A Comparison of Anthropometric Measures with Bioelectrical Impedance Analysis in the Classification of Overweight and Obesity in U.S. Military Personnel (P21-050-19)

Given that the prevalence of obesity in the military is a growing national security concern, we sought to determine: 1) the level of agreement between body mass index (BMI), circumference-based equation (CBE), waist circumference (WC), and bioelectrical impedance analysis (BIA) measures, and 2) if BMI, CBE, or WC measures alone or in combination adequately reflect adiposity in military personnel compared to BIA measures. Measurements taken by trained personnel were used to estimate BMI (using overweight cut-offs in males from the military (BMI, ≥27.5 kg/m²) and World Health Organization (WHO BMI, ≥25.0 kg/m²)), BMI % body fat (BF%), WC, CBE BF%, and BIA BF%. Respondents were categorized as normal vs. overweight/obese or overfat. Anthropometric measures were compared to BIA to examine levels of agreement and standard screening performance measures. Among the 389 participants (78% male), WHO BMI and BIA BF% classified the most males (61.1% and 42.6%, respectively) and females (both 51.2%) as overweight/obese, whereas WC with BMI (BMI + WC) and WC alone were the least likely to classify males (10.9% and 11.6%, respectively) and females (both 9.3%) as overweight/obese. The levels of agreement were all statistically significant and highest for BMI (males Cohen's kappa (Cₖ) = 0.711, females Cₖ = 0.814) and WHO BMI (males Cₖ = 0.578, females Cₖ = 0.814); moderate for BMI + CBE (males Cₖ = 0.447, females Cₖ = 0.676); and lowest for WC and WC + BMI (all Cₖ ≤ 0.270). To maximize sensitivity and minimize false discovery rate, BMI + CBE performed best overall (sensitivity = 50.9%, false discovery rate = 5.4%). Valid measures of adiposity are needed to ensure that military recruitment and job security are not jeopardized by inaccurate body fat standards. Our findings support BMI + CBE as an easy-to-implement combination to assess adiposity in the military. Future studies need to consider overall goals, including associated health risks, before expanding on these findings. All authors contributed their efforts without receiving funding or salary support. The research study was supported by a Grant to the Consortium for Health and Military Performance at the Uniformed Services University from the US Army's Comprehensive Soldier and Family Fitness program.