Doctor of Philosophy (PhD)



Document Type



Psychosocial factors are thought to influence health through primarily direct physiological mechanisms or the alteration of health related behaviors. Three factors hypothesized to negatively impact health include arousal, life stress, and depressive symptomatology. One recent theorist suggests that the interaction between psychological stress and stress hormones on the neuroendocrine system may result in adverse changes to body composition, most notably the increased deposition of visceral adipose tissue (Bjorntorp, 1993). The current study prospectively examined the relationship between self-reported stressful life events, depressive symptoms and trait arousal on the deposition of visceral fat, as measured by computerized tomography (CT). Subjects were obtained from a sample of middle-aged males and females (n = 120). Stress measures included the Weekly Stress Inventory (WSI), a life-events measure of minor stressors, and the Life Events Survey (LES) a measure of major life events. Depression symptoms were measured with the Center for Epidemiological Studies Depression Scale (CES-D). Stress and depression were assessed at baseline, 6 and 12 months, and the CT images of visceral fat were obtained at baseline and 12 months. Trait arousal was measured with the Arousal Predisposition Scale at baseline. Arousal, stress and depression scores over 12 months were then standardized and averaged, and entered into a hierarchical multiple regression model in order to predict changes in visceral adiposity from baseline to 12 months. The model was significant in predicting visceral fat, accounting for 16.9% of the variance. Further examination of the model indicated the presence of a significant 3-way interaction between arousal, stress and depression, such that visceral fat was predicted by the interaction of low arousal, high stress and high depression. When the interaction terms were added to the regression analysis as additional steps, the model continued to be significant, accounting for 20.9% of the variance. Interestingly, these models were significant in predicting visceral adiposity despite the fact that the relationships observed were not all in the expected directions. These findings have implications for both researchers and clinicians, who may wish to incorporate more specific psychosocial measures and interventions in the study and treatment of overweight and obesity.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Phillip J. Brantley



Included in

Psychology Commons