Prevalence of diabetes mellitus type 2 and predictors of glycated hemoglobin in schoolchildren from marginalized communities of Gqeberha, South Africa

Background: In low-and-middle income countries prevalence of diabetes type 2 is increasing at an incomparable rate. Sedentarism and insufficient physical activity are major contributors to this rise. Assessment of glycated hemoglobin is considered as current gold standard for diagnosis of diabetes. Various biomarkers and anthropometric indices are assumed to be associated with glycated hemoglobin and could therefore be beneficial for identifying subjects with diabetes or at risk. The aim of the present study was to calculate the prevalence of diabetes and prediabetes as well as predictors of glycated hemoglobin among a sample of South African schoolchildren. Methods: Cross-sectional data was collected from February to April 2019 in schoolchildren (n = 1,372, age = 6−12 years) living in marginalized neighborhoods of Gqeberha, South Africa. Accelerometry, examination of capillary blood, anthropometric measurements and questionnaires were applied to obtain data on volume of physical activity and sedentary behavior, levels of glycated hemoglobin, hemoglobin, vitamin D, blood lipids, body mass index and socioeconomic status. Results: Prevalence of diabetes and prediabetes corresponded to 0%, 16.7% respectively. Coefficients of determination were small in multiple linear regression models, indicating that the selected explanatory variables only predicted a marginal fraction of variance in glycated hemoglobin. Only hemoglobin levels were considered statistically and practically significantly associated with glycated hemoglobin. Conclusion: In our sample glycated hemoglobin was associated with hemoglobin. Association of glycated hemoglobin with other biomarkers and anthropometric indices was marginal to non-existent. Therefore, caution is advised when interpreting glycated hemoglobin in populations with deviating hemoglobin levels