Gender differences in sociodemographic and behavioural factors associated with BMI in an adult population in rural Burkina Faso - an AWI-Gen sub-study.

Background: The global health transition is linked with an increased burden of non-communicable diseases with cardiovascular diseases leading the epidemic. In sub-Saharan Africa (SSA), the prevalence of obesity has increased during the past decades and there is a need to investigate the associated driving factors. In Burkina Faso obesity remains low, especially in rural areas. In this study we recruited middle-aged adults, as part of a larger study on genetic and environmental contributions to cardiometabolic disease among Africans.

Objectives: To investigate the distribution of BMI and prevalence of obesity in a cross-sectional population-based study and to determine the sociodemographic and behavioural correlates with BMI.

Methods: Participants (N = 2,076) were recruited from the Nanoro Health and Demographic Surveillance System area and were aged 40-60 years. We applied hierarchical modelling to identify factors associated with BMI and structural equation modelling to identify mediated effects of sociodemographic and behavioural variables on BMI.

Results: Data are presented on 2,076 participants (49.9% female). Men had significantly higher BMI than women with medians of 21.1 (19.2 - 23.4) vs 19.8 (18.1 - 21.6) (p < 0.001), and there were significantly more underweight women compared to men (31.0% vs 17.4%) (p < 0.001). More men were overweight and obese than women (11.9% vs 5.2% and 2.2% vs 1.4%). Socioeconomic status was the major contributor to increased BMI for men, and education was the main contributor in women. Tobacco smoking and chewing, and problematic alcohol consumption were associated with a decrease in BMI in men and women.

Conclusion: Overweight and obesity are relatively low among adults in rural Burkina Faso, and men had a higher median BMI than women. Behavioural factors, including tobacco use and alcohol consumption, contributed to a decrease in BMI, whereas socioeconomic status and education (which were both generally low in this community) contributed to an increase in BMI.