and those with chronic diseases of
interest. As the CCHS contains a high degree
of detail with respect to risk behaviours and
personal characteristics, our analysis sheds
light on system-wide economic issues
related to health risk behaviour and chronic
disease. Our results indicate that per-person
incremental costs rise within the spectrum of
risk behaviours prior to the development of
TABLE 2
Average unit costs for family physician visit, specialist visit and hospitalization
day by age category in Alberta, 2000–01
Family physician
a
visit
Age
Specialist
b
visit
Hospitalization
c
day
20 – 44
$32.11
$122.29
$918.78
45 – 64
$36.98
$114.20
$884.42
65 +
$33.63
$92.40
$826.80
a
Source: Alberta Health Care Insurance Plan Payment database (AHW) for physician visits, and
provincial fee schedule for laboratory services and diagnostic radiology.
b
Source: Alberta Health Care Insurance Plan Payment database (AHW) for physician visits, Alberta
Ambulatory Care Classification System outpatient facility fee (AHW) where applicable, and provincial
fee schedule for laboratory services and diagnostic radiology.
c
Source: Inpatient Database (AHW) and Management Information System data from Alberta.
20
Chronic Diseases in Canada
Chronic Diseases in Canada
21
Vol 27, No 1, 2006
4.15
6.52
8.06
Only heart disease
Only COPD
Only diabetes
7.85
5.63
5.93
Only heart disease
Only COPD
Only diabetes
4.15
4.44
6.86
8.10
5.76
8.00
One risk behaviour
More than 1 risk behaviour
Only heart disease
Only COPD
Only diabetes
More than one diseasec
6.406
5.866
8.391
6.035
5.267
4.922
3.450
8.188
5.012
4.728
7.420
5.446
4.028
3.959
10.473
7.926
3.843
5.264
5.243
4.959
3.874
SD
7.97–8.03
5.69–5.83
7.91–8.29
6.80–6.93
4.40–4.48
4.15–4.19
3.33–3.40
9.06–9.17
5.87–5.99
5.50–5.77
7.76–7.95
3.60–3.64
3.21–3.24
3.08–3.13
11.68–12.44
7.90–8.22
6.26–6.78
4.03–4.28
3.51–3.54
3.18–3.22
2.84–2.88
95% CI
4.25–4.33
1.27–1.39
4.98–5.39
3.74–3.92
1.09–1.15
0.93–0.99
0.34–0.39
4.17–4.28
1.35–1.44
3.25–3.65
2.27–2.42
0.31–0.33
0.24–0.25
0.19–0.22
4.96–5.33
1.00–1.14
NA
0.26–0.35
0.42–0.44
0.29–0.31
0.18–0.20
95% CI
Means are based on the responses from the entire Canadian population due to irregularities in the utilization of Albertans with more than one of the specified diseases.
8.292
4.796
8.943
8.650
3.790
3.835
2.279
8.593
3.958
6.811
5.614
2.191
1.602
1.730
8.781
3.500
0.000
2.005
2.564
1.927
1.495
SD
Excluding long-term chronic care facility stay
4.29
1.33
5.19
3.83
1.12
0.96
0.37
4.22
1.40
3.45
2.34
0.32
0.24
0.21
5.15
1.07
0.00
0.30
0.43
0.30
0.19
Mean
c
1.92–1.94
0.59–0.62
1.85–1.96
1.33–1.37
0.61–0.64
0.67–0.70
0.48–0.50
2.52–2.57
1.10–1.15
1.47–1.65
1.40–1.46
0.56–0.57
0.66–0.68
0.84–0.87
4.10–4.26
1.45–1.57
0.61–0.79
0.96–1.07
0.62–0.63
0.58–0.59
0.53–0.54
95% CI
Including outpatient hospital visits
2.903
1.168
2.378
2.089
1,863
1.583
1.093
3.250
2.246
3.109
2.459
1.744
1.625
2.028
3.890
2.995
1.269
2.239
1.931
1.721
1.558
SD
b
Hospitalization days
b
1.93
0.61
1.90
1.35
0.62
0.69
0.49
2.55
1.13
1.56
1.43
0.56
0.67
0.85
4.18
1.51
0.70
1.02
0.63
0.58
0.53
Mean
a
Specialist visits
a
Note: Means are based on the responses of valid cases (i.e. non-missing cases).
3.37
No risk behaviour
Age group 65+
9.11
3.62
More than 1 risk behavior
More than one disease
3.22
One risk behaviour
c
3.11
No risk behaviour
Age group 45-64
11.89
3.52
More than 1 risk behaviour
c
3.20
More than one disease
2.86
One risk behavior
Mean
No risk behavior
Age group 20-44
Health status
Family physician visits
TABLE 3
Average (per capita) health care utilization by age in different risk behaviour and chronic disease groups in Alberta, 2000–01
the specified chronic diseases, a finding
which to our knowledge has not previously
been reported on a population basis. Furthermore, costs increase markedly when
chronic disease occurs, especially so for
those with multiple co-existing diseases.
The WHO has predicted a shift in the prevalence of chronic diseases because of the
widespread prevalence of risk behaviours.1
Our results suggest that an increase in the
disease burden will have substantial
economic consequences, because of both
the sheer number of persons currently exhibiting unhealthy behaviours who are at risk of
developing the diseases, and the significant
cost implications of developing a chronic
disease. While the number of persons who
have chronic diseases is relatively small at
present, such persons are very costly, especially so in younger age groups. The number
of persons without chronic disease who
exhibit high risk behaviours is very large—
especially in the younger age groups though
they are not yet costly (although clearly
more costly than their no-risk counterparts).
As such, the potential for large increases in
economic burden due to chronic disease is
substantial, although the proportion of subjects who will go on to develop chronic disease and at what age the disease will
manifest are not known.
While the incremental per-person costs for
those exhibiting unhealthy behaviours
(without the chronic diseases in question)
are relatively small, the large number of persons accounts for a large fraction of health
care costs. While this health care use may be
advantageous if it is addressing risk factor
modification, the relatively low health care
expenditure on prevention in Alberta suggests that this may not be the case.21 The
confluence of multiple risk factors may lead
to opportunities to provide prevention strategies efficiently. The nature of the health care
services used needs to be clarified in future
studies, and opportunities for multiple
simultaneous risk factor modification (similar to those provided in disease management
clinics), as well as improved efficiency of
health care resource use, should be promulgated.
Several limitations merit specific mention.
The scope of risk behaviours considered was
limited to smoking, nutrition and physical
activities, and while the WHO has identified
these as major modifiable risks, other
genetic, environmental, and person-level
risks were not included in this analysis. In
addition, the self-reported nature of risk
behaviours and disease states may lead to
systematic error, although evidence suggests
this error is likely to be small.15,22
Our utilization analysis is based on selfreported data, and investigators have questioned whether , as such, is subject to a recall
bias.23,24 We conducted a validity check to
determine the degree of correspondence
between costs as estimated by our method,
and as those reported through provincial
expenditure data, and found a difference of
18 percent. There may be several reasons
for tthis differences. For example, First
Nations persons who live on reserves were
not captured. This population has an increased prevalence of chronic diseases including diabetes and respiratory disease, and
incurs higher health care utilization than a
matched population.25 Approximately 3.8
percent of the Alberta population in 2000
was of First Nations status,25 and it is estimated that 60 percent live on reserves.26
Additionally, institutionalized adults are also
excluded from the CCHS. Approximately
five percent of Canadians reside in nursing
homes. The vast majority of these are over
the age of 65 years and on average consume
more health care resources.27 Also, persons
of very low socioeconomic status who may
utilize several times more health care
resources than the general population may
not be captured in population health
surveys. When accounting for this incomplete capture of approximately seven percent
of the population who are very likely to
exhibit greater-than-average health care utilization, the range of error is not wide. This
adds confidence and face validity to our
estimates based on individual-level data,
and establishes this methodology as a credible approach to population-based costing.
The estimation of the total population was
done by multiplying the age- and risk/
disease-category-specific rate in the CCHS by
the Statistics Canada population estimate.
The validity of this estimate depends on the
accuracy of the sampling and the prevalence
of the condition or behaviour. The sampling
methodology of the CCHS has been demonstrated to be very accurate in representing
the Canadian population characteristics,28
and as such even a relatively small prevalence of some diseases in the lowest age
group (20–44) is likely to lead to accurate
population estimates at a global level. However, the small numbers of subjects with
disease in younger age groups may increase
the uncertainty of the cost estimates to a
FIGURE 3
Total health care cost in different chronic disease groups, and in their
combinations, among three adult-Albertan age groups, 2000–01
(millions of dollars)
Millions
$160
$140
$120
$100
20 - 44
45 - 64
65 +
$80
$60
$40
$20
$0
Only heart disease
Vol 27, No 1, 2006
22
Only COPD
Only diabetes
More than 1
chronic condition
Chronic Diseases in Canada
certain amount, mainly in the group with
multiple chronic conditions. To minimize
this risk of small numbers, we used
Canada-wide health care utilization estimates in all multiple chronic disease groups.
Lastly, disease classification errors in causeof-death reporting may also lead to some
inaccuracies.
One of the benefits of our approach has been
the ability to generate relatively complete
costs for each of the components of utilization that we studied. Doctors’ office visits
contained both direct fees and diagnostic
costs. Doctors’ hospital outpatient (including
emergency room) visits considered doctors’
fees and facility costs. Hospital inpatient
stays calculated the doctor and facility components, including overheads relating to
administrative, diagnostic and support services. The comprehensiveness of our unitcost measure is a partial explanation for the
correspondence between, on one hand,
costs as we calculated them, and provincially budgeted expenditures on the other.
Several components of care are not attainable with our method. The most obvious
omission is outpatient drug costs, which
cannot be estimated from CCHS. We were
able to estimate from Alberta administrative
data only the prescription drug costs of diabetes, heart disease and COPD for population over 65 years of age ($134 million).
Most of these costs were related to heart disease drugs (85.5 percent), and a much
smaller proportion to diabetes (9.1 percent)
and COPD (5.4 percent). We also omitted
home care because it is vaguely reported and
its current economic impact is of small magnitude. Our analysis also does not include
the indirect costs associated with lost productivity caused by disability and mortality.
Indirect costs are usually included in burden
of disease calculations, although methodologically they are a controversial topic due
to difficulties in accurately defining and
measuring the “‘opportunity cost” of future
lost work.29,30 In addition, CCHS determines
the long-term disability for only the
12-month period prior to the interview, thus
making the estimation of future or past disability/lost-productivity costs difficult.
Chronic Diseases in Canada
We have demonstrated and described in
detail the gradient of increasing health care
costs across the risk behaviour and chronic
disease spectrum using a framework advocated by the World Health Organization. Our
analyses indicate that person-level data from
large, population-based health surveys can
be used to accurately estimate bottom-up
global health care costs, thus offering new
possibilities to examine the impact on
resources and costs of demographics, risk
behaviours, and major chronic diseases in
isolation or in combination. This information can also be used to determine the size
and characteristics of the target populations
of preventive interventions. Our findings
using this approach demonstrate increased
resource utilization by those who exhibit
risk factors but who have not yet developed
the diseases of interest, as well as by those
with multiple co-existing chronic diseases of
interest. This may have important implications for identification of persons exhibiting
risk behaviours, since modification of their
behaviours may provide an opportunity to
attenuate resource utilization before chronic
disease sets in.
Acknowledgements
We would like to thank Health Canada and
Alberta Health and Wellness for their
financial support of this study.
References
1.
World Health Organization. Global strategy
for the prevention and control of noncommunicable disease. Geneva, Switzerland:
World Health Organization, A53/14;3–
22–2000.
5.
Chan B, Coyte P, Heick C. Economic impact
of cardiovascular disease in Canada. Can J
Cardiol Cardiol. 1996;12(10):1000–6.
6.
National Institute of Cancer of Canada.
Canadian cancer statistics 2004. Toronto,
Ontario: Canadian Cancer Society; 2004.
7.
Kaiserman M. The cost of smoking in
Canada, 1991. Chronic Diseases in Canada
1997;18(1):1–12.
8.
Katzmarzyk PT, Gledhill N, Shephard RJ.
The economic burden of physical inactivity
in Canada. CMAJ. 2000;163(11):1435–40.
9.
Simpson SH, Corabian P, Jacobs P, et al. The
cost of major co-morbidity in people with
diabetes mellitus. CMAJ. 2003;168(13):
1661–7.
10. Ramsey SD, Newton K, Blough D, et al.
Patient-level estimates of the cost of complications in diabetes in a managed-care population. Pharmacoeconomics. 1999;16(3):2
85–95.
11. Rapoport J, Jacobs P et al. (2004). Refining
the measurement of the economic burden of
chronic diseases in Canada. Chronic Dis Can.
2004;25(1):13–21.
12. Canadian Community Health Survey 2000–
2001. [Database] Ottawa: Statistics Canada,
2003.
13. Statistics Canada. Canadian Community
Health Survey Methodological Overview.
Health Reports 2002 Mar;13(3):9–14.
14. Stephens TM, Craig CL. The well-being of
Canadians: Highlights of the 1988 Campbell’s Survey. Ottawa, ON: Canadia Fitness
and Lifestyle Research Institute, 1993;
91–118.
15. Craig C, Russell S, and Cameron C. Reliability and validity of Canada’s Physical Activity
Monitor for assessing trends. Med Sci Sports
Exerc. 2002 Sep;34(9):1462–7.
2.
Health Canada. Economic burden of illness
in Canada, 1998. Ottawa, Ontario: Population and Public Health Branch, Health
Canada; 2002.
16. Jacobs P. A National list of provincial costs
for health care: Canada 1997/8. Institute of
Health Economics Working Paper, 2000
3.
Colman R. The cost of chronic disease in
Nova Scotia. Halifax: Nova Scotia, GPI
Atlantic; 2002.
17. Jacobs P, Roos NP. Standard cost lists for
health care in Canada. Issues in validity and
inter-provincial consolidation. Pharmacoeconomics. 1999;15:551–60.
4.
Dawson KG, Gomes D, Gerstein H, et al. The
economic cost of diabetes in Canada, 1998.
