Available online at www.sciencedirect.

com

General Hospital Psychiatry 34 (2012) 345 351

PsychiatricMedical Comorbidity
The PsychiatricMedical Comorbidity section will focus on the prevalence and impact of psychiatric disorders in patients with chronic medical illness as well
as the prevalence and impact of medical disorders in patients with chronic psychiatric illness.

Changes in cardiovascular disease burden associated with

psychopathology in Australian adults 20042008
Evan Atlantis, Ph.D. a, b,, Thomas Sullivan, B.Ma. & Comp.Sc.(Hons) c
a

Royal Adelaide Hospital/Institute of Medical and Veterinary Science, SA Health, Government of South Australia, Adelaide, South Australia 5000, Australia
b
School of Medicine, The University of Adelaide, Adelaide, SA 5005, Australia
c
Data Management & Analysis Centre Discipline of Public Health, The University of Adelaide, Adelaide, SA 5005, Australia
Received 1 November 2011; accepted 13 February 2012

Abstract
Objective: The objective was to investigate changes in self-reported cardiovascular disease (CVD) burden associated with psychopathology
for Australia from 2004 to 2008.
Method: Data analyzed were from 32,073 participants aged 25 years from the 20042005 or 20072008 National Health Surveys.
Lifetime diagnosis of CVD (heart attack or stroke) was by self-report. Psychopathology was determined by the 10-item Kessler Psychological
Distress Scale (using scores 30) and use of antidepressants or antianxiety (AD/AA) medications.
Results: The prevalence of CVD (4.1% to 4.5%, P=.045) had increased slightly from 2004 to 2008 for the general population, but not among
those with psychopathology. On average, psychological distress only [odds ratio (OR) 2.00; 95% confidence interval, 1.522.62] and AD/
AA medications with (OR 2.02; 1.412.88) and without psychological distress (OR 1.24; 1.001.55) were associated with increased odds of
CVD over the 4-year period, independent of sociodemographic, lifestyle and chronic disease covariates. Both psychological distress only
(OR 1.61; 1.152.25) and AD/AA medications with psychological distress (OR 1.62; 1.082.44) conferred higher odds of CVD than AD/
AA medications without psychological distress.
Conclusion: In comparison to those without psychopathology, the odds of self-reported CVD were persistently higher among people with
psychopathology from 2004 to 2008, particularly for psychological distress.
2012 Elsevier Inc. All rights reserved.
Keywords: Psychological distress; Depression; Anxiety; Antidepressants; Trends

1. Introduction
The prevalence of cardiovascular disease (CVD) is
disproportionately high in people with clinically significant
depression, anxiety or psychological distress (psychopathology) [13]. Psychopathology could significantly worsen the
health and economic burden of CVD. For instance,
psychopathology in people with CVD is frequently associated with nonadherence to treatment and healthy lifestyle

Conflict of interest: There are no conflicts of interest to disclose.

Corresponding author. School of Medicine, The University of
Adelaide, Adelaide, SA 5000, Australia. Tel.: +61 468323636; fax: +61
8 8313 0355.
E-mail address: evan.atlantis@adelaide.edu.au (E. Atlantis).
0163-8343/$ see front matter 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.genhosppsych.2012.02.006

guidelines [4], low health-related quality of life scores [5],

increased CVD mortality [2] and health resource utilization,
lost productivity and functional disability [6].
Depression and CVD (ischemic heart disease and
cerebrovascular disease) were among the top six leading
causes of global burden of disease for 2004 and are projected
to move up in the rankings by 2030 [7]. However, the global
burden of CVD prevalence is likely to vary considerably
across regions and countries given the existing populationspecific data for trends in risk factors, including psychopathology. On average, global population mean total cholesterol and systolic blood pressure have now stabilized or
decreased slightly since 1980, but across regions, they fell
most in high-income countries and had increased within Asia
and Africa [8,9]. Conversely, glycemia and diabetes have

