Depressive Symptoms and Risks of Coronary Heart Disease
and Mortality in Elderly Americans
Abraham A. Ariyo, MD, MPH; Mary Haan, MPH, PhD; Catherine M. Tangen, PhD;
John C. Rutledge, MD; Mary Cushman, MD, MS; Adrian Dobs, MD, MHS;
Curt D. Furberg, MD, PhD; for the Cardiovascular Health Study Collaborative Research Group
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Background—Several epidemiological studies have associated depressive symptoms with cardiovascular disease. We
investigated whether depressive symptoms constituted a risk for coronary heart disease (CHD) and total mortality
among an apparently healthy elderly cohort.
Methods and Results—In a prospective cohort of 5888 elderly Americans (ⱖ65 years) who were enrolled in the
Cardiovascular Health Study, 4493 participants who were free of cardiovascular disease at baseline provided annual
information on their depressive status, which was assessed using the Depression Scale of the Center for Epidemiological
Studies. These 4493 subjects were followed for 6 years for the development of CHD and mortality. The cumulative
mean depression score was assessed for each participant up to the time of event (maximum 6-year follow-up). Using
time-dependent, proportional-hazards models, the unadjusted hazard ratio associated with every 5-unit increase in mean
depression score for the development of CHD was 1.15 (P⫽0.006); the ratio for all-cause mortality was 1.29
(P⬍0.0001). In multivariate analyses adjusted for age, race, sex, education, diabetes, hypertension, cigarette smoking,
total cholesterol, triglyceride level, congestive heart failure, and physical inactivity, the hazard ratio for CHD was 1.15
(P⫽0.006) and that for all-cause mortality was 1.16 (P⫽0.006). Among participants with the highest cumulative mean
depression scores, the risk of CHD increased by 40% and risk of death by 60% compared with those who had the lowest
mean scores.
Conclusions—Among elderly Americans, depressive symptoms constitute an independent risk factor for the development
of CHD and total mortality. (Circulation. 2000;102:1773-1779.)
Key Words: risk factors 䡲 epidemiology 䡲 coronary disease 䡲 mortality 䡲 depression
D
epressive symptoms occur in 19% to 30% of the elderly,
but only 1% of those affected receive the necessary
treatment.1,2 Several epidemiological studies3,4 have suggested that high depression scores may predispose an individual to an increased risk of developing cardiovascular
disease (CVD). Although most,3–11 but not all12–14 prospective studies of middle-aged populations addressing this issue
have reported positive associations, data regarding the relationship between depressive symptoms and cardiovascular
risk in the elderly are sparse.
In this article, we report the results of a 6-year study that
prospectively investigated the relationship between depressive symptoms and subsequent risk of coronary heart disease
(CHD) and mortality in 4493 elderly Americans.
was started in 1989 with an initial cohort of 5201 subjects. An
additional ethnic minority cohort of 687 subjects was later recruited,
which brought the total to 5888 subjects. The participants were
recruited from random samples of the Medicare eligibility lists
provided by the Health Care Financing Administration from the
following 4 communities: Washington County, Md (Johns Hopkins
University); Sacramento County, Calif (University of CaliforniaDavis); Forsyth County, NC (Wake Forest University); and Allegheny County, Pa (University of Pittsburgh). Participants were noninstitutionalized persons who signed the informed consent forms to
participate in the study; persons were eligible for enrollment whether
or not they had a history of CVD. The full details of the CHS
recruitment process have been published elsewhere.16
In brief, participants were initially given a 90-minute home interview,
during which a questionnaire was given and information was sought
regarding their health, depressive status, and medications. All participants were given a 4- to 5-hour medical examination.
Our study cohort comprised 4493 subjects who were free of CVD;
CVD included angina, CHD, myocardial infarction, angioplasty,
coronary artery bypass surgery, congestive heart failure, or stroke at
Methods
The Cardiovascular Health Study (CHS)15 is a multicenter study of
cardiovascular risk factors in Americans aged 65 years and older. It
Received February 17, 2000; revision received May 8, 2000; accepted May 11, 2000.
