1°C – the silent killer

From the Harvard School of Public Health , news that an extra 1°C temperature swing in summer will kill the elderly.

Summer temperature variability may increase mortality risk for elderly with chronic disease

Large day-to-day variations in temperature could result in thousands more deaths per year

Boston, MA – New research from Harvard School of Public Health (HSPH) suggests that seemingly small changes in summer temperature swings—as little as 1°C more than usual—may shorten life expectancy for elderly people with chronic medical conditions, and could result in thousands of additional deaths each year. While previous studies have focused on the short-term effects of heat waves, this is the first study to examine the longer-term effects of climate change on life expectancy.

The study will be published online April 9, 2012 in the Proceedings of the National Academy of Sciences.

“The effect of temperature patterns on long-term mortality has not been clear to this point. We found that, independent of heat waves, high day to day variability in summer temperatures shortens life expectancy,” said Antonella Zanobetti, senior research scientist in the Department of Environmental Health at HSPH and lead author of the study. “This variability can be harmful for susceptible people.”

In recent years, scientists have predicted that climate change will not only increase overall world temperatures but will also increase summer temperature variability, particularly in mid-latitude regions such as the mid-Atlantic states of the U.S. and sections of countries such as France, Spain, and Italy. These more volatile temperature swings could pose a major public health problem, the authors note.

Previous studies have confirmed the association between heat waves and higher death rates. But this new research goes a step further. Although heat waves can kill in the short term, the authors say, even minor temperature variations caused by climate change may also increase death rates over time among elderly people with diabetes, heart failure, chronic lung disease, or those who have survived a previous heart attack.

The researchers used Medicare data from 1985 to 2006 to follow the long-term health of 3.7 million chronically ill people over age 65 living in 135 U.S. cities. They evaluated whether mortality among these people was related to variability in summer temperature, allowing for other things that might influence the comparison, such as individual risk factors, winter temperature variance, and ozone levels. They compiled results for individual cities, then pooled the results.

They found that, within each city, years when the summer temperature swings were larger had higher death rates than years with smaller swings. Each 1°C increase in summer temperature variability increased the death rate for elderly with chronic conditions between 2.8% and 4.0%, depending on the condition. Mortality risk increased 4.0% for those with diabetes; 3.8% for those who’d had a previous heart attack; 3.7% for those with chronic lung disease; and 2.8% for those with heart failure. Based on these increases in mortality risk, the researchers estimate that greater summer temperature variability in the U.S. could result in more than 10,000 additional deaths per year.

In addition, the researchers found the mortality risk was 1% to 2% greater for those living in poverty and for African Americans. The risk was 1% to 2% lower for people living in cities with more green space.

Mortality risk was higher in hotter regions, the researchers found. Noting that physiological studies suggest that the elderly and those with chronic conditions have a harder time than others adjusting to extreme heat, they say it’s likely these groups may also be less resilient than others to bigger-than-usual temperature swings.

“People adapt to the usual temperature in their city. That is why we don’t expect higher mortality rates in Miami than in Minneapolis, despite the higher temperatures,” said Joel Schwartz, professor of environmental epidemiology at HSPH and senior author of the paper. “But people do not adapt as well to increased fluctuations around the usual temperature. That finding, combined with the increasing age of the population, the increasing prevalence of chronic conditions such as diabetes, and possible increases in temperature fluctuations due to climate change, means that this public health problem is likely to grow in importance in the future.”


Support for the study was provided by the U.S. Environmental Protection Agency and the National Institute of Environmental Health Sciences.

“Summer Temperature Variability and Long-term Survival Among Elderly People with Chronic Disease,” Antonella Zanobetti, Marie S. O’Neill, Carina J. Gronlund, and Joel D. Schwartz, Proceedings of the National Academy of Sciences, online April 9, 2012.

Visit the HSPH website for the latest news, press releases and multimedia offerings.

Harvard School of Public Health (http://www.hsph.harvard.edu ) is dedicated to advancing the public’s health through learning, discovery, and communication. More than 400 faculty members are engaged in teaching and training the 1,000-plus student body in a broad spectrum of disciplines crucial to the health and well being of individuals and populations around the world. Programs and projects range from the molecular biology of AIDS vaccines to the epidemiology of cancer; from risk analysis to violence prevention; from maternal and children’s health to quality of care measurement; from health care management to international health and human rights. For more information on the school visit: http://www.hsph.harvard.edu

HSPH on Twitter:


HSPH on Facebook:


HSPH on You Tube:


HSPH home page:



newest oldest most voted
Notify of

1C higher will kill more elderly? Then why in the hell do they all move from the mid-west and northeast to Florida and Arizona?


