Project 7 (2017-2024) - How Health Insurance Affects Health

Understanding the consequences of health insurance coverage is central to evaluating proposals to expand or modify health insurance coverage in the U.S. Yet there is remarkably little convincing evidence on the impact of insuring the uninsured on their medical utilization, health outcomes, health behaviors, or overall well-being, largely because enrolling in health insurance is a choice that is made jointly with other choices that determine these outcomes.

This project takes advantage of a remarkable opportunity to provide just such evidence. For a limited window in early 2008, Oregon opened a waiting list for enrollment in its public health insurance program for low income adults, and then randomly drew names from the list to determine who would be given the opportunity to enroll. This unique policy environment provided researchers with a rare occasion to bring the strengths of random assignment – the standard in medical trials – to address a critical social policy question. This proposal builds on our prior work assessing the effects of Medicaid on health care use, health, and well-being, including a focus on the near-elderly (those aged 50-64) whose health risks are in many ways similar to those aged 65 and up and who will soon age onto Medicare themselves. We found that Medicaid substantially increased health care use across settings (including primary care, hospitals, prescription drugs, and emergency departments (EDs)); reduced financial strain (including bills sent to collection and catastrophic out-of-pocket expenses); and reduced the prevalence of depression; but had no detectable effects on several measures of physical health (including blood pressure, cholesterol, and diabetic blood sugar control).

This project will leverage our work on the Oregon Health Insurance Experiment to build several new data sets and conduct new analyses to assess multi-faceted effects of Medicaid coverage on health and other outcomes, including: care and health outcomes for chronic physical conditions and mental health, including diabetes, asthma, and depression; patterns of ED and hospital use over time and substitutability of different sites of care; and dental care and outcomes. We will also develop a novel econometric method for assessing and synthesizing the multidimensional effects of insurance on health. As policy-makers and stakeholders assess the broad consequences of expanded access to health insurance, we believe these analyses will add crucial evidence on the many potential effects of Medicaid on health care use, quality, and long-term health outcomes
of an aging population.