346

E. Atlantis, T. Sullivan / General Hospital Psychiatry 34 (2012) 345351

risen globally since 1980, but changes in the prevalence of

diabetes varied across regions and ranged from 2% in east
and southeast Asia to 60% in Oceania [10].
Trends in these intermediate risk factors are partly
attributable to the regional trends in body mass index
(BMI) since 1980, which also varied substantially between
countries [11]. In addition, global variability in psychopathology [12] and other well-established lifestyle risk factors
such as smoking [13], inadequate fruit and vegetable
consumption [14], and physical inactivity [15] could
undoubtedly confound CVD trends across populations.
Reliable population-specific data for trends in CVD are
therefore needed to better inform national health care policy
makers and researchers regarding these conditions. Recent
epidemiological evidence based on nationally representative
data sources shows that the prevalence of obesity doubled
and that of diabetes and high cholesterol increased from
3.9% to 6.0% and 11.4% to 14.1%, whereas the prevalence
of high blood pressure and CVD (heart attack, stroke or
angina) decreased from 21.3% to 20.1% and 6.2% to 5.3%,
in the Australian adult population between 1989 and 2005
[16]. Findings from several studies also suggest that there
have been increasing trends in the prevalence of major
depression [17], comorbid major depression and diabetes
[18], and antidepressant prescribing [19] during this period.
In a recent study, however, we showed that the prevalence of
psychological distress changed little for Australian adults
between 2004 and 2008, whereas the reported use of
antidepressant or antianxiety (AD/AA) medications decreased from 8.1% to 4.8% [3]. To our knowledge, recent
trends in CVD burden associated with psychopathology (use
of AD/AA medications and psychological distress) independent of changes in chronic disease and lifestyle risk factors
have yet to be documented at a national level. This
information would help develop health care policies and
set national health priorities for interventions. To address this
knowledge gap, we analyzed two of the most recent
nationally representative survey data sources to investigate
comprehensively changes in the odds of CVD associated
with psychopathology in Australia from 2004 to 2008.
2. Methods
2.1. Survey design and sampling procedure
Data were obtained from the 20042005 [20] and 2007
2008 [21] Australian National Health Surveys conducted by
the Australian Bureau of Statistics every 3 years; i.e., the data
come from two cross-sectional studies (surveys) of independent samples. Each survey was conducted using a stratified
multistage area sample of private dwellings. Information
about health status, use of health services, health-related
lifestyle factors, and demographic and socioeconomic
characteristics of participants were obtained by trained
interviewers from residents in urban and rural areas of
Australia. A private dwelling was predefined as a house,

flat, home unit, caravan, garage, tent and any other structure
being used as a private place of residence. Information of
interest in this study was collected using the same questions
and methods in both 20042005 and 20072008 surveys.
2.2. Study populations
The 20042005 and 20072008 surveys achieved overall
response rates of 89%, and 91% (fully responding households,
after sample loss). After excluding participants aged less than
25 years, the final crude (nonweighted) samples studied
consisted of n=17,665 and n=14,408 adults in the 20042005
and 20072008 surveys. The decision to exclude those aged
less than 25 years was based on predefined age groups across
the surveys to allow for merging of the separate data files.
2.3. Outcome measure
2.3.1. Cardiovascular disease
Lifetime diagnosis of CVD was determined by selfreport. Participants were asked if they had ever been told by a
doctor or nurse that they had a heart or circulatory condition
(using a prompt card). Classification of prevalent CVD was
for self-reported Heart attack or Stroke. Self-reported
CVD events identified using similar questions have been
shown to be accurate in 87.5% of cases compared with
adjudication of medical records and/or linkage to a hospital
morbidity database [22].
2.4. Predictors and covariates
2.4.1. Psychopathology predictors
Current symptoms of psychological distress were measured with the 10-item Kessler Psychological Distress Scale
(K10), which was developed to monitor population prevalence and trends in nonspecific psychological distress [23].
Participants were asked to rate how often, in the past 4 weeks,
they felt negative emotions on 10 question items. The K10 has
a five-value response option for each question (1) all of the
time, (2) most of the time, (3) some of the time, (4) a little of
the time and (5) none of the time that is scored in reverse.
Thus, summed scores can range from 10 (indicating no
distress) to 50 (indicating severe distress). Clinically significant psychological distress was defined using a cutoff point
of 30 for the K10, which has been shown to predict any
current anxiety or affective disorder group membership
accurately according to the Composite International Diagnostic Interview classification system and corresponds to an
increased frequency of mental health service utilization [24].
Participants were asked whether they had taken any
sleeping tablets or capsules, tablets or capsules for anxiety or
nerves, or tranquilizers, antidepressants, mood stabilizers or
other medications for mental health (using a prompt card) in
the last 2 weeks to determine contemporaneous use of
medications for mental health. This information was
collected with respect to mental well-being along with the
K10 questions. The broad groups of medications of interest
in this study were confined to AD/AA medications.

