Cost-Related Nonadherence in the Medicare Program

The Impact of Part D

Bibliographic Data

ID9099752
AuthorsJames J Kennedy (Washington State University Spokane, corresponding author), Matthew L Maciejewski (0000-0003-1765-5938, Durham VA Medical Center), Matthew Maciejewski (0000-0002-6336-2135), Darren Liu (0000-0003-0301-4080, University of Central Florida), Elizabeth Blodgett, Elizabeth G Blodgett (University of North Carolina at Chapel Hill)
Year2011
Volume49
Issue5
Pages522-526
Publication date2011-05-01
Peer ReviewedYes
Open AccessNo
TypeARTICLE
VenueMedical Care (JOURNAL)
Journal identifiersISSN: 0025-7079 • E-ISSN: 1537-1948
PublisherOvid Technologies (Wolters Kluwer Health) (PUBLISHER)
DOI10.1097/mlr.0b013e318210443d
PMID21430574
OpenAlexW2085315717
LanguageEN
Citations received8

BACKGROUND: Cost-related nonadherence (CRN) among Medicare beneficiaries declined after the implementation of the Part D program, but it is unknown whether CRN changes varied on the basis of beneficiaries' change in drug coverage. OBJECTIVE: To determine how CRN changed with the introduction of Part D, and whether CRN changes from 2005 to 2006 varied between newly insured beneficiaries, continuously insured beneficiaries, and continuously uninsured beneficiaries. METHODS: CRN, drug coverage, and beneficiary characteristics in 2005 and 2006 were constructed from merged Medicare Current Beneficiary Survey Access to Care files (sample, n=8935). Change in CRN was modeled using multinomial logistic regression to identify predictors of resolved CRN (reported in 2005 but not in 2006), unresolved CRN (reported in 2005 and 2006), and new CRN (reported in 2006 but not in 2005), relative to no CRN (not reported in 2005 or 2006). RESULTS: Rates of CRN declined from 2005 to 2006 for all beneficiaries, with the greatest reductions (from 22.1% in 2005 to 14.3% in 2006) for newly insured beneficiaries who gained drug coverage through Part D. In adjusted analyses, newly insured beneficiaries were more likely to have resolved CRN (adjusted odds ratio [AOR] =1.7; 95% confidence interval, 1.3-2.2). Younger beneficiaries (under the age of 65 years) and beneficiaries with multiple chronic conditions, poor health, and depression were significantly more likely to report CRN. CONCLUSIONS: Part D coverage reduced but did not eliminate CRN for newly insured beneficiaries. Unresolved CRN persisted for newly insured and continuously uninsured beneficiaries, particularly among disabled beneficiaries

Beneficiary · Business · Confidence interval · Logistic regression · Odds · Odds ratio · Demography · Finance · Healthcare Policy and Management · Internal Medicine · Medication Adherence and Compliance · Medicine · Primary Care and Health Outcomes