Prevalence of obesity using the body mass index and equivalent age- and sex-specific thresholds for percentage body fat mass

Abstract

The body mass index (BMI, kg/m2) of at least 30 kg/m2 is used to identify obesity. However, the BMI indicates weight adjusted for height without considering differences in body composition and the contribution of body fat mass to an individual’s weight. Consequently, the BMI as an indicator of adiposity tends to over-estimate adiposity in muscular body builds and under-estimate adiposity in the elderly. The aim of this study was to identify ageand sex-specific values for percentage body fat mass (%BFM), measured using whole body dual energy Xray absorptiometry (DXA), that correspond to BMI 30 kg/m2 [1], and to compare the prevalence of obesity in the adult Australian population using BMI≥ 30 kg/m2 and the equivalent cut-points for %BFM. This analysis utilises data from randomlyselected men (n = 1467) and women (n = 1076), age 20—96 years, who had BMI and DXAmeasures as part of the Geelong Osteoporosis Study, 2001—2008. Age-specific %BFM cut-points for obesity in men and women, respectively, were calculated as: 20—29 years 27.5%, 43.4%; 30—39 years 28.1%, 43.6%; 40—49 years 28.7%, 43.8%; 50—59 years 29.3%, 44.0%; 60—69 years 29.9%, 44.2%; 70—79 years 30.5%, 44.4%; 80+ years 31.1%, 44.6%. Using these cut-points, the age-standardised prevalence of obesity for men is 25.0% (95% CI 22.6—27.4) and for women is 29.8% (95% CI 27.0—32.6). Prevalence using BMI≥ 30 kg/m2 is 20.2% (95% CI 18.0—22.4) for men and 28.6% (95% CI 25.8—31.3) for women. Estimating the prevalence of obesity using one BMI threshold may under-estimate the true extent of obesity in the population, particularly among men. We recognise that optimal thresholds for defining obesity will depend on risk assessment for morbidity and mortality.