Federal and state governments combined are the largest healthcare purchasers in the United States, with nearly 50 percent of national expenditures paid by taxpayers. Medicaid alone covers 44 million people and accounts for 21.5 percent of state spending, leading to intense pressure to control escalating costs. Medicaid costs are driven by the sickest of the sick; 60 percent of patients suffer from a chronic or disabling condition, with average monthly expenditures of 556 dollars compared to 36 dollars for others. Disease management (DM) services aim to improve patients' self management of chronic conditions and, for many states, DM programs are considered the long awaited panacea expected to control costs. However, the growth in Medicaid agencies contracting with DM vendors has been based on the promise of lower costs and higher quality - despite a lack of research evidence of program effectiveness. This dissertation addresses that gap. It examines the causal relationship between DM interventions and outcomes, introduces a new measurement methodology to quantify the scope and intensity of interventions, and then applies this in an evaluation of DM in a Medicaid population with persistent asthma. It answers the question, how do the scope and intensity of DM program interventions impact the frequency of emergency room visits for Medicaid patients with asthma?

A study group was selected and matched to an age/gender equivalent control group, each comprised of 3,922 Medicaid patients with asthma. The study showed a small but statistically significant reduction in adverse events for the study group compared with the control group, despite minimal intervention. There were positive correlations between a reduction in adverse events and increasing total contacts, the number of telephone calls, tailored interventions or the number of physician contacts. The use of a matched control group in a DM evaluation is new, and shows the real impact of regression to the mean; on average those with adverse events in the six months before the program had fewer in the six months following, even without interventions. The study highlights the lack of DM program oversight by state Medicaid agencies and discusses implications for public policy and management.