E. Atlantis, T. Sullivan / General Hospital Psychiatry 34 (2012) 345351

Psychopathology predictors (exposure vs. reference) were

classified accordingly for analysis:
1. AD/AA medications: yes vs. no (reference)
2. Psychological distress: yes vs. no (reference)
3. A composite AD/AA medication and psychological
distress variable was modeled as a three-level
predictor:
i. AD/AA medications with psychological distress vs.
none (reference)
ii. AD/AA medications without psychological distress
vs. none (reference)
iii. Psychological distress only vs. none (reference)
2.4.2. Sociodemographic and lifestyle covariates
Sociodemographic information regarding sex, age, relationship status, country of birth, usual work hours and
workforce status was collected using standard questions.
BMI was derived from self-report height and weight values
and computed as weight in kilograms divided by height in
meters squared. Standard international cutoff points were
used to define underweight (BMI b18.5 kg/m 2), healthy
weight (BMI 18.524.9 kg/m 2), overweight (BMI 2529.9
kg/m 2 ) and obesity (BMI 30 kg/m 2 ). Leisure-time
physical activity data were derived from self-report duration
and intensity of physical exercise (including walking) for
recreation, sport or health/fitness during the past 2 weeks at
the time of the interview. Total time spent in leisure-time
physical activity was multiplied by intensity weights (3.3 for
walking, 4.0 for moderate and 8.0 for vigorous intensity
exercise) to compute metabolic equivalent-minutes per week
(MET-min/wk). A MET-min/wk score of 540 is equivalent to 135 min or more per week of at least moderateintensity physical activity. Participants were asked questions
about current and historical smoking behaviors to assess
smoking status.
A detailed description of methods used to collect
information about diet and alcohol consumption has been
described elsewhere [3]. Briefly, daily fruit and vegetable
intake was assessed using short semiquantitative dietary
questions. Information about past-week alcohol consumption was recorded against 10 general categories of alcoholic
drinks, and the size and number of drinks consumed were
used to obtain the amount of pure alcohol consumed.
2.4.3. Chronic disease covariates
Lifetime diagnosis of diabetes (type 1 or type 2), high
blood pressure and high cholesterol was determined by selfreport. As with CVD, participants were asked if they had
ever been told by a doctor or nurse that they had any of these
specific conditions.
2.5. Statistical analyses
Statistical analyses were completed using SPSS version
17.0 (SPSS Inc., Chicago, IL, US). All data were age within
sex standardized to the Australian population distributions
according to the 2001 Census of Population and Housing.

347

Changes over time in the standardized prevalence of CVD

(outcome measure) and associations with psychopathology,
sociodemographic, lifestyle and chronic disease variables
were assessed using logistic regression models, with effect
estimates expressed as odds ratios (ORs) with 95%
confidence intervals (CIs). Models were fitted with survey
year as a categorical variable and included tests for
interaction with survey year to determine whether associations differed across surveys by sociodemographic and
lifestyle variables (Table 1) and by psychopathology
predictors (Model 1 in Table 2). Models in Table 2 were
further adjusted for sets of covariates to explore the extent to
which sociodemographic and lifestyle variables explained
the associations between CVD and psychopathology predictors (Model 2). Finally, models were further adjusted for
chronic disease variables and were fitted with a survey year
interaction to determine whether associations with psychopathology predictors differed across surveys independent of
covariates (Model 3). To estimate how much of the
association of CVD with psychopathology predictors was
explained by the covariates, the percentage change in the
effect estimate after adjustment for these variables was
computed as: 100(log OR Model [without covariate]log
OR Model [with covariate])/log OR Model [without
covariate]. Statistical tests were two-sided, and significance
was set at Pb.05.

