Three Essays in Applied Health Economics

Abstract

In this dissertation study, I examine three issues in applied health economics within the context of the Medicare program. Medicare provides health insurance coverage to the elderly and the disabled. Although the core components of Medicare have remained largely intact since its introduction in 1965, the program has undergone a number of changes in recent decades. During this period, program enrollment and costs have increased dramatically, and Medicare continues to play an increasingly prominent role in the U.S. health care system. In this study, I examine several integral aspects of the Medicare program, and, in doing so, I contribute to our understanding of health economics more broadly. In Chapters 1 and 2, I focus on the interaction between the incentives of private health insurers that operate in the Medicare program and externalities from prescription drugs. More specifically, I test the hypothesis that integrated plans that provide coverage for drug and non-drug expenditures within the Medicare program internalize negative externalities from prescription opioids; because of the breadth of coverage that these plans provide, they have an incentive to consider adverse health outcomes that are linked to opioid use. Using Medicare Part D drug utilization data, I find evidence that supports this hypothesis in Chapter 1; relative to enrollment in a stand-alone drug plan, enrollment in an integrated plan lowered the probability of high dosage opioid use linked to hospitalizations by 32 percent in 2008 and 2009. In Chapter 2, I extend my research question towards benefit design for prescription opioids. Because benefit design directly impacts enrollee drug use, I hypothesize that integrated plans that operate within the Medicare Part D program structure benefits in a way that limits enrollees’ use of high dosage opioids. To examine this issue, I test for differences in benefit design for opioids across integrated plans and stand-alone drug plans. I find that, relative to opioids covered by stand-alone drug plans, opioids covered by integrated drug plans are more likely to have a quantity limit restriction. Furthermore, conditional on a quantity limit restriction, opioids covered by integrated plans have lower opioid dosage allowances relative to opioids covered by stand-alone plans. These results reinforce the finding that integrated plans internalize negative externalities from prescription opioids, and they provide evidence of a mechanism through which this occurs. In Chapter 3, I shift gears and focus on the interaction between the onset of Medicare at age 65 and mental health care utilization and mental health outcomes. Using data from the National Health Insurance Survey, I examine whether the changes in health insurance coverage rates that occur at age 65 are accompanied by changes in mental health outcomes. I employ a regression discontinuity design to test for changes in perceived financial barriers to mental health care, visits with mental health professionals, and self-reported mental health. I find that the onset of Medicare at age 65 is accompanied by a substantial decline in self-reported barriers to receiving mental health care, especially among individuals with lower levels of educational attainment. However, I find no changes in visits with mental health care professionals or measures of self-reported mental health.