Insights on the Dynamic Effects of Heat Alerts on Heat-Related Morbidity in the US Southeast

• In Event: Resilience in Action: Insights into Private and Public Sector Responses to Climate Risks

Abstract

Climate change is leading to warmer weather overall and increasing the frequency and intensity of heat waves in the US. Exposure to high ambient temperatures is associated with a wide range of adverse health outcomes in humans. In addition to acute heat-related illnesses like heat stroke, heat exposure can exacerbate existing health issues, such as cardiovascular and respiratory disease, and sometimes result in premature mortality. As heat-health risks continue to increase, it is more important than ever to improve local heat governance systems in the US. Heat warning systems are a key policy tool during extreme heat events, encouraging people to engage in behavioral changes to avoid heat exposure, although there is mixed evidence on the effectiveness of these systems in reducing heat-related health outcomes. Drawing on the results of previous studies, I hypothesize that heat alerts do encourage protective behavioral change, although this effect may be stronger in higher-risk settings and among certain heat-vulnerable populations. The availability of adaptive resources during events and alert fatigue may play also a role in the heterogeneity of findings. In this study, I investigate the various mechanisms and moderators of heat alert effectiveness using data on EMS encounters in Georgia and North Carolina from 2018-2023.

The analysis relies on a matching approach with controls to estimate the impact of heat-alerts on health outcomes. Because heat alerts are issued based on forecasted heat index, there exists a set of days with and without heat alerts where similar actual weather conditions were experienced. For each heat alert day, I will identify matched comparison days with similar weather conditions where a heat alert was not issued for the same affected geographic area. In EMS data, location types enable us to assess the tradeoffs between two potentially offsetting mechanisms – increased “awareness” of heat-related illness symptoms and protective “behavioral change”. Additionally, I will evaluate if the results vary across sub-populations by age group, sex, and race, as well as the potentially moderating effect of alert fatigue and other policies that are triggered by alerts. The results of this study can inform the improvement of heat warning systems by illuminating the underlying processes through which heat alerts affect public audiences. Preliminary results indicate that heat alerts are associated with an overall increase in EMS encounters for heat-related illness. I find that the effects of heat alerts may vary temporally, with increased “awareness” of heat-related illness symptoms leading to immediate increases in heat-related EMS encounters and delayed “behavioral change” effects occurring in the days following the alert. The study will contribute to the growing literature on heat governance and help inform how heat management strategies can be improved to reduce adverse heat-health outcomes.

Author

• Katherine Burley Farr, University of North Carolina at Chapel Hill
  Presenting Author