Ten thousand additional deaths per year. Every year? Or maybe you can only die once and there is a limited supply of old, poor and vulnerable people. Anyhow, how are those cold winters working out, death-wise?
Sometimes you need to look at these studies. Other times one’s initial scornful reaction is sufficient. This is one of those times.


In the era of air-conditioned homes how can any one tell whether the seasonal weather has anything to do with increased mortality.


Some background on air-conditioned America.


What are the figures for 1C colder?
I would have thought that cold might be worse than heat.


So, how many fewer people will die in winter if temperatures are less cold?
OTOH, it’s the same old same old – junk science again.

So what happens if the temp drops by the same amount?


And assuming you believe in AGW, where is the study showing the reduction in deaths due to warmer winters. Usual biased rubbish.


So what is the death rate increase for cooling?
I bet it is much higher!
Jumping on AGW funding bandwagon, next stop Biasville, all aboard….

No indication they controlled for actual temperatures. IE greater variability is associated with higher temperatures. And as they note extreme high temperatures for a locality is associated with high mortality.
Otherwise, no real evidence ‘climate change’ causes greater temperature variability.

Harold Ambler

These people appear to believe that the ocean-atmosphere system dished out ideal weather worldwide until about 30 years ago. The ignorance of history and science is so extreme as to leave one amazed.
I’ve done a few interviews of late on my book, btw, two of which are here:

I can only find a press release, but a couple of obvious questions that come to mind:
(1) Why are they citing an increase in temperature fluctuations rather than just a general warming trend? Since general warming has no detectable ill effect, it seems critical that their claim is supportable. What citations do they use to support this assertion?
(1a) Wasn’t the most significant effect of AGW an increase in winter temperatures/reduction of lows. Wouldn’t this decrease fluctuations in temperature overall?
(2) Did they offset increased deaths due to heat fluctuation (stronger potential for heat waves I suppose) against reductions in winter extreme lows? Because cold kills more than warmth, did they correctly deduct reduced deaths from cold before considering increased deaths from heat?


As I understand it, it says 1C extra swing. So could be hotter OR colder. Makes sense, I guess, that people have a harder time coping with a higher variability in temperature. That people living in warmer climates are more susceptible…..dunno.

Is this a reiteration of long-known facts masquerading as a new insight for someone’s PhD or a poorly-disguised and crass appeal for more funding from an Ivy League institution that should really know better? It’s just that those of us at the tea-time of our lives know that we will die of old age with some spurious reason entered on the Death Certificate, and that cause of death is subject to the medical fad of the moment. While not being conversant with either Minneapolis or Florida, one could reasonably expect that sub-tropical, maritime florida would have less violent temperature swings than a city buried in the northern great plains, even without taking into account that millions of Americans work all their lives to be able to retire to Florida, and that personalwealth is the major contributor to an extended life?


And reading their conclusion reveals:
‘In conclusion, we confirmed in a large sample of cities that subpopulations such as the elderly, diabetics, and black subjects are especially susceptible to temperature extremes. According to our results, susceptibility of populations may vary according to the primary cause of death, suggesting that future studies on susceptibility will benefit from taking this into consideration. Finally, we found evidence that cardiovascular deaths, especially cardiac arrest deaths, show much larger increases on extremely cold days than other mortality causes.’
So cold does kill more (as we all know) but give us a grant anyway….

This study shows without any doubt that people from Northern states moving South run a higher risk of dying due to temperature differences. Statistics show that the majority of elderly retirees who have moved south to florida die within 10 years of moving South.
Statistics don’t lie, the science is settled.


Summer temperature variability may increase mortality risk for elderly with chronic disease
Though there is no “may” when driving up the cost of Winter heating via insane green schemes.

Tom in Florida

Perhaps it’s just nature’s way of evening things out. Most of those chronically ill people would have already died if it wasn’t for extensive medical care that all the rest of us pay for.


Did I miss the /sarc at the end of the article?
This study can’t be for real for the reasons cited in the first half-dozen or so comments.

Old Goat

Here, in France, I was experiencing daily temperatures of about 24 degrees centigrade a week or so ago. These last few days, we’re down to single figures again, so the temperature has risen and dropped 15 degrees or so within the short space of three weeks. In a fwe weeks it will have gained at least 15 degrees, again.
I have to say, that I have witnessed no deaths or apart from whingeing, any serious temperature related problems amongst friends, and my elderly neighbours, nor have I any other year since I’ve been here, and it does get pretty hot (and pretty cold, too).
So why should I be concerned about propaganda that these idiots are continually throwing at us? ONE degree? Come on, do they think we’re as stupid as they obviously are?
Who in their right minds believes this rubbish? They must be really thick if they think we fall for it – we’re not buying. I just wish they’d go and find something useful to do with their lives, instead of preaching, taxing, and trying to scare us to death with things that we (and, hopefully, they) all know aren’t true. Sheer, utter madness.

