| Previous literature finds that the uninsured have fewer doctor and preventive visits, while using the emergency room “inappropriately,” relative to the privately insured; the invariable policy conclusion is to extend insurance to the uninsured. However, insurance is a complex policy instrument and policymakers may target either side of the market independently. Nevertheless, the role of the supply-side, i.e. of access barriers faced by the uninsured, has largely been ignored. This dissertation explores differences in supply by patient's insurance status and the impact of such differences on observable health care use differentials. The advantages of studying utilization differences within a diagnostic category are also explored.; The first chapter formalizes the care-seeking decision across two provider types by insurance status. All individuals suffer from one condition though of varying degrees and can seek treatment either at the doctor's office or the emergency room. I find that when the underlying demand for care is the same for both the insured and uninsured, the empirical utilization patterns could be entirely driven by the supply-side, where doctors accept only p < 1 of the uninsured, but all insured patients.; The second chapter empirically explores the role of barriers to access in uninsured care-seeking. Access barriers by primary care physicians take three forms: locating practices away from uninsured populations; refusing to accept uninsured patients; and providing less intense services to the uninsured. I find that access barriers do not impact the privately insured, but induce more emergency room visits and fewer doctor visits for the uninsured, mainly through lower service intensity.; In the third chapter, I develop my own health care use disparity estimate using the treatment of those with a substance abuse diagnosis. I tackle two fundamental problems not previously addressed, which vary systematically by insurance status: variations in health conditions motivating use; and prior treatment history. The uninsured are substantially more likely to choose hospitals than both the privately insured and Medicaid-insured, among publicly-funded facility users. I find support for extending Medicaid to the uninsured to reduce relatively more expensive hospital use and encourage treatment seeking. |
