Importance: Studies of nursing home (NH) residents show lower than expected antiretroviral therapy (ART) use, but it is unclear whether ART use changes across the transition from the community to long-term NH stay.
Objective: To examine changes in ART use across the transition from the community to long-term NH stay.
Design, setting, and participants: This retrospective cohort study examined long stays in US NHs for people with HIV in a sample of 5% of Medicare claims from 2014 to 2019. Stays were at least 30 days long, had at least 3 months between multiple stays, and were for those continuously enrolled in Medicare for the stay and 6 months before. Analysis was completed in May 2025.
Exposures: Admission year demographics included age, race and ethnicity (non-Hispanic Black, non-Hispanic White, and other [American Indian or Alaska Native, Asian or Pacific Islander, other, and unknown]), binary sex, Medicaid eligibility, whether the stay was preceded by a skilled nursing stay, and whether disability was Medicare original eligibility. NH characteristics included for-profit status, census region, and facility quality rating.
Main outcomes and measures: Linear regression estimated changes in the proportion of days covered by 3-drug ART, and hierarchical multinomial logistic regression estimated the risk of never having, losing, or gaining ART vs always having it, across the transition from the community to long-term NH stay.
Results: There were 713 long NH stays for 657 people with HIV (mean [SD] age, 61.0 [11.4] years) across 598 facilities; 271 stays (38%) were for people aged 65 years and older. Only 23 individuals lost ART (3%), 97 individuals (14%) gained ART, 185 individuals (26%) never had ART, and 408 individuals (57%) always had ART across the transition. Excluding those who lost ART, all other groups were mostly men (never, 132 men [71%]; always, 289 men [71%]; gained, 72 men [74%]) and Black (never, 85 individuals [46%]; always, 237 individuals [58%]; gained, 58 individuals [60%]). There was an increase in the proportion of days covered (mean intercept α = 13.92; 95% CI, 9.57-18.29). Compared with always having ART, Black race (relative risk [RR], 0.52; 95% CI, 0.35-0.77), polypharmacy (RR, 0.41; 95% CI, 0.23-0.74), and disability as original Medicare eligibility (RR, 0.47; 95% CI, 0.29-0.77) were associated with lower risk of never having ART. For-profit facilities were associated with higher risk (RR, 1.63; 95% CI, 1.03-2.59) of never having ART. Polypharmacy was associated with lower risk of gaining ART (RR, 0.15; 95% CI, 0.05-0.49).
Conclusions and relevance: These findings suggest that long-term NH stays may be associated with improved ART use among people with HIV because most stays without ART never had ART before admission.