Impact of Part D low-income subsidies on medication patterns for Medicare beneficiaries with diabetes.

BACKGROUND: It is not known whether low-income subsidies (LIS) under Medicare Part D help beneficiaries overcome impediments to medication use associated with poor socioeconomic status and high disease burden.

OBJECTIVES: To compare Medicare beneficiaries with LIS and Medicaid (duals), LIS without dual eligibility, and non-LIS recipients on use of medications recommended in diabetes treatment.

RESEARCH DESIGN: Fixed-effect comparisons among beneficiaries in the same Part D plans in 2006-2007.

SUBJECTS: Nationally representative sample of enrollees in Part D prescription drug plans. A total of 109,292 beneficiaries were in 204 prescription drug plans; 47.5% non-LIS, 44.4% duals, and 8.1% nondual LIS recipients.

MEASURES: Medications included antidiabetic agents, renin-angiotensin-aldosterone system inhibitors, and antihyperlipidemics. Drug use was measured by exposure, duration of therapy, and medication possession ratio.

RESULTS: The LIS dual cohort had significantly higher comorbidity compared with non-LIS comparisons, LIS nonduals were significantly more likely to take medications in all 3 drug classes compared with non-LIS recipients, but differences were small (between 2% and 4%; P<0.05). Non-LIS recipients and duals had equivalent exposure to any antidiabetic drug and antihyperlipidemics, but duals were 3% less likely to receive renin-angiotensin-aldosterone system inhibitors compared with non-LIS recipients (P<0.05). Small differences in adjusted values for duration of therapy and medication possession ratio among the 3 cohorts were also observed, none of which were clinically meaningful.

CONCLUSIONS: Similarities in medication utilization among Part D enrollees with and without LIS coverage supports the program objective of providing enhanced access to needed medications for diverse groups of Medicare beneficiaries.