Cardiovascular Anthropometry: What Is Best Suited for Large-Scale Population Screening in Sub-Saharan Africa?

Abstract

Background : Body mass index (BMI) measures overweight/obesity. It, however, especially in sub-Saharan Africa (SSA), misclassifies cardiometabolic risk. Central obesity measures are superior. We therefore sought to compare BMI, waist-to-hip ratio (WHR) and abdominal height (AH) in predicting cardiovascular disease risk in sub-Saharan Africa. Methods : Subjects had blood pressures, BMI, and WHR determined. Blood pressure was taken, weight and height measured to generate BMI, and AH measured with a new locally fabricated abdominometer. The ability of the anthropometric indices in identifying abnormal individuals needing intervention was assessed with sensitivity, specificity, and area under the receiver operator characteristic curve. Results : Adults totaling 1,508 (728 M/780 F) adults were studied. For BMI, 985 (65.3%) were normal, while 375 (24.9%), consisting of 233 males and 142 females, had normal WHR. Blood pressure was normal in 525 (34.8%) and 317 (21.0%) for systolic and diastolic blood pressures, respectively. Using BMI as gold standard, sensitivity, specificity, positive, and negative predictive values for WHR in males were 80.7, 37.5, 62.5, and 19.3%, respectively. For females and in the same order, they were 62.0, 34.3, 65.7, and 38.0%. For AH, it was equal in both genders at 82.6, 39.2, 60.8, and 17.4%. By receiver operating curves comparing AH, WHR, and BMI against blood pressure detection, the area under the curve was 0.745, 0.604, and 0.554 for AH, BMI, and WHR, respectively. Conclusion : Abdominometer-derived AH has a better sensitivity and greater area under the receiver operator curve compared with BMI and WHR in this sub-Sahara African population; implying superiority as a cardiovascular anthropometric index.