(1346) Determinants of Medication Adherence Problems Among Rural, Older Adult Participants of an Area Agency on Aging Community Care Transition Initiative

Objectives: Our objective was to examine the relationship between participant demographics and barriers to care and self-management needs with medication adherence problems among participants in the Michigan Region VII Area Agency on Aging (AAA) pharmacist-led community care transition initiative (CCTI). Between 30 to 50% of older adults have two or more comorbidities and are susceptible to fragmented healthcare delivery after hospital discharge. Fragmented care transitions after hospitalizations are associated with preventable hospital readmissions and mortality, costing Medicare approximately $12 billion annually. The Michigan Region VII AAA implemented a pharmacist-led CCTI (AAA CCTI) program for rural, low-income older adults to reduce hospital readmission rates and enhance older adults’ quality of life. The AAA CCTI included a home visit by a community health worker and a telehealth comprehensive medication review conducted by the pharmacist, with follow-up for up to one year. Medication-related problems were commonly identified during the pharmacist medication review with medication adherence problems being highly prevalent. Understanding the determinants of adherence problems may enable the modification of the pharmacist-led intervention and improve medication use by AAA CCTI participants.

Methods: A retrospective, cross-sectional, secondary analysis of information previously collected from AAA CCTI chart reviews was conducted. Subjects were participants of the AAA CCTI discharged between January 2018–December 2019. Data were abstracted from the AAA CCTI pharmacist visit note and included demographics, assessment of barriers to care and self-management needs, and medication-related problems (adherence, indication, effectiveness, and safety). Medication-related problems were categorized by the Pharmacy Quality Alliance Medication Therapy Problem Categories Framework. The Behavioral Model for Vulnerable Populations informed the selection of predictor variables including predisposing characteristics (e.g., sex, age, housing barriers, and legal or criminal barriers), enabling resources (e.g., health skills; financial or insurance benefits barriers; consumer engagement, awareness, and skills barriers; services and supports barriers; Medicaid waiver programs barriers; facility-related barriers; and social support barriers), and need characteristics (e.g., LACE score (i.e., Length of hospital stay, Acuity of admission, Comorbidity, Emergency department use), physical health barriers, and cognitive status barriers). The primary outcome was having at least one medication adherence problem. A t-test or chi-square test was used to examine the differences in means or proportions between participants with and without medication adherence problems. Logistic regression was used to test the unadjusted and adjusted association between demographics and barriers to care and self-management needs with a medication adherence problem.

Results: A total of 477 AAA CCTI participants were included. There were approximately equal numbers of males and females. The mean age of AAA CCTI participants was 73.72 years (SD = 9.69). Medication adherence problems were identified in 2 out of 5 participants (n = 193, 40.5%). Enabling resources of consumer engagement, awareness, and skills barriers (7.0% vs. 14.2% p = 0.03) and Medicaid waiver program barriers (26.3% vs. 36.3%, p = 0.04) were different between the participants with and without medication adherence problems. Participants with Medicaid waiver program barriers had lower odds of medication adherence problems than those without the barriers (adjusted odds ratio: 0.42, 95% confidence interval 0.19-0.90). No predisposing or need characteristics were significantly associated with having a medication adherence problem in our multivariable logistic regression model.

Conclusions: In our study, medication adherence problems after care transitions were common. Potential AAA CCTI program revisions could include interventions to address medication adherence. We found that AAA CCTI participants with Medicaid waiver program barriers were less likely to have medication adherence problems, which was unexpected. Future research to understand how the benefits of the Medicaid waiver program influence medication adherence in low-income, rural older adults after discharge is warranted.