A Dual-Event Analysis of Rural and Nonrural Differences in Nursing Home Disaster Resilience.

OBJECTIVES: We evaluated whether rural nursing home residence was associated with adverse postdisaster outcomes, overall and among evacuated residents. DESIGN, SETTING, AND PARTICIPANTS: We conducted a dual-cohort, retrospective study of 72,992 Medicare beneficiaries aged ≥65 years residing in nursing homes exposed to Hurricane Michael (2018) or Hurricane Ian (2022). METHODS: We evaluated the association between rural nursing home residence and 30-day mortality and hospitalization, adjusting for resident-, facility-, and area-level characteristics. We further evaluated whether rurality modified the effect of evacuation. RESULTS: The full, dual-cohort sample comprised 72,992 residents across 1176 nursing homes, with an average age of 81.2-years. Rural residence was not independently associated with postdisaster mortality (hazard ratio [HR], 0.91; 95% CI, 0.79-1.05) or hospitalization (HR, 0.97; 95% CI, 0.87-1.08). However, rurality modified the association between evacuation and mortality. If evacuated, rural residents incurred a greater than 2-fold increase in mortality risk relative to nonrural residents (HR, 2.62; 95% CI, 1.66-4.13). CONCLUSIONS AND IMPLICATIONS: Rural nursing homes exhibited disaster resilience under baseline disaster conditions but not during evacuation. It is possible that evacuation exposes rural nursing home residents to structural and operational constraints that may differentially compromise postdisaster survival. Future federal and state guidance may benefit from aligning preparedness standards with the distinct constraints of rural nursing homes.