From the Divisions of Cardiovascular Medicine (A.A.A.) and Endocrine (A.D), Johns Hopkins University, Baltimore, Md; the Divisions of
Cardiovascular Medicine (J.C.R), Epidemiology and Prevention (M.H.), University of California-Davis; the Department of Biostatistics (C.M.T.),
University of Washington, Seattle, Wash; Department of Pathology (M.C.), University of Vermont, Burlington; and the Department of Public Health
Sciences (C.D.F.), Wake Forest University, Winston-Salem, NC.
Correspondence to Dr Abraham A. Ariyo, Cardiovascular Division, Johns Hopkins Hospital, 600 N Wolfe St, Carnegie 568, Baltimore, MD 21287.
E-mail [email protected]
Reprint requests to CHS Coordinating Center, Century Square, 1501 Fourth Ave, Suite 2105, Seattle, WA 98101.
© 2000 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org
1773
1774
Circulation
October 10, 2000
TABLE 1. Center for Epidemiological Studies—
Depression Scale
TABLE 2.
Distribution of Depression Scores Among the Cohort
CHD
Using the scale below, indicate the number that best describes how often
you felt or behaved this way during the past week.
1. I was bothered by things that usually don’t bother me.
2. I had trouble keeping my mind on what I was doing.
3. I felt that everything I did was an effort.
4. I felt depressed.
5. I felt hopeful about the future.*
6. I felt fearful.
7. My sleep was restless.
8. I was happy.*
9. I felt lonely.
10. I could not get going.
The four point scale is as follows:
0⫽Rarely or none of the time (⬍1 day)
Death
Yes
(n⫽606)
No
(n⫽3887)
Yes
(n⫽614)
No
(n⫽3879)
Baseline score
Mean (SD)
4.53 (0.18)
4.42 (0.07)
4.77 (0.18)
4.38 (0.07)
Range
0–25
0–26
0–26
0–26
Interquartile
range
1–6
1–6
1–7
1–6
5.07 (0.17)
5.02 (0.06)
5.63 (0.17)
4.98 (0.06)
Cumulative mean at
last visit before
event
Mean (SD)
Range
0–24
25–75%
2–7
50.7
Subjects with ⱖ1
scoreⱖ8, %
0–24
0–22
0–24
2.2–7
2.5–7.8
2.2–6.8
48.4
47.9
48.8
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
1⫽Some or a little of the time (1 to 2 days)
2⫽A moderate amount of time (3 to 4 days)
3⫽Most of the time.
*Reverse scoring
study entry. Baseline status was initially based on a self-report that
was confirmed by clinical examination, a review of prior hospital
records, or both. Depression was assessed at baseline and annually
thereafter for an average of 6 years for the initial cohort and for 3
years for the minority cohort.
Ascertainment of Cardiovascular Events
The cardiovascular event of interest was CHD, which was defined as
first occurrence of angina, myocardial infarction, angioplasty, coronary artery bypass grafting, or coronary death. All events were
assessed semiannually. The surveillance and ascertainment of cardiovascular events in CHS have been described elsewhere.17 In brief,
all morbid events after the baseline clinic visit were classified as
incident events. Mortality was investigated on the basis of death
code, which was graded according to the International Classification
of Disease code.18 Information on death was obtained through
reviews of obituaries, medical records, death certificates, and interviews of contacts and proxies. The CHS has nearly 100% ascertainment of mortality status.17 All provisional diagnoses of CHD and
fatal events were reviewed and adjudicated at periodic meetings of
the Morbidity and Mortality Subcommittee. This Subcommittee was
comprised of an investigator from each center, the coordinating
center, and the project office of the National Heart, Lung, and Blood
institute. The events and deaths discussed in this article are those that
were adjudicated from 1989 through June 1996.
Assessment of Depressive Status
We used the modified, shorter version of the Center for Epidemiological Studies’ Depression Scale,19 which is a questionnaire widely
used as a screening tool to assess depression in the elderly, especially
in primary care or outpatient settings (Table 1). The reliability of the
modified version of this scale has been validated by other studies.20,21 It is a 10-item scale, with participants’ depression scores
recorded as a continuous measure from 0 to 30. Traditionally, a score
ⱖ8 is recognized as “at risk of clinical depression”; however, we
used continuous depression scores in our analyses instead so as not
to lose any information by categorization.
