Tailoring Body Mass Index for Prediction of Obesity in Young Adults: A Multi-Centric Study on MBBS Students of Southeast India

Introduction: Body mass index (MI) has been used for a long period as a surrogative measure for obesity. But BMI does not differentiate between fat and nonfat tissue (blood, bone, and muscle) due to which it is not considered accurate anymore. But since BMI is easier to estimate and used widely for assessment of obesity, it is better if it is re-standardized according to the body fat percentage (BFP) of a specific population, community, and their ethnicity. Objective: To estimate and propose the BMI cut-off values in young Indian population especially MBBS students taking BFP as a standard. Design: This is a cross-sectional study. Anthropometric data (age, gender, height, weight, waist circumference, and hip circumference) were collected from the participants after taking consent. BMI was calculated using Quetelet's Rule. BFP was estimated using Omron Body fat Monitor (HBF 385). It measures the BFP by the bioelectrical impedance (BI) method. Data were analyzed with appropriate statistical tests and receiver operating curve (ROC) curves were drawn to find the cut-off values of BMI to determine obesity. Setting: The present study is a multi-centric study conducted in four medical colleges (two in each state; Odisha and Andhra Pradesh, India). Participants: Apparently healthy MIII3S students aged 18-24 years were included in this study. Students having any chronic or acute illnesses were excluded from the study. Out of 904 students contacted from four medical colleges, 863 (430 males and 433 females) consented and participated. Results: Some 863 MBBS students have participated in this study. After adjusting for age, BMI was found to he higher in males. BMI was found to be 29.33 for males and in females it was 29.06. BFP was higher in females (34.23) as compared to males (20.77). Waist hip ratio was found to be higher in females (0.92) than in males (0.84). Whereas, fat free mass (FFM) and fat free mass index (FFMI) are higher in males, i.e., 56.24 and 18.48 respectively. Most appropriate cut-off value for obesity on ROC curve was found to be 22.09 (sensitivity 84.5%, specificity 83.46%) in males and that of females was 23.73 (sensitivity 85.26, specificity 81.23). Whereas, the conventional cut-off of 25 for males had sensitivity of only 46% and that of females was 70.5%. For total population BMI cut-off value was found to be 22.2 with 81% sensitivity and 74% specificity. Conclusion: We propose the cut-off value for overweight/obesity in males to be 22.09 kg/m and for females to be 23.73 kg/m(2) in young adult Indian population. These values were found to have more sensitivity and specificity than current BMI cut-off value.