Risk of Hospitalization Among Assisted Living Residents Dually Enrolled in Medicare and Medicaid

Objectives
To examine how risk of hospitalization among assisted living (AL) residents differs by dual enrollment in Medicare and Medicaid and by the percent of dually enrolled individuals in an AL community.
Design
Retrospective cohort study.
Setting and Participants
We used Medicare data from 2008 to 2018 and a national directory of licensed AL communities to identify Medicare beneficiaries with a change in their ZIP+4 code suggesting a new residence in an AL.
Methods
We estimated linear regression models of hospitalization onto interactions of residents’ dual enrollment status and categories of the AL community’s percentage of dually enrolled residents. In the models, we adjusted for person-level clinical and demographic characteristics, year-fixed effects, and fixed effects for the AL residents’ prior ZIP code.
Results
Among 620,542 Medicare beneficiaries who moved to an AL community, the 1-year risk of hospitalization was higher for dually enrolled residents compared with Medicare-only residents. In adjusted models, dually enrolled residents in high-dual AL communities (>50% dually enrolled) had an 7.4% higher risk of hospital admission compared with dually enrolled residents in low-dual AL communities. Medicare-only beneficiaries in high-dual AL communities had a 9.4% higher risk of hospitalization than Medicare-only beneficiaries in low-dual ALs.
Conclusions and Implications
The proportion of residents in an AL community who were dually enrolled was associated with residents’ risk of hospitalization, regardless of their dual enrollment status. Additional research is needed to understand whether differences observed in residents’ risk of hospitalization are due to differences in the types of services provided, unmeasured resident acuity, or the quality of care delivered in these settings.