Statistical Analysis
We used the nonparametric Wilcoxon-Mann-Whitney test for categorical covariates and the Spearman correlation for continuous
covariates to assess the association between covariate measures and
the baseline depression scores.
For a missing depression score for a given yearly visit, the last
prior non-missing score was carried forward. Thus, depression scores
for every visit were cumulatively averaged after missing depression
values were replaced. Scores after an event did not contribute to the
exposure measure. For individuals without an event, mean depression scores of all visits were calculated up to the last visit. Although
depression was assessed annually, to reduce possible biases, depression scores were considered to take effect 6 months before the clinic
visit (the halfway-point between visits). This approach is more
representative of the exposure window for a given depressive state
than having the depression score take effect on the clinic visit day.
Participants with missing data ranged from 5% in year 1 to 13% in
year 6. A missing value carry-forward method was used to replace
these missing values.
The following variables were forced into the covariate-adjusted
model: age at entry, race, sex, education, hypertension, diabetes,
smoking status, total cholesterol levels, physical inactivity, triglycerides, marital status, and alcohol consumption. Time-dependent
covariates for congestive heart failure, chronic obstructive pulmonary disease, and cancer diagnosis were constructed for all-cause
mortality outcome. There was no indication of collinearity between
the covariates and depressive symptoms or between the covariates
themselves. Proportional hazards models were fit to the outcomes of
time to angina, myocardial infarction, CHD, and death. All analyses
were performed using the SAS system.22
Results
Table 2 shows the distribution of depression scores among
the cohort at baseline according to associated events. Few
subjects had depression scores ⬎15. Tables 3 and 4 depict the
characteristics of the 4493 participants in this analysis. Age
range at entry was 65 to 98 years, with a mean age of 72 years
for women and 73 years for men. Depressive symptoms were
not associated with age at study entry. White women and men
composed 84% and 85% of the cohort, respectively. At study
entry, nonwhites had higher depression scores than whites,
and women had higher baseline depression scores than men
(4.9 versus 3.7; P⬍0.001). Depression was more prevalent
among the less educated, those with lower incomes, and those
with a history of diabetes. Smokers had the highest depressive
scores, with previous smokers having higher scores than
Ariyo et al
TABLE 3.
Depressive Symptoms and Cardiovascular Disease
1775
Baseline Characteristics of the 4493 Participants
Women (n⫽2737)
Covariates
Mean Depression
Score (SD)
Men (n⫽1756)
P
Mean Depression
Score (SD)
⬍0.001
Race
⬍0.001
White
4.7 (0.1)
3.4 (0.1)
Nonwhite
6.0 (0.2)
5.2 (0.3)
⬍0.004
Hypertension
Yes
5.2 (0.1)
No
4.8 (0.1)
Yes
5.5 (0.2)
No
4.8 (0.1)
3.6 (0.1)
⬍0.03
4.0 (0.2)
3.6 (0.1)
⬍0.011
Smoking
Never
⬍0.29
3.8 (0.2)
⬍0.007
History of DM
4.8 (0.1)
⬍0.26
3.4 (0.2)
Current
5.7 (0.3)
3.8 (0.3)
Previous
5.0 (0.2)
3.7 (0.1)
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
⬍0.001
Income
$5000–15 999
5.5 (0.1)
⬍0.001
4.5 (0.2)
$16 000–34 999
4.6 (0.2)
3.5 (0.2)
$35 000–49 999
4.2 (0.3)
3.2 (0.2)
ⱖ$50 000
4.3 (0.2)
2.9 (0.2)
⬍0.001
Education
⬍0.001
Not high school graduate
5.7 (0.2)
4.4 (0.2)
High school graduate
4.9 (0.2)
3.7 (0.2)
College degree
4.3 (0.2)
3.2 (0.2)
⬍0.001
Marital Status
⬍0.001
Never married
4.3 (0.4)
Married
4.6 (0.1)
3.4 (0.1)
Widowed/separated/
divorced
5.5 (0.1)
5.0 (0.3)
4.6 (0.7)
⬍0.001
Live alone
⬍0.001
Yes
5.6 (0.2)
6.0 (0.5)
No
4.7 (0.1)
3.4 (0.1)
⬍0.001
Self-perceived health
Better than others
4.3 (0.1)
Not better
5.6 (0.2)
⬍0.001
3.0 (0.1)
4.6 (0.2)
⬍0.001
Activities of daily living
0
4.7 (0.1)
1⫹
8.1 (0.4)
⬍0.001
3.5 (0.1)
6.7 (0.6)
⬍0.001
Instrumental activity of
daily living
⬍0.001
0
4.2 (0.1)
3.3 (0.1)
1⫹
7.0 (0.2)
5.6 (0.3)
persons who had never smoked. Married participants and
those who lived with others had lower depression scores than
those who were widowed, separated, or divorced. Those who
had problems with the instrumental activities of daily living
or lower social network scores had higher depression scores.
