Improving attribution of extreme heat deaths through interagency cooperation.

Attributing individual deaths to extreme heat events (EHE) in Canada and elsewhere is important for understanding the risk factors, protective interventions, and burden of mortality associated with climate change. However, there is currently no single mechanism for identifying individual deaths due to EHE and different agencies have taken different approaches, including (1) vital statistics coding based on medical certificates of death, (2) probabilistic methods, and (3) enhanced surveillance. The 2018 EHE in Montréal provides an excellent case study to compare EHE deaths identified by these different approaches. There were 353 deaths recorded in the vital statistics data over an 8-day period, of which 102 were potentially attributed to the EHE by at least one approach and 251 were not attributed by any approach. Only nine of the 102 deaths were attributed to the EHE by all three approaches, 23 were attributed by two approaches, and 70 were attributed by only one approach. Given that there were approximately 50 excess deaths during the EHE, it remains unclear exactly which of the total 353 deaths should be attributed to the extreme temperatures. These results highlight the need for a more systematic and cooperative approach to EHE mortality in Canada, which will continue to increase as the climate changes.