Hostility now, depression later? Longitudinal associations among emotional risk factors for coronary artery disease

Abstract

Background/purpose: Given that emotional risk factors for coronary artery disease (CAD) tend to cluster within individuals, surprisingly little is known about how these negative emotions might influence one another over time. We examined the longitudinal associations among measures of depressive symptoms and hostility/anger in a cohort of 296 healthy, older adults.

Methods: Participants completed the Beck Depression Inventory-II (BDI-II), Cook-Medley Hostility (Ho) scale, and Anger-In and Anger-Out subscales of the State-Trait Anger Expression Inventory at baseline and 6-year follow-up. We conducted a series of path analyses to evaluate the directionality of the depression-hostility/anger relationship.

Results: Baseline Ho scale was a predictor of 6-year increases in BDI-II (beta = 0.15, p = 0.004), Anger-In (beta = 0.14, p = 0.002), and Anger-Out (beta = 0.11, p = 0.01). In contrast, baseline BDI-II, Anger-In, and Anger-Out did not predict change in any of the emotional variables. Additional path analytic models revealed that the pattern of relationships was not altered after controlling for demographic, biomedical, and behavioral covariates; anxiety symptoms; social support; and subjective sleep quality.

Conclusions: The present results suggest that the cognitive aspects of hostility/anger may precede and independently predict future increases in depressive symptoms but not vice versa. Our findings lead us to speculate that (a) hostility may exert part of its cardiotoxic influence by acting to precipitate and/or maintain symptoms of depression and that (b) the potency of depression interventions designed to improve cardiovascular outcomes might be enhanced by incorporating treatments addressing hostility.

Figure 1

Results of the path analysis examining the longitudinal associations among measures of depressive symptoms and hostility/anger. Values associated with unidirectional arrows (i.e., structural paths) are standardized regression coefficients. Only the significant structural paths are shown in the figure; correlations and nonsignificant structural paths were omitted. N = 296.