Figure 2. Subgroup Analyses of Heat Waves and Heat-Related Emergency Department Visits Among Dually Eligible Beneficiaries, 2016-2019.

A single day of extreme heat was defined as a day with a maximum temperature of 90 °F (32.2 °C) or higher and in the 97th percentile of daily maximum temperatures for that zip code tabulation area (ZCTA) during the study period. A heat wave was defined as 3 or more consecutive extreme heat days. Incidence rate ratio estimates indicate the ratio of incidence rates for heat-related emergency department visits between heat wave and non–heat wave days for each subgroup. An incidence rate ratio higher than 1 indicates a higher incidence rate during heat wave days compared to non–heat wave days. The Poisson regression models adjusted for binary variables of ZCTA, day of the week, federal holidays, week of the study period, daily ZCTA-level relative humidity, and annual ZCTA-level demographics (proportion of female beneficiaries; age groups of 65-74, 75-84, and ≥85 years). Standard errors were clustered at the ZCTA level and included an offset for annual ZCTA-level population. Alzheimer disease and related dementias (yes/no) analyses excluded nursing home residents due to the high proportion of this population residing in facilities with mandated temperature control. Long-term services and supports (LTSS) analyses excluded 5 states (Arkansas, Idaho, Mississippi, New Hampshire, and Rhode Island) with unreliable long-term care claim volume, as indicated by the DQ Atlas.²⁶ Regions were defined according to the National Centers for Environmental Information classification. HCBS indicates home- and community-based services; MA, Medicare Advantage.