Do end‐of‐life outcomes differ by assisted living memory‐care designation?

• Published in: Journal of the American Geriatrics Society (JAGS) Vol. 72; no. 8; pp. 2491 - 2499
• Main Authors: Wang, Xiao(Joyce), Cornell, Portia Y., Belanger, Emmanuelle, Thomas, Kali S.
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01-08-2024; Wiley Subscription Services, Inc
• Subjects: Alzheimer's disease; Dementia disorders; end‐of‐life; Hospice care; Memory; memory care; Mortality; Neurodegenerative diseases; Older people; related dementia; residential care/assisted living; memory care; end‐of‐life; Alzheimer's disease; related dementia; residential care/assisted living
• ISSN: 0002-8614; 1532-5415; 1532-5415
• DOI: 10.1111/jgs.18899

Background
Residential care/assisted living (RC/AL) is an increasingly common place of end‐of‐life care for persons with Alzheimer's disease and related dementia (ADRD), who have unique care needs as their health declines. Approximately 22% of RC/ALs provide specialized memory care (memory‐care RC/AL). Understanding how end‐of‐life outcomes differ by memory care among residents with ADRD could facilitate aging/dying in place for this population. The objective of this paper is to examine if end‐of‐life outcomes (i.e., mortality, hospice use, and number of days receiving hospice in the last month of life) differ between residents with ADRD who moved to memory‐care RC/AL, compared with residents with ADRD who moved to RC/AL without memory care (general RC/AL).
Methods
Prospective cohort of 15,152 fee‐for‐service Medicare beneficiaries with ADRD who moved to large RC/AL (> = 25 beds) between 2016 and 2018. We used inverse probability treatment weighting to account for observable differences between memory‐care and general RC/AL residents. Two‐part models estimated the difference by memory care in the number of days receiving hospice care in the last months of life among RC/AL decedents.
Results
The unadjusted mortality rates were 13.4% in general RC/AL and 15.8% in memory‐care RC/AL with an adjusted difference of 1.3 percentage points higher mortality among memory‐care RC/AL residents (p = 0.04). Hospice use was 8% and 10.6% among general and memory‐care RC/AL residents, respectively, with an adjusted difference of 1.4 percentage points (p = 0.01) higher in memory care. Two‐part models showed that decedents in memory‐care RC/AL spent about 1.4 more days receiving hospice care in the last month of life (p = 0.02).
Conclusion
We find a higher mortality rate and higher rate of hospice use among memory‐care RC/AL residents. These findings suggest that memory care may attract residents closer to the end of life and/or promote hospice use at the end of life.