Determinants of U.S. Health Insurance Charges: Evidence from Multivariate Regression

U.S. national health expenditures accounted for 17.3% of GDP, raising persistent concerns among insurers, policymakers, and households. Using the Medical Cost Personal Dataset (n=1,338), which includes demographic and lifestyle variables, this study examines the determinants of individual health insurance charges. A multivariate regression framework was used to estimate the effects of age, body mass index (BMI), smoking status, number of children, sex, and region. The results indicate that smoking increased the predicted charges by more than $23,000, while age and BMI also had statistically significant positive effects. In contrast, sex and regional differences were not significant after controlling for other factors. The model explained approximately 74% of the variance in charges. These findings align with prior evidence on lifestyle-related risks and provide support for risk-adjusted premium design, highlighting the policy relevance of smoking cessation and obesity prevention.