Metabolic risk patterns and associations of anthropometric indices with elevated random blood glucose in urban Somalia: a cross-sectional study

Background Rapid urbanization in Somalia is driving an undocumented rise in metabolic disorders, yet context-specific data remain scarce.Aim To characterize metabolic risk patterns and evaluate the association between body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR), and elevated random blood glucose among urban adults in Bosaso, Somalia.Methods A community-based cross-sectional study was conducted with 160 adults aged ≥18 years in Bosaso. Participants underwent anthropometric measurements and random blood glucose (RBG) testing. Screen-positive elevated RBG was defined as RBG ≥140 mg/dL (7.8 mmol/L). K-means clustering was used to identify exploratory metabolic patterns. Receiver operating characteristic (ROC) analysis assessed the discriminative ability of anthropometric indices.Results The prevalence of screen-positive elevated RBG was 13.8%. Participants with elevated RBG had significantly higher BMI (31.5 vs. 26.2 kg/m2, p < 0.001) and WC (100.9 vs. 94.3 cm, p = 0.003) than those with normal RBG. WHR did not differ significantly between groups (p = 0.761). In ROC analysis, BMI demonstrated the highest discriminatory ability (AUC = 0.76, 95% CI 0.64–0.88), followed by WC (AUC = 0.71). WHR showed no meaningful discriminatory ability (AUC = 0.51). Cluster analysis identified a distinct ‘High Sugar–Normal BMI’ subgroup, representing 30.0% of the sample.Conclusions In this urban Somali population, BMI showed the highest exploratory discrimination for screen-positive elevated RBG, followed by WC, while WHR showed no meaningful discriminatory value. These findings should be interpreted as preliminary. The presence of non-obese individuals with high sugar intake suggests a need for comprehensive screening strategies beyond anthropometry alone.