LINEAR REGRESSION MODEL FOR ESTIMATING PERCENT BODY FAT IN YOUNG ADULT MEDICAL STUDENTS: FINDINGS FROM A CROSS-SECTIONAL STUDY

Abstract

Objectives: To identify factors associated with percent body fat (PBF) and to develop a predictive model in healthy young adult medical students. Methods: A cross-sectional study was conducted among 76 medical students aged 22 - 24 years. Body composition was assessed using multi-frequency bioelectrical impedance analysis (BIA). Dietary intake was collected using three non-consecutive 24-hour dietary recalls, and physical activity was assessed using the International Physical Activity Questionnaire - Short Form (IPAQ-SF) with total activity expressed as MET-min/week. Block-wise multivariable linear regression was applied to determine independent predictors of PBF. Results: PBF was correlated positively with fat mass and visceral fat area, and negatively with skeletal muscle mass, skeletal muscle index, and phase angle. In the final multivariable model, sex, height, waist-to-hip ratio (WHR), lipid intake (g/kg/day), and physical activity (MET-min/week) were independently associated with PBF. The model equation was: PBF (%) = 53.80 - 7.02 x sex (male = 1) - 0.2 x height (cm) + 18.24 x WHR - 5.53 x lipid intake (g/kg/day) - 0.001 x physical activity (MET-min/week). Conclusion: In young adult medical students, a simple linear model incorporating WHR, sex, height, lipid intake, and physical activity may help estimate PBF and support screening of adiposity patterns in similar student populations. Given the cross-sectional, single-center design, self-reported measures, and lack of validation, larger multicenter studies are needed.