Health Insurance and Mortality Outcomes - Extending Observational Methodologies to the Publicly Insured

abstract: Existing research into the health benefits of insurance fall into two major categories \u2014 observational and experimental. Observational studies have centered on data sets from before 2000 and focus on the mortality differences between the privately insured and the uninsured. Experimental studies began with Massachusetts' 2006 health reform and continued after the passage of the Affordable Care Act. These studies measure the effects of public insurance among the coverage expansion populations. These two bodies of literature come to ambiguous and contradictory conclusions to the mortality effects and health value of insurance. This study extends the observational methodologies to the publicly insured in samples from the National Health and Nutrition Examination Survey in both the 1988-1994 survey and the 2001-2002 survey. Using the Cox Proportional Hazard model, this study estimates the hazard ratios faced by the privately and publicly insured compared to the uninsured. This study finds the publicly insured face hazards 1.5 times those of the uninsured (p<.001), while the privately insured do not face hazards significantly different from those of the uninsured. Literature suggests that some unobserved characteristic of the publicly insured are influencing their mortality. Interacting with participants health reveals that these differences across groups shrink as health declines. Experimental literature suggests that public insurance lowers the uninsured risk from "healthcare amenable" conditions. Treatment of these conditions may explain the hazard reductions among the uninsured in non-excellent health. The high risk of the publicly insured in excellent health defies explanation