3. Results
Changes in the standardized prevalence of CVD by
sociodemographic, lifestyle and chronic disease variables
across the two surveys appear in Table 1. The prevalence of
CVD increased from 4.1% to 4.5% (OR 1.11; 1.001.23;
P=.045) between the 20042005 and 20072008 surveys.
Between 2004 and 2008, the prevalence of CVD was highest
for men, older age groups, not being in the workforce, former
smokers, low physical activity and alcohol consumption
groups, lowest and highest BMI groups, highest fruit and
vegetable intake groups, and those with diabetes, high blood
pressure and high cholesterol. An absence of a significant
interaction indicates that the odds of CVD associated with
sociodemographic, lifestyle and chronic disease variables
were consistent across the two surveys (i.e., there were no
significant differences in the above described associations
between the two surveys).
Changes in the standardized prevalence of CVD associated with psychopathology predictors across the two surveys
appear in Table 2. On average, use of AD/AA medications
(OR 1.68; 1.412.00) was associated with increased odds of
CVD (see Model 1). CVD prevalence remained significantly
associated with AD/AA medications after adjustments for
sociodemographic and lifestyle variables (see Model 2) and
chronic diseases (see Model 3), but the strength of these
associations (percentage change in the log OR) decreased by
24% and 25%, respectively. An absence of a significant

348

E. Atlantis, T. Sullivan / General Hospital Psychiatry 34 (2012) 345351

Table 1
Standardized a prevalence of CVD in Australian adults by sociodemographic, lifestyle and chronic disease variables from the National Health Surveys 2004
2005 and 20072008
Weighted n
20042005

Total
Sex
Female
Male
Age (years)
2549
5059
60
Relationship status
Partnered
Not partnered
Australian born
Yes
No
Usual work hours/wk
Not in workforce
24
2539
4049/50
49/50
BMI (kg/m 2)
b18.5
18.524.9
2529.9
30
Missing
MET-min/wk
0
1539
540
Usual daily servings of fruit
1
23
4
Usual daily servings of vegetables
1
23
4
Past week alcohol intake (ml)
0
1139
140
Smoker status
Current smoker
Former smoker
Never smoker
Diabetes
Yes
No
High blood pressure
Yes
No
High cholesterol
Yes
No
a

CVD
20072008

20042005
%

20072008

17,665

14,276

4.1

(3.84.4)

(95% CI)

4.5

(4.24.9)

9639
8026

7588
6688

3.2
5.1

(2.93.6)
(4.75.6)

3.3
5.9

(2.93.7)
(5.46.4)

9958
3185
4522

8024
2583
3669

0.7
3.4
12.1

0.8
4.4
12.8

(0.61.0)
(3.65.2)
(11.813.8)

9710
7955

7768
6508

3.9
4.3

(3.54.3)
(3.94.8)

4.5
4.6

(4.04.9)
(4.15.1)

12,920
4744

10,287
3989

3.8
4.8

(3.54.2)
(4.25.4)

4.5
4.5

(4.14.9)
(4.05.2)

6736
2040
3003
3297
2588

4964
1596
2678
2873
2165

8.6
2.2
1.0
0.9
1.4

(8.09.3)
(1.72.9)
(0.71.4)
(0.61.3)
(1.01.9)

9.8
2.1
1.5
1.3
2.2

(9.010.6)
(1.52.9)
(1.12.0)
(1.01.8)
(1.72.9)

349
6859
5706
3149
1515

229
4909
4311
2771
1964

4.8
3.2
4.5
4.4
5.4

(3.17.4)
(2.93.7)
(4.05.1)
(3.85.2)
(4.46.5)

5.0
3.7
4.6
5.8
4.5

(2.88.6)
(3.34.3)
(4.05.2)
(5.06.7)
(3.75.4)

5917
5247
6501

5017
4305
4953

6.1
3.2
3.0

(5.56.7)
(2.83.7)
(2.63.4)

6.1
3.6
3.7

(5.56.8)
(3.14.2)
(3.24.2)

8098
7799
1768

6783
6419
1074

3.9
4.1
4.9

(3.54.3)
(3.74.6)
(4.05.9)

3.8
4.9
6.4

(3.44.3)
(4.45.5)
(5.18.0)

3340
8186
6139

3606
7422
3248

3.8
3.9
4.5

(3.24.5)
(3.54.3)
(4.15.1)

4.0
4.3
5.5

(3.54.7)
(3.94.8)
(4.86.3)

6597
6561
4507

5362
5378
3536

5.2
3.5
3.3

(4.85.8)
(3.13.9)
(2.83.8)

5.9
4.0
3.2

(5.36.5)
(3.54.6)
(2.73.8)

