Predictors of Antiamyloid Medication Utilization and Adherence Among People With Alzheimer Disease and Mild Cognitive Impairment

BACKGROUND AND OBJECTIVES: From 2021 to 2024, 3 amyloid monoclonal antibody medications (aducanumab, lecanemab, and donanemab) were approved by the US Food and Drug Administration for treatment of mild cognitive impairment (MCI) and mild dementia due to Alzheimer disease (AD). We determined the demographics and social drivers of health associated with utilization and initial adherence of these medications. METHODS: We used the Vizient database to identify patients with MCI or dementia using an ICD-10 coding definition. Propensity score matching was used to match patients who received an antiamyloid medication to those who did not, based on date of MCI/dementia diagnosis and follow-up. Social drivers of health were assessed using the Vizient Vulnerability Index (VVI), a neighborhood-level measure of socioeconomic vulnerability. Logistic regression determined associations between demographics, Elixhauser comorbidity, and VVI, with utilization of antiamyloid medications. Poisson regression determined associations between demographic, Elixhauser comorbidity, and VVI and initial adherence (measured as days supplied) of antiamyloid medications. RESULTS: We identified 1,198,362 participants with MCI/dementia, of whom 2,427 (0.2%) were prescribed an antiamyloid medication. Of those prescribed an antiamyloid medication, 1,696 (69.9%) were matched to participants with MCI/dementia that were not prescribed antiamyloid medications (mean age 75.5 [SD 9.5], 56.1% female, 80.1% non-Hispanic White). We found that higher age (odds ratio [OR] 0.93, 95% CI 0.92-0.94), higher Elixhauser comorbidity index (OR 0.56, 95% CI 0.52-0.59), being non-Hispanic Black (OR 0.14, 95% CI 0.09-0.23), and being Hispanic (OR 0.21, 95% CI 0.13-0.36) associated with lower odds of receiving an antiamyloid medication. Furthermore, having greater VVI was associated with lower odds of receiving an antiamyloid medication. By contrast, only higher Elixhauser comorbidity score was associated with lower initial medication adherence (rate ratio 0.94, 95% CI 0.93-0.96), but not age, sex, race/ethnicity, and overall vulnerability. DISCUSSION: We found that younger, non-Hispanic White patients, with fewer comorbidities, and lower vulnerability, were more likely to receive antiamyloid medications. By contrast, among patients receiving an antiamyloid medication, only higher comorbidity score associated with lower initial medication adherence. These findings indicate a potential disparity in medication access for US patients with MCI/dementia that could become problematic barriers to care as medications for dementia continue to evolve.