The impact of Medicare prescription drug coverage on the use of antidementia drugs.

BACKGROUND: Cholinesterase inhibitors and memantine are prescribed to slow the progression dementia. Although the efficacy of these drugs has been demonstrated, their effectiveness, from the perspective of patients and caregivers, has been questioned. Little is known about whether the demand for cholinesterase inhibitors and memantine are sensitive to out-of-pocket cost. Using the 2006 implementation of Medicare Part D as a natural experiment, this study examines the impact of changes in drug coverage on use of cholinesterase inhibitors and memantine by comparing use before and after Medicare Part D implementation among older adults who did and did not experience a change in coverage.

METHODS: Retrospective analyses of claims data from 35,102 community-dwelling Medicare beneficiaries in Pennsylvania aged 65 or older. Beneficiaries were continuously enrolled in a Medicare Advantage plan from 2004 to 2007. Outcome variables were any use of donepezil (Aricept(®)), galantamine (Razadyne(®)), rivastigmine (Exelon(®)), tacrine (Cognex(®)), or memantine (Namenda(®)) each year and the number of 30-day prescriptions filled for these drugs. Independent variables included type of drug benefit pre-Part D (No coverage, $150 cap, $350 cap, and No cap as the reference group), time period, and their interaction. Sensitivity analyses were conducted to test if there are differences in use by drug class or if beneficiaries with a diagnosis of dementia pre-Part D experienced an increase in use post-Part D.

RESULTS: The No coverage group had a 38% increase in the odds ratio of any use of antidementia medications (P = 0.0008) post-Part D relative to the No cap group. All four coverage groups had significant increases in number of 30-day prescriptions (P < 0.001) over the study period. In adjusted models that included the sub-sample with any use pre-Part D, the No coverage group had a 36% increase in prescriptions (P = 0.002) and the $350 cap group had a 15% increase (P = 0.003) after adjusting for trends in the No cap group. Results from the sensitivity analysis for the sub-sample with a diagnosis of dementia pre-Part D show that each group had significant increases in 30-day prescriptions compared to the No cap control group (P < 0.05).

CONCLUSIONS: Use of cholinesterase inhibitors and memantine in our sample increased and a greater increase in use was observed among Medicare beneficiaries who experienced improvements in drug coverage under Medicare Part D.