4156
5739
7770

3070
4622
6584

3.1
6.3
3.0

(2.63.7)
(5.76.9)
(2.73.4)

3.3
7.0
3.4

(2.74.0)
(6.37.7)
(3.03.8)

952
16,713

809
13,467

15.6
3.4

(13.517.9)
(3.23.7)

18.0
3.7

(15.720.7)
(3.44.0)

3873
13,791

3056
11,221

10.3
2.3

(9.511.3)
(2.12.6)

11.7
2.6

(10.612.8)
(2.32.9)

2656
15,009

2084
12,193

13.0
2.5

(11.814.3)
(2.32.8)

13.2
3.0

(11.914.6)
(2.83.3)

(0.50.9)
(2.84.0)
(11.213.0)

Age within sex standardized to the Australian population distribution using the 2001 Census of Population and Housing.
Significant (at Pb0.05) for main effect of each variable adjusted for survey; no significant interactions with survey group were found.

(95% CI)

E. Atlantis, T. Sullivan / General Hospital Psychiatry 34 (2012) 345351

349

Table 2
Standardized a prevalence and odds of CVD associated with psychopathology in Australian adults aged 25 years from the National Health Surveys 20042005
and 20072008
Psychopathology predictors

Weighted n

CVD

20042005 20072008 20042005

20072008

AD/AA medications
Yes
1425
No
16,240
Psychological distress b
Yes
730
No
16,934
Composite AD/AA and
psychological distress b
Psychological distress only
443
AD/AA with
288
psychological distress
AD/AA without
1137
psychological distress
None
15,797

(95% CI)

Model 1

Model 2

Model 3

OR

OR

OR

(95% CI)

(95% CI)

(95% CI)

(95% CI)

6.3 (4.78.3)
4.4 (4.14.8)

690
13,587

6.8 (5.68.2)
3.8 (3.64.1)

530
13,747

9.6 (7.711.8) 8.3 (6.310.8) , 2.34 (1.932.84) 2.18 (1.762.7) 2.05 (1.642.57)
3.8 (3.64.1) 4.4 (4.14.7)
1
(Reference) 1
(Reference) 1
(Reference)

351
178

8.6 (6.411.4) 9.6 (7.013.1)

11.1 (8.115.2) 5.6 (3.19.8)

1.68 (1.412.00) 1.48 (1.231.79) 1.35 (1.111.63)

1
(Reference) 1
(Reference) 1
(Reference)

2.41 (1.893.06) 2.20 (1.692.86) 2.00 (1.522.62)

2.41 (1.773.30) 2.30 (1.653.22) 2.02 (1.412.88)

512

5.7 (4.57.1)

6.5 (4.78.9)

1.56 (1.271.91) 1.36 (1.091.69) 1.24 (1.001.55)

13,235

3.7 (3.44.0)

4.3 (4.04.6)

(Reference) 1

(Reference) 1

(Reference)

Model 1, each psychopathology predictor adjusted for survey group; Model 2, further adjustment for sociodemographic and lifestyle variables (as described in
Table 1); Model 3, further adjustment for diabetes, high blood pressure and high cholesterol.
a
Age within sex standardized to the Australian population distribution using the 2001 Census of Population and Housing.
b
Defined with the K10 (using scores 30).
Significant (at Pb0.05) for main effect of each psychopathology predictor modeled separately and adjusted for survey group; **significant for interaction
effect of each predictor with survey group fitted in fully adjusted models (Model 3).

interaction with survey year (P=.342) adjusted for all

covariates (as in Model 3) indicates that the increased odds
of CVD associated with AD/AA medications were relatively
consistent over the 4-year period.
On average, psychological distress (OR 2.34; 1.932.84)
was associated with increased odds of CVD (see Model 1).
CVD prevalence remained significantly associated with
psychological distress after adjustments for sociodemographic and lifestyle variables (see Model 2) and chronic
diseases (see Model 3), but the strength of these associations
decreased by 8%, respectively. A significant interaction with
survey year (P=.044) adjusted for all covariates indicates that
the increased odds of CVD associated with psychological
distress were relatively inconsistent over the 4-year period. In
a post hoc analysis, the odds of CVD in 2008 compared to
2004 (reference) significantly increased among those without
psychological distress (OR 1.19; 1.061.34), but not among
those with psychological distress (OR 0.75; 0.471.20).
On average, psychological distress only (OR 2.41; 1.89
3.06) and use of AD/AA medications with (OR 2.41; 1.77
3.30) and without (OR 1.56; 1.271.91) psychological
distress were associated with increased odds of CVD (see
Model 1). CVD prevalence remained significantly associated
with all the composite AD/AA and psychological distress
predictors after adjustments for sociodemographic and
lifestyle variables (see Model 2) and chronic diseases (see
Model 3), but the strength of these associations decreased by
5% to 31% and 12% to 29%, respectively. In a post hoc
analysis, both psychological distress only (OR 1.61; 1.15
2.25) and AD/AA medications with psychological distress
(OR 1.62; 1.082.44) conferred significantly higher odds of

