198 - Heat Education and Alert Tools (HEAT): Developing a Prevention-Focused Unhealthy Heat Temperature Threshold for Massachusetts

Publication Number: 2190.198

Darien Mather, Massachusetts Department of Public Health, Boston, MA, United States; Alicia Fraser, Massachusetts Department of Public Health, Boston, MA, United States; Kate Adams, Massachusetts Department of Public Health, Boston, MA, United States; Nalina Narain, MDPH, Boston, Massachusetts, MA, United States; Marissa Hauptman, Boston Children's Hospital, Boston, MA, United States

Background: Climate change has made extreme heat a worsening problem for Massachusetts and across the globe.

Objective: The Massachusetts Department of Public Health sought to define and implement a data-driven health-protective extreme heat temperature threshold that, when triggered, prompts local boards of health, clinicians, and community partners to implement heat mitigation efforts to reduce impacts of extreme heat on health and health systems.

Design/Methods: General linear models of 2017-2019 and 2021 Massachusetts heat stress ED visits from the Massachusetts Center for Health Information and Analysis (ICD10 CM T67, X30, or X32) and various heat wave definitions were compared. Heat wave definitions were based on daily maximum temperature and consecutive days of such heat from Oregon State's PRISM data that were population-weighted to the state level using the 2020 Decennial Census. Binary threshold models were also compared to a multi-tiered risk - none, low, moderate, and high - model. A final temperature-based threshold was compared to a threshold using the National Weather Service and Centers for Disease Control and Prevention HeatRisk tool categories for 2024 heat-related illness data. A sensitivity analysis was run using all-cause ED visits. All model comparisons were made using the Akaike Information Criterion fit statistic.

Results: Three consecutive days of 85°F or higher maximum temperature was the best fit binary threshold of heat stress ED visit data. A tiered risk system - with risk levels more protective of vulnerable groups - improved model fit. A sensitivity analysis showed similar results for all-cause ED visits. This unhealthy heat threshold model showed there were 25.0 additional heat stress and 225.8 additional all-cause ED visits per day during heatwaves (95% CI: 23.3 to 26.7 and 87.9 to 363.7, respectively). In our statewide analysis, both the binary unhealthy heat threshold and four-tiered expanded model were better fits to the data compared to the respective binary and multi-tiered models using HeatRisk.

Conclusion(s): Our state-wide tiered Unhealthy Heat Risk System, with unhealthy heat alert triggered at 85°F for 3 or more days, identified predictably significant increase in emergency room visits for heat-related illnesses and all-causes across the tiered categories and was consistent across five years of heat season data. Leveraging this data-driven approach to trigger unhealthy heat alert notifications for healthcare professionals, local governments and other community and state organizations is critical so they and respective constituents can plan, prepare, and act.

Heat Related Illness by Daily Maximum Temperature
Heat-Related Illness ED Visits by Daily Maximum Temperature as classified by MDPH Unhealthy Heat Categories, May-September 2024

Additional ED Visits per Day by Unhealthy Heat Category
This chart shows additional ED visits for Heat Stress and All-Cause by Unhealthy Heat Category compared to baseline ED visits on days 77°F or less during heat seasons 2017-2019, 2021.