Application of Causal Inference Techniques to Examine the Determinants and Outcomes of Obesity

Background Obesity is a multifaceted and chronic health condition with major implications for global public health and socio-economic systems. Guided by the socio-ecological model, this study frames obesity as the outcome of a complex interaction between biological, behavioral, and environmental determinants. Methods To empirically examine these determinants, the study employs advanced causal inference techniques, including Ordinary Least Squares (OLS), Instrumental Variables (IV), and Propensity Score Matching (PSM). A dataset of 2,111 records from Colombia, Peru, and Mexico is analyzed to identify factors associated with obesity while addressing confounding and endogeneity concerns. Results The study identifies key variables linked to obesity, such as age, gender, family history, and high-calorie food consumption. The overall prevalence of obesity is found to be 46.1%. Significant associations also emerge between obesity and factors including family history, physical activity, and mode of transportation. Conclusion By applying robust causal inference methods, the study provides reliable evidence on the determinants of obesity and offers actionable insights to guide targeted interventions and public health policies. The findings highlight the importance of combining theoretical frameworks with causal techniques to better understand and address the complexity of obesity.