Background: There is increasing evidence that depressive symptoms and obesity are positively related, but the mechanisms that explain the association between them are unclear.
Objective: We examined direct and indirect associations between depressive symptoms, emotional eating, physical activity (PA) self-efficacy (ie, an individual's confidence in his or her ability to overcome barriers to maintain PA behaviors), and adiposity indicators.
Design: Structural equation modeling was used to test the hypothesized mediation model in Finnish men (n = 2312) and women (n = 2674) aged 25-74 y from the National Cardiovascular Risk Factor Survey conducted in 2007. The Center for Epidemiologic Studies-Depression Scale, the Three-Factor Eating Questionnaire-R18, and a PA barriers self-efficacy scale were used. Body mass index (BMI), waist circumference (WC), and percentage body fat of participants were measured in a health examination.
Results: Depressive symptoms and emotional eating had positive correlations and PA self-efficacy had negative correlations with BMI, WC, and percentage body fat. Elevated depressive symptoms were related to higher emotional eating (β = 0.38 for men and 0.31 for women) and lower PA self-efficacy (β = -0.41 for men and -0.31 for women), whereas emotional eating and PA self-efficacy were inversely correlated (r = -0.12 and -0.18, respectively). The positive bivariate associations between depressive symptoms and adiposity indicators became nonsignificant in models that included emotional eating and PA self-efficacy, and both of these factors significantly mediated the effects of depressive symptoms on adiposity indicators.
Conclusions: Psychological factors related to both eating and PA may be relevant in explaining the positive relation between depressive symptoms and adiposity. Interventions that target obesity should take into account the effects of these factors on weight regulation.