Diabetes Care. 2002;25(8):1303–7.
23
Vol 27, No 1, 2006
18. Philip Jacobs P, Scott Klarenbach S, Arto
Ohinmaa A, Kamran Golmohammadi K,
Sandor Demeter S, Donald Schopflocher D.
Chronic diseases in Alberta: Cost of treatment and investment in prevention. Alberta
Healthy Living Network. www.health-inaction.org/library/pdf/AHLN/whatsnew/
Chronic_Diseases_in_AB.pdf
19. Alberta Government Services. Alberta Vital
Statistics. Annual review 2000. Edmonton:
Alberta, Alberta Government Services; 2001.
20. Menec V, Lix L, Steinbach C et al. Patterns of
health care use and cost at the end of life.
Winnipeg: Manitoba Centre for Health Policy; February 2004.
21. Jacobs P, Ohinmaa A, Golmohammadi K,
Demeter S, Schopflocher D, Klarenbach S.
Public investment in providing information
for chronic disease prevention for adults in
Alberta, 2003: A cross-sector analysis.
Canadian Journal of Public Health. 2005 (in
press)
22. Haapanen N, Miilunpalo S, Pasanen M, Oja
P, Vuori I: Agreement between questionnaire
data and medical records of chronic diseases
in middle-aged and elderly Finnish men and
women. Am J Epidemiol. 1997;145:762–9
23. Harlow S, Linet M. Agreement between
questionnaire data and medical records. The
evidence for accuracy of recall. Am J
Epidemiol. 1989;129(2):233–48.
25. Cardinal J, Schopflocher D, Svenson L,
Morrison K, Laing L. First Nations in Alberta:
A focus on health service use. Edmonton:
AB, Alberta Health and Wellness, 2004
Available at: http://www.health.gov.ab.ca/
resources/publications/index.html#Nations
Accessed November 15, 2005
26. Alberta: First Nations Population – Summary
(December, 2002). AB, Alberta Aboriginal
Affairs and Northern Development, June
2003
27. Donna Dryden D. Health services utilization
in the population aged 65 and older: Review
of the literature. Alberta Centre for Health
Services Utilization Research. March 1999.
Available at: http://www.health.gov.ab.ca/
key/research/01_summary.PDF (accessed
Nov 24, 2005)
28. Canadian Community Health Survey (CCHS)
[Statistics Canada Web site]. December 18,
2003. Available at: http://www.statcan.ca/
english/concepts/health/. Accessed June 24,
2004.
29. Koopmanschap M, Rutten F. Indirect costs in
economic studies: Confronting the confusion.
Pharmacoeconomics. 1993;4(6):446–54.
30. Oostenbrink J, Koopmanschap M, Rutten F.
Standardisation of costs: The Dutch manual
for costing in economic evaluations.
Pharmacoeconomics. 2002;20(7):443–54.
24. Linet M, Harlow S, McLaughlin J, McCaffrey
L. A comparison of interview data and medical records for previous medical conditions
and surgery. J Clin Epidemiol. 1989;42(12):
1207–13.
Vol 27, No 1, 2006
24
Chronic Diseases in Canada
Multiple exposures to smoking, alcohol, physical
inactivity and overweight: Prevalences according to
the Canadian Community Health Survey Cycle 1.1
Julia E Klein-Geltink, Bernard CK Choi and Richard N Fry
Abstract
The objective of this study was to calculate the prevalence of multiple exposures to four
modifiable risk factors (smoking, alcohol, physical inactivity and overweight) and to establish
whether there are more Canadians with multiple risk factor exposures than those with
singular ones. Weighted estimates of the prevalence of mutually exclusive risk factor clusters
were calculated according to the Canadian Community Health Survey, Cycle 1.1 (2000).
Confidence limits were estimated by bootstrap techniques. Findings indicate that 21.0 percent
of Canadians have no risk factor exposures, 53.5 percent are physically inactive, 21.5 percent
currently smoke, 44.8 percent are overweight, and 6.0 percent are high-risk drinkers.
Compared to females, males are less physically inactive but more likely to smoke, have high
alcohol intake and be overweight, across all age groups. At least one risk factor was present in
79.0 percent of Canadians and 39.0 percent have at least two coexistent exposures. The
distribution of risk factor prevalences differed significantly by age, most peaking among those
between age 35 and 64, with the exception of physical inactivity. Those who smoke and are
physically inactive account for the highest proportion of the population with two or more
coexistent risk factors. Canadians who are free of the four risk factors for chronic disease
examined in this paper constitute the minority. Future studies are recommended to examine
other risk factors, as well as interactions of multiple exposures in association with chronic
disease.
Key words: chronic diseases, epidemiology, multiple exposures, prevalence
Introduction
Chronic illness represents a major disease
burden to society and is to a large extent preventable.1–3 The major chronic diseases
causing death in Canada are cardiovascular
disease (CVD), cancer, chronic respiratory
disease (CRD) and diabetes.4 Several of
these diseases share common preventable
risk factors, including smoking, high alcohol
intake, physical inactivity and overweight.1,4–8 It is incumbent upon public
health professionals to determine if—and
potentially to what extent—unhealthy behaviours can be modified to reduce the risk
of disease.2,3
Much of the research relating risk factors to
chronic diseases has focussed on singular
independent risk factors. Yet these factors
are known not to occur in isolation. Smoking, high alcohol intake, physical inactivity
and overweight coexist within individuals.
Dawson notes the literature has established
that, within individuals, drinking is
associated with long-term smoking beha-
viour.9 Within-person associations between
physical inactivity and overweight,10–12 and
alcohol intake and overweight are reported.13 Similar relationships between
smoking and physical inactivity,14,15 smoking and overweight,16–18 and alcohol intake
and physical inactivity have also been
found.19
These risk factors are also known to coexist,
or cluster, with respect to disease, allowing
researchers to identify those who are at an
especially high risk for a disease based on
risk factor profiles. Research has focussed
primarily on Syndrome X, a cluster of metabolic risk factors including insulin resistance, abnormal blood fats, overweight and
high blood pressure, which increase risk for
CVD and diabetes.20,21 Past studies have
looked at the clustering of the major behavioural risk factors for CVD in relation to
Syndrome X. Particularly, Twisk et al. found
clustering with respect to CVD among
Syndrome X, physical inactivity and, in
males, heavy alcohol consumption.22 Genest
et al. reviewed the research on clustering of
behavioural and metabolic risk factors for
CVD in an effort to identify those with highrisk profiles.23 A similar study quantified the
extent of clustering in the American Indian
and Alaskan Native population.24 It was also
found that as the number of risk factors increases in young people, so too does severity
of asymptomatic coronary and aortic atherosclerosis.25 Similar clustering relationships
with respect to other chronic diseases have
been noted.7,26,27 One recent study quantified
Author References
Julia E Klein-Geltink, Centre for Chronic Disease Prevention and Control, Public Health Agency of Canada; Department of Public Health Sciences, Faculty of Medicine,
University of Toronto
Bernard CK Choi, Centre for Chronic Disease Prevention and Control, Public Health Agency of Canada; Department of Public Health Sciences, Faculty of Medecine,
University of Toronto; Department of Epidemiology and Community Medicine, Faculty of Medicine, University of Ottawa
Richard N Fry, Centre for Chronic Disease Prevention and Control, Public Health Agency of Canada
Correspondence: Bernard CK Choi, Centre for Chronic Disease Prevention and Control, Public Health Agency of Canada, AL 6701A, 120 Colonnade Road, Ottawa,
Ontario, Canada K1A 1B4; e-mail: [email protected]
Chronic Diseases in Canada
25
Vol 27, No 1, 2006
the prevalence and clustering of cigarette
smoking, risky drinking of alcoholic beverages, physical inactivity and overweight in
the U.S. population based on 2001 data.28
Despite these reports of risk factor coexistence,28,29 the prevalence of such coexistence
and resultant impacts on risk for the major
chronic diseases has not been described systematically in the Canadian population. The
primary objective of this study was to estimate the prevalences30 within the Canadian
population of various clusters of risk factors
of interest (smoking, high alcohol intake,
physical inactivity and overweight). Our
research question was to quantify what proportion of the Canadian population have
multiple coexistent rather than singular
independent risk factor exposures.
TABLE 1
Weighted prevalence and confidence limits of selected chronic disease risk
factors in Canada, Canadian Community Health Survey, Cycle 1.1 (2000)
Number in
sample
Risk factor
None
Smoking
‡
High alcohol intake
Physical inactivity
High BMI
§
¶
#
Total
*
Number in
population
Prevalence
*
Lower
†
95% CL
Upper
95% CL
23,186
4,863,489
21.0
20.7
21.4
29,011
5,539,648
21.5
21.2
21.9
7,277
1,524,372
6.0
5.8
6.2
61,444
12,662,515
53.5
53.0
53.9
58,258
11,352,881
44.8
44.3
45.2
125,574
25,801,718
Prevalence of risk factor was calculated by weighted methods and expressed in terms of percentage
of total Canadian population. Total prevalence equals more than 100% because some respondents
may be counted in more than one risk factor category.
†
CL = Confidence limit
‡
Smoking = current smoker; Non-smoking = never been smoker or former smoker.
§
High alcohol intake = consuming more than 14.0 drinks per week (male) or 9.0 drinks per week
(female); Low alcohol intake = consuming 14.0 or less drinks per week (male) or 9.0 or less drinks
per week (female)
Materials and methods
¶
Physically inactive = having an energy expenditure level of less than 1.5 kcal/kg/day; Physically
active = having an energy expenditure level of 1.5 or more
Prevalences for various risk factor clusters
were estimated from the Canadian Community Health Survey (CCHS) Cycle 1.1 (2000)
data file, using a methodology based on
binary risk factor variables, stratified by age
and sex.
#
High BMI = overweight or having a BMI of greater than or equal to 25; Low BMI = BMI of less
than 25
Data source
The CCHS is a cross-sectional survey of
health determinants, health status and
health care system utilization in Canada.31, 32
Data collection began in September 2000
and follows a two-year cycle in which
health-region-level data is collected in the
first year (Cycle 1.1), and provincial-level
data is collected in the second (Cycle 1.2).
Data from the first year, with a sample size
of approximately 130,000, were used in this
study. The sampling frame covered approximately 98 percent of the Canadian population over age 12. The sample included one
randomly selected respondent per selected
household in all provinces and territories.
First Nations reserves, Canadian Forces
Bases and some remote areas were
excluded.
Risk factor definitions
“Smoking” was defined as a current daily
smoker; “non-smoking” denoted never having smoked or being a former or current
Vol 27, No 1, 2006
occasional smoker. Smoking status was
derived from four CCHS questions, which
assessed quantity of cigarettes smoked during a lifetime, the smoking status of the
respondent at the present time (daily, occasional or not at all) and if the respondent had
ever smoked cigarettes daily.33
“High alcohol intake” was defined as having
more than 14 drinks per week (males) or
nine drinks per week (females); “low
alcohol intake” meant 14 or fewer drinks per
week (males) or nine or fewer drinks per
week (females). These cutoffs were based on
the recommendations by Statistics Canada
(2002),34 Bondy et al.35 and the metaanalysis findings of English et al. which indicate that weekly consumption of more than
14 drinks per week for males and nine drinks
per week for females begin to link with an
increased risk of overall mortality.36 Alcohol
intake was assessed using a measure for
derived, continuous alcohol consumption
over the previous week.33
“Physical inactivity” was assessed using a
derived CCHS variable—energy expenditure—and was calculated using the frequency and duration of a respondent’s
physical activity sessions self-report as well
26
as its metabolic equivalent (MET) value. The
MET is a value of metabolic energy cost
expressed as a multiple of the resting metabolic rate.33 Expressed as kilocalories per kilogram of body weight per day (kcal/kg/
day), an energy expenditure value of less
than 1.5 (moderate activity) was considered
physically inactive. This approach is consistent with other literature reporting the prevalence of physical inactivity in Canada.37
Body mass index (BMI)30 was calculated
based on weight and height variables using
the equation BMI = weight (kg)/height
(m)2. Among adults 19 years of age and
older, a BMI equal to or greater than 25
(overweight) was considered high, while a
BMI less than 25 was considered low. For
those 18 years and under, we used overweight cutoff levels by age and sex, as suggested by Cole et al.38
Risk factor cluster variables
When data were analysed with respect to the
four single risk factors, each risk factor was
coded binary (YES = 1, NO = 0), irrespective of the status of the other risk factors. In
order to measure prevalence of risk factor
clusters in the population, a total of 15 risk
Chronic Diseases in Canada
TABLE 2
Weighted prevalence and 95% confidence limits of selected chronic disease
risk factors in Canada, by age and sex, Canadian Community Health Survey,
Cycle 1.1 (2000)
Age groups (years)
Risk factor
12–19
20–34
35–49
50–64
65+
Total
49.2
(47.5, 50.9)
18.0
(16.9, 19.1)
12.0
(11.3, 12.7)
11.7
(10.9, 12.6)
18.5
(17.2, 19.9)
18.9
(18.4, 19.4)
12.2
(11.3, 13.2)
29.6
(28.3, 30.8)
29.4
(28.4, 30.3)
22.2
(21.1, 23.3)
11.8
(10.8, 12.8)
23.6
(23.1, 24.1)
3.8
(3.2, 4.5)
11.3
(10.5, 12.1)
7.8
(7.2, 8.4)
7.6
(6.9, 8.4)
4.9
(4.3, 5.6)
7.8
(7.5, 8.1)
24.7
(23.3, 26.0)
48.0
(46.6, 49.4)
56.0
(54.9, 57.2)
55.5
(54.1, 56.9)
53.1
(51.4, 54.8)
49.6
(48.9, 50.3)
25.6
(24.1, 27.1)
45.5
(44.3, 46.8)
59.2
(58.1, 60.3)
64.3
(62.9, 65.6)
54.5
(53.0, 56.1)
51.8
(51.2, 52.4)
1,662,580
3,193,934
3,826,542
2,427,992
1,594,367
12,705,415
None
45.0
(43.3, 46.6)
25.2
(24.1, 26.4)
20.2
(19.3, 21.2)
17.0
(16.0, 18.0)
16.6
(15.5, 17.7)
23.1
(22.5, 23.6)
Smoking ‡
13.7
(12.7, 14.7)
22.7
(21.7, 23.7)
24.5
(23.5, 25.4)
19.6
(18.6, 20.5)
9.5
(8.8, 10.2)
19.5
(19.0, 20.0)
3.4
(2.8, 3.9)
5.2
(4.7, 5.6)
5.1
(4.7, 5.6)
3.7
(3.2, 4.2)
2.7
(2.2, 3.2)
4.3
(4.0, 4.5)
38.6
(37.0, 40.2)
54.7
(53.5, 55.9)
58.6
(57.5, 59.6)
59.1
(57.8, 60.3)
68.2
(66.9, 69.5)
57.0
(56.4, 57.5)
High BMI#
16.4
(15.4, 17.5)
27.7
(26.7, 28.8)
39.5
(38.4, 40.6)
53.3
(51.9, 54.7)
47.9
(46.6, 49.1)
37.8
(37.3, 38.4)
Total Canadian population
1,580,702
3,107,188
3,897,071
2,457,918
2,053,423
13,096,302
Males
None
Smoking
‡
High alcohol intake
Physical inactivity
High BMI
§
¶
#
Total Canadian population
Females
High alcohol intake
Physical inactivity
§
¶
*
Prevalence of risk factor was calculated by weighted methods and expressed in terms of percentage of total Canadian population.
