Prevalence and Predictors of Obesity among Undergraduate Medical Students A Hierarchical Modelling Approach

Abstract

Background: Obesity is an escalating public health challenge in India, particularly among medical students who navigate high-stress "obesogenic" environments. This study assessed obesity prevalence and the relative impact of non-modifiable traits versus modifiable lifestyle behaviours among undergraduate medical students in northern West Bengal.

Methods: An observational, cross-sectional study was conducted in January 2026 at Jalpaiguri Government Medical College. Data were collected from 151 MBBS students via complete enumeration. Anthropometric measurements were taken, with obesity defined as a BMI ≥ 25 kg/m² as per NCDC India guidelines. Hierarchical logistic regression was performed: Model 1 assessed non-modifiable factors (age, sex, family history), and Model 2 added modifiable lifestyle factors (diet, sleep, exercise).

Results: The prevalence of obesity was 27.8%. Initial bivariate analyses showed no significant associations between obesity and individual predictors. In the hierarchical analysis, Model 1 demonstrated low explanatory power (Nagelkerke R² = 0.033) and poor fit. The addition of lifestyle variables in Model 2 improved model calibration (Hosmer–Lemeshow p = 0.353) and increased the explained variance to 19.1%. Notably, students participating in outdoor games 4–5 times weekly had significantly higher obesity odds (OR = 4.945; 95% CI: 1.060–23.068) than those playing less than once weekly, suggesting possible reverse causality. Other lifestyle factors did not reach levels of statistical significance.

Conclusion: A substantial proportion of medical students are obese. While most demographic and lifestyle variables lacked linear statistical significance, the improvement in model fit with behavioural factors suggests a more nuanced influence. Implementing campus-level wellness initiatives, including structured physical activity and sleep hygiene, is necessary to mitigate long-term health risks.