Physical activity was inversely related to depression
scores. Those with a higher body mass index had higher
scores than those who had a lower body mass index. Among
P
women, total cholesterol, HDL cholesterol, and factor VII
levels were not associated with depression scores. However,
LDL cholesterol levels were negatively correlated with depression, and fibrinogen was positively correlated with depression. In men, total cholesterol, LDL cholesterol, triglycerides, platelet count, and factor VII levels were not
significantly associated with depressive scores. Of the 4493
study participants, 188 (143 women and 45 men) were on
1776
Circulation
October 10, 2000
TABLE 4.
Association of Covariates and Depression Score at Study Entry
Women
Men
Spearman Correlation
With Depression
Score
Covariates
Age at study entry
P
0.03
0.13
Spearman Correlation
With Depression
Score
P
0.05
0.04
Lipids
Total cholesterol
⫺0.03
0.14
⫺0.03
0.19
LDL
⫺0.04
0.05
⫺0.04
0.08
HDL
⫺0.01
0.70
0.06
0.02
Triglycerides
0.03
No. of blocks walked in previous
week
⫺0.15
0.09
⫺0.03
0.15
0.0001
⫺0.15
0.0001
0.80
Clotting factors
Factor VII
0.03
0.19
⫺0.01
Fibrinogen
0.04
0.04
0.04
0.09
Platelets
0.03
0.10
⫺0.04
0.17
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Body mass index
0.05
⫺0.14
Social network score
0.008
⫺0.01
0.67
0.0001
⫺0.15
⬍0.001
antidepressants. No differences existed in baseline characteristics of the few men on antidepressants and the rest of the
cohort. However, women on antidepressants had higher
platelet counts (P⫽0.001) and were more likely to smoke
than those not on medication (P⫽0.001).
increasing levels of depressive scores. It shows a good
correlation between the depression scores and vascular risks;
for example, among the participants with some of the highest
depression scores, the CHD risk increased by 40% and that of
death by 60%.
Cumulative Mean Depression Scores and Events
Baseline Depression Scores and Events
Table 5 depicts the hazard ratios (HRs) associated with each
5-unit increase in mean depression score before an event. In
both univariate and multivariate analyses, every 5-unit increase in mean score was associated with a 15% increased
risk of developing CHD (P⫽0.006). For mortality, 16% and
29% increased risks were observed in univariate and adjusted
models, respectively.
We further explored the possibility of a sex interaction for
each event of interest but found no evidence of such effect
modification, indicating that the HR estimates shown in Table
5 hold equally for both sexes. However, some indication
existed for a higher HR for new angina among women than
among men (P⫽0.09). Analyses excluding those who were
on antidepressants produced no change in results. The Figure
further illustrates the increased risk of CHD and death with
Table 6 displays the HRs associated with 5-unit increases in
baseline depression scores and cardiovascular events. Persons
with higher baseline scores had a significantly higher risk of
dying (P⫽0.004), although the risk for CHD was nonsignificant (P⫽0.162). However, after adjusting for demographic
and cardiovascular risk factors, baseline depression was a
significant predictor of CHD (P⫽0.032) and death
(P⫽0.012).
