Health insurance design has distinct and well-documented effects on choice of insurance plan and health care utilization and expenditures. The question of how to design benefits for newer technologies such as specialty drugs presents new challenges in balancing adverse selection and moral hazard concerns with beneficiary coverage. The Medicare Prescription Drug Benefit Program adds complexity via additional governmental regulation of the market and coverage design for specialty drugs. This dissertation presents a conceptual model of the supply-side tier placement decision on the part of Part D plans, and uses Part D formulary and claims data to estimate the demand-side response by Medicare beneficiaries. An instrumental variables approach is employed to address the endogeneity present in the specialty tier placement (supply-side) and enrollment, utilization, and spending (demand-side) decisions. Results are mixed when applying the instrumental variable, showing that a $100 increase in the out-of-pocket costs imposed on beneficiaries results in a significant increase in the likelihood of utilization for those beneficiaries with rheumatoid arthritis or multiple sclerosis, but a significant reduction of 16.2% (2007) and 35.1% (2008) for beneficiaries diagnosed with one of four types of cancer. Results suggest that beneficiaries do respond to higher cost sharing, raising questions about the imposition of higher cost sharing amounts for such expensive drugs.