Application of Medicare Part D data in Alzheimer’s disease and related dementias: a scoping review

Abstract

Although Medicare Part D data are extensively used in pharmacoepidemiologic research, no previous synthesis has assessed their role in studies of Alzheimer’s disease and related dementias (ADRD). This scoping review aims to map how Medicare Part D data have been employed to examine medication use, adherence, disparities, and outcomes among individuals with ADRD. We systematically searched PubMed/MEDLINE, CINAHL, Scopus, and Web of Science for peer-reviewed studies published between January 1, 2006, and March 31, 2025, with no age restrictions applied. Eligible studies evaluated medication use (antidementia therapies, antipsychotics, or symptomatic management) in ADRD populations using Medicare Part D data. Fifty-three studies met the inclusion criteria. Most were nationwide in scope (86.8%) and employed cohort design (56.6%). Antipsychotics were the most examined medication class (60.4%), followed by dementia-specific therapies (52.8%) and symptomatic management agents (34.0%). The most frequently assessed outcomes involved patterns of medication use (71.7%), followed by clinical and safety outcomes (28.3%), disparities (20.8%), adherence and persistence (9.4%), and healthcare costs or utilization (5.7%). Just over half of the studies (50.9%) utilized Part D Prescription Drug Event files, whereas a small fraction incorporated formulary data (1.9%) or Medicare Current Beneficiary Survey linkages (9.4%). Regression models were used in 60.4% of studies, while survival analysis (20.8%), causal inference (13.2%), and policy evaluation (11.3%) were applied less frequently. This review lays the groundwork for Medicare Part D research in ADRD, highlighting the need to expand studies on adherence, treatment effectiveness, policy, equity, and advanced analytic approaches.