: After a period of relative stability in the mid-1990s, medical costs are again rising at double-digit rates. Pharmaceutical spending is responsible for a large proportion of this growth. Employers and health insurers have pursued multiple strategies to control increasing pharmaceutical costs, including the use of """"""""incentive formularies"""""""" that promote the use of more cost efficient agents by consumers of care through the use of tiered levels of copayments. The rise of incentive formularies is part of a broader trend in health care whose goal is to shift more of the financial burden of medical decision making onto consumers of care. This proposal takes advantage of a comprehensive current data set derived from a unique relationship that we have established with a large national health plan that will give us access to data on 1.25 million people under the age of 65 who were continuously enrolled in the health plan for the two-year period of January 2000-December 2001, when incentive formularies started to become popular. We will use these data to examine the consequences of shifting decisions about pharmaceutical agents and the associated economic burden to beneficiaries. In particular, we will study variation in both pharmaceutical and total medical spending and in quality of care associated with changes in cost sharing for drugs. We propose to use two different and complementary approaches. The first design will be a pre/post intervention with a concurrent control group (differences-in-differences). The second approach will utilize a regression format to estimate the effects on spending and quality of care implied by the pharmacy benefit designs. The regression approach will allow us to estimate effects of a wide variety of benefit designs. The enrollees are clustered in a smaller set of markets where the health plan has a large presence. This clustering will allow us to include substantial numbers of employers (and enrollees) in the same market that have stable benefit structures as well as others that have changed their benefit structure thereby controlling for local market effects. ? ?