Discussion
In this prospective study of elderly (ⱖ65 years) Americans
who were free of cardiovascular disease at baseline and who
were followed for 6 years, we found that depressive symptoms constituted an independent risk factor for the development of CHD and total mortality. This risk increased with
TABLE 5. Relationship Between HRs Associated With Every 5-Unit Increase in Cumulative Mean
Depression Score and Cardiovascular Events
Incident
No. of Events
Unadjusted HR (95% CI)
P
*Adjusted HR (95% CI)
P
Death†
614
1.29 (1.18, 1.41)
0.0001
1.16 (1.04, 1.28)
0.006
CHD
606
1.15 (1.04, 1.26)
0.006
1.15 (1.04, 1.27)
0.006
MI
270
1.12 (0.96, 1.29)
0.141
1.14 (0.98, 1.34)
0.088
Angina without concurrent MI
298
1.18 (1.03, 1.35)
0.018
1.20 (1.05, 1.38)
0.009
MI indicates myocardial infarction; CI, confidence interval.
*Adjusted covariates are age, race, sex, education, diabetes, hypertension, smoking status, physical activity, total cholesterol,
physical inactivity, marital status, alcohol consumption, and time-dependent covariates for congestive heart failure and angina.
†For all-cause mortality, the baseline indicators for chronic obstruction pulmonary disease and cancer diagnosis were also
included, in addition to those listed above.
Ariyo et al
Depressive Symptoms and Cardiovascular Disease
Covariate-adjusted hazard ratios for having an event based on
grouped 5-unit categories for average cumulative depression
score estimated before the event of interest. Covariates were
the same as those listed in Table 5.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
higher depression scores. Our study provides new evidence
and adds to the growing body of data indicating that depressive status is a risk factor for CVD.
Our findings are consistent with those of other studies
finding that depressive symptoms constitute a risk factor for
CHD3–7 and mortality.4 – 8 However, our data differ from those
in other studies because the present study focused exclusively
on the elderly. It provides a large amount of prospective data
on depressive symptoms as risk factors for CHD in older
adults, as well as information on the vascular risk of depressive symptoms in women. We demonstrated the magnitude
and incremental risks of CHD associated with increasing
levels of depressive scores. These findings are particularly
important for the elderly, in whom the applicability of some
traditional risk factors has been challenged.23–26
Cumulative mean assessment of depression scores was
more predictive of vascular events than baseline scores. The
fact that baseline depression scores did not adequately predict
vascular events as cumulative mean scores may be due to the
following. (1) A one-time assessment of depression at study
entry could not capture the chronic depressive state of an
individual as adequately as a 6-year follow-up. (2) A healthy
cohort like ours may have produced lower initial depression
scores; therefore, a longitudinal assessment of depression that
is also more stable may do a better job of predicting vascular
events.
It is conceivable that depressive symptoms could have
occurred as a result of disease rather than being a precursor of
CVD, because any life-threatening illness could potentially
cause a depressive state and cloud the cause-and-effect
1777
relationship between depressive symptoms and the occurrence of cardiovascular events. However, because our study
design excluded participants with prior cardiac disease, the
prospective data collection of antecedent depression scores
before cardiovascular events and the use of apparently
healthy, noninstitutionalized elderly subjects argues against
this notion. Second, the demonstration of an independent
relationship between higher depression scores and increasing
vascular risks in this apparently healthy elderly cohort
strongly favors a relationship between depressive symptoms
and cardiovascular events. Third, because the last clinic visit
scores were carried forward for participants who missed
clinics, our analyses would have probably underestimated the
effects of depression, because sick individuals (more depressed) tend to stay at home. Finally, the results shown in
Table 6, which used only baseline assessment of depression,
validated the time-dependent analysis results and lent support
to the argument that it is not changing disease assessment that
is being captured in Table 5 but depressive status.
It is also possible that our findings may have occurred by
chance. However, the prospective nature of this study, the
large sample size, the duration of follow-up, the blinded
ascertainment of events, and our analyses, which used cumulative mean depression scores, argue against chance as a
primary explanation for the result. Second, although we
evaluated depressive symptoms annually in our cohort, we
used cumulative mean depression scores in our model, which
measures chronic or persistent depressive state rather than an
acute or a brief one-time episodic depressive mood. The
“halfway point” (mid-visit) depression scores that were used
in these analyses better assess the day-to-day depressive state
rather than the office visit scores.
