Problem drinking as intentional risky behavior: Examining the association between state health insurance coverage and excessive alcohol consumption.

The moral hazard theory asserts that having health insurance may increase individual risk-taking behaviors. We examined the association between state health insurance coverage and excessive alcohol use among U.S. adults. We used 2001-2017 Behavioral Risk Factor Surveillance System data to estimate annual binge and heavy drinking rates for each state. In a multivariable regression analysis, we used difference-in-difference (DID) models to assess the association between state-level insurance coverage and binge and heavy drinking. Additionally, we assess the potential asymmetric effect and whether economic recessions (2001, 2008-09) had a moderation effect. In the multivariable DID analysis, aggregate state insurance coverage was not significantly associated with binge drinking rates in baseline analysis with state-fixed effects (Model 1), and in the analysis that extends the baseline model to include state unique time trend (Model 2). A similar result was found for heavy drinking in Model 1. In contrast, the result showed a significant association between health insurance coverage and heavy drinking rates in Model 2. However, we found no significant association for binge and heavy drinking rates in both models in the analyses restricting data to periods before the methodological change in the BRFSS sampling frame. The results did not show asymmetric effects, and the association between health insurance and excessive alcohol use did not differ during economic recessions. These findings largely do not support assertions that health insurance may lead to moral hazards (risk-taking behaviors), especially binge drinking.