†
Confidence limit
‡
Smoking = current smoker; Non-smoking = never been smoker or former smoker.
§
High alcohol intake = consuming more than 14.0 drinks per week (male) or 9.0 drinks per week (female); Low alcohol intake = consuming 14.0 or
less drinks per week (male) or 9.0 or less drinks per week (female)
¶
Physically inactive = having an energy expenditure level of less than 1.5 kcal/kg/day; Physically active = having an energy expenditure level of 1.5 or more
#
High BMI = overweight or having a BMI of greater than or equal to 25; Low BMI = BMI of less than 25
factor categories (four singular risk factors,
six risk factor pairs, four risk factor trios, one
category with all four risk factors) were
encoded based on the four risk factors under
study. The group of respondents with no risk
factors (i.e., smoking = NO; high alcohol
intake = NO; physical inactivity = NO; high
BMI = NO) was defined for analyses as the
baseline comparison group, named “None”.
Fifteen categorical variables for the risk
factor clusters were created. For example,
individuals were counted in the “Smoking”
category if they were smoking, but had none
of the other risk factors. Individuals were
Chronic Diseases in Canada
counted in the “Smoking and High alcohol
intake” category if they currently smoked
and had high alcohol intake, but were physically active and had low BMI; otherwise,
they were not counted. Finally, individuals
were counted in the “Smoking, High alcohol
intake, Physically inactivity and High BMI”
category if they had exposure to all four risk
factors; otherwise, they were not counted.
Statistical analysis
the stratified complex design of CCHS and
expressed as a percentage of the total
Canadian population. Its 95 percent confidence limits (CL) were estimated by bootstrap techniques.39 Cases for whom data on
any of these variables were missing were not
included in the respective prevalence calculations. The Statistical Analysis System, version 8.01 for Windows (SAS Institute, Inc.,
Cary, North Carolina), was used for all analyses including bootstrapping. All differences
discussed are statistically significant.
Prevalence of risk factor clusters was estimated by weighted methods appropriate for
27
Vol 27, No 1, 2006
Results
TABLE 3
Weighted prevalence of selected chronic diseases risk factors and risk factor
clusters in Canada, Canadian Community Health Survey,
Cycle 1.1 (2000)
Tables 1 and 2 show prevalences of single
risk factor exposures. Here, prevalence refers to weighted prevalence of a single factor
in the population, regardless of any of its
co-occurrences with other factors.
Table 1 shows the weighted prevalence of
the four selected risk factors expressed as
percentages of the total Canadian population. Our results indicate that 21.0 percent of
Canadians have no risk factor exposures,
21.5 percent currently smoke, 6.0 percent
are high-risk drinkers, 53.5 percent are physically inactive, and 44.8 percent are overweight.
Table 3 presents risk factor cluster prevalences as percentages of the Canadian population and their associated 95 percent confidence limits in the Canadian population. For
example, 4.3 percent of Canadians (Sample
N = 5,555) were current smokers who had
low alcohol intake, were not physically inactive and had low BMI. Whereas 0.8 percent
of Canadians (Sample N = 923) were current smokers who had high alcohol intake,
but were not physically inactive and had low
BMI (Table 3).
Vol 27, No 1, 2006
Number in
sample
None
23,186
4,863,489
‡
Smoking (only)
Number in
population
Prevalence
21.0
*
Lower
†
95% CL
20.7
Upper
95% CL
21.4
5,555
1,004,329
4.3
4.2
4.5
1,146
258,008
1.1
1.0
1.2
19,712
4,488,975
19.4
19.0
19.8
18,628
3,510,975
15.2
14.9
15.5
923
174,284
0.8
0.7
0.8
Smoking and physical
inactivity
7,660
1,523,506
6.6
6.4
6.8
Smoking and high BMI
3,793
644,632
2.8
2.7
2.9
High alcohol intake and
physical inactivity
669
157,492
0.7
0.6
0.8
High alcohol intake and
high BMI
1,102
229,186
1.0
0.9
1.1
Physical inactivity and
high BMI
22,143
4,394,103
19.0
18.7
19.4
Smoking, high alcohol
intake and physical
inactivity
1,019
214,517
0.9
0.9
1.0
Smoking, high alcohol
intake and high BMI
543
101,418
0.4
0.4
0.5
6,583
1,229,995
5.3
5.1
5.5
High alcohol intake,
physical inactivity and
high BMI
894
179,685
0.8
0.7
0.9
Smoking, high alcohol
intake, physical inactivity
and high BMI
730
142,894
0.6
0.6
0.7
25,801,718
(2,684,230
missing)
99.9
§
High alcohol intake (only)
¶
Physical inactivity (only)
#
High BMI (only)
Smoking and high alcohol
intake
Table 2 shows that, across most age groups,
males were found to be significantly less
physically inactive, but more likely to
smoke, have high alcohol intake and be
overweight when compared to females. Additionally, the proportion of males with none
of the four risk factors was significantly
lower than that of females. Prevalence values for high-risk drinking and overweight
peaked in the same age groups for both
males and females.
Tables 3 to 5 show prevalences of multiple
risk factor exposures. The prevalence refers to
the weighted prevalence of the population,
counting only those risk factors specified. For
example, smoking prevalence specifically refers to the prevalence of current smokers
(smoking=YES) who had low alcohol intake
(high alcohol intake=NO), were physically
active (physical inactivity=NO) and had low
BMI (high BMI=NO). Thus, the risk factors
specified in Tables 3–5 are discrete (i.e. nonoverlapping).
Risk factors present within
individual**
Smoking, physical
inactivity and high BMI
Total
*
125,574
(11,288
missing)
Prevalence of risk factor cluster was calculated by weighted methods and expressed in terms of
percentage of total Canadian population.
** All categories are discrete and non-overlapping.
†
Confidence limit
‡
Smoking = current smoker; Non-smoking = having never smoked or being a former smoker
§
High alcohol intake = consuming more than 14.0 drinks per week (male) or 9.0 drinks per week
(female); Low alcohol intake = consuming 14.0 or less drinks per week (male) or 9.0 or less drinks
per week (female)
¶
Physical inactivity = having an energy expenditure level of less than 1.5 kcal/kg/day; Physical
activity = having an energy expenditure level of equal to or greater than 1.5
#
High BMI = overweight or having a BMI of greater than or equal to 25; Low BMI = having a
BMI of less than 25
From Table 3, 79.0 percent (or 100 percent
minus 21.0 percent) of the population had at
least one of the four risk factors, 39.0 percent
had at least two, 8.1 percent had at least
three and 0.6 percent had all four. The
28
cluster with none of the risk factors
accounted for the highest cluster prevalence
(21.0 percent), followed by those with physical inactivity (19.4 percent); physical inactivity and overweight (19.0 percent); and
Chronic Diseases in Canada
TABLE 4
Weighted prevalence and 95% confidence limits of selected chronic disease risk factors and risk
factor clusters in Canadian males, by age, Canadian Community Health Survey,
Cycle 1.1 (2000)
Age groups (years)
Risk factor present within individual**
12–19
20–34
35–49
50–64
65+
Total
49.2
(47.5, 50.9)
18.0
(16.9, 19.1)
12.0
(11.3, 12.7)
11.7
(10.9, 12.6)
18.5
(17.2, 19.9)
18.9
(18.4, 19.4)
5.6
(4.8, 6.3)
5.7
(5.1, 6.3)
4.3
(3.8, 4.7)
3.3
(2.8, 3.8)
2.0
(1.6, 2.4)
4.3
(4.1, 4.6)
1.4
(0.9, 1.8)
2.0
(1.6, 2.4)
0.7
(0.5, 0.9)
0.9
(0.6, 1.1)
0.7
(0.4, 1.0)
1.1
(1.0, 1.3)
Physical inactivity (only)
14.6
(13.4, 15.8)
15.6
(14.5, 16.7)
13.2
(12.3, 14.0)
11.7
(10.7, 12.7)
18.0
(16.8, 19.3)
14.3
(13.7, 14.8)
High BMI# (only)
15.9
(14.7, 17.0)
17.1
(16.0, 18.2)
19.3
(18.4, 20.2)
22.7
(21.6, 23.9)
22.7
(21.4, 24.1)
19.4
(18.9, 19.9)
Smoking and high alcohol intake
1.2
(0.8, 1.5)
2.0
(1.6, 2.4)
0.8
(0.6, 1.0)
0.5
(0.3, 0.7)
0.2
(0.1, 0.4)
1.0
(0.9, 1.1)
Smoking and physical inactivity
2.5
(2.0, 3.0)
8.1
(7.3, 9.0)
7.5
(6.9, 8.2)
5.6
(5.0, 6.3)
3.9
(3.3, 4.5)
6.3
(5.9, 6.6)
Smoking and high BMI
1.6
(1.2, 2.0)
3.8
(3.3, 4.3)
5.0
(4.5, 5.4)
3.2
(2.8, 3.6)
1.3
(1.0, 1.6)
3.5
(3.3, 3.7)
High alcohol intake and physical
inactivity
0.4
(0.2, 0.6)
1.1
(0.8, 1.3)
0.5
(0.3, 0.6)
0.7
(0.5, 0.9)
0.5
(0.3, 0.7)
0.7
(0.6, 0.8)
High alcohol intake and high BMI
0.5
(0.3, 0.8)
2.2
(1.8, 2.6)
1.4
(1.2, 1.7)
2.0
(1.5, 2.4)
1.3
(1.0, 1.7)
1.6
(1.4, 1.8)
Physical inactivity and high BMI
5.3
(4.6, 6.1)
13.5
(12.5, 14.5)
22.3
(21.3, 23.3)
26.7
(25.5, 28.0)
25.2
(23.8, 26.7)
19.2
(18.7, 19.7)
Smoking, high alcohol intake and
physical inactivity
0.3
(0.2, 0.4)
1.5
(1.2, 1.8)
1.5
(1.2, 1.8)
0.9
(0.6, 1.2)
0.5
(0.3, 0.8)
1.1
(1.0, 1.2)
Smoking, high alcohol intake and high
BMI
0.4
(0.3, 0.6)
1.1
(0.9, 1.4)
0.7
(0.5, 0.9)
0.5
(0.3, 0.6)
0.2
(0.0, 0.4)
0.7
(0.6, 0.8)
Smoking, physical inactivity and high
BMI
1.0
(0.6, 1.4)
6.0
(5.4, 6.6)
8.3
(7.7, 8.9)
7.2
(6.4, 7.9)
3.0
(2.5, 3.6)
6.0
(5.7, 6.2)
High alcohol intake, physical inactivity
and high BMI
0.1
(0.0, 0.2)
1.2
(0.9, 1.5)
1.4
(1.1, 1.7)
1.4
(1.1, 1.7)
1.4
(1.0, 1.8)
1.2
(1.1, 1.3)
Smoking, high alcohol intake, physical
inactivity and high BMI
–
1.2
(0.9, 1.4)
1.3
(1.1, 1.5)
1.0
(0.8, 1.3))
0.4
(0.2, 0.6)
0.9
(0.8, 1.1)
1,662,580
3,193,934
3,826,542
2,427,992
1,594,367
12,705,415
None
‡
Smoking (only)
§
High alcohol intake (only)
¶
Total Canadian population
*
Prevalence of risk factor cluster was calculated by weighted methods and expressed in terms of percentage of total Canadian population.
