Quasi-Poisson Distributed Lag Nonlinear Regression Model to Investigate the Association Between Population-Weighted Daily Mean Temperature and Mortality ¡ª Maricopa County, Arizona, 2019¨C2023

What did we do?

• During 2019–2023, the Maricopa County Department of Public Health (MCDPH) reported a yearly average of 386 heat-related deaths from heat surveillance, representing ~25% of all heat-related deaths reported nationwide. Although Maricopa County has a robust heat surveillance system, it might underrepresent total effects of heat on mortality. MCDPH investigated the association between heat exposure and mortality and estimated heat-attributable deaths that occurred during the 2019–2023 heat seasons (April 1–October 31).

What did we find?

• Mortality risk increased with daily mean temperature. At 99°F (95th percentile) and 102°F (99th percentile), the cumulative relative risk (RR) was 1.11 and 1.17, respectively.
• Among 126,465 deaths, an estimated 3,036 deaths were heat-attributable with a yearly average of 607 (95% eCI: 194–977); 57% more deaths than were captured by the heat surveillance system over the 5-year period.

Why does it matter?

• Mortality risk increased with daily mean temperature during the heat season, suggesting effects of heat on mortality are multifaceted and likely undercaptured, especially ones that are indirectly related to heat.
• Statistical heat-attribution methods are one way to estimate mortality effects, complementing robust, countywide surveillance efforts to provide actionable insights for enhancing heat prevention strategies.

***This presentation has updated data that will be shared at the EIS Conference.

Abstract Category: Environmental Health