BMC Public Health. 2026 Jan 26. doi: 10.1186/s12889-026-26247-1. Online ahead of print.
ABSTRACT
BACKGROUND: Obesity is a growing public health concern, particularly in Sub-Saharan Africa, where undernutrition and obesity are in conjuction. In Botswana, the urban transition and environmental factors are likely influencing obesity risk. This study aimed to investigate the spatial distribution of central obesity across selected rural and peri-urban communities in Botswana and elucidate on likely demographics influencing the identified spatial variations.
METHODS: Data were drawn from 2,039 individuals (aged 16-65 + years) who participated in the Botswana Combination Prevention Project (BCPP), a community-based (n = 22) cluster-randomized trial (NCT01965470). Central obesity was defined using waist-to-hip ratio > 0.90 in men, > 0.85 in women. Generalized linear mixed models (GLMM) were estimated to predict central obesity prevalence. A Global Moran’s I was used to evaluate if there was clustering of central obesity prevalence across communities. To further explore local patterns, Local Indicators of Spatial Association (LISA) was applied to identify specific communities where central obesity prevalence was not random. This analysis distinguished spatial clusters (high-high [HH] and low-low [LL]) as well as spatial outliers. Finally, bivariate local Moran’s I was performed to determine the spatial heterogeneity of central obesity and specific predictors (female and median age).
RESULTS: Central obesity prevalence was estimated at 27.6% using GLMM (AoR 0.276 95% CI 0.131-0.582), (n = 887) and of these, 44.8% (n = 645) participants, resided in rural areas. Spatial analysis revealed significant clustering across rural and peri-urban communities (Moran’s I = 0.158, p = 0.04;I = 0.171 p = 0.05). The bivariate Local Moran’s I results showed significantly high clusters of central obesity and female (sex) (0.071 p < .001) and central obesity and median age (I = 0.225 p < .001) in both rural and peri-urban communities.
CONCLUSIONS: Central obesity in Botswana demonstrates spatial clustering and is significantly influenced by socio-demographic factors. These findings underscore the need for geographically targeted interventions to address central obesity, especially in peri-urban communities.
TRIAL REGISTRATION: NCT01965470. Registered on October 2013. More information: https://www.
CLINICALTRIALS: gov/study/NCT01965470?tab=history&a=4.
PMID:41588385 | DOI:10.1186/s12889-026-26247-1