I’ve noticed that Harvard pumps out an enormous amount of junk science and this appears to be one more example. As is always the case with observational studies, correlation does not mean causation.


That’s why we need technology and energy for air-conditioning! But even then, no air-conditioning in summer won’t kill as easily as no heating during winter.

Paul Bell

What with the sun going up and down all the time we are experiencing 10degC temperature swings every single day! I don’t know how the old folks here are hanging on.

R Barker

It seems to me that there is some flaw in the logic of this study which claims that “for elderly people with chronic medical conditions…… could result in thousands of additional deaths each year. ”
Everyone dies eventually. Elderly persons with chronic disease would be near term candidates for death. Large temperature swings may be the trigger for death at a particular time, sooner rather than later for those at risk. It does not follow that 10,000 more will die each year without an increase in the number of people in that subset of the population (3.7 million) at near term risk.
If the temperature swings did not actually cause an increase in the number of elderly with chronic disease, then while the death rate might fluctuate with temperature swings, I would not expect it to be changed among that population subset over the longer period of time.


“even minor temperature variations caused by climate change may also increase death rates…”
Are minor temperature variations not caused by climate change (assumed AGW) okay?

Living causes death


How many people will die when the cost of electricity goes up making air conditioning less accessible to people living on a fixed income.
Let’s say that there are 20 million chronically ill folks in the country with air conditioning. If 10% of those folks have to forego air conditioning due to the higher cost, their bodies will be forced to experience uncharacteristic swings of let’s say 5 degrees C.
We should expect a one time increase in the death rate of 4% * 5 = 20%, or 400,000 dead grammys and grampys.
I sure the EPA will view this massacre as an investment, offset by all the people saved from breathing poisonous CO2 in the future.

In hot weather my favourite occupation is sitting in the garden with a cold beer or a G&T. Thoughts of my daily stroll will be totally forgotten. It won’t be the heat that is responsible for my final demise, but the alcohol and lack of exercise
But then I suppose you can blame the heat for increasing my alcohol consumption and my lack of any desire to go out for a walk, so my death, one day,will be added to the statistics as heat related..

I would like to see a study on the increased number of heart attacks and strokes brought on by an increased number of stressed and angry people because they have to live under the regimes of politicians implementing costly ineffective policies.

Disko Troop

I would like to have seen the paper that came before this…researched by the Bursar of HSPH which said that a 1% increase in the mentions of Global Warming in medical research papers result in a 15% increase in Government Grants, EPA grants and bungs from Soros and Suzuki. The memo that went out to his professors must have been a revelation. I can palpably taste my contempt for poeple who put crap like this into the public domain.


They want us to give up fossil fuels — which means giving up industrial cvilzation. How many more deaths will that cause each year? Have the geniuses at Harvard done any studies on whether life expectancy was longer or shorter in pre-industrial times? If this doesn’t convince people that there is no science in climate “science,” just anti-industrial ideology disguised as science, then nothing will — and our civilization is doomed.


I see we’ve finally managed to work in the old racist angle…. If you don’t combat global warming, you’re racist because it adversely affects African-Americans disproportionately. The “hyphenation effect” has long been known here but skeptics have continuously refused to acknowledge it.
The hyphenation effect only impacts hyphenated races, such as African-Americans, but strangely not Africans, like those living in equatorial or desert regions of Africa where summer temperatures can be higher than in Chicago or Nashville.


Look at insurance statistics. More elderly people die when it is cold!

Eric Dailey

Where is that big red button when I need it?


The quote posted earlier, “If your experiment needs statistics, you ought to have done a better experiment.” seems relevent.
In global terms, of those aged over 65, there is a greater proportion of women to men, at present something like 1.5 to 1 (women:men). Over 85 it is 3 to 1. The elderly also have lower incomes and a higher proportion, compared to the under 65’s, live in poverty with less access to healthcare.
The medical conditions mentioned are as much associated with rising obesity as with those aged over 65. Infact all studies of the ageing population show a rapid increase in those aged 65 and over and a similar increase in those aged 80 and over.
Elderly people overall are actually becoming healthier, they have less disease and lead more active lives often continuing to work long past the age of 65.
The projected increase in the ageing population of the entire world is from about 7 % of the population in 2000 to a projected 20% in 2050. Even in Africa the increase is likely to be a rise in the elderly of from 3% in 2000 to 7% in 2050. In the US the rise is from 12% in 2000 to a projected 21% in 2050.
These statistics represent much healthier, rapidly growing elderly populations. In 2009 in the US there were 39 million Americans aged over 65. If the rise continues at its present rate, It is projected that by 2030 there will be 72 mllion Americans aged over 65, many of whom will be over 80 and still leading active lives.
I suppose next they will try to correlate the rise in obesity with climate change. :o(