CVD than AD/AA medications without psychological

distress (reference). An absence of a significant interaction
with survey year (P=.088) adjusted for all covariates suggests
that the increased odds of CVD associated with the composite
AD/AA medications and psychological distress predictors
were relatively consistent over the 4-year period.

4. Discussion
Our comprehensive study of two serial and representative
population-based data sources show that the prevalence of
CVD increased slightly in the adult population from 2004 to
2008 (4.1% to 4.5%). Furthermore and contrary to our
expectations, results show that the prevalence of CVD had
increased among people without psychopathology, but not
among those with psychopathology. The diminished odds of
CVD associated with psychopathology in 2008 compared to
2004 might have been partially explained by increasing
trends in mental health literacy due to national initiatives to
enhance public knowledge and diminish stigma associated
with symptoms of depression, and health service use [18].
Psychopathology is associated with significant increases in
general practitioner visits and use of other health services,
particularly among people with comorbid chronic diseases
[18,24,25]. More health service use and treatment could have
resulted in better quality of care and outcomes that attenuated
the association between psychopathology and CVD [26,27].
The current results may help inform national health care
policy makers and researchers regarding these conditions in
the future. Such information would be particularly important

350

E. Atlantis, T. Sullivan / General Hospital Psychiatry 34 (2012) 345351

for high-income countries, which accounted for the greatest

proportion of the global CVD burden in 2004 [7], where
depression, antidepressant prescribing and mental health
care spending have risen markedly over the recent decade
[17,2830].
On average, the reported use of AD/AA medications
(1.35-fold), psychological distress (2.05-fold) and all the
composite AD/AA medications and psychological distress
(1.24- to 2.02-fold) predictors were significantly associated
with increased odds of CVD prevalence over the 4-year
period, independent of an extensive range of covariates.
However, the unadjusted effect estimates (presented in
Model 1) were progressively weakened after stepwise
adjustments for sociodemographic and lifestyle variables
(by 5% to 31% in Model 2) and chronic diseases (by 8% to
29% in Model 3). This suggests that the rising burden of
CVD associated with psychopathology was partially driven,
and confounded, by the rising prevalence of obesity and
diabetes in Australia [16].
In addition, both people with psychological distress only
(1.61-fold) and reported users of AD/AA medications with
psychological distress (1.62-fold) conferred significantly
higher odds of CVD than reported users of AD/AA
medications without psychological distress. An absence of
clinically significant psychological distress among users of
AD/AA medications may reflect better response to treatment
and adherence to treatment and healthy lifestyle guidelines
[4], which could have been due to more health service use
resulting in improved quality of care and outcomes
associated with CVD risk [26,27].
The strengths and innovations of this study include the
analysis of prevalence data across two large and representative population-based surveys, which had used almost
identical methodologies; participant information was collected via in person interview; psychological distress was
defined using symptom scores shown to predict any current
anxiety or affective disorder accurately according to the
Composite International Diagnostic Interview classification
system [24]; contemporaneous use of medications for mental
well-being was inventoried; population-based changes in
self-reported CVD prevalence associated with psychopathology were reliably estimated; and results were age/sex
standardized to the 2001 Australian reference population to
account for age-related changes in the population distribution and nonresponse and noncoverage biases.
The results of this study should also be balanced against
noteworthy limitations. Because data were derived from
serial surveys (cross-sectional studies), the issue of directionality of associations between CVD prevalence and
psychopathology predictors and covariates remains unclear
in this study. On the other hand, issues of directionality of
association are attenuated somewhat because depression is
likely to have preceded the onset of CVD in more than 80%
of comorbid cases [31]. Other limitations, including selfreport outcome measures, missing information on nonrespondents and misclassification errors for covariates, could

