Abstract

This study employs quantitative modeling to examine the causal relationship between socioeconomic status, specifically low income, and Body Mass Index (BMI), utilizing Double Machine Learning (DML) as a robust econometric framework. Using nationally representative data from the National Health and Nutrition Examination Survey (NHANES), the analysis integrates demographic, socioeconomic, and lifestyle covariates to isolate the impact of income levels on BMI outcomes. Descriptive statistics indicate only marginal differences in average BMI between low-income and non-low-income groups; however, substantial within-group variability is observed. Through causal inference using DML, the study identifies a statistically significant positive effect of low-income status on BMI, with an estimated causal effect of 0.4856 (95% CI: [0.4221, 0.5491], p < 0.0001). Subgroup analyses reveal that this effect is compounded by variables such as education level, household size, and age, particularly in individuals categorized as both low-income and high BMI. These findings provide strong evidence of the socioeconomic gradient in health and underscore the value of advanced quantitative modeling in informing public health policy, targeted interventions, and resource allocation.

This work is licensed under a Creative Commons Attribution 4.0 License.