Women reportedly suffer more than men from the adverse
vascular effects of a depressive state.27 This may be due to the
higher prevalence of depressive symptoms in women than in
men. Nonetheless, we found that the HRs associated with
5-unit increases in depression scores were similar in men and
women but, because women had higher scores to begin with,
more vascular risks may have been expected for them.
The exact mechanism by which depressive symptoms may
predispose some individuals to increased vascular risk is
unknown; however, 3 plausible mechanisms have been proposed. (1) Depressive state is associated with poor physical
activity, less exercise, more smoking, and a high likelihood of
indulging in behavioral patterns that may increase vascular
TABLE 6. Relationship Between HRs Associated With Every 5-Unit Increase in
the Baseline Depression Scores in the Cohort
Event
Unadjusted HR
(95% CI)
P
*Adjusted HR
(95% CI)
P
Death
1.13 (1.04, 1.23)
0.004
1.13 (1.03, 1.23)
0.012
CHD
1.07 (0.98, 1.17)
0.162
1.11 (1.01, 1.22)
0.032
MI
1.06 (0.93, 1.21)
0.385
1.12 (0.97, 1.29)
0.116
Angina without
concurrent MI
1.08 (0.95, 1.22)
0.264
1.13 (0.99, 1.29)
0.074
MI indicates myocardial infarction; CI, confidence interval.
*Adjusted covariates in the models are the same as those listed in Table 5, except for CHF or
cancer diagnosis on the study.
1778
Circulation
October 10, 2000
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
risk. (2) Depressive state as a mental stress increases autonomic sympathetic activation.28,29 This activation may result
in increased levels of circulating platelets (with enhanced
activation and aggregability), fibrinogen, and thromboxane
A2.30 (3) Depressive state is related to lipid metabolism, such
that the increased production of steroid and free fatty acids
and a reduced glucose use ensues when depressed.13,31–34 This
combined sympathoadrenal activation stimulates platelets via
␣2-adrenoceptor activation13,33 and augments arterial thrombosis. Interestingly, we observed higher circulating platelet
counts and a correlation between fibrinogen levels and
depression scores among women on antidepressant medications but not among those not taking antidepressants. Thus,
platelet activation and increased fibrinogen may be the major
links between depression and CHD. Also, the effects of
antidepressant medications on platelets, vascular events, and
mortality in patients in depressive states remain unclear.
The increased mortality observed in our study is similar to
that reported by other investigators.4 – 8 Deaths associated
with depressive state have been attributed to the imbalance
between the autonomic parasympathetic and sympathetic
systems, with an overriding effect of the latter.35 Animal and
human studies have shown a correlation between increased
sympathetic activity and induction of ventricular arrhythmias36 that can lead to sudden death. The serotonergic
system, a key player in the depressive mechanism, also plays
an important role in the genesis of arrhythmias.36 A depressive state is associated with social isolation, less functional
capacity, and less exercise, and it may be a surrogate of other
diseases. High suicide rates, violent deaths, and deaths from
cancer and chronic diseases also compound the effects of
depressive symptoms on excess mortality.
The first limitation of this study relates to mortality.
Because the CHS was designed to assess cardiovascular risk
factors, complete ascertainment of all possible risk factors for
mortality is impossible; therefore, we cannot draw strong
conclusions about depressive symptoms being a major risk
factor for total mortality. Further, high depressive scores may
be a surrogate measure for deteriorating health that could not
be assessed in this study. However, we did control for known
risk factors for CHD. Second, a close association exists
between depressive state and chronic medical conditions.
Although our cohort was free of disease at baseline, it is
difficult to ascertain with certainty the presence or absence of
subclinical disease in an elderly population. Third, antidepressant medications are also used for medical conditions
other than clinical depression, and we could not ascertain the
specific indications for their use in individual patients in our
database.