** All categories are discrete and non-overlapping.
‡
Smoking = current smoker; Non-smoking = having never smoked or being a former smoker
§
High alcohol intake = consuming more than 14.0 drinks per week (male) or 9.0 drinks per week (female); Low alcohol intake = consuming 14.0 or
less drinks per week (male) or 9.0 or less drinks per week (female)
¶
Physical inactivity = having an energy expenditure level of less than 1.5 kcal/kg/day; Physical activity = having an energy expenditure level of equal
to or greater than 1.5
#
High BMI = overweight or having a BMI of greater than or equal to 25; Low BMI = having a BMI of less than 25
Chronic Diseases in Canada
29
Vol 27, No 1, 2006
TABLE 5
Weighted prevalence and 95% confidence limits of selected chronic disease risk factors and risk factor
clusters in Canadian females, by age, Canadian Community Health Survey,
Cycle 1.1 (2000)
Age groups (years)
Risk factor present within individual**
12–19
20–34
35–49
50–64
65+
Total
45.0
(43.3, 46.6)
25.2
(24.1, 26.4)
20.2
(19.3, 21.2)
17.0
(16.0, 18.0)
16.6
(15.5, 17.7)
23.0
(22.5, 23.6)
5.3
(4.6, 6.0)
5.8
(5.3, 6.3)
5.0
(4.5, 5.5)
3.2
(2.8, 3.7)
1.6
(1.3, 2.0)
4.4
(4.1, 4.6)
1.1
(0.7, 1.5)
1.5
(1.2, 1.7)
1.1
(0.9, 1.4)
0.8
(0.6, 1.0)
0.8
(0.5, 1.1)
1.1
(1.0, 1.2)
25.7
(24.3, 27.1)
28.3
(27.1, 29.5)
23.1
(22.1, 24.1)
17.1
(16.1, 18.2)
27.7
(26.6, 28.9)
24.2
(23.7, 24.7)
High BMI# (only)
7.9
(7.1, 8.7)
8.7
(8.1, 9.4)
11.1
(10.5, 11.7)
16.3
(15.3, 17.2)
12.0
(11.2, 12.7)
11.3
(11.0, 11.6)
Smoking and high alcohol intake
0.6
(0.4, 0.8)
0.9
0.7, 1.1)
0.6
(0.4, 0.7)
0.3
(0.1, 0.4)
0.1
(0.0, 0.2)
0.5
(0.5, 0.6)
Smoking and physical inactivity
4.9
(4.2, 5.5)
8.4
(7.7, 9.1)
8.3
(7.8, 8.9)
6.4
(5.8, 7.0)
4.0
(3.6, 4.5)
6.9
(6.6, 7.2)
Smoking and high BMI
1.2
(0.9, 1.5)
2.3
(2.0, 2.7)
2.8
(2.4, 3.1)
2.7
(2.3, 3.2)
0.7
(0.5, 0.9)
2.2
(2.0, 2.3)
High alcohol intake and physical
inactivity
0.4
(0.2, 0.6)
0.8
(0.6, 1.1)
0.8
(0.6, 1.0)
0.7
(0.4, 0.9)
0.6
(0.4, 0.8)
0.7
(0.6, 0.8)
High alcohol intake and high BMI
0.1
(0.0, 0.2)
0.5
(0.3, 0.6)
0.5
(0.3, 0.6)
0.6
(0.4, 0.8)
0.3
(0.1, 0.4)
0.4
(0.4, 0.5)
Physical inactivity and high BMI
5.1
(4.4, 5.8)
11.3
(10.5, 12.2)
18.2
(17.3, 19.1)
27.2
(25.9, 28.4)
31.6
(30.3, 32.8)
18.9
(18.4, 19.3)
Smoking, high alcohol intake and
physical inactivity
0.9
(0.6, 1.1)
0.9
(0.7, 1.1)
1.1
(0.8, 1.3)
0.4
(0.2, 0.5)
0.4
(0.2, 0.5)
0.8
(0.7, 0.9)
Smoking, high alcohol intake and high
BMI
0.2
(0.1, 0.3)
0.4
(0.2, 0.5)
0.3
(0.2, 0.5)
0.1
(0.0, 0.1)
–
0.2
(0.2, 0.3)
–
4.6
(4.1, 5.1)
6.0
(5.5, 6.6)
6.5
(5.9, 7.1)
2.9
(2.5, 3.4)
4.7
(4.5, 5.0)
High alcohol intake, physical inactivity
and high BMI
0.1
(0.0, 0.2)
0.3
(0.2, 0.4)
0.4
(0.3, 0.6)
0.5
(0.3, 0.6)
0.6
(0.4, 0.8)
0.6
(0.4, 0.8)
Smoking, high alcohol intake, physical
inactivity and high BMI
0.4
(0.1, 0.7)
0.2
(0.1, 0.3)
0.5
(0.4, 0.6)
0.4
(0.2, 0.5)
0.1
(0.0, 0.1)
0.3
(0.3, 0.4)
1,580,702
3,107,188
3,897,071
2,457,918
2,053,423
3,096,302
None
‡
Smoking (only)
§
High alcohol intake (only)
¶
Physical inactivity (only)
Smoking, physical inactivity and high
BMI
Total Canadian population
*
Prevalence of risk factor cluster was calculated by weighted methods and expressed in terms of percentage of total Canadian population.
** All categories are discrete and non-overlapping.
‡
Smoking = current smoker; Non-smoking = having never smoked or being a former smoker
§
High alcohol intake = consuming more than 14.0 drinks per week (male) or 9.0 drinks per week (female); Low alcohol intake = consuming 14.0 or
less drinks per week (male) or 9.0 or less drinks per week (female)
¶
Physical inactivity = having an energy expenditure level of less than 1.5 kcal/kg/day; Physical activity = having an energy expenditure level of equal
to or greater than 1.5
#
High BMI = overweight or having a BMI of greater than or equal to 25; Low BMI = having a BMI of less than 25
Vol 27, No 1, 2006
30
Chronic Diseases in Canada
overweight (15.2 percent). Respondents
who are physically inactive and overweight
account for the highest proportion of the
population with two or more coexistent risk
factors.
Tables 4 and 5 show the age distribution of
risk factor clusters for males and females,
respectively. With the exception of the clusters for smoking (only), high-risk drinking
(only), smoking and physical inactivity,
high-risk drinking and physical inactivity,
and physical inactivity and overweight, all
cluster prevalence values (for the “all-age”
comparisons) were statistically different
between males and females. Further, when
compared to females, the prevalence values
were higher in males for all-risk factor clusters, except for the no-risk factor, smoking,
physical inactivity, and smoking and physical inactivity clusters. Except for four risk
factor clusters (physical inactivity; physical
inactivity and overweight; high alcohol
intake and overweight; and smoking, physical inactivity and overweight), prevalence
values peaked in the same age groups for
both males and females. Specifically, for
both sexes, the prevalence for the no-risk
factor cluster peaked among those aged 12 to
19. Prevalence figures for most of the clusters peaked among those aged 20 to 34 years
or those 35 to 49 years of age. Exceptionally,
high BMI-only peaked among those aged 50
to 64 years; the high alcohol intake, physical
inactivity and high BMI cluster peaked
among those over 65 years of age.
Discussion
The CCHS represents the most recent and
the largest population health survey to date
in Canada. The findings of this study therefore closely reflect the current risk factor situations of Canadians. It provides insights
not previously available on the question of
chronic disease risk factor coexistence in
Canada and sets the stage for renewed clinical, policy and research directions.
Our research question was to determine
whether or not Canadians have multiple
rather than singular risk factor exposures.
Based on our study, 40 percent of Canadians
had one independent risk factor, while 39
percent had multiple coexistent risk factors,
Chronic Diseases in Canada
and the remaining 21 percent had none. This
distribution differs somewhat from that
found by Fine et al. for the U.S. population,
where 9.7 percent had no risk factors, 32.6
percent had one independent risk factor and
57.7 percent had multiple coexistent risk
factors.28 This difference is mainly due to the
difference in how physical inactivity and
high-risk drinking are defined. Our definition
for physical inactivity made use of an energy
expenditure cut-point of 1.5, while Fine et al.
looked at individuals who reported engaging
in light/moderate physical activities for less
than 30 minutes at a time for five or more
times a week, or who reported engaging in
vigorous physical activity for less than 20
minutes at a time for three or more times a
week. Our definition for high-risk drinking
used weekly consumption cutoff values of at
least 15 drinks for men and at least 10 drinks
for women. Fine et al. defined risky drinking
for men as the average weekly consumption
of more than 14 drinks, or five or more
drinks per day at least twice in the last year,
or four or more drinks per day for at least
three times in the last year. For women it
was defined as the average weekly consumption of more than seven drinks, or four
or more drinks per day at least twice in the
last year, or three or more drinks per day at
least three times in the last year.
According to our results, males are expected
to be more at risk of chronic disease outcomes than females due to increased smoking, alcohol intake and overweight.
Similarly, in the U.S. population, men were
found to have more risk factors than
women.28 The gender differences in health
behaviours, including modifiable chronic
disease risk factors, are consistent with the
literature. Particularly, it has been noted that
males are more likely to partake in “risky”
behaviours40 and that females are more
likely to be physically inactive.41,42
According to our analyses, the group with
none of the risk factors was the most common (21.0 percent), followed by those with
physical inactivity only (19.4 percent), physical inactivity and overweight (19.0 percent),
and overweight (15.2 percent). Those who
are physically inactive and overweight
account for the highest proportion of the
population with two or more coexistent risk
31
factors. Our findings are comparable to those
found in the U.S. population, where the most
common risk factor clusters were physical
inactivity and overweight (26.4 percent),
physically inactive (16.4 percent), overweight (11.7 percent) and the no-risk factor
cluster (9.7 percent).28 The slight differences
are due to different risk factor definitions.
Although best efforts were taken to define
chronic disease risk factor presence in terms
of cut-points that are meaningful to chronic
disease outcomes, we were reliant upon literature to inform our decisions. One particular definition at issue is that of physical
inactivity. The definition used by Fine et al.,
while technically different from ours (therefore potentially explaining prevalence differences from the two studies) did incorporate
exercise duration and intensity measures,
something that the CCHS definition intended
to do, although in a different way. We chose
a cut-point of 1.5 kcal/kg/day, which is consistent with the definition of physical inactivity used throughout the Canadian chronic
disease risk factor literature.33, 37 In choosing
this cut-point, we are assuming that the population is healthy and has no physical activity limitations. As a result, the prevalence of
physical inactivity in the population would
be high, especially among elderly females.
However, using a standard definition such
as this does allow for comparisons across
populations and time periods from various
Canadian studies.
Our study has certain limitations, though we
employed techniques to deal with some of
them. Because there were missing data, and
cases with missing data were automatically
excluded from prevalence calculations,
weighted numbers in the population would
have been underestimated. However, this
was corrected in our study by programming.
Cases with missing data were programmed
to be excluded from both the numerator and
denominator so that prevalence estimates
were corrected. Because of the complexity of
the sampling design, sampling error for
prevalence estimates was calculated using
the bootstrap re-sampling technique.30 In the
CCHS, training and use of skilled interviewers, monitoring of interviewers and use of
various quality assurance protocols reduced
the amount of non-sampling error.30
Vol 27, No 1, 2006
Non-response was rare as a result of the use
of computer-assisted telephone interviews
as the data collection instrument.30 Lastly,
the CCHS is based on self-report; thus, the
true prevalences of risk factor clusters are
most likely underestimated, a phenomenon
known as social desirability bias.43
This study has described a new approach
that examines multiple coexistent risk factor
clusters to assess the corresponding prevalence rates in the Canadian population. Its
results are important in that they quantify
the level of coexistent risks in the population. The existence of multiple risk factors is
known to elevate the risk for chronic disease
outcome beyond that which would exist,
due to the presence of a single risk factor.
The impacts of risk factor clusters on the risk
for the major chronic diseases experienced
by the Canadian population can now, with
these data, be more accurately assessed and
so further research into the coexistence of
multiple chronic disease risk factors is warranted. As well, other risk factors (such as
nutritional status, ethnicity and family history of disease) for other chronic disease outcomes should be studied using the
methodology described in this study. Different definitions of the risk factors under study
(e.g., light, moderate and heavy levels for
physical activity) could be used to arrive at
different prevalence figures which might allow for more accurate assessment of population attributable risks. Additionally, other
demographic groups, defined on the basis of
income and immigrant status, could be examined to determine which are at an especially high risk to chronic disease outcomes
in Canada. Similar methodologies used
across studies will allow for comparisons between populations, both nationally and internationally. Lastly, better systematic and
ongoing surveillance of the risk factors for
chronic disease in the Canadian population
via a longitudinal database over time would
allow for more definitive results.
Vol 27, No 1, 2006
Acknowledgments
The authors wish to thank Peter Walsh for
his assistance with SAS programming;
Seema Nagpal and Howard Morrison for
their assistance in the conceptualization of
this project; Shirley Bryan for her content
expertise; Wei Luo, Lisa Pogany and Jay
Onysko for their discussion input; and
Monique Haan and Ineke Neutel for their
editorial support.
References
1.
Heart and Stroke Foundation of Canada. The
changing face of heart disease and stroke in
Canada. Ottawa: (Canada): Heart and Stroke
Foundation of Canada; 1999.
2.
Katzmarzyk PT, Gledhill N, Shephard RJ.
The economic burden of physical inactivity
in Canada. CMAJ. 2000;163:1435–40.
3.
Birmingham CL, Muller JL, Palepu A, et al.
The cost of overweight in Canada. CMAJ.
1999;160:483–8.
4.
Centres for Disease Control. Chronic diseases
and their risk factors: The nation’s leading
causes of death. 1999. Available from:
http://www.cdc.gov/nccdphp/statbook/
statbook.htm.
5.
Choi BCK, Shi F. Risk factors for diabetes
mellitus by age and sex: Results of the
National
Population
Health
Survey.
Diabetologia. 2001;44:1221–31.
6.
Magnus P. The real contribution of the major
risk factors to the coronary epidemics: time
to end the “only-50%” myth. Arch Intern
Med. 2001;161:2657–60.
7.
Rennard SI. COPD: Overview of definitions,
epidemiology, and factors influencing its
development. Chest. 1998;113:235–41S.
8.
Health Canada. Cancer: What’s your risk?
Health Canada Magazine; 2001.
9.
Dawson DA. Drinking as a risk factor for
sustained smoking. Drug Alcohol Depend.
2000;59:235–49.
10. Wing RR, Hill JO. Successful weight loss
maintenance. Annu Rev Nutr. 2001; 21:
323–41.
32
11. Grundy SM, Blackburn G, Higgins M, et al.
Physical activity in the prevention and treatment of Overweight and its comorbidities.
Med Sci Sports Exerc. 1999;31:S502–8.
12. Godsland IF, Leyva F, Walton C, et al.
Associations of smoking, alcohol and physical activity with risk factors for coronary
heart disease and diabetes in the first
follow-up of the Heart Disease and Diabetes
Risk Indicators in a Screened Cohort study
(HDDRISC-1). J Intern Med. 1998;244:
33–41.
13. Doucet E, Tremblay A. Food intake, energy
balance and body weight control. J Clin
Nutr. 1997;51:846–55.
