In a two-year period more than two million elderly Medicare beneficiaries did not adhere to drug treatment regimens because of cost. This poor adherence tended to be more common among beneficiaries with no or partial medication coverage and was associated with poorer health and higher rates of hospitalization. The risk for cost-related poor adherence was especially pronounced among lower-income beneficiaries with high out-ofpocket drug spending. We argue that this pattern of cost-related poor medication adherence should inform the design of Medicare prescription drug benefit legislation.
R a pi d ly r i s i n g p r e s c r i p t i o n d ru g c o s tshave raised concerns over access for older Americans. Although Medicare covers most of the elderly, the traditional Medicare benefit package does not cover outpatient prescription drugs. Many beneficiaries have some drug coverage through employersponsored supplemental insurance, individually purchased Medigap plans, Medicare health maintenance organization (HMO) plans, or public programs such as Medicaid and state pharmacy programs. However, there has been a recent decline in these sources of supplemental coverage.1 Approximately ten million Medicare beneficiaries have no prescription drug coverage.2 Without governmental intervention, this number is likely to grow.Despite widespread public attention to this issue, little information exists concerning the use of medications and health outcomes among Medicare beneficiaries without prescription drug coverage. Some evidence suggests that elderly beneficiaries who do not have drug coverage use fewer prescription drugs, forgo filling their prescriptions, skip doses, or use lower doses than prescribed, because of the cost of medications.3 However, the association of cost-related poor medication adherence with health outcomes in representative samples of Medicare beneficiaries remains unexamined. In this paper we use data from a recent national household survey of older Americans to examine the association of prescription drug coverage with adherence to medications prescribed for various chronic conditions and the association of cost-related poor adherence with health outcomes. We also examine the prevalence of poor adherence among beneficiaries at different income levels and with different levels of out-of-pocket spending.
Study Methodsn Sample. The data come from the most recent wave (2000) of the Health and Retirement Study (HRS), an ongoing longitudinal survey of community-dwelling older Americans. 4 The HRS sampled household residents in the forty-eight contiguous states using a multistage area probability sampling design. The first wave of the HRS included participants born in 1931-1941. It was conducted in 1992 and has been repeated every two years since. In addition to the original cohort, new cohorts are added to provide coverage of the whole age range of U.S. elderly. 5 The response rate at the 1992 interview was 78 percent. Nonrespondent households were more likely to be white, married, in good physical...