Trends in detection of dementia among beneficiaries in Medicare advantage and traditional Medicare
Year: 2026
Type: article
Study design: Observational Study
Abstract: INTRODUCTION: Timely dementia detection is critical for care planning and supportive services, yet whether claims-based detection has improved over time remains unknown. METHODS: This retrospective cohort study used Health and Retirement Study (HRS) survey data linked to Medicare claims (1998 to 2020). Persons with probable dementia (PWPD) were identified using the Langa-Weir algorithm. Covariate-adjusted logistic regression with predictive margins estimated the proportions of PWPD receiving a claims-based dementia diagnosis, overall and among early-stage PWPD. RESULTS: Among 3144 PWPD from 2004 to 2019, fewer than half of PWPD in Traditional Medicare (TM) had ever been diagnosed, with no significant improvement. Annual detection from 2015 to 2019 was consistently higher in TM than in Medicare Advantage (MA), with adjusted proportions rising from 36.7% versus 32.1% to 48.8% versus 39.2%, but remained low and unchanged among early-stage PWPD. Detection was highest in the Northeast and lowest in the West in both TM (50.0% vs 39.5%) and MA (39.8% vs 25.7%). DISCUSSION: Detection of dementia among PWPD remained incomplete and largely unchanged, particularly at the early stage when diagnosis is most actionable. These persistent gaps underscore the urgent need for targeted strategies to improve dementia identification across plan types and regions.
DOI: 10.1002/dad2.70503
Authors: Yuting Qian, Heather Allore, Xi Chen
Institutions: Yale University, National Bureau of Economic Research
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $2,540
OpenAlex ID: W7223168109
