Gender-Related Differential Effects of Obesity on Health-Related Quality of Life via Obesity-Related Comorbidities: A Mediation Analysis of a French Nationwide Survey.

<bold>Background: </bold>Negative effects of obesity on health-related quality of life (HRQoL) have been reported, especially in women, but the relative contribution of cardiometabolic and other obesity-related comorbidities to such effects remains unclear. Our objective was to model the association by sex between body mass index and HRQoL and to precisely quantify the indirect effects mediated by obesity-related comorbidities.<bold>Methods and Results: </bold>Data were drawn from the latest French Decennial Health Survey, a nationwide cross-sectional study conducted in 2003 (21 239 adults aged 25-64 years analyzed). HRQoL was measured by the 36-item short-form health survey questionnaire. A mediation analysis based on the counterfactual framework was performed to quantify the proportion of obesity effects on HRQoL mediated by related comorbidities, including cardiometabolic risk factors (diabetes mellitus, hypertension, dyslipidemia) and diseases (ischemic heart disease, cerebrovascular, and peripheral vascular disease), musculoskeletal disorders, and asthma. After multiple linear regression, inverse associations were found between increasing body mass index category and physically oriented and most mentally oriented 36-item short-form health survey dimensions, with evidence of greater effects in women. Mediation analysis revealed that obesity effects were significantly mediated by several comorbidities, more apparently in men (eg, proportion of obesity class II total effect mediated via cardiometabolic factors: general health 27.0% [men] versus 13.6% [women]; proportion of obesity class II total effect mediated via total count of comorbidities: physical functioning 17.8% [men] versus 7.7% [women] and general health 37.1% [men] versus 20.3% [women]).<bold>Conclusions: </bold>Women have a greater overall impact of obesity on HRQoL, but with proportionally lower effects mediated by cardiometabolic and other obesity-related conditions, suggesting the possible role of other specific psychosocial processes. [ABSTRACT FROM AUTHOR]