Optimal body fat percentage cut-off values for identifying cardiovascular risk factors in Mongolian and Han adults: a population-based cross-sectional study in Inner Mongolia, China

OBJECTIVE The present study was designed to determine the optimal cut-off values of body fat percentage (BF%) for the detection of cardiovascular disease (CVD) risk factors in Mongolian and Han adults. METHOD This cross-sectional study involving 3221 Chinese adults (2308 Han and 913 Mongolian) aged 20-80 years was conducted in Inner Mongolia Autonomous Region, China, in 2014. Data from a standardised questionnaire, physical examination and blood sample were obtained. The BF% was estimated using bioelectrical impedance analysis. Optimal BF% cut-offs were analysed by receiver operating characteristic curves to predict the risk of diabetes, hypertension and dyslipidaemia. Binary logistic regression analysis was performed to evaluate the OR of each CVD risk factor according to obesity defined by BF%. RESULTS Mean BF% levels were lower in men than in women (22.54±5.77 vs 32.95±6.18 in Han, 23.86±5.72 vs 33.98±6.40 in Mongolian population, respectively; p<0.001). In Han population, the area under curve (AUC) values for BF% ranged from 0.589 to 0.699 for men and from 0.711 to 0.763 for women. Compared with men, AUCs for diabetes and clustering of ≥2 risk factors in women were significantly higher (p<0.05). The AUCs for BF% in women (0.685-0.783) were similar with those in men (0.686-0.736) for CVD risk factors in Mongolian population. In Han adults, the optimal BF% cut-off values to detect CVD risk factors varied from 18.7% to 24.2% in men and 32.7% to 35.4% in women. In Mongolian population, the optimal cut-off values of BF% for men and women ranged from 21.0% to 24.6% and from 35.7% to 40.0%, respectively. Subjects with high BF% (≥24% in men, ≥34% in women) had higher risk of CVD risk factors in Han (age-adjusted ORs from 1.479 to 3.680, 2.660 to 4.016, respectively). In Mongolia, adults with high BF% (≥25% in men, ≥35% in women) had higher risk of CVD risk factors (age-adjusted ORs from 2.587 to 3.772, 2.061 to 4.882, respectively). CONCLUSIONS The optimal BF% cut-offs for obesity for the prediction of CVD risk factors in Chinese men and women were approximately 24% and 34% for Han adults and 25% and 35% for Mongolian population of Inner Mongolia, China, respectively.