It's Just Hot: What Is the Impact of Heat-Waves in Florida?

BACKGROUND: Florida has a humid subtropical/tropical environment. Although the population is acclimated, heat-related adverse health outcomes (e.g. heat-related illness [HRI]) still occur. This analysis examines the burden of work-related (WR) and non-work-related (NWR) HRI during heat-waves. We attempt to develop a specific heat-wave definition for use in Florida and assess the added impact of heat-waves above the general effect of daily temperature/humidity.

METHODS: An analysis was conducted of emergency department visits and hospitalizations among Florida residents for the period May-October (2005-2012). HRI cases were identified via ICD-9-CM codes. Heat-waves were defined as ≥2 days, or ≥4 days, with a sustained heat index ≥108°F or above the 95^th or 99^thpercentile of the daily summer heat index. Using these six heat-wave definitions, we compared rates of HRI on heat-wave days and non-heat-wave days for each of the seven Florida regions. A time-series analysis was also conducted with exposure defined as daily heat index and a heat-wave indicator term included to represent the heat-wave effect. Statewide rate ratios (RR) were estimated using meta-regression techniques to summarize region-specific estimates.

RESULTS: A small proportion of HRI cases occurred on heat-wave days with the broadest definition (≥2 days above 95^th percentile) having the largest proportion of cases (WR=6.7%; NWR=5.8%). The largest effects were observed when a heatwave was defined as ≥2 days at or above a threshold of 108°F (WR RR=2.24; NWR RR=1.93) or above the 95^th percentile (WR RR=2.06; NWR RR=1.81) or 99^thpercentile (WR RR=2.41; NWR RR=1.66) of daily summer heat index.  However, there was little contribution of an indicator for heatwaves to model fit above the general effect of daily temperature/humidity. A potential reduction in HRI rates was observed on heat-wave days with intensity ≥108°F compared to non-heat-wave days (WR: RR=0.77 [95%CI=0.49, 1.24]; NWR: RR=0.90 [95%CI=0.76, 1.06]).

CONCLUSIONS: A heat-wave in Florida should be defined as ≥2 days while the threshold value may vary dependent on the public health application. However, given that a heat-wave effect was not observed and that the heat-wave HRI burden represents only a small fraction of the HRI cases, heat-related prevention activities and surveillance in Florida (and potentially the Southeastern U.S.) should occur throughout the warm season. Based on these findings, a key message should be that heat-related outcomes do occur on a typical summer day. Nationally, there needs to be recognition that a focus on heat-waves may not be appropriate for all communities.