Numerous studies have shown network assessments of social contact predict mortality in patients with coronary artery disease (CAD). Fewer studies have demonstrated an association between perceived social support and longevity in patient samples. It has been suggested that 1 of the mechanisms linking social support with elevated risk for mortality is the association between social support and other risk factors associated with decreased longevity such as smoking, failure to exercise, and depressive symptoms. The present study examined an assessment of perceived support as a predictor of all-cause and CAD mortality and examined the hypothesis that smoking, sedentary behavior, and depressive symptoms may mediate and/or moderate this association.
Ratings of social support and the risk factors of smoking, sedentary behavior, and depressive symptoms were examined as predictors of survival in 2711 patients with CAD, and associations between support and these risk factors were assessed. Smoking, sedentary behavior, and depressive symptoms were examined as mediators and/or moderators of the association between social support and mortality.
Social support, smoking, sedentary behavior, and depressive symptoms were predictors of mortality (p’s <.01). Results also indicated that sedentary behavior, but not smoking status or depressive symptoms, may substantially mediate the relationship between support and mortality. No evidence for moderation was found.
The relation between social support and longevity may be partially accounted for by the association between support and sedentary behavior.
CAD = coronary artery disease; MOSS = Mediators Of Social Support; ISEL = Interpersonal Support Evaluation List; CHF = congestive heart failure; HR = hazard ratios.
From the Department of Psychiatry and Behavioral Sciences (B.H.B., I.C.S., R.B.W., M.A.B., J.C.B.), Duke Clinical Research Institute (D.B.M., N.E.C.-C.), and the Division of Cardiology, Department of Medicine (D.B.M.), Duke University Medical Center, Durham, North Carolina.
Address correspondence and reprint requests to Beverly H. Brummett, PhD, Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Box 2969, Durham, NC 27710. E-mail: firstname.lastname@example.org.
Received for publication March 29, 2004; revision received August 9, 2004.
This research was supported in part by grants P01-HL36587 and R01- HL54780 from the National Heart, Lung and Blood Institute, and by R01-HL55356 with cofunding by the National Institutes on Aging.