Transformations in Health Policy: An Analysis of Alzheimer's Disease Testing, Medicaid Enrollment, and Insurance Market Concentration

Abstract

This dissertation consists of three quantitative papers addressing contemporary issues in health policy. The first paper draws on a survey of 2,678 adults from the United States and four European countries to assess demand for a hypothetical early medical test for Alzheimer's disease (AD). Overall, 67% of respondents reported that they would be "very" or "somewhat" likely to get the test if it were available. Through logistic regression analysis, we find that interest was higher among those worried about developing AD, with an immediate blood relative with AD, and who have provided care for AD patients. Knowing that AD is fatal did not influence demand, except among those with an affected blood relative. We expect that a test becoming available could precipitate the creation of a large constituency of asymptomatic, diagnosed adults, affecting a range of health policy decisions. The second paper utilizes Current Population Survey data to explore state-level Medicaid enrollment rates among eligible parents between 2003 and 2010, focusing on the interaction of race and ethnicity and political ideology. Using logistic regression analysis, we find that average take-up for Hispanics in conservative states was 23%, whereas take-up was 38% for both whites and blacks in those states, adjusting for state and individual demographics. These differences abated in liberal and moderate states. Among eligible Hispanics, enrollment rates were less than half as high in conservative states than in liberal states (23% versus 61%). Adjusting for differences in state Medicaid policies narrowed these disparities significantly, highlighting the importance of new provisions aimed at streamlining enrollment procedures across all states. The last paper draws on public and private data from 2007 to 2010 to analyze how administrative spending by health insurers and providers varied across states with different levels of insurance and hospital market concentration. Using regression analysis, we find that in provider offices, high levels of insurance concentration were associated with lower administrative costs. If all states were as concentrated as the most concentrated state in our sample, we would expect nationwide savings of $3.6 billion in administrative expenses. However, market concentration did not reduce administrative spending by insurers or hospitals.