14. Stamford BA, Matter S, Fell RD, et al.
Cigarette smoking, physical activity and
alcohol consumption: relationship to blood
lipids and lipoproteins in premenopausal
females. Metabolism. 1984;7:585–90.
15. Paffenbarger RS, Hyde RT, Wing AL, et al.
Physical activity, all-cause mortality and
longevity of college alumni. New Engl J Med.
1986;314:605–13.
16. Williamson DF, Madans J, Anda RF, et al.
Smoking cessation and severity of weight
gain in a national cohort. New Engl J Med.
1991;324:739–45.
17. Klesges RC, Meyers AW, Klesges LM, et al.
Smoking, body weight, and their effects on
smoking behaviour: a comprehensive review
of the literature. Psychol Bull. 1989;106:
204–30.
18. Flegal KM, Troiano RP, Pamuk ER, et al
(1995). The influence of smoking cessation
on the prevalence of overweight in the
United States. New Engl J Med. 1995;33:
165–75.
19. Smothers B, Bertolucci D. Alcohol consumption and health-promoting behaviour in a
U.S. household sample: leisure-time physical
activity. J Stud Alcohol. 2001;62:467–76.
20. Timar O, Sestier F, Levy E. Metabolic syndrome X: A review. Can J Cardiol. 2000;16:
779–89.
21. Saito I, Folsom AR, Brancati, FL, et al. Nontraditional risk factors for coronary heart
disease incidence among persons with diabetes: the atherosclerosis risk in communities (ARIC) study. Ann Intern Med. 2000;
133:81–91.
Chronic Diseases in Canada
22. Twisk JWR, Kemper HCG, Van Mechelen W,
et al. Clustering of risk factors for coronary
heart disease: the longitudinal relationship
with lifestyle. Ann Epidemiol. 2001;11:
157–65.
29. Hahn RA, Heath GW, Chang MH. Cardiovascular disease risk factors and preventive
practices among adults – United States, 1994:
a behavioural risk factor atlas. MMWR Morb
Mortal Wkly Rep. 1998;47(SS–5):35–69.
23. Genest J, Cohn JS. Clustering of cardiovascular risk factors: targeting high-risk individuals. Am J Cardiol. 1995;76:8A–20A.
30. Last JM. A dictionary of epidemiology. 4th
ed. New York (NY): Oxford University Press,
Inc.; 2001.
24. Centers for Disease Control. Prevalence of
selected cardiovascular disease risk factors
among American Indians and Alaska Natives
– United States, 1997. MMWR Morb Mortal
Wkly Rep. 2000;49:461–5.
31. Beland Y. Canadian Community Health Survey – Methodological overview. Health Rep.
2002;13:9–14.
25
Berenson GS, Srinivasan SR, Bao W, et al.
Association between multiple cardiovascular
risk factors and atherosclerosis in children
and young adults. N Engl J Med. 1998;
338:1650–6.
26. Bryant H, Murphy E, Fayers C, et al. A snapshot of cancer in Alberta 2001. Calgary (AB):
Alberta Cancer Board; 2002.
27. Chang M, Hahn RA, Teutsch SM, et al. Multiple risk factors and population attributable
risk for ischemic heart disease mortality in
the United States, 1971–1992. J Clin
Epidemiol. 2001;54:634–44.
28. Fine LJ, Philogene GS, Gramling R, et al.
Prevalence of multiple chronic disease risk
factors: 2001 National Health Interview
Survey. Am J Prev Med. 2004; 27(2S).
Chronic Diseases in Canada
32. Statistics Canada. The Canadian community
health survey (CCHS). 2000. Available from:
http://www.statcan.ca/english/conepts/
health/cchsinfo.htm.
33. Canadian community health survey (CCHS),
Cycle 1.1, Derived variable (DV) specifications. Health Statistics Division, Statistics
Canada 2002.
34. Wilkins K. Moderate alcohol consumption
and heart disease. Health Rep. 2002;14:9–24.
35. Bondy SJ, Rehm J, Ashley MJ, et al. Lowrisk drinking guidelines: The scientific evidence. CJPH. 1999;90:264–270.
36. English, DR, Holman, CDJ, Milne, E, et al.
1995. The quantification of drug caused
morbidity and mortality in Australia, 1995
edition. Canberra: Commonwealth Department of Human Services and Health.
33
37. Heart and Stroke Foundation of Canada. The
growing burden of heart disease and stroke
in Canada 2003. Ottawa (Canada): Heart and
Stroke Foundation of Canada; 2003.
38. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH.
Establishing a standard definition for child
overweight and overweight worldwide:
international survey. BMJ. 2000;320:
1240–3.
39. Efron B, Tibshirani, RJ. An introduction to
the bootstrap. NewYork (NY): Chapman and
Hall; 1993.
40. Doyal L. Sex, gender, and health: The need
for a new approach. BMJ. 2001;323:1061–3.
41. Meisinger C, Thorand B, Schneider A, et al.
Sex Differences in Risk Factors for Incident
Type 2 Diabetes Mellitus: The MONICA
Augsburg Cohort Study. Arch Intern Med
2002;162:82–9.
42. Health Canada. Women’s Health Surveillance Report: A multi-dimensional look at
the health of Canadian women. Ottawa,
Canada: Canadian Institute for Health Information, 2003.
43. Choi BCK, Pak AWP. Bias, Overview. In:
Armitage P, Colton T, eds. Encyclopedia of
Biostatistics. Chinchester: John Wiley &
Sons, 1:331–338.
Vol 27, No 1, 2006
Letters
Chronic or non-communicable?
Dear Editor,
Re: Choi et al.’s Situational
analysis of chronic disease
surveillance in Canada
In the spirit of contributing to a collegial discussion about implications for practice and
surveillance, I would like to comment on the
difficult question of choosing a definition of
“chronic disease”.
The definition cited in the above-cited status
report 1 is only one of two definitions offered
by McKenna, et al., in Brownson, et al.’s,
Chronic Disease Epidemiology and Control 2.
“Disease that has a prolonged course, that
does not resolve spontaneously and for
which a complete cure is rarely achieved” is
a standard clinical definition which could as
easily refer to tuberculosis or marginal
blepharitis as it could to heart disease,
cancer or diabetes.
McKenna et al. offer a more specific definition in the first paragraph of the section
“Definition of Chronic Disease”. They write
that chronic diseases “... are generally characterized by uncertain etiology, multiple risk
factors, a long latency period, a prolonged
course of illness, non-contagious origin,
functional impairment or disability, and
incurability.” This definition is much more
in keeping with how public health practitioners use the term and is, in my opinion,
more relevant to the content and context of
the status report.
Vol 27, No 1, 2006
Even though we are finding out more about
the infectious origins of some of what we
traditionally refer to as the “chronic” diseases, I feel it is still useful to make explicit
the non-communicable aspect, as is exemplified in the name of the working group that
co-authored the aforementioned status
report. Using the broader definition is
appealing semantically and it does appear
more inclusive. However, it makes the scope
of conditions to be addressed, their risk
factors, control methods, etc., unsuitably
broad. Among other things, it can make the
entire focus of the risk factor reduction-end
of the control spectrum (and its surveillance)
very fuzzy and almost impracticable.
References
1.
Choi CK, Wright E, Ulrick A and the Chronic
non-communicable disease surveillance
infostructure sub-group, Health surveillance
working group. Situational analysis of
chronic disease surveillance in Canada:
Results of a stakeholder interview. Chron
Dis Can 2005;26(4):127-9.
2.
McKenna MT, Taylor WR, Marks JS, Koplan
JP. Current issues and challenges in chronic
disease control. In: Brownson RC,
Remington PL, Davis JR, editors. Chronic
disease epidemiology and control. Second
edition. Washington DC: American Public
Health Association; 1998.
Christina Mills
University of Waterloo
34
Chronic Diseases in Canada
Dear Editor,
I would like to thank Dr. Mills for correctly
pointing out the difficulty in choosing a definition for “chronic disease”. At the start of
the situational analysis project, we identified
a large number of definitions for chronic
disease, none of which was considered perfect. We finally chose one of two definitions
offered by McKenna et al. because it is more
concise and more correctly describes the key
word “chronic”. In the literature, we found
that “chronic” refers to “prolonged course”
and not the “long latency period”.
I would also like to mention that there is currently debate on the use of the terms
“chronic disease” and “non-communicable
disease” to describe conditions such as cardiovascular diseases, cancers, asthma and
diabetes. Some have challenged the use of
the term “non-communicable”, positing that
these diseases are also communicable.
Chronic non-communicable diseases are, in
fact, transferable by virtue of their underlying risk factors.1 Unhealthy risk behaviours
such as smoking, physical inactivity and
cooking style can be passed on through
families, communities and populations, and
are therefore “communicable”.2
Other authors point out a current confusion
in the classification system: while noncommunicable disease is based on cause,
chronic disease is based on effect.3 Thus
while certain chronic diseases have an infectious origin, certain communicable diseases
require chronic, ongoing care.
Chronic Diseases in Canada
Another issue is that the terms “chronic disease” and “non-communicable disease”
may contribute to the lack of a perceived
need among decision makers to pay attention to chronic diseases. These terms may
not be adequately conveying the importance
and urgency of chronic disease surveillance,
prevention and control to public health decision makers. hronic conveys the idea of a
disease being always present and, therefore,
non-urgent. on-communicable conveys the
idea of non-infectiousness and implies that
these diseases are safe. A jurisdiction may
not realize the need to dedicate scarce resources toward preventing and controlling
diseases that are long term (chronic) and
where causation is unclear (non-communicable).4
References
1.
Choi BCK, Bonita R, McQueen DV. The need
for global risk factor surveillance. J
Epidemiol Community Health 2001;55:370.
2.
Ackland M, Choi BCK, Puska P. Rethinking
the terms non-communicable disease and
chronic disease. J Epidemiol Community
Health 2003;57:838-39.
3.
Unwin N, Epping Jordan J, Bonita R. Rethinking the terms non-communicable disease and chronic disease. (Letter). J
Epidemiol Community Health 2004;58:801.
4.
Ackland M, Choi BCK, Puska P. Rethinking
the terms non-communicable disease and
chronic disease. (Authors’ reply). J
Epidemiol Community Health 2004;58:801.
It is hoped that with further discussion
among and efforts from public health
researchers and practitioners, a more appropriate term will be available to describe the
true nature of a group of diseases that
include cardiovascular diseases, cancers,
asthmas and diabetes.
Bernard CK Choi
Public Health Agency of Canada
35
Vol 27, No 1, 2006
Status Report
Two to three percent of infants are born with a
congenital anomaly, but who’s counting?
A national survey of congenital anomalies surveillance
in Canada
Dana Paquette, R Brian Lowry and Reg Sauvé
Introduction
The thalidomide tragedy, recognized in
1962,1 led to the development of congenital
anomalies surveillance systems in many
jurisdictions. Today, identifying potential
teratogens is one of many important public
health functions served by congenital
anomalies surveillance.
Major congenital anomalies are detected in
two to three percent of births every year in
Canada,2 and surveillance systems offer a
way of evaluating the impact of prevention
strategies (e.g., food fortification with folic
acid). The systems are also useful in hypotheses generation, in describing the epidemiology of specific anomalies and in identifying infants in need of special services or
programs. Existing systems have also been
used for follow-up studies of survival and
economic impact.3–5
In Canada, the Canadian Congenital
Anomalies Surveillance Network (CCASN)
was established in 2002 by Health Canada
(now the Public Health Agency of Canada
[PHAC]) under the umbrella of the Canadian
Perinatal Surveillance System (CPSS). The
CCASN is made up of clinicians, academics
and public health professionals from across
the country and its goal is to enhance the
quality of surveillance data. It achieves this
by advising PHAC on strategies that encourage provinces/territories to develop surveillance systems where there are none, and by
maintaining and enhancing existing surveillance systems.
In December 2004, the CCASN undertook a
national survey of congenital anomalies
surveillance systems across the country. The
goal of the survey was to gain a better understanding of existing surveillance systems
and to determine how best to fulfill the
CCASN’s mission of supporting the development and maintenance of those that are both
population based and of high quality.
Methods
A list of 37 potential respondents was compiled, which included representatives from
provincial/territorial ministries of health, reproductive care programs, maternal serum
screening, medical genetics programs and
university departments of medical genetics. A
questionnaire, based on a similar survey conducted by Miller and Kirby6 in the United
States, was modified and approved by the
CCASN advisory group. The questionnaire
asked respondents whether they conduct
congenital anomalies surveillance, for what
time periods congenital anomalies data are
available, whether these data include prenatal diagnostic data, which coding/classification system is used, and how data were
used in the previous year.
A survey package was mailed, which included a stamped, return envelope. Two
reminders were sent following the original
mailing, after two and four weeks, respectively.
TABLE 1
Suvey of congenital anomaly surveillance systems in Canada (2004).
