Longitudinal associations between depressive symptoms and body mass index in a 20-year follow-up

Abstract

Longitudinal studies have rarely investigated changes in depressive symptoms and indicators of obesity simultaneously, although it is often proposed that the positive relationship between depression and obesity is bidirectional. The present study examined the reciprocal nature of the relationship between depressive symptoms and body mass index (BMI) in a 20-year follow-up survey. Participants of a Finnish cohort study in 1989 at 22 years (N=1656) were followed up at ages 32 (N=1262) and 42 (N=1155) with postal questionnaires. BMI was calculated on the basis of self-reported weight and height, and depressive symptoms were assessed using the short form of the Beck Depression Inventory. Latent growth models (LGM) and cross-lagged autoregressive models were used to determine prospective associations between depressive symptoms and BMI. LGM analyses indicated that men with higher initial levels of depressive symptoms experienced a faster rate of increase in BMI (β=0.20, P<0.01). Among women, change in BMI or depressive symptoms was not predicted by the other construct, but initial levels of BMI and depressive symptoms as well as their rate of change correlated positively with each other (r=0.15 and 0.37, respectively). In cross-lagged models, depressive symptoms at age 32 predicted greater BMI at 42 (β=0.10, P<0.001) among men, whereas women with higher BMI at age 32 were more likely to have more depressive symptoms at 42 (β=0.08, P<0.05). Elevated depressive symptoms predicted weight gain in men, while changes in depressive symptoms and body weight occurred concurrently in women. Tentative evidence showed that women with excess body weight were more likely to have increased symptoms of depression 10 years later. More emphasis should be placed on depressive symptoms in weight control programs as well as on reducing weight-based stigmatization and discrimination in society.