UCare generously supports MinnPost’s Second Opinion coverage; learn why.

Cold weather linked to 20 times more deaths than hot weather, study finds

Photo by Craig Lassig
The study found that cold temperatures are associated with about 20 times more excess deaths than hot temperatures.

Both cold and hot temperatures can have a negative effect on human health — and cause death. But is one more potentially harmful to our health than the other?

An international team of researchers recently set out to answer that question. They conducted a statistical meta-analysis of more than 74 million deaths that occurred between 1985 and 2012 in 384 locations in 13 countries, including in 135 U.S. cities.

Their findings, published late last week in the journal The Lancet, are somewhat surprising — and not exactly what those of us living on the Northern Plains will want to hear. For the study found that cold temperatures are associated with about 20 times more excess deaths than hot temperatures. 

Specifically, the analysis revealed that about 7.71 percent of the deaths in the study could be attributed directly to weather. But of those excess deaths, most — 7.3 percent — were associated with cold weather. Only 0.4 percent of them were linked to hot weather. 

The analysis also found that extreme weather (days when temperatures fell in either the lowest or the highest 2.5 percentile for each location) was responsible for only 11 percent of the weather-related deaths in the study, or 0.86 percent of all the deaths.  

Most of the deaths attributed to weather occurred on days that were only mildly colder than the average day in each particular location.

Possible explanations

“It is often assumed that extreme weather causes the majority of deaths, with most previous research focusing on the effects of extreme heat waves,” said lead author Antonio Gasparrini, a biostatistician and epidemiologist at the London School of Hygiene & Tropical Medicine, in a released statement. “Our findings, from an analysis of the largest dataset of temperature-related deaths every collected, show that the majority of these deaths actually happen on moderately hot and cold days, with most deaths caused by moderately cold temperatures.”

Gasparrini and his colleagues suggest that the increase in deaths on colder-than-normal days may be due to the extra stress that such temperatures place on the cardiovascular system, raising the risk of heart attack and stroke. Cold temperatures may also weaken the immune system, leaving the body more susceptible to respiratory illnesses.

The results of the study have important implications for public health policy, they add, because the policies of most countries tend to focus primarily on minimizing the health consequences of heat waves.

“Our findings suggest that these measures need to be refocused and extended to take account of a whole range of effects associated with temperature,” said Gasparrini.


This study’s statistical modeling is impressive, and its findings are interesting. But, as the experts who reviewed the study for the “Behind the Headlines” website of Great Britain’s National Health Service (NHS) point out, the findings are somewhat skewed by the fact that extreme temperatures occur on very few days. 

So, “even though the relative risk of death is increased on those days,” the NHS experts explain, “the absolute number of deaths is nowhere near as high as on the majority of days.”

In other words, all of us should still take steps to protect ourselves from the health risks of both heat waves and brutal cold spells

The study had other limitations as well. For example, although the researchers adjusted their findings for some confounding variables (such as air pollution and humidity), they didn’t do it for other potentially important ones, such as whether people had access to shelter and heating.

“Since high or low temperatures affect susceptible groups such as unwell, young, and elderly people the most,” write the authors of a commentary that accompanies the study, “ attempts to mitigate the risk associated with temperature would benefit from in-depth studies of the interaction between attributable mortality and socioeconomic factors, to avoid averse policy outcomes and achieve effective adaptation.”

You can read the study and the commentary in full on The Lancet website.

You can also learn about all our free newsletter options.

Comments (6)

  1. Submitted by Ray Schoch on 05/26/2015 - 08:48 am.

    More reasons

    …for snowbirds to flee to Tucson or Tampa from November to May…

  2. Submitted by frank watson on 05/26/2015 - 10:44 am.

    In Minnesota

    …. most cold weather deaths are due to excessive drinking.

  3. Submitted by Ilya Gutman on 05/26/2015 - 09:09 pm.

    Can this be true?

    Here is a possible clear positive result of global warming – it can save lives…

    • Submitted by John Clark on 05/27/2015 - 09:07 am.

      But on the other hand . . .

      Here is a comment in today’s Strib by meteroligist Paul Douglas that just might be relevant too:


      On Memorial Day the National Weather Service issued a rare “Flash Flood Emergency” for life- threatening conditions near Austin, Texas. NOAA has been using this terminology for only 7 years. Like a “Tornado Emergency” it implies an even greater danger and urgency than a warning.

      Much of Texas is experiencing historic flooding; the same counties that were in a severe drought just 3 weeks ago.

      When climate skeptics say “What’s a few degrees among friends?” or “CO2 is plant food, Paul!” I point them to new research showing a growing link between rapid Arctic warming and weather-whiplash, along with shifts in the jet stream.

      There’s early evidence that weather systems are deeper, moving slower, especially during the summer months. Storms and fronts seem more prone to getting “stuck”. When weather stalls bad things can unfold, debilitating droughts, stronger heatwaves and mega-floods.”

      • Submitted by Ilya Gutman on 05/27/2015 - 08:35 pm.

        And on the other hand, Minnesota is in the midst of the third year in a row of colder than usual weather… So we should be really careful when we try to correlate weather and climate – it works both ways.

Leave a Reply