Previous studies indicate that body adiposity is associated with diabetes, myocardial infarction, stroke, hypertension, certain cancers, and mortality. BMI has limitations in assessing body adiposity, as it does not distinguish between fat mass and fat-free mass, nor does it take into account sex-related differences in fat composition. Thus, alternative methods are needed.
For a study published in Scientific Reports, Richard Bergman, PhD, and I sought to identify a simple and more accurate anthropometric equation to estimate whole-body fat percentage–a measure of body adiposity–among adult individuals. Using data from the National Health and Nutrition Examination Survey (NHANES) 1999-2004 (~12,600 individuals), we developed more than 350 indices using common anthropometrics. Among these indices, height/waist circumference ratio was the simplest index among women and men, showing a good correlation with whole-body fat percentage measured by DXA scan, superior to BMI and waist/height ratio. Next, we developed a simple linear equation based on height/waist ratio, called relative fat mass (RFM). We validated our findings in nearly 3,500 participants of the NHANES 2005-2006. RFM was a better predictor and estimator of body fat percentage than BMI among women and men, independently. RFM performance was also more consistent than that of BMI across ethnic groups and age categories. Obesity total misclassification error was lower with RFM than with BMI among women (~13% vs 57%) and men (~9% vs 13%). RFM had a better diagnostic accuracy for diabetes than BMI overall and DXA-measured central body adiposity among women.
Obesity is a major health problem worldwide. The use of RFM could better identify individuals with actual body adiposity who would benefit of prompt counselling, lifestyle intervention, or even pharmacologic treatment. However, longitudinal studies are required to determine the recommended healthy and unhealthy ranges for body fat percentage based on health outcomes and mortality. RFM performance in other populations also warrants further studies, and we are currently working on the performance of the RFM index in children and adolescents.