Causal Effects of Low Income on Obesity: Business and Health Insights From a National Survey and Machine Learning Analysis With Applied Econometrics Technique

Abstract

This study investigates the causal impact of low income on Body Mass Index (BMI) using data from the 2017–2018 National Health and Nutrition Examination Survey (NHANES). While previous research has established a correlation between socioeconomic status and obesity, this study employs Double Machine Learning (DML) to identify causal effects, controlling for confounders such as age, gender, education, ethnicity, and household size. The full sample (n = 8,005) and two subgroups, high BMI and high BMI + low income, were analyzed. Results from DML indicate a statistically significant causal effect, with low-income status increasing BMI by approximately 0.49 units (p < 0.001). Subgroup analyses reveal that low-income individuals, especially older adults and females, face disproportionately higher obesity risks. These findings underscore the need for equity-centered public health strategies targeting the socioeconomic roots of obesity, including nutritional support, education, and community-based interventions.