Modifiable determinants of central obesity among the rural black population in the DIMAMO HDSS, Limpopo, South Africa

Abstract

Background: Central obesity is a leading risk factor for cardiometabolic diseases, in which body fat accumulates to a particular extent, and may negatively impact on health. The prevalence of abdominal obesity has increased over the last 10 years and currently surpasses that of overall obesity. There is a scarcity of data on the determinants of central obesity, especially among populations residing in rural Africa. The aim of the present study was thus to determine sociodemographic and lifestyle factors that are associated with central obesity.

Methods: This was a cross-sectional, retrospective study. The present study used secondary data from the AWI-Gen phase 1 study. The study comprised 791 participants, of which 242 were men and 549 were women aged 40 years and above. The participants were selected by convenient sampling. Data were analyzed using the Statistical Package for Social Sciences version 27. A comparison of proportions was performed using the chi-square test, while a comparison of means was performed using an unpaired Student t-test. The association between sociodemographic and lifestyle factors with central obesity was analyzed using bivariate correlation, partial correlation, and binary regression analysis, and the statistical significance was set at a p-value of <0.05.

Results: The proportion of central obesity in the total population was 59.9%, and significantly more women were centrally obese (79.6 vs. 15.3%, p = <0.001) as compared to men. Married status correlated positively and significantly with central obesity in both bivariate and partial correlations. Moreover, binary logistic regression further confirmed the positive association between married status and central obesity. Single status correlated negatively and significantly with central obesity. The correlation remained unchanged even after controlling for age and gender. Binary logistic regression showed that unemployment correlated significantly with central obesity. The proportion of smokers was also significantly higher in participants without central obesity than in those with central obesity (87.2 vs. 34.0%, p = <0.001). Smoking correlated negatively and significantly with central obesity in bivariate and partial correlations. In addition, binary logistic regression further confirmed the negative association between smoking and central obesity.

Conclusion: The present study shows that in this population, central obesity is determined by gender, unemployment, and marital status.

Keywords: central obesity; educational status; marital status; smoking; unemployed.

Figure 1

Forest plot illustrating the binary logistic regression of central obesity and sociodemographic factors. Normal = 1, OR>1 = positive relationship, OR < 1 = negative relationship. *P = <0.05.