OBJECTIVES: To assess the impact of Medicare Part D in the nursing home (NH) setting.

DESIGN: A population-based study using 2005/06 prescription dispensing records, Poisson regressions with generalized estimating equations, and interrupted times series estimation with segmented regression methods.

SETTING: Nursing Homes.

PARTICIPANTS: A nationwide sample of long-stay Medicare enrollees in NHs (N=861,082).

MEASUREMENTS: Probability of Part D enrollment, changes in source of drug payments, changes in average number of monthly prescriptions dispensed per resident.

RESULTS: In 2006, 81.0% of NH residents were enrolled in Part D, 16.1% had other drug coverage, and 3.0% (n=11,000) remained without drug coverage, which was the same rate of no drug coverage as in 2005. NH residents who did not enroll in Part D were the oldest (relative risk (RR)=0.82, P<.001), had no drug coverage in 2005 (RR=0.84, P<.001), and had high comorbidity burden (RR=0.94, P<.001). The proportion of prescription drugs paid out of pocket decreased from 11.0% in 2005 to 8.1% in 2006 (P<.001). Average monthly prescription use per resident in 2006 decreased by half a prescription from 2005 levels (9.6 vs 10.1, P=.003).

CONCLUSION: Part D decreased some out-of-pocket drug costs but did not expand drug coverage in the NH population or reach some vulnerable segments. Part D was also associated with some disruption in NH drug use, especially right after implementation.