Importance: Extreme heat poses a serious health risk to older adults because they are generally more susceptible to heat-related mortality. Many nursing homes in the US and elsewhere lack air conditioning (AC). Ecological studies have evaluated the protective role of AC against mortality on extreme heat days in prisons and large urban settings; however, the nursing home setting remains understudied.
Objective: To assess mortality rates during extreme heat days in nursing homes with AC compared to those without AC.
Design, setting, and participants: This was a case-crossover study conducted in Ontario, Canada, to determine the odds of mortality associated with extreme days separated for nursing homes with and without AC using conditional logistic regression. Nursing home residents who died during the warm months (June to September) from 2010 to 2023 were included. Data were analyzed from June 2024 to April 2025.
Exposures: An extreme heat day was defined as any day in the ≥90th percentile of the heat index (ambient temperature and relative humidity) for any given nursing home location during the study period.
Main outcomes and measures: All-cause mortality during extreme heat days, by AC status of nursing homes.
Results: Of the 73 578 deaths of nursing home residents from 2010 to 2023, 40 255 residents (mean [SD] age, 86.8 [8.8] years; 65.6% women) died at 276 homes with AC and 33 323 residents (mean [SD] age, 87.2 [8.7] years; 64.8% women) at 339 homes without AC. Before the AC mandate was announced in July 2020, nursing homes without AC (55.1%) were predominantly investor owned (ie, for profit; standardized mean difference [SMD], 0.47), were built to older design standards (SMD, 0.57), and had more residents per room (SMD, 0.58) compared with nursing homes with AC. Overall, 4889 deaths (13.8%) in nursing homes without AC occurred on extreme heat days compared with 4611 deaths (12.1%) in those with AC. Extreme heat was associated with significantly increased odds of mortality in nursing homes without AC (odds ratio [OR], 1.11; 95% CI, 1.06-1.16) but not in those with AC (OR, 1.03; 95% CI, 0.98-1.07). Compared to nursing homes with AC, those without AC were associated with significantly higher relative odds of mortality on extreme heat days (relative OR, 1.08; 95% CI, 1.01-1.15). Lagged analyses suggest that the associated effects of extreme heat persisted for 3 days beyond the initial exposure.
Conclusion and relevance: In this case-crossover study, mortality was lower during extreme heat days in nursing homes with AC compared to those without AC. These findings suggest that AC provision in nursing homes and other congregate care settings may be important for preventing mortality among older adults during extreme heat days.