In Florida, Medicaid expenditures have grown more quickly than Medicaid caseloads. Since the Medicaid program comprises a large share of all state budgets, controlling Medicaid spending has been the first priority for state governments. To manage spending and improve access to healthcare, states have adopted a managed care approach to Medicaid services. When states pursue Medicaid managed care, they believe that managed care can control Medicaid spending and maintain budget predictability through a capitated payment system. Medicaid beneficiaries are also guaranteed access to care through contracts with managed care health plans and/or providers. Several studies that have investigated the impact of Medicaid managed care using nationally representative data have often yielded inconclusive or contradictory results. Despite the growing interest in Medicaid managed care in Florida, only a few studies have examined the impact of managed care programs on preventable hospitalization, length of stay, and cost of inpatient care for Florida's Medicaid enrollees. In addition, only a few studies have examined the impact of Medicaid managed care in the context of different market structures. The purpose of this dissertation is three-fold. First, this study examines the effects of a Medicaid managed care program on preventable hospitalization, length of stay, and cost of inpatient care for Medicaid patients. Second, the impact of market structure on preventable hospitalization, length of stay, and cost of inpatient care is a subject of interest. Finally, this dissertation extends the existing literature methodologically. This study uses multilevel modeling to estimate variables at different levels simultaneously. Agency theory and transaction cost economics provide a useful framework for understanding the effects of different insurance schemes on incentives that affect the behavior of both patients and providers. In addition, the model of market structure provides a theoretical framework for predicting the market behavior of Medicaid managed care organizations under market contexts ranging from monopoly to perfect competition. The study included 1,957,072 Medicaid patients aged 18 to 64 years who were discharged from a short-term general hospital in Florida from January 1, 2006 to December 31, 2012. The researcher used two-level hierarchal models to predict the probability of preventable hospitalization, length of stay, and inpatient cost in Medicaid patients. The results show that Medicaid managed care patients had a higher probability of preventable hospitalization than Medicaid fee-for-service patients. In addition, Medicaid managed care patients were more likely to have shorter lengths of stay and lower inpatient costs compared with fee-for-service Medicaid patients. This study found that market competition moderates the relationship between Medicaid insurance type and dependent variables. The study supports a spillover effect of Medicaid managed care on inpatient costs for Medicaid fee-for- service patients. In conclusion, the dissertation discusses several policy implications.
