A causal model of chronic socioeconomic stress and psychological distress as predictors of obesity for Black and White women in NHANES I.

Abstract

Obesity is an important risk factor for chronic diseases and early mortality. Disturbingly, its prevalence varies by socioeconomic class and ethnicity, with low status individuals and African-Americans at higher risk. Severe obesity affects about half of older Black women. For women, the relationship between socioeconomic status and the prevalence of obesity is generally inverse and linear. Biochemical and physiologic studies of adipose tissue metabolism, neurohormonal regulation of energy metabolism, sympathetic nervous system reactivity, and the impact of cognitive stress on those responses, longitudinal studies of adiposity and its correlates, and the social and economic correlates of subjective well-being and psychological distress in women provide an outline of mechanisms by which chronic stress as experienced by women in their social context can promote a pathogenic process leading to major adult weight gain and obesity. The direct effects of chronic psychological distress, as well as its indirect effects through four intermediate, "lifestyle" variables (smoking, alcohol consumption, physical activity and dietary intake), were analyzed in a LISREL structural equation model of obesity among 1790 Black and White women, aged 25 to 75 years, in the first National Health and Nutrition Examination Survey. High levels of chronic psychological distress (modeled by depressed subjective well-being, a latent construct indicated by responses to the General Well-Being Questionnaire) were hypothesized to be causally related to increased adiposity. The causal model estimates were contrasted to multiple linear regression results. In both approaches, psychological distress was found to be modestly, but significantly predictive of greater adiposity. Structural equation modeling not only found a direct effect of psychological distress on adiposity similar to that in linear regression modeling, but the influences of socioeconomic resources and African-American race as exogenous factors were adequately explained by their indirect effects on the intervening, endogenous factors of the model, most importantly physical activity and secondarily distress. The epidemiologic implications of the model for the primary prevention of obesity and related chronic diseases, a hypothesis regarding the cumulative effects of signal recognition problems and response mismatch on the regulatory control of biological systems, and future research possibilities for a molecular biology of stress were discussed.