Response rate by respondent type
Respondent type
Number of
Number of
questionnaires questionnaires Response rate
sent
received
(percentage)
Provincial/territorial ministries of health
13
12
92.3
Reproductive care programs
10
8
80.0
University departments of medical
genetics
10
5
50.0
4
3
75.0
37
28
75.7
Maternal serum screening and medical
genetics programs
Total
Author References
Dana Paquette, Public Health Agency of Canada, Ottawa, Ontario, Canada
R Brian Lowry, Department of Medical Genetics, Alberta Children’s Hospital, Calgary, Alberta, Canada
Reg Sauvé, Department of Community Health Sciences, Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada
Correspondence: Reg Sauvé, Department of Community Health Sciences, Faculty of Medicine, University of Calgary, 3330 Hospital Drive NW, Calgary, Alberta,
Canada T2N 4N1; fax: (403) 270-7307; e-mail: [email protected]
Vol 27, No 1, 2006
36
Chronic Diseases in Canada
TABLE 2
Description of congenital anomalies surveillance systems in Canada (2004)
Earliest year of
available
data
Estimated number
of live births
covered annually
Province/
territory-wide
coverage
Coding*
Captures
terminations
of pregnancies
British Columbia Health Status
Registry
1952
40,000
Yes
ICD-9/10
OMIM#
Yes (starting in
2006)
Alberta Congenital Anomalies
Surveillance System
1980
37,000
Yes
BPA, ICD-9/10,
OMIM #
Yes
mid-1990s
350
Yes
N/A
N/A
British Columbia Reproductive
Care Program
2000
40,000
Yes
ICD-9/10
No
Ontario Niday Perinatal Database
2004
100,000
No†
Niday
definitions
No
1980 (Atlee)
1992 (FAD)
9,500
Yes
FAD and Atlee
definitions
Yes
1990
1,400
Yes
ICD-9/10
No
Manitoba Maternal Serum Screening
Program
1985
14,000
No§
ICD-9/10
Yes
Ontario Maternal Serum Screening
Program
1993
70,000
No§
ICD-9/10
Yes
Newfoundland and Labrador's
Medical Genetics Program
1976
4,800
Yes
ICD-9/10
Yes
Name of surveillance system
Provincial ministry of health
Fetal Alcohol Spectrum Disorder
Registry (Yukon)
Reproductive care program
Nova Scotia Atlee Perinatal Database
and the Fetal Anomaly Database
(FAD)‡
Prince Edward Island Reproductive
Care Program
Maternal serum screening and medical
genetics programs
*
ICD-9/10: International Classification of Diseases, ninth or tenth revision;
BPA: British Paediatric Association Classification of Diseases;
OMIM #: Online Mendelian Inheritance in Man six-digit number.
†
Covers 85 percent of births.
‡
The Fetal Anomaly Database is an IWK Health Centre Department of Obstetrics and Gynaecology database and is used in combination with the
Atlee Perinatal Database to report on congenital anomalies in Nova Scotia.
§
Limited to women undergoing prenatal screening. (~70 percent of pregnant women).
Results
The response rate to the survey was 76 percent (28/37). A breakdown by type of
respondent is provided in Table 1.
According to the responses, ten surveillance
systems in eight provinces/territories collect
congenital anomalies data. Four reproductive care programs (RCPs), three maternal
serum screening/medical genetics programs
and three provincial/territorial ministries of
health operate surveillance systems.
Chronic Diseases in Canada
The surveillance systems employ multiple
sources of data, with the exception of the
Ontario RCP, which uses only hospital
records, and Yukon’s Fetal Alcohol Spectrum Disorder (FASD) registry, which relies
only on physician reports.
Seven of the ten surveillance systems collect
data on all major birth defects, while three
are more limited in the anomalies they monitor. The Yukon registry collects data on
FASD, Newfoundland and Labrador's
Medical Genetics Program collects data on
neural tube defects (NTDs), and the Ontario
Maternal Serum Screening Program focuses
37
on NTDs, trisomies 18 and 21 and other
cytogenetic and ultrasound abnormalities.
Reproductive care programs, maternal
serum screening programs and the medical
genetics programs gather data until discharge from hospital or shortly thereafter.
This is unlike the surveillance systems run
by provincial and territorial Ministries of
Health (i.e., Alberta, Yukon and British
Columbia), which capture data on infants up
to one year of age, up to school age, and up
to 19 years of age, respectively.
Respondents were asked to list the ways that
they had used congenital anomalies data in
Vol 27, No 1, 2006
the previous year. Eight (80 percent) replied
that they conducted routine statistical monitoring, five (50 percent) used the data for
epidemiological studies, and three (30 percent) used the data for monitoring outbreaks
and cluster investigation. Other uses of the
data included identifying cases for other
epidemiological studies, evaluating public
health programs and identifying individuals
for referral to specialized services.
Further details on the surveillance systems
are presented in Table 2.
Discussion
At the time of this survey, seven provinces
and one territory had congenital anomalies
surveillance systems. However, variations in
coding, outcomes captured and case ascertainment make it difficult to compare rates
across the country.
The ability to compare numbers and rates
across provinces and territories is valuable,
especially in regards to congenital anomalies. When rare events are studied, the sample size must often be increased to beyond
that which is captured by one province or
territory. If a new teratogen appears, its
effects may be more rapidly detected if comparisons can be made across several
jurisdictions.
A national surveillance system, the
Canadian Congenital Anomalies Surveillance System (CCASS), does exist. This is the
only population-based surveillance system
in Canada which provides national data on
congenital anomalies. However, it has
several limitations that hinder its usefulness.
CCASS relies primarily on hospital separations to calculate congenital anomaly rates.
This reliance on administrative databases
results in issues with timeliness and
representativity (i.e., prenatal diagnoses of
congenital anomalies that result in a termination of pregnancy are not captured). As
well, key data elements are not available,
such as the gestational age of the infant.
Major congenital anomalies are a leading
cause of death in infants,2 and create a considerable emotional and economic burden
for families and society.7–8 Surveillance
systems make vital contributions to our
Vol 27, No 1, 2006
knowledge of causative factors and to the
evaluation of preventive measures.
Congenital anomalies surveillance is important to public health and should be promoted within all provinces and territories.
The Canadian Congenital Anomalies Surveillance Network is taking the lead by
working to develop guidelines for coding, a
list of suggested congenital anomalies that
should be captured, and recommended data
collection practices.
7.
Hunfeld JA, Tempels A, Passchier J,
Hazebroek FW, Tibboel D. Brief report:
parental burden and grief one year after the
birth of a child with a congenital anomaly. J
Pediatr Psychol. 1999 Dec;24(6):515–20.
8.
Waitzman NJ, Romano PS, Scheffler RM.
Estimates of the economic costs of birth
defects. Inquiry. 1994 Summer;31(2):188–
205.
A review of existing case definitions has
already begun and preliminary recommendations have been developed. Once finalized, these guidelines and recommendations
will be distributed to provincial and territorial representatives, and posted on the
CCASN Web site. (http://www.phac-aspc.
gc.ca/ccasn-rcsac/index.html)
Acknowledgements
We would like to thank all of our survey respondents for providing us with the information on their congenital anomalies registries
and surveillance systems.
References
1.
Speirs AL. Thalidomide and congenital
abnormalities. Lancet. 1962 Feb 10;1:303–
305.
2.
Health Canada. Canadian perinatal health
report, 2003. Ottawa: Minister of Public
Works and Government Services Canada,
2003.
3.
Edmonds LD, Layde PM, James LM, Flynt
JW, Erickson JD, Oakley GP Jr. Congenital
malformations surveillance: two American
systems. Int J Epidemiol. 1981 Sep;10(3):
247–52.
4.
Cordero JF. Registries of birth defects and
genetic diseases. Pediatr Clin North Am.
1992 Feb;39(1):65–77.
5.
Lechat MF, Dolk H. Registries of congenital
anomalies: EUROCAT. Environ Health
Perspect. 1993 Jul;101 Suppl 2:153–7.
6.
Miller LA, Kirby RS. Neural tube defects surveillance: a national survey. Teratology.
2000 Jan-Feb;61(1–2):28–32.
38
Chronic Diseases in Canada
Status Report
Easy access to chronic disease surveillance information:
The NCD Surveillance Infobase
What is the percentage of current smokers in
the Durham region of Ontario?
Which gender has a higher age-standardized
hospital discharge rate for chronic, obstructive pulmonary disease in Alberta?
What is the age-standardized mortality rate
for ischemic heart disease in the St. John’s
region relative to all the other Newfoundland
and Labrador health regions?
Has the incidence trend for stomach cancer in
Nova Scotia been going up or down?
If you want to find answers to these and
many other surveillance questions, why not
try out the NCD Infobase? The Non-Communicable Diseases (NCD) Surveillance
Infobase is one of a number of Internetbased Web tools used to disseminate surveillance information at the Public Health
Agency of Canada.
Chronic Diseases in Canada
The Infobase profiles the epidemiology of
chronic diseases in Canada— including most
current rates for cancers, cardiovascular and
respiratory diseases—and provides analysis
by province/territory and by regional health
unit. Demographic, mortality, morbidity,
risk factor and related health care data are
currently available. Infobase is designed
with advanced Internet technology to provide users with interactive, dynamic access
to an extensive database of chronic disease
statistics and allows for their presentation as
tables, graphs or maps. Multiple-area comparisons, morbidity and mortality time
trends, birth cohort mortality trends and proportional mortality trends are just some of
the options available.
The NCD Infobase evolved from its predecessor, the Global Cardiovascular Disease
(CVD) Infobase. The CVD was developed
seven years ago by the Ottawa Hospital as
39
part of its role as a Canadian Collaborating
Centre for Cardiovascular Disease for the
World Health Organization.
The NCD Infobase is under constant development, with new data being added as they
become available. Future enhancements will
include facilitated user-interfaces and healthregion-specific summary pages. Feedback
and suggestions are welcome through the
“contact us” link.
You can bookmark the NCD and CVD Web
sites using the links below:
The Non-Communicable Disesase Infobase:
http://www.cvdinfobase.ca/surveillance
The Global Cardiovascular
http://www.cvdinfobase.ca
Infobase:
Vol 27, No 1, 2006
Calendar of Events
16–19 May 2006
Denver, Colorado, USA
Centers for Chronic Disease Control and
Prevention
2006 CDC Diabetes and Obesity Conference
<http://www.cdc.gov/diabetes/conferences/>
28–31 May 2006
Vancouver, British Columbia,
Canada
Canadian Public Health Association
97th Annual Conference
e-mail: [email protected]
<http://www.cpha.ca/english/conf/conf97/97confe.htm>
21–22 April 2006
Halifax, Nova Scotia, Canada
11th Annual Atlantic Canada
Cardiovascular Congress
e-mail: [email protected]
21–24 June 2006
Seattle, Washington, USA
2nd North American Congress of
Epidemiology
8–12 July 2006
Washington, DC, USA
UICC World Cancer Congress
e-mail: [email protected]
<http://www.2006conferences.org/u-index.php>
11-18 August 2006
Vancouver, British Columbia,
Canada
Cancer in Women
e-mail: [email protected]
21–25 August 2006
Rio de Janeiro, Brazil
World Federation of Public Health
Associations (WFPHA)
11th World Congress on Public Health
<http://www.saudecoletiva2006.com.br>
2–6 September 2006
Paris, France
Joint ISEE/ISEA International Conference
on Environmental Epidemiology and
Exposure
<http://www.paris2006.afsse.fr/>
3–8 September 2006
Sydney, Australia
International Association for the Study of
Obesity
10th International Conference on Obesity
<http://www.ico2006.com>
17–21 September 2006
Geneva, Switzerland
International Society of Paediatric
Oncology 38th SIOP Congress
<http://www.siop.nl>
26–29 October 2006
Berlin, Germany
The World Congress on Controversies in
Obesity, Diabetes and Hypertension
e-mail: [email protected]
<http://www.codhy.com>
3–6 December 2006
Winnipeg, Manitoba, Canada
7th Canadian Immunization Conference
<http://www.phac-aspc.gc.ca/cnic-ccni/index.html>
3–7 December 2006
Cape Town, South Africa
International Diabetes Federation
19th World Diabetes Congress
e-mail: [email protected]
<http://www.idf2006.org>
Vol 27, No 1, 2006
40
<http://www.epicongress2006.org>
Chronic Diseases in Canada
2005 Peer Reviewers
We are extremely grateful to the following people for their enormous contribution to Chronic Diseases in
Canada as peer reviewers in 2005.
Tye Arbuckle
Nicole Hébert-Croteau
Marie-Élise Parent
Sten Ardal
Greg Hislop
Mike Patterson
Lynne Baillie
Michael Jerrett
Stuart Peacock
Chris Bajdik
Michel Joffres
Linda Pederson
Linda Bartlett
Ken Johnson
Pierre Philippe
Ugis Bickis
Julia Knight
William Rickert
Sue Bondy
Betsy Kristjansson
Will Rickett
Larry Chambers
Fabrice Larribe
L Dawn Satterfield
Yue Chen
Adrian Levy
Jean Shoveller
Mary Chipman
Francine Lortie-Monette
Jack Siemiatycki
Bernard Choi
Sora Ludwig
Richard Stanwick
Linda Cook
Lynne MacLean
Tom Stephens
John Cunningham
Doug Manuel
David L. Streiner
Carl D’Arcy
Yang Mao
Larry Svenson
Philippe de Wals
Loraine Marrett
Valerie Tarasuk
Alain Demers
Christopher Martin
Gilles Thériault
Linda Dodds
Catherine McCourt
Jean-Pierre Thouez
Sheila Dubois
Ian McDowell
Jeff Whitehead
David Feeny
Steven McFaull
Andy Wielgosz
Christine Friedenreich
David McLean
Cam Wild
Linda Geiss
Wayne Millar
Kathryn Wilkins
Katherine Gray-Donald
Howard Morrison
Noreen Willows
Eva Grunfeld
Cameron Mustard
Christina Wolfson
Brian Habbick
Stephen Newman
Jill Hamilton
Robert Pampalon
Chronic Diseases in Canada
41
Vol 27, No 1, 2006
Indexes for Volume 26, 2005
Volume 26 Contents
No 1, 2005
Influential observations in weighted analyses: Examples
from the National Longitudinal Survey of Children and
Youth (NLSCY) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
An analysis of the effect of selection bias on the association of
hormone replacement therapy and breast cancer risk . . . .
73
Ilona Csizmadi, Christine M Friedenreich,
Heather E Bryant and Kerry S Courneya
1
A comparison of measures of socioeconomic status for
adolescents in a Canadian national health survey . . . . . . .
IJennifer J Macnab, JJ Koval, KN Speechley, and
MK Campbell
The estimation of heritability for twin data based on
concordances of sex and disease . . . . . . . . . . . . . . . . . . .
9
Workshop Report
Hongzhuan Tan, Mark Walker, France Gagnon, and
Shi Wu Wen
Breast cancer trends in Manitoba: 40 years of follow-up . . .
Occupational cancer surveillance . . . . . . . . . . . . . . . . . . .
