Abstract
BACKGROUND: To date, no definitive waist circumference (WC) cut-off values for abdominal obesity (AO) have been established for sub-Saharan Africa, including Botswana. Therefore, the classification of AO among these populations is based on European values. For accurate diagnosis of the metabolic syndrome (MetS), cut-off values reflective of the population investigated must be used. OBJECTIVE: The study was an attempt to determine optimal cut-off values for AO among Batswana adults. METHODS: The receiver operating characteristic curve was used to determine the optimal cut-off values for predicting at least two other risk factors of the MetS. Data were used from a descriptive cross-sectional study employing a complex multi-stage cluster sampling. Demographic and anthropometric measurements (weight and height, waist and hip circumferences), blood pressure, and blood glucose, triglycerides, high-density lipoprotein cholesterol and total cholesterol levels were collected from 384 men and 416 women in Gaborone and the surrounding villages. RESULTS: The ability of waist circumference to predict at least two other risk factors of the MetS gave cut-off values of ≥ 91.0 cm (sensitivity 69.1% and specificity of 90.8%, area under the curve 0.85) for men and ≥ 82.3 cm (sensitivity of 88.6% and specificity of 58.9%, area under the curve of 0.76) for women. CONCLUSIONS: There is a difference between the cut-off values for Europeans with those determined for Batswana adults. Inconsistencies in cut-off values used have the potential for undesirable consequences for cardiovascular risk stratification and prioritisation of preventative strategies for AO and the MetS. The need to determine population-, ethnic- and gender-based cut-off values for AO for Batswana adults has never been more paramount.