Of the 31 millions Americans ⱖ65 years of age, 5 million
have depressive symptoms, with a lifetime risk of major
depressive disorder estimated at 7% to 12% for men and 20%
to 25% for women.37 Depressive disorder is second only to
CVD in time lost from work and lost productivity, and it has
been estimated to cost society $44 billion per year.38 In a
recent study,1 elderly persons with depressive symptoms
accrued 50% higher healthcare costs from more frequent use
of medical services than nondepressed elderly people. Although available therapy improves symptoms and quality of
life,39 most physicians under-recognize depressive symptoms
in elderly patients and, thus, the elderly are underdiagnosed
and undertreated.39 Increased awareness and early screening
for depressive symptoms may help reduce the associated
healthcare costs and human suffering associated with depressive symptoms.
Acknowledgments
This study was supported by National Heart, Lung, and Blood
Institute contracts N01-HC-850079-85086 and NO1-HC-15103. The
authors thank the Publication and Presentation and Steering Committee members for their critical review of this manuscript. They
further thank Ms Shirley Cable for her assistance in the preparation
of the manuscript.
References
1. Unutzer J, Patrick DL, Simon G, et al. Depressive symptoms and the cost
of health services in HMO patients aged 65 years and older: a 4-year
prospective study. JAMA. 1997;28:277:1618 –1623.
2. Judd LL. Surveys show that 20 –30% of elderly patients exhibit SSD
symptomatology. Int Med World Rep. 1997;12:33–34.
3. Ford DF, Mead LA, Chang PP, et al. Depression predicts cardiovascular
disease in men: the Precursors Study. Arch Intern Med. 1998:158:
1422–1426.
4. Pratt LA, Ford DE, Crum RM, et al. Depression, psychotropic medication, and risk of myocardial infarction: prospective data from the
Baltimore ECA follow-up. Circulation. 1996;94:3123–3129.
5. Anda R, Williamson D, Jones D, et al. Depressed affect, hopelessness,
and the risk of ischemic heart disease in a cohort of US adults. Epidemiology. 1993;4:285–294.
6. Barefoot JC, Schroll M. Symptoms of depression, acute myocardial
infarction, and total mortality in a community sample. Circulation. 1996;
93:1976 –1980.
7. Everson SA, Goldberg DE, Kaplan GA, et al. Hopelessness and risk of
mortality and incidence of myocardial infarction and cancer. Psychosom
Med. 1996;58:113–121.
8. Wassertheil-Smoller S, Applegate WB, Berge K, et al, for the SHEP
Cooperative Research Group. Change in depression as a precursor of
cardiovascular events. Arch Intern Med. 1996;156:553–561.
9. Penninx BW, Guralnik JM, Mendes de Leon CF, et al. Cardiovascular
events and mortality in newly and chronically depressed persons ⬎70
years of age. Am J Cardiol. 1998;81:988 –994.
10. Everson SA, Roberts ER, Goldberg DE, et al. Depressive symptoms and
increased risk of stroke mortality over a 29-year period. Arch Intern Med.
1998;158:1133–1138.
11. Whooley MA, Browner WS, for the Study of Osteoporotic Fractures
Research Group. Association between depressive symptoms and mortality in older women. Arch Intern Med. 1998;158:2129 –2135.
12. Colantonio A, Kasi SV, Ostfeld AM. Depressive symptoms and other
psychosocial factors as predictors of stroke in the elderly. Am J Epidemiol. 1992;136:884 – 894.
13. Vogt T, Pope C, Mullooly J. Mental health status as a predictor of
morbidity and mortality: a 15 year follow-up of members of a health
maintenance organization. Am J Public Health. 1994;84:227–231.
14. Thomas C, Kelman HR, Kennedy GR, et al. Depressive symptoms and
mortality in elderly persons. J Gerontol. 1992;47(suppl 2):S80 –S87.
15. Fried LP, Borhani NO, Enright P, et al. The Cardiovascular Health Study:
design and rationale. Ann Epidemiol. 1991;3:263–276.
16. Tell GS, Fried LP, Hermanson B, et al, for the Cardiovascular Health
Study (CHS) collaborative group. Recruitment of adults 65 years and
older as participants in the Cardiovascular Health Study. Ann Epidemiol.
1993;3:358 –366.
17. Ives GD, Fitzpatrick AL, Bild DE, et al. Surveillance and ascertainment
of cardiovascular events: the Cardiovascular Health Study. Ann Epidemiol. 1995;5:278 –285.