13
2004 Peer Reviewers. . . . . . . . . . . . . . . . . . . . . . . . . . . .
91
Calendar of Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
92
20
No 4, 2005
Michael O Chaiton, Neil E Collishaw, and Aaron J Callard
Trends in mortality from diabetes mellitus in Canada,
1986–2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Prevalence and predictors of depression in elderly Canadians:
The Canadian Study of Health and Aging . . . . . . . . . . . . . 93
25
Truls qstbye, Betsy Kristjansson, Gerry Hill,
Stephen C Newman, Rebecca N Brouwer and
Ian McDowell
Jinfu Hu, Glenn Robbins, Anne-Marie Ugnat, and
Chris Waters
Status Report
The Canadian Incidence Study of Reported Child Abuse and
Neglect (CIS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Measurement issues related to the evaluation and monitoring of
major depression prevalence in Canada . . . . . . . . . . . . . . 100
Scott B Patten, Jian Li Wang, Cynthia A Beck and
Colleen J Maxwell
30
Ambika Dewan and Lil Tonmyr
Calendar of Events . . . . . . . . . . . . . . . . . . . . . . . . . . . .
32
Indexes for Volume 25, 2004 . . . . . . . . . . . . . . . . . . . . . .
33
A descriptive analysis of Canadian youth treated in emergency
departments for work-related injuries . . . . . . . . . . . . . . . . 107
Tammy Lipskie and F Curtis Breslin
Acceptability of micronutrient sprinkles: A new food-based
approach for delivering iron to First Nations and Inuit children
in Northern Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
No 2–3, 2005
Mediterranean dietary components and body mass index in
adults: The Peel Nutrition and Heart Health Survey . . . . . .
Anna Christofides, Claudia Schauer, Waseem Sharieff
and Stanley H Zlotkin
43
Mamdouh M Shubair, R Stephen McColl and
Rhona M Hanning
Screening mammography participation and invitational
strategy: The Quebec Breast Cancer Screening Program,
1998–2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Letters
The seasonality of SIDS in Canada between 1985–1989 and
1994–1998. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
52
Sonia Jean, Diane Major, Louise Rochette and
Jacques Brisson
Status Reports
Child maltreatment in different populations: Investigated cases
and Canadian street youth. . . . . . . . . . . . . . . . . . . . . . . . 124
An observational study of sun and heat protection during Canada
Day outdoor celebrations, 2003 . . . . . . . . . . . . . . . . . . . .
59
Richard DeMarco, Michelle Wesley, Cara Bowman,
Susanne Shields and Tom Wong
ST David, U Chandran,D Paquette, D Scholten, J Wilson,
E Galanis, M Becker, F Crane, R Lester, T Mersereau,
E Wong and D Carr
Situational analysis of chronic disease surveillance in Canada:
Results of a stakeholder interview . . . . . . . . . . . . . . . . . . 127
Validity of a 12-item version of the CES-D used in the National
Longitudinal Study of Children and Youth . . . . . . . . . . . .
65
Bernard CK Choi, Elizabeth Wright and Ulrick Auguste
Calendar of Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Christiane Poulin, Denise Hand and Brock Boudreau
Vol 27, No 1, 2006
90
Jennifer I Payne
Alain A Demers, Donna Turner, Daojun Mo, and
Erich V Kliewer
Smoker preference for “elastic cigarettes” in the Canadian
cigarette market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
80
Beth K Potter, Kathy N Speechley, Iris A Gutmanis,
M Karen Campbell, John J Koval and Douglas Manuel
42
Chronic Diseases in Canada
Volume 26 Subject Index
ABORIGINAL HEALTH
HEALTH SURVEYS
Acceptability of micronutrient sprinkles: A new food-based
approach for delivering iron to First Nations and Inuit children in
Northern Canada. 26(4):114-120
Influential observations in weighted analyses: Examples from the
National Longitudinal Survey of Children and Youth (NLSCY).
26(1):1-8
Mediterranean dietary components and body mass index in
adults: The Peel Nutrition and Heart Health Survey. 26(2-3):43-51
ALTERNATIVE MEDICINE
BOOK REVIEWS
CANCER
Validity of a 12-item version of the CES-D used in the National
Longitudinal Study of Children and Youth. 26(2-3):65-72
Breast cancer trends in Manitoba: 40 years of follow-up.
26(1):13-19
A comparison of measures of socioeconomic status for
adolescents in a Canadian national health survey. 26(2-3):80-89
Screening mammography participation and invitational strategy:
The Quebec Breast Cancer Screening Program, 1998 – 2000.
26(2-3):52-58
Prevalence and predictors of depression in elderly Canadians:
The Canadian Study of Health and Aging. 26(2-3):93-99
An analysis of the effect of selection bias on the association of
hormone replacement therapy and breast cancer risk.
26(2-3):73-79
HEART DISEASE
Mediterranean dietary components and body mass index in
adults: The Peel Nutrition and Heart Health Survey. 26(2-3):43-51
Occupational cancer surveillance. 26(2-3):90-91
INFANT AND CHILD HEALTH
CEREBROVASCULAR DISEASES
DIABETES
The estimation of heritability for twin data based on
concordances of sex and disease. 26(1):9-12
Trends in mortality from diabetes mellitus in Canada, 1986–2000.
26(1):25-29
The Canadian Incidence Study of Reported Child Abuse and
Neglect (CIS). 26(1):30-31
ECONOMIC BURDEN
ENVIRONMENTAL HEALTH
A descriptive analysis of Canadian youth treated in emergency
departments for work-related injuries. 26(4):107-113
An observational study of sun and heat protection during Canada
Day outdoor celebrations, 2003. 26(2-3):59-64
Child maltreatment in different populations: Investigated cases
and Canadian street youth. 26(4):124-126.
FOOD ISSUES
The seasonality of SIDS in Canada between 1985–1989 and
1994–1998. 26(4):121-123
Mediterranean dietary components and body mass index in
adults: The Peel Nutrition and Heart Health Survey. 26(2-3):43-51
INJURIES
Acceptability of micronutrient sprinkles: A new food-based
approach for delivering iron to First Nations and Inuit children in
Northern Canada. 26(4):114-120
A descriptive analysis of Canadian youth treated in emergency
departments for work-related injuries. 26(4):107-113
GEOGRAPHIC VARIATIONS
MENTAL HEALTH
Trends in mortality from diabetes mellitus in Canada, 1986–2000.
26(1):25-29
Prevalence and predictors of depression in elderly Canadians:
The Canadian Study of Health and Aging. 26(4):93-99
Measurement issues related to the evaluation and monitoring of
major depression prevalence in Canada. 26(4):100-106
Chronic Diseases in Canada
43
Vol 27, No 1, 2006
OCCUPATIONAL HEALTH
TOBACCO ISSUES
A descriptive analysis of Canadian youth treated in emergency
departments for work-related injuries. 26(4):107-113
Smoker preference for “elastic cigarettes” in the Canadian
cigarette market. 26(1):20-24
Occupational cancer surveillance. 26(2-3):90-91
TRAUMA
POPULATION SURVEILLANCE
WOMEN’S HEALTH
Trends in mortality from diabetes mellitus in Canada, 1986–2000.
26(1):25-29
Screening mammography participation and invitational strategy:
The Quebec Breast Cancer Screening Program, 1998 – 2000.
26(2-3):52-58
The Canadian Incidence Study of Reported Child Abuse and
Neglect (CIS). 26(1):30-31
An analysis of the effect of selection bias on the association of
hormone replacement therapy and breast cancer risk.
26(2-3):73-79
Child maltreatment in different populations: Investigated cases
and Canadian street youth. 26(4):124-126.
Occupational cancer surveillance. 26(2-3):90-91
Situational analysis of chronic disease surveillance in Canada:
Results of a stakeholder interview. 26(4):127-129
Vol 27, No 1, 2006
44
Chronic Diseases in Canada
Volume 26 Author Index
Auguste, Ulrick
Bernard CK Choi, Elizabeth Wright and Ulrick Auguste.
Situational analysis of chronic disease surveillance in
Canada: Results of a stakeholder interview. 26(4):127-129
Callard, Aaron J
Michael O Chaiton, Neil E Collishaw and Aaron J Callard.
Smoker preference for “elastic cigarettes” in the Canadian
cigarette market. 26(1):20-24
Beck, Cynthia A
Scott B Patten, Jian Li Wang, Cynthia A Beck and Colleen J
Maxwell. Measurement issues related to the evaluation and
monitoring of major depression prevalence in Canada.
26(4):100-106
Campbell, MK
Jennifer J Macnab, JJ Koval, KN Speechley and MK Campbell.
Influential observations in weighted analyses: Examples
from the National Longitudinal Survey of Children and
Youth (NLSCY). 26(1):1-8
Becker, M
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr.An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Campbell, M Karen
Beth K Potter, Kathy N Speechley, Iris A Gutmanis, M Karen
Campbell, John J Koval and Douglas Manuel. A comparison
of measures of socioeconomic status for adolescents in a
Canadian national health survey. 26(2-3):80-89
Carr, D
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Boudreau, Brock
Christiane Poulin, Denise Hand and Brock Boudreau. Validity of
a 12-item version of the CES-D used in the National
Longitudinal Study of Children and Youth. 26(2-3):65-72
Bowman, Cara
Richard De Marco, Michelle Wesley, Cara Bowman, Susanne
Shields and Tom Wong. Child maltreatment in different
populations: Investigated cases and Canadian street youth.
26(4):124-126
Chaiton, Michael O
Michael O Chaiton, Neil E Collishaw and Aaron J Callard.
Smoker preference for “elastic cigarettes” in the Canadian
cigarette market. 26(1):20-24
Breslin, F Curtis
Shaw Amanda K, Morrison Howard I, Speechley Kathy N,
Maunsell Elizabeth, Barrera Maru, Schanzer Dena, Pogany
Lisa and Desmeules Marie. The Late Effects Study: Design
and subject representativeness of a Canadian, multi-centre
study of the late effects of childhood cancer. 25(3/4):119-26.
Collinshaw, Neil E
Michael O Chaiton, Neil E Collishaw and Aaron J Callard.
Smoker preference for “elastic cigarettes” in the Canadian
cigarette market. 26(1):20-24
Chandran, U
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Bell, Neil R
Tammy Lipskie and F Curtis Breslin. A descriptive analysis of
Canadian youth treated in emergency departments for
work-related injuries. 26(4):107-113
Brouwer, Rebecca
Truls qstbye, Betsy Kristjansson, Gerry Hill, Stephen C Newman,
Rebecca N Brouwer and Ian McDowell. Prevalence and
predictors of depression in elderly Canadians: The Canadian
Study of Health and Aging. 26(4):93-99
Choi, Bernard CK
Bernard CK Choi, Elizabeth Wright and Ulrick Auguste.
Situational analysis of chronic disease surveillance in
Canada: Results of a stakeholder interview. 26(4):127-129
Bryant, Heather E
Ilona Csizmadi, Christine M Friedenreich, Heather E Bryant and
Kerry S Courneya. An analysis of the effect of selection bias
on the association of hormone replacement therapy and
breast cancer risk. 26(2-3):73-79
Christofides, Anna
Anna Christofides, Claudia Schauer, Waseem Sharieff and
Stanley H Zlotkin. Acceptability of micronutrient sprinkles:
A new foodbased approach for delivering iron to First
Nations and Inuit children in Northern Canada.
26(4):114-120
Chronic Diseases in Canada
45
Vol 27, No 1, 2006
Courneya, Kerry S
Ilona Csizmadi, Christine M Friedenreich, Heather E Bryant and
Kerry S Courneya. An analysis of the effect of selection bias
on the association of hormone replacement therapy and
breast cancer risk. 26(2-3):73-79
Galanis, E
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Crane, F
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Gutmanis, Iris A
Beth K Potter, Kathy N Speechley, Iris A Gutmanis, M Karen
Campbell, John J Koval and Douglas Manuel. A comparison
of measures of socioeconomic status for adolescents in a
Canadian national health survey. 26(2-3):80-89
Csizmadi, Ilona
Ilona Csizmadi, Christine M Friedenreich, Heather E Bryant and
Kerry S Courneya. An analysis of the effect of selection bias
on the association of hormone replacement therapy and
breast cancer risk. 26(2-3):73-79
Hand, Denise
Christiane Poulin, Denise Hand and Brock Boudreau. Validity of
a 12-item version of the CES-D used in the National
Longitudinal Study of Children and Youth. 26(2-3):65-72
Hunter, Duncan JW
Hunter Duncan JW, Grant Heather J, Purdue Mark PH, Spasoff
Robert A, Dorland John L and Bains Nam. An
epidemiology-based needs assessment for stroke services.
25(3/4):138-46.
David, ST
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Hanning, Rhonda M
Mamdouh M Shubair, R Stephen McColl and Rhona M Hanning.
Contents
De Marco, Richard
Richard De Marco, Michelle Wesley, Cara Bowman, Susanne
Shields and Tom Wong. Child maltreatment in different
populations: Investigated cases and Canadian street youth.
26(4):124-126
Hill, Gerry
Truls qstbye, Betsy Kristjansson, Gerry Hill, Stephen C Newman,
Rebecca N Brouwer and Ian McDowell. Prevalence and
predictors of depression in elderly Canadians: The Canadian
Study of Health and Aging. 26(4):93-99
Demers, Alain A
Alain A Demers, Donna Turner, Daojun Mo and Erich V Kliewer.
Breast cancer trends in Manitoba: 40 years of follow-up.
26(1):13-19
Hu, Jinfu
Jinfu Hu, Glenn Robbins, Anne-Marie Ugnat and Chris Waters.
Trends in mortality from diabetes mellitus in Canada,
1986–2000. 26(1):25-29
Dewan, Ambika
Ambika Dewan and Lil Tonmyr. The Canadian Incidence Study
of Reported Child Abuse and Neglect (CIS). 26(1):1-8:30-31
Jean, Sonia
Sonia Jean, Diane Major, Louise Rochette and Jacques Brisson.
Screening mammography participation and invitational
strategy: The Quebec Breast Cancer Screening Program,
1998 – 2000. 26(2-3):52-58
Friedenreich, Christine M
Ilona Csizmadi, Christine M Friedenreich, Heather E Bryant and
Kerry S Courneya. An analysis of the effect of selection bias
on the association of hormone replacement therapy and
breast cancer risk. 26(2-3):73-79
Kliewer, Erich V
Alain A Demers, Donna Turner, Daojun Mo and Erich V Kliewer.
