Percentage Body Fat Corresponding to Korean BMI Classification for Obesity in a Nationally Representative Population.

Background/

Objectives

Body mass index (BMI) does not distinguish fat mass from lean mass. Gallagher et al. proposed percentage body fat (%BF) ranges at NIH/WHO BMI cutoffs, but no study has derived %BF ranges for the full set of Asian-specific BMI cutoffs adopted in Korea. This study aimed to develop and externally validate prediction equations for %BF from BMI in Korean adults and to derive %BF ranges corresponding to Korean Society for the Study of Obesity (KSSO) BMI cutoffs.

Methods

Sex-specific equations were developed using survey-weighted linear regression with the 1/BMI transformation in adults aged ≥ 19 years from KNHANES 2022 (n = 4466) using InBody 970 BIA, and externally validated in KNHANES 2023 (n = 4778). Predicted %BF values were calculated at each KSSO BMI cutoff (18.5, 23.0, 25.0, 30.0, and 35.0 kg/m2) across age-group midpoints.

Results

Final equations were: Men, %BF = 51.98 - 733.89 × (1/BMI) + 0.058 × age (R2 = 0.544; SEE = 3.95%); Women, %BF = 64.92 - 731.24 × (1/BMI) (R2 = 0.638; SEE = 3.66%). Age was not significant in women (p = 0.82). External validation showed mean bias of -0.10% and calibration slope of 1.02. The predicted %BF at the pre-obese cutoff (BMI 23.0) was 21.5-24.4% in men and 33.1% in women; at class I obesity (BMI 25.0), 24.1-27.0% and 35.7%, respectively.

Conclusions

This study provides the first nationally representative %BF ranges for all five KSSO BMI cutoffs. The predicted %BF at BMI 25.0 corresponds to %BF-defined obesity in both sexes, and is consistent with the clinical relevance of the lower obesity cutoff adopted for Asian populations.