18. US Department of Health and Human Services. International Classification of Diseases. 9th rev. Vol 1. 2nd ed. Washington, DC: US
Department of Health and Human Services; 1980. DHHS publication:
(PHS) 80 –1260.
19. Radloff LS. The CES-D scale: a self-report depression scale for research
in the general population. Appl Psychol Meas. 1977;3:385– 401.
Ariyo et al
Depressive Symptoms and Cardiovascular Disease
20. Shrout PE, Yager TJ. Reliability and validation of screening scales: effect
of reducing scale length. J Clin Epidemiol. 1989;42:69 –78.
21. Whooley MA, Avins AL, Miranda J, et al. Case-finding instruments for
depression: two questions are as good as many. J Gen Intern Med.
1997;12:439 – 445.
22. SAS Institute Inc. Phreg, Corr, and NPAR1WAY procedures. In:
SAS/STAT User’s Guide. Version 6. Vol 1 and 2. 4th ed. Cary, NC: SAS
Institute Inc; 1992.
23. Kannel WB, Vokonas PS. Primary risk factors for coronary heart disease in
the elderly: the Framingham Study. In: Wenger NK, Furberg CD, Pitt E, eds.
Coronary Heart Disease in the Elderly. New York: Elsevier; 1986:60–95.
24. Rose G, Shipley M. Plasma cholesterol concentration and death from
coronary heart disease: 10 year results of the Whitehall study. BMJ.
1986;293:306 –7.
25. Langer RD, Ganiats TG, Barrett-Connor E. Paradoxical survival of
elderly men with high blood pressure. BMJ. 1989;298:1356 –1358.
26. Mattila K, Haavisto M, Rajala S, et al. Blood pressure and five year
survival in the very old. BMJ. 1988;296:887– 889.
27. Dimsdale JE. Coronary heart disease in women: personality and stressinduced biological responses. Ann Behav Med. 1993;15:119 –123.
28. Fielding R. Depression and acute myocardial infarction: a review and
reinterpretation. Sci Med. 1991;32:1017–1027.
29. Haft JI. Cardiovascular injury induced by sympathetic catecholamines.
Prog Cardiovasc Dis. 1974;17:73– 86.
1779
30. Hjemdahl P, Larsson T, Wallen NH. Effects of stress and ␤-blockade on
platelet function. Circulation. 1991;84(suppl VI):44 – 61.
31. Pryce IG. The relationship between 17-hydroxycorticosteroid
excretion and glucose utilization in depressions. Br J Psychiatry.
1964;110:90 –94.
32. Rahe RH, Rubin RT, Gunderson EKE, et al. Psychologic correlates of serum
cholesterol in man: a longitudinal study. Psychosom Med. 1971;33:399–408.
33. Appels A, Mulder P. Fatigue and heart disease: the association between
vital exhaustion and past, present and future coronary heart disease.
J Psychosom Res. 1989;33:727–738.
34. Pryce IG. Melancholia, glucose tolerance and body weight. J Ment Sci.
1958;104:421– 427.
35. Appels A, Mulder P. Excess fatigue as a precursor of myocardial
infarction. Eur Heart J. 1988;9:758 –764.
36. Verrier RL, Lown B. Behavioral stress and cardiac arrhythmias. Ann Rev
Physiol. 1984;46:155–176.
37. American Psychiatry Association. Diagnostic and Statistical Manual of
Mental Disorders. 4th ed. Washington, DC: American Psychiatry Association; 1994.
38. Greenberg PE, Stiglin LE, Finkelstein SN, et al. The economic burden of
depression in 1990. J Clin Psychiatry. 1993;54:405– 418.
39. Lamberg L. Treating depression in the medical community may improve
quality of life. JAMA. 1996;276:857– 858.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Depressive Symptoms and Risks of Coronary Heart Disease and Mortality in Elderly
Americans
Abraham A. Ariyo, Mary Haan, Catherine M. Tangen, John C. Rutledge, Mary Cushman,
Adrian Dobs and Curt D. Furberg
for the Cardiovascular Health Study Collaborative Research Group
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Circulation. 2000;102:1773-1779
doi: 10.1161/01.CIR.102.15.1773
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2000 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/102/15/1773
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/