Assessment of the effect of cold and hot temperatures on mortality in Ontario, Canada: a population-based study

Background — Ambient high temperature is associated with mortality; however, heat-related mortality risk has not been quantified systematically in Ontario, the largest province in Canada. Less is known about cold-related risk in this population. Our objective was to quantify the health impact from cold and hot temperatures in Ontario.

Methods — This study comprised all residents of Ontario who died during 1996-2010. A case crossover analysis was applied to assess the relation between daily temperature fluctuation and deaths from non-accidental and selected causes in cold (December-February) and warm season (June-August), respectively, adjusting for various potential confounders. Risk estimates were obtained for each census division, and then pooled across Ontario. We examined potential effect modification for selected comorbidities and sociodemographic characteristics.

Results — In warm season, each 5°C increase in daily mean temperature was associated with a 2.5% increase in non-accidental deaths (95% confidence interval (CI)=1.3%-3.8%) on the day of exposure (lag 0). In cold season, each 5°C decrease in daily temperature was associated with a 3.0% (95%CI=1.8%-4.2%) increase in non-accidental deaths, which persisted over seven days (lag 06). The cold-related effects (lag 06) were stronger for cardiovascular-related deaths (any cardiovascular: 4.1%; 95%CI=2.3-5.9% and ischemic heart: 5.8%; 95%CI=3.6%-8.1%), especially among individuals aged <65 years (8.0%; 95%CI=3.0%-13.0%). Conversely, heat most strongly increased respiratory-related deaths during hospitalization (26.0%; 95%CI=0-61.4%). Across Ontario, each 5°C change in daily temperature was estimated to induce 7 excess deaths per day in cold season and 4 excess deaths in warm season.

Interpretation — Heat contributed to excess deaths in Ontario, although cold-related impact appeared greater. Further work is required to better define high-risk subgroups, which might include the homeless and individuals with inadequately-heated housing.