One thing we know for sure: Thousands of people are unnecessarily dying in Britain because they can’t afford to create large differences of temperature between inside and outside. When they go in their houses, it’s just as cold as outside. Carbon tax should make them live forever, but for some mysterious reason it doesn’t.


Perhaps next week the National Academy of Sciences will publish a study out of Yale demonstrating that if only we ban automobiles, we will save 40,000 lives each year in the U.S. alone! And it’s only the evil car industry that prevents us from doing it!

Ian W

Note that this article is cleverly written to imply that 1degC rise in temperature will kill people especially the poor and Black communities. However, that is NOT what the study actually appears to show which is higher variability MAY increase death rates.
Now we also have to add the ineptness of medical researchers in statistics and their application. I presume by variability they are looking at temperature range over what period of time? As has been pointed out the daily range of temperatures in some places is up to 15 or 20degC. Then they admit in the conclusion “Finally, we found evidence that cardiovascular deaths, especially cardiac arrest deaths, show much larger increases on extremely cold days than other mortality causes.” and as everyone already knows cold kills – but were the cold snaps accounted for in their statistics? I doubt it as a research paper stating that cold kills even if its only a few cold nights would hardly merit attention at Rio+20 – whereas a paper saying 1degC can kill especially the poor…. means a guaranteed research grant and possibly an all expenses paid trip to Brazil.


Like everything on Climate Change in PNAS, this latest is a blatant perversion of the truth. For millennia more people have died in cold winters than in hot summers (amongst them my 4 grandparents who all died in the bitter English winters of 1946-7 (aggravated by the globalwarmists’ coal strike?), 1956-7, 1960-61, and 1962-3). Even today across northern Europe and America more die in winter than in summer. Get real! – impossible at PNAS, which has published papers implying rice does better in Scotland than in Thailand, wheat in Greenland than in Australia, and cane sugar in Alaska than in the Sudan (I have the PNAS citations claiming crop yields are higher in cold regimes than in warmer).

Peter Dunford

The study period had lots of other tbings changing over that time period. We could suggest that climate alarmism increased during tgat period, perhaps that was the cause. Or income inequality. Energy costs. It is also just possible that people with chronic conditions have not benefitted from increasing life spans over the period because of their, er, chronic conditions. They only apear to have shortened lives because they haven’t had longer ones.
Are they implying the being African American is like having diabetes or asthma? Bit harsh.


this will play perfectly to decrease the costs of obamacare. after all, its all those damn sick old people that use all the resources


JunkScience.com skewers this “study”:
“This study is, of course, an exercise in statistics that is devoid of any relationship to the real world of morbidity and mortality.”

quack quack quack…if it looks like a duck it probably is. I don’t want of those quacks near me…ever.
Harvard School of Public Health? Sure…


The warmists I know think the world is over-populated with humans anyway. They should be glad if death rates rise, shouldn’t they?? :-\


That is why we don’t expect higher mortality rates in Miami than in Minneapolis, despite the higher temperatures,”
wrong……Minneapolis gets a lot hotter and if they are talking about temp swings Miami changes very little, Minneapolis changes a lot


Follow the money:
“Support for the study was provided by the U.S. Environmental Protection Agency and the National Institute of Environmental Health Sciences.”


Now let them do a study on winter deaths and how manhy elderly with chronic conditions are likely to suffer with a 1c colder and benefit with 1c warmer winters caused by global warming.

Winter deaths on the riseHypothermia cases double in five years
14 February 2012
Hypothermia-related deaths and hospital admissions have almost doubled over the past five years.
Almost 2000 people were admitted to hospital and 260 died from the condition in 2010/2011, according to figures from the NHS information centre.
Hypothermia occurs when body temperature falls below 35C from its usual 37C.
Saving money on energy bills
Last week a survey by the charity Age UK revealed that half of pensioners – the age group which makes up three quarters of hypothermia-related hospital admissions – had turned down their heating to save money.

Dodgy Geezer

This is NOT about temperature increase, it’s about temperature VARIABILITY.
I