Psychosocial factors and obesity in 17 high-, middle- andlow-income countries: the Prospective Urban RuralEpidemiologic study

Psychosocial stress has been proposed to contribute to obesity, particularly abdominal, or centralobesity, through chronic activation of the neuroendocrine systems. However, these putative relationships are complex anddependent on country and cultural context. We investigated the association between psychosocial factors and general andabdominal obesity in the Prospective Urban Rural Epidemiologic study.SUBJECTS/METHODS: This observational, cross-sectional study enrolled 151 966 individuals aged 35–70 years from 628 urban andrural communities in 17 high-, middle- and low-income countries. Data were collected for 125 290 individuals regarding education,anthropometrics, hypertension/diabetes, tobacco/alcohol use, diet and psychosocial factors (self-perceived stress and depression).RESULTS: After standardization for age, sex, country income and urban/rural location, the proportion with obesity (body massindex ⩾ 30 kgm−2) increased from 15.7% in 40 831 individuals with no stress to 20.5% in 7720 individuals with permanent stress,with corresponding proportions for ethnicity- and sex-specific central obesity of 48.6% and 53.5%, respectively (Po0.0001 forboth). Associations between stress and hypertension/diabetes tended to be inverse. Estimating the total effect of permanent stresswith age, sex, physical activity, education and region as confounders, no relationship between stress and obesity persisted(adjusted prevalence ratio (PR) for obesity 1.04 (95% confidence interval: 0.99–1.10)). There was no relationship between ethnicityandsex-specific central obesity (adjusted PR 1.00 (0.97–1.02)). Stratification by region yielded inconsistent associations. Depressionwas weakly but independently linked to obesity (PR 1.08 (1.04–1.12)), and very marginally to abdominal obesity (PR 1.01(1.00–1.03)).