The Heat Is On: Decision-Maker Perspectives on When and How to Issue a Heat Warning

To better understand how to prevent illness and deaths during hot weather, particularly among at-risk populations, we conducted a study in Detroit, Michigan; Phoenix, Arizona; New York, New York, and Philadelphia, Pennsylvania. Our aims were to characterize and better understand how heatwave and health early warning systems (HHWS) and related prevention and sustainability programs can be more widely and effectively implemented. Specifically, we here report on the scientific evidence, expert judgments and the process used in deciding to trigger a HHWS and activate public health and social services interventions. We conducted interviews with public officials who decide if and when heat advisories/warnings are issued. After transcribing the interviews, we used a qualitative analysis software, QSR NVivo 9.0, to assign codes to portions of text from each transcript and allow analysis of information with common themes across the data. For example, several sentences in a transcript discussing a heat index might be coded as 'definition of heat wave'. A common theme across cities was that deciding what type of weather is dangerous to health is not straightforward. The time in season that heat occurs; the duration of the heat; the level of humidity and other meteorological factors; the extent to which temperatures drop at night, allowing people to cool off; and prevailing weather conditions all play a role. A single 'safe' threshold is unrealistic because people's individual sensitivity, housing, surrounding environments, behaviors, and access to air conditioning can differ greatly. However, choices must be made as to the trigger for the HHWS. Although quantitative analysis with health data (mortality, hospital admissions) can inform the design of the triggers, historical analysis has limitations, and decisions to issue heat warnings are sometimes related to planned activities, such as parades or fairs, that may expose large numbers of people to heat. The HHWS approach designed by Lawrence Kalkstein and colleagues using synoptic air mass forecasts and mortality data has been used by some cities. Other cities use National Weather Service products that are built on a variety of data inputs and approaches, including calculation of season-specific thresholds. More than one respondent mentioned distaste for 'black box' approaches that were not easily communicated to end-users. The decision to issue a heat warning can save lives, through such activities as opening cooling centers, distributing water to the homeless, and assisting elderly residents. A relatively simple triggering system that is easily understood by the media and public may facilitate more widespread adoption of HHWS. Funding: U.S. Centers for Disease Control and Prevention Grant R18-EH000348