have partially confounded our findings. Finally, we

acknowledge that our results are based on psychopathology
predictors not classified according to a clinical diagnosis of
depression or anxiety according to the Diagnostic and
Statistical Manual of Mental Disorders, Fourth Edition or
International Classification of Diseases, 10th Revision
criteria with little knowledge about the duration, severity
or treatment response of the symptoms, and therefore, they
should be interpreted with caution.
This comprehensive study of serial and representative
population-based data sources shows that the prevalence of
CVD increased slightly for adult Australians from 2004 to
2008, but not among those with psychopathology when
analyzed separately. On average, psychopathology was
persistently associated with increased odds of CVD
prevalence over the 4-year period. However, the rising
burden of CVD associated with psychopathology was
partially explained by sociodemographic and lifestyle variables, as well as chronic diseases. Future studies in Australia
and other countries should confirm whether the excess
burden of CVD associated with psychopathology is indeed
diminishing. The current results may help inform national
health care policy makers and researchers regarding these
conditions in the future.

References
[1] Vogelzangs N, Seldenrijk A, Beekman AT, van Hout HP, de Jonge P,
Penninx BW. Cardiovascular disease in persons with depressive and
anxiety disorders. J Affect Disord 2010;125:2418.
[2] Atlantis E, Shi Z, Penninx BJ, Wittert GA, Taylor A, Almeida OP.
Chronic medical conditions mediate the association between depression
and cardiovascular disease mortality. Soc Psychiatry Psychiatr
Epidemiol 2011 Mar 8. [Epub ahead of print].
[3] Atlantis E, Sullivan T, Sartorius N, Almeida OP. Changes in the prevalence
of psychological distress and use of antidepressants or anti-anxiety
medications associated with comorbid chronic diseases in the adult
Australian population, 20012008. Aust N Z J Psychiatry 2012 Jan 3.
[Epub ahead of print].
[4] Gonzalez JS, Peyrot M, McCarl LA, Collins EM, Serpa L, Mimiaga
MJ, et al. Depression and diabetes treatment nonadherence: a metaanalysis. Diabetes Care 2008;31:2398403.
[5] Alonso J, Ferrer M, Gandek B, Ware Jr JE, Aaronson NK, Mosconi P,
et al. Health-related quality of life associated with chronic conditions in
eight countries: results from the International Quality of Life
Assessment (IQOLA) Project. Qual Life Res 2004;13:28398.
[6] Egede LE. Major depression in individuals with chronic medical
disorders: prevalence, correlates and association with health resource
utilization, lost productivity and functional disability. Gen Hosp
Psychiatry 2007;29:40916.
[7] World Health Organization. The global burden of disease: 2004
update. Geneva, Switzerland: WHO Press; 2008.
[8] Danaei G, Finucane MM, Lin JK, Singh GM, Paciorek CJ, Cowan MJ,
et al. National, regional, and global trends in systolic blood pressure
since 1980: systematic analysis of health examination surveys and
epidemiological studies with 786 country-years and 54 million
participants. Lancet 2011;377:56877.
[9] Farzadfar F, Finucane MM, Danaei G, Pelizzari PM, Cowan MJ,
Paciorek CJ, et al. National, regional, and global trends in serum total
cholesterol since 1980: systematic analysis of health examination

E. Atlantis, T. Sullivan / General Hospital Psychiatry 34 (2012) 345351

[10]

[11]

[12]

[13]
[14]
[15]

[16]
[17]

[18]

[19]

[20]

