Comprehensive Medication Reviews in Medicare Were Not Associated With Reduced Central Nervous System-Active Polypharmacy in 2021.

BACKGROUND: Central nervous system (CNS)-active polypharmacy is associated with increased risks such as impaired cognition and falls. In 2021, CNS-active polypharmacy was added as a Medicare Part D display measure to monitor for this risk. Enrollees in the Medicare Part D Medication Therapy Management program are at increased risk of CNS-active polypharmacy and are offered comprehensive medication reviews (CMRs) to optimize their medication management and reduce medication-related safety risks. OBJECTIVE: Evaluate the association of CMRs with CNS-active medication discontinuation among Medication Therapy Management enrollees in 2021. METHODS: Observational study applying inverse probability of treatment weights to compare the time until discontinuation of at least one medication contributing to CNS-active polypharmacy in CMR recipients versus non-recipients in 2021 using 5% Medicare fee-for-service claims and enrollment data. RESULTS: Of 2702 community-dwelling, Medication Therapy Management program enrollees ≥ 66 years of age with CNS-active polypharmacy, 969 (35.9%) were CMR recipients. Both CMR recipients and non-recipients were taking a median of four CNS-active medications. As compared to non-recipients pre-weighting, CMR recipients were more likely to use certain CNS-active medications, such as antidepressants, antiseizure medications, benzodiazepines, and nonbenzodiazepine sedative hypnotics and opioids. Compared to non-recipients pre-weighting, CMR recipients were also more likely to have more prescribers contributing to the CNS-active polypharmacy and to have a mix of prescriber types involved. Comparable numbers of CMR and non-CMR patients discontinued at least one CNS-active medication within 1 year (11.5% vs. 13.2%). In the weighted analyses, there was no difference in likelihood of discontinuation of at least one CNS-active medication between CMR recipients and non-recipients (hazard ratio = 1.03, 95% confidence interval = 0.94-1.12). CONCLUSIONS: CMRs were not associated with reduced CNS-active polypharmacy in older adults in the first year that it served as a Part D Display measure. Future research is needed to better understand why and whether this continues.

Duke Scholars

Author Anna Hung Population Health Sciences

Author Juliessa Pavon Medicine, Geriatrics and Palliative Care

Author Susan Nicole Hastings Medicine, General Internal Medicine

Author Valerie A. Smith Population Health Sciences

Author Matthew Leonard Maciejewski Population Health Sciences