The Relationship Between the Low-Income Subsidy and Cost-Related Nonadherence to Drug Therapies in Medicare Part D

Authors

  • Iris I. Wei DrPH,

    Corresponding author
    • Research and Rapid-Cycle Evaluation Group, Centers for Medicare and Medicaid Services, Center for Medicare and Medicaid Innovation, Baltimore, Maryland
    Search for more papers by this author
  • Jennifer T. Lloyd PhD,

    1. Research and Rapid-Cycle Evaluation Group, Centers for Medicare and Medicaid Services, Center for Medicare and Medicaid Innovation, Baltimore, Maryland
    Search for more papers by this author
  • William H. Shrank MD, MSHS

    1. Research and Rapid-Cycle Evaluation Group, Centers for Medicare and Medicaid Services, Center for Medicare and Medicaid Innovation, Baltimore, Maryland
    Search for more papers by this author

Address correspondence to Iris I. Wei, Centers for Medicare and Medicaid Services, Center for Medicare and Medicaid Innovation, 7500 Security Boulevard, Mailstop WB-06–05, Baltimore, MD 21244. E-mail: wei.iris.999@gmail.com

Abstract

Objectives

To examine the relationship between receiving the Medicare Part D low-income subsidy (LIS) and cost-related medication nonadherence (CRN).

Design

Cross-sectional.

Setting

Medicare Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey collected in spring 2007.

Participants

Part D–enrolled Medicare beneficiaries who responded to the CAHPS survey.

Measurements

Respondents were categorized into three LIS groups: deemed LIS (Medicare and Medicaid dual-eligible and individuals receiving Supplemental Security Income), LIS applicants (other low-income individuals who applied for and received LIS), and non-LIS. Adjusted logistic models were used to assess the likelihood of CRN according to LIS status. Sample weights were applied in all analyses to account for complex sampling design.

Results

Of 171,573 Part D-enrolled respondents (weighted N = 14,572,827; response rate 48%), 17.2% reported CRN. Specifically, 14.7% of non-LIS respondents, 22.2% of deemed-LIS respondents, and 24.0% of LIS applicants reported CRN. LIS groups had higher unadjusted odds of CRN than the non-LIS respondents, but fully adjusted odds of CRN were lower in the deemed-LIS (adjusted odds ratio = 0.66, 95% confidence interval = 0.59, 0.74) than the LIS applicants or the non-LIS respondents. Subgroup analyses revealed that sociodemographic and health-related characteristics were associated with higher CRN in all three groups.

Conclusion

The lower adjusted odds of CRN in deemed-LIS is reassuring, suggesting that autoenrollment provides meaningful assistance in removing cost-related barriers to medication use, but certain sociodemographic characteristics were associated with higher odds of CRN. Efforts to improving outreach to these subgroups and tracking of CRN are warranted.

Ancillary