Interestingly, the research found that most weather-related deaths were not associated with times of temperature extremes — polar vortexes or wilting heat waves — but to moderately cold days.
In fact, extreme temperatures were responsible for less than one per cent of deaths in the countries studied while moderately poor temperature conditions were responsible for about seven per cent of deaths. And most of the deaths attributed to moderately poor temperatures were related to moderately cold conditions.
"What we're basically seeing in this study is that moderate days of cold or heat are producing more deaths than days of extremes," said Eric Lavigne, an adjunct professor at the University of Ottawa and one of the authors of the paper.
"So when we're seeing these very extreme temperatures — let's say in some cities in Canada it could be under -30 C, but the rest of the winter it could be between -10 C and -30 C — well, this range of temperatures ... is creating more mortality cases than the extreme."
The study, led by Antonio Gasparrini of the London School of Hygiene and Tropical Medicine, is published in the British medical journal the Lancet.
An international team of scientists analyzed data on more than 74 million deaths in 384 locations across 13 countries which represent a variety of climates. Countries studied included Australia, Thailand, China, Sweden and Spain. The deaths occurred over more than two decades, from 1985 to 2012.
The authors said periods of high heat often attract attention and concern.
"It's often assumed that extreme weather causes the majority of deaths, with most previous research focusing on the effects of extreme heat waves," Gasparrini said in a statement.
Lavigne, who teaches in the University of Ottawa's Interdisciplinary School of Health Sciences, said he and his co-authors wanted to put that assumption to the test.
"We wanted to have a clearer picture of the relationships between cold and heat on the health impacts," he said.
The research suggests about 7.7 per cent of deaths are caused by "non-optimal" temperatures, though that is an average across the countries studied.
The percentage is higher in some countries — China came in at 11 per cent — than others. Brazil, for instance, came in near the low end of the spectrum, with only 3.5 per cent of deaths attributable to less-than-ideal temperatures.
The study said five per cent of deaths in Canada were attributable to non-optimal temperatures.
"In the Canadian context I would say that cold weather kills nine times as many people as hot weather," Lavigne said.
"For (all) the countries we investigated, we are seeing that cold weather kills 20 times as many people as hot weather."
The authors said public health officials should keep the impact of moderately cold weather in mind when devising policies aimed at lowering the health risks posed by weather.
A commentary accompanying the paper agrees that the findings may point to a need to rethink public health programs aimed at temperature risks.
"A focus on extreme weather (such as heat waves) might ignore the incremental risk of moderately unusual temperatures,'' wrote Keith Dear and Zhan Wang of the Global Health Research Center at Duke Kunshan University in Jiangsu, China.Suggest a correction