Background: Depression was positively associated with cardiovascular disease (CVD) or mortality in previous studies. However, whether the observed association can be explained by health status is not clear.
Objectives: To study the association of depressive symptoms with CVD, stroke, and coronary heart disease (CHD) mortality in older Chinese in Hong Kong, and whether the associations varied by gender or health status.
Design: Prospective population-based study.
Setting: Elderly Health Centers.
Participants: A total of 62,839 people age 65 or older (21,473 men and 41,366 women) enrolled during July 1998 to December 2001 at all 18 Elderly Health Centers of the Department of Health of Hong Kong.
Measurements: Fifteen-item Geriatric Depression Scale (GDS) was used and presence of depressive symptoms was defined by GDS score 8 or more. The cohort was followed up for mortality till March 31, 2009.
Results: Depressive symptoms were only associated with CHD mortality in men (hazard ratio [HR] 1.41, 95% confidence interval [CI]: 1.08-1.84; p for gender interaction = 0.02) adjusted for age, education, monthly expenditure, smoking, alcohol use, physical activity, body mass index, health status, and self-rated health. GDS score was associated with stroke mortality (similarly adjusted HR 1.02 per score, 95% CI: 1.00-1.04) in all subjects (adjusted also for gender), and CHD mortality (1.04 [1.01-1.07]) in men. Health status attenuated but did not modify any associations.
Conclusion: Depressive symptoms were independently associated with higher CHD mortality in older Chinese men, and with higher stroke mortality in both genders. However, attenuation by health status, and lack of consistency by gender indicate that these associations could be noncausal and further studies by treatment trials and Mendelian randomization are needed.
Keywords: Cardiovascular mortality; Geriatric Depression Scale; coronary heart disease mortality; depressive symptoms; stroke mortality.
Copyright © 2013 American Association for Geriatric Psychiatry. Published by Elsevier Inc. All rights reserved.