Breast cancer trends in Manitoba: 40 years of follow-up.
26(1):13-19
Gagnon, France
Hongzhuan Tan, Mark Walker, France Gagnon and Shi Wu Wen.
The estimation of heritability for twin data based on
concordances of sex and disease. 26(1):9-12
Vol 27, No 1, 2006
46
Chronic Diseases in Canada
Koval, John J
Beth K Potter, Kathy N Speechley, Iris A Gutmanis, M Karen
Campbell, John J Koval and Douglas Manuel. A comparison
of measures of socioeconomic status for adolescents in a
Canadian national health survey. 26(2-3):80-89
McColl, R Stephen
Mamdouh M Shubair, R Stephen McColl and Rhona M Hanning.
Mediterranean dietary components and body mass index in
adults: The Peel Nutrition and Heart Health Survey.
26(2-3):43-51
Koval, JJ
Jennifer J Macnab, JJ Koval, KN Speechley and MK Campbell.
Influential observations in weighted analyses: Examples
from the National Longitudinal Survey of Children and
Youth (NLSCY). 26(1):1-8
McDowell, Ian
Truls qstbye, Betsy Kristjansson, Gerry Hill, Stephen C Newman,
Rebecca N Brouwer and Ian McDowell. Prevalence and
predictors of depression in elderly Canadians: The Canadian
Study of Health and Aging. 26(4):93-99
Kristjansson, Betsy
Truls qstbye, Betsy Kristjansson, Gerry Hill, Stephen C Newman,
Rebecca N Brouwer and Ian McDowell. Prevalence and
predictors of depression in elderly Canadians: The Canadian
Study of Health and Aging. 26(4):93-99
Mersereau, T
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Lester, F
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Mo, Daojun
Alain A Demers, Donna Turner, Daojun Mo and Erich V Kliewer.
Breast cancer trends in Manitoba: 40 years of follow-up.
26(1):13-19
Lipskie, Tammy
Tammy Lipskie and F Curtis Breslin. A descriptive analysis of
Canadian youth treated in emergency departments for
work-related injuries. 26(4):107-113
Newman, Stephen C
Truls qstbye, Betsy Kristjansson, Gerry Hill, Stephen C Newman,
Rebecca N Brouwer and Ian McDowell. Prevalence and
predictors of depression in elderly Canadians: The Canadian
Study of Health and Aging. 26(4):93-99
Macnab, Jennifer J
Jennifer J Macnab, JJ Koval, KN Speechley and MK Campbell.
Influential observations in weighted analyses: Examples
from the National Longitudinal Survey of Children and
Youth (NLSCY). 26(1):1-8
qstbye, Truls
Truls qstbye, Betsy Kristjansson, Gerry Hill, Stephen C Newman,
Rebecca N Brouwer and Ian McDowell. Prevalence and
predictors of depression in elderly Canadians: The Canadian
Study of Health and Aging. 26(4):93-99
Major, Diane
Sonia Jean, Diane Major, Louise Rochette and Jacques Brisson.
Screening mammography participation and invitational
strategy: The Quebec Breast Cancer Screening Program,
1998 – 2000. 26(2-3):52-58
Paquette, D
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Manuel, Douglas
Beth K Potter, Kathy N Speechley, Iris A Gutmanis, M Karen
Campbell, John J Koval and Douglas Manuel. A comparison
of measures of socioeconomic status for adolescents in a
Canadian national health survey. 26(2-3):80-89
Patten, Scott B
Scott B Patten, Jian Li Wang, Cynthia A Beck and Colleen J
Maxwell. Measurement issues related to the evaluation and
monitoring of major depression prevalence in Canada.
26(4):100-106
Maxwell, Colleen J
Scott B Patten, Jian Li Wang, Cynthia A Beck and Colleen J
Maxwell. Measurement issues related to the evaluation and
monitoring of major depression prevalence in Canada.
26(4):100-106
Chronic Diseases in Canada
Payne, Jennifer I
Jennifer I Payne. Occupational cancer surveillance. 26(2-3):90-91
47
Vol 27, No 1, 2006
Potter, Beth K
Beth K Potter, Kathy N Speechley, Iris A Gutmanis, M Karen
Campbell, John J Koval and Douglas Manuel. A comparison
of measures of socioeconomic status for adolescents in a
Canadian national health survey. 26(2-3):80-89
Shubair, Mamdouh M
Mamdouh M Shubair, R Stephen McColl and Rhona M Hanning.
Mediterranean dietary components and body mass index in
adults: The Peel Nutrition and Heart Health Survey.
26(2-3):43-51
Poulin, Christiane
Christiane Poulin, Denise Hand and Brock Boudreau. Validity of
a 12-item version of the CES-D used in the National
Longitudinal Study of Children and Youth. 26(2-3):65-72
Speechley, KN
Jennifer J Macnab, JJ Koval, KN Speechley and MK Campbell.
Influential observations in weighted analyses: Examples
from the National Longitudinal Survey of Children and
Youth (NLSCY). 26(1):1-8
Robbins, Glenn
Jinfu Hu, Glenn Robbins, Anne-Marie Ugnat and Chris Waters.
Trends in mortality from diabetes mellitus in Canada,
1986–2000. 26(1):25-29
Speechley, Kathy N
Beth K Potter, Kathy N Speechley, Iris A Gutmanis, M Karen
Campbell, John J Koval and Douglas Manuel. A comparison
of measures of socioeconomic status for adolescents in a
Canadian national health survey. 26(2-3):80-89
Rochette, Louise
Sonia Jean, Diane Major, Louise Rochette and Jacques Brisson.
Screening mammography participation and invitational
strategy: The Quebec Breast Cancer Screening Program,
1998 – 2000. 26(2-3):52-58
Tan, Hongzhuan
Hongzhuan Tan, Mark Walker, France Gagnon and Shi Wu Wen.
The estimation of heritability for twin data based on
concordances of sex and disease. 26(1):9-12
Brisson, Jacques
Sonia Jean, Diane Major, Louise Rochette and Jacques Brisson.
Screening mammography participation and invitational
strategy: The Quebec Breast Cancer Screening Program,
1998 – 2000. 26(2-3):52-58
Tonmyr, Lil
Ambika Dewan and Lil Tonmyr. The Canadian Incidence Study
of Reported Child Abuse and Neglect (CIS). 26(1):1-8:30-31
Turner, Donna
Alain A Demers, Donna Turner, Daojun Mo and Erich V Kliewer.
Breast cancer trends in Manitoba: 40 years of follow-up.
26(1):13-19
Schauer, Claudia
Anna Christofides, Claudia Schauer, Waseem Sharieff and
Stanley H Zlotkin. Acceptability of micronutrient sprinkles:
A new foodbased approach for delivering iron to First
Nations and Inuit children in Northern Canada.
26(4):114-120
Ugnat, Anne-Marie
Jinfu Hu, Glenn Robbins, Anne-Marie Ugnat and Chris Waters.
Trends in mortality from diabetes mellitus in Canada,
1986–2000. 26(1):25-29
Scholten, D
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Walker, Mark
Hongzhuan Tan, Mark Walker, France Gagnon and Shi Wu Wen.
The estimation of heritability for twin data based on
concordances of sex and disease. 26(1):9-12
Sharieff, Waseem
Anna Christofides, Claudia Schauer, Waseem Sharieff and
Stanley H Zlotkin. Acceptability of micronutrient sprinkles:
A new foodbased approach for delivering iron to First
Nations and Inuit children in Northern Canada.
26(4):114-120
Wang, Jian Li
Scott B Patten, Jian Li Wang, Cynthia A Beck and Colleen J
Maxwell. Measurement issues related to the evaluation and
monitoring of major depression prevalence in Canada.
26(4):100-106
Waters, Chris
Jinfu Hu, Glenn Robbins, Anne-Marie Ugnat and Chris Waters.
Trends in mortality from diabetes mellitus in Canada,
1986–2000. 26(1):25-29
Shields, Susanne
Richard De Marco, Michelle Wesley, Cara Bowman, Susanne
Shields and Tom Wong. Child maltreatment in different
populations: Investigated cases and Canadian street youth.
26(4):124-126
Vol 27, No 1, 2006
48
Chronic Diseases in Canada
Wen, Shi Wu
Hongzhuan Tan, Mark Walker, France Gagnon and Shi Wu Wen.
The estimation of heritability for twin data based on
concordances of sex and disease. 26(1):9-12
Wong, E
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Wesley, Michelle
Richard De Marco, Michelle Wesley, Cara Bowman, Susanne
Shields and Tom Wong. Child maltreatment in different
populations: Investigated cases and Canadian street youth.
26(4):124-126
Wright, Elizabeth
Bernard CK Choi, Elizabeth Wright and Ulrick Auguste.
Situational analysis of chronic disease surveillance in
Canada: Results of a stakeholder interview. 26(4):127-129
Wilson, J
ST David, U Chandran, D Paquette, D Scholten, J Wilson, E
Galanis, M Becker, F Crane, R Lester, T Mersereau, E Wong
and D Carr. An observational study of sun and heat
protection during Canada Day outdoor celebrations, 2003.
26(2-3):59-64
Zlotkin, Stanley H
Anna Christofides, Claudia Schauer, Waseem Sharieff and
Stanley H Zlotkin.
Acceptability of micronutrient sprinkles: A new foodbased
approach for delivering iron to First Nations and Inuit
children in Northern Canada. 26(4):114-120
Wong, Tom
Richard De Marco, Michelle Wesley, Cara Bowman, Susanne
Shields and Tom Wong. Child maltreatment in different
populations: Investigated cases and Canadian street youth.
26(4):124-126
Chronic Diseases in Canada
49
Vol 27, No 1, 2006
CDIC: Information for Authors
Chronic Diseases in Canada (CDIC) is a
quarterly scientific journal focussing on
the prevention and control of noncommunicable diseases and injuries in
Canada. Its feature articles are peer
reviewed. The content of articles may
include research from such fields as epidemiology, public/community health,
bio- statistics, the behavioural sciences,
and health services or economics. CDIC
endeavours to foster communication on
chronic diseases and injuries among
public health practitioners, epidemiologists and researchers, health policy
planners and health educators. Submissions are selected based on scientific
quality, public health relevance, clarity,
conciseness and technical accuracy.
Although CDIC is a publication of the
Public Health Agency of Canada, contributions are welcomed from both the
public and private sectors. Authors retain
responsibility for the contents of their
papers, and opinions expressed are not
necessarily those of the CDIC editorial
committee nor of the Public Health
Agency of Canada.
Article Types
Peer-reviewed Feature Article: Maximum 4,000 words for main text body
(excluding abstract, tables, figures, references) in the form of original research,
surveillance reports, meta-analyses or
methodological papers.
Status Report: Describe ongoing national
programs, studies or information systems
bearing on
Canadian public health
(maximum 3,000 words). Abstract not
required.
Workshop/Conference Report: Summarize significant, recently held events relating to national public health (maximum
1,200 words). Abstract not required.
(maximum 3,000 words). Abstract not Abstract: Unstructured (one paragraph,
no headings), maximum 175 words (100
required.
for short reports); include 3–8 key words
Letter to the Editor: Comments on arti(preferably from the Medical Subject
cles recently published in CDIC will be
Headings (MeSH) of Index Medicus).
considered for publication (maximum
Text: Double-spaced, 1 inch (25 mm)
500 words). Abstract not required.
margins, 12 point font size.
Book/Software Review: Usually solicited by the editors (500–1,300 words), Acknowledgements: Include disclosure
but requests to review are welcomed. of financial and material support in
acknowledgements; if anyone is credited
Abstract not required.
in acknowledgements with substantive
scientific contributions, authors should
state in cover letter that they have
Submitting Manuscripts
obtained written permission.
Submit manuscripts to the Editor-in-Chief,
Chronic Diseases in Canada, Public References: In “Vancouver style” (consult
Health Agency of Canada 130 Colonnade a recent CDIC issue for examples); numRoad, CDIC Address Locator: 6501G, bered in superscript in the order cited in
Ottawa, Ontario
K1A 0K9, e-mail: text, tables and figures; listing up to six
authors (first three and “et al.” if more);
[email protected]
without any automatic reference numberSince CDIC adheres in general (section on ing feature used in word processing; any
illustrations not applicable) to the “Uni- unpublished observations/data or perform Requirements for Manuscripts sonal communications used (discouraged)
Submitted to Biomedical Journals” as to be cited in the text in parentheses
approved by the International Committee (authors responsible for obtaining writof Medical Journal Editors, authors ten permission); authors are responsible
should refer to this document for com- for verifying accuracy of references.
plete details before submitting a manuscript to CDIC (see <www.cma.ca> or Tables and Figures: Send vector
graphics only. Each on a separate page
Can Med Assoc J 1997;156(2): 270–7).
and in electronic file(s) separate from the
text (not imported into the text body); as
self- explanatory and succinct as possible;
Checklist for Submitting
not too numerous; numbered in the order
Manuscripts
that they are mentioned in the text;
Cover letter: Signed by all authors, stating explanatory material for tables in footthat all have seen and approved the final notes, identified by lower-case superscript
manuscript and have met the authorship letters in alphabetical order; figures limincluding a full statement regarding ited to graphs or flow charts/templates
any prior or duplicate publication or sub- (no photographs), with software used
specified and titles/footnotes on a sepamission for publication.
rate page.
First title page: Concise title; full names of Number of copies: If submitting by mail,
all authors and institutional affiliations; one complete copy, including tables and
name, postal and e-mail addresses, tele- figures; one copy of any related supplephone and fax numbers for corresponding mentary material, and a copy of the
author; separate word counts for abstract manuscript on diskette. If submitting by
and text.
e-mail to [email protected],
Cross-Canada Forum: For authors to
present or exchange information and
opinions on regional or national surveilplease fax or mail the covering letter to
lance findings, programs under develop- Second title page: Title only; start page the address on the inside front cover.
ment or public health policy initiatives numbering here as page 1.
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