surveys and epidemiological studies with 321 country-years and 30

million participants. Lancet 2011;377:57886.
Danaei G, Finucane MM, Lu Y, Singh GM, Cowan MJ, Paciorek CJ,
et al. National, regional, and global trends in fasting plasma glucose
and diabetes prevalence since 1980: systematic analysis of health
examination surveys and epidemiological studies with 370 countryyears and 2.7 million participants. Lancet 2011;378:3140.
Finucane MM, Stevens GA, Cowan MJ, Danaei G, Lin JK, Paciorek
CJ, et al. National, regional, and global trends in body-mass index
since 1980: systematic analysis of health examination surveys and
epidemiological studies with 960 country-years and 9.1 million
participants. Lancet 2011;377:55767.
Kessler RC, Aguilar-Gaxiola S, Alonso J, Chatterji S, Lee S, Ormel J,
et al. The global burden of mental disorders: an update from the WHO
World Mental Health (WMH) surveys. Epidemiol Psichiatr Soc
2009;18:2333.
Ezzati M, Henley SJ, Thun MJ, Lopez AD. Role of smoking in global
and regional cardiovascular mortality. Circulation 2005;112:48997.
Hall JN, Moore S, Harper SB, Lynch JW. Global variability in fruit and
vegetable consumption. Am J Prev Med 2009;36:4029 e405.
Guthold R, Ono T, Strong KL, Chatterji S, Morabia A. Worldwide
variability in physical inactivity a 51-country survey. Am J Prev Med
2008;34:48694.
Atlantis E, Lange K, Wittert GA. Chronic disease trends due to excess
body weight in Australia. Obes Rev 2009;10:54353.
Goldney RD, Eckert KA, Hawthorne G, Taylor AW. Changes in the
prevalence of major depression in an Australian community sample
between 1998 and 2008. Aust N Z J Psychiatry 2010;44:90110.
Atlantis E, Goldney RD, Eckert KA, Taylor AW, Phillips P. Trends in
health-related quality of life and health service use associated with
comorbid diabetes and major depression in South Australia, 19982008.
Soc Psychiatry Psychiatr Epidemiol 2011 May 18. [Epub ahead of print].
Mant A, Rendle VA, Hall WD, Mitchell PB, Montgomery WS,
McManus PR, et al. Making new choices about antidepressants in
Australia: the long view 19752002. Med J Aust 2004;181:S214.
Australian Bureau of Statistics b National Health Survey: 2005.
Authorised accessed via basic CD-ROM. Confidentialised Unit Record
File No: 1778.

351

[21] Australian Bureau of Statistics c National Health Survey: 200708.

Authorised accessed via basic CD-ROM. Confidentialised Unit Record
File No: 4324.
[22] Barr EL, Tonkin AM, Welborn TA, Shaw JE. Validity of self-reported
cardiovascular disease events in comparison to medical record
adjudication and a statewide hospital morbidity database: the AusDiab
study. Intern Med J 2009;39:4953.
[23] Kessler RC, Andrews G, Colpe LJ, Hiripi E, Mroczek DK, Normand
SL, et al. Short screening scales to monitor population prevalences and
trends in non-specific psychological distress. Psychol Med
2002;32:95976.
[24] Andrews G, Slade T. Interpreting scores on the Kessler Psychological
Distress Scale (K10). Aust N Z J Public Health 2001;25:4947.
[25] Atlantis E, Lange K, Goldney RD, Martin S, Haren MT, Taylor A,
et al. Specific medical conditions associated with clinically
significant depressive symptoms in men. Soc Psychiatry Psychiatr
Epidemiol 2011;46:130312.
[26] Cully JA, Zimmer M, Khan MM, Petersen LA. Quality of depression
care and its impact on health service use and mortality among veterans.
Psychiatr Serv 2008;59:1399405.
[27] Harris MF, Priddin D, Ruscoe W, Infante FA, O'Toole BI. Quality of
care provided by general practitioners using or not using divisionbased diabetes registers. Med J Aust 2002;177:2502.
[28] Andersen I, Thielen K, Bech P, Nygaard E, Diderichsen F. Increasing
prevalence of depression from 2000 to 2006. Scand J Public Health
2011;39:85763.
[29] Marcus SC, Olfson M. National trends in the treatment for depression
from 1998 to 2007. Arch Gen Psychiatry 2010;67:126573.
[30] Moore M, Yuen HM, Dunn N, Mullee MA, Maskell J, Kendrick
T. Explaining the rise in antidepressant prescribing: a descriptive
study using the general practice research database. BMJ 2009;339:
b3999.
[31] O'Neil A, Williams ED, Stevenson CE, Oldenburg B, Berk M,
Sanderson K. Co-morbid cardiovascular disease and depression:
sequence of disease onset is linked to mental but not physical
self-rated health. Results from a cross-sectional, population-based
study. Soc Psychiatry Psychiatr Epidemiol 2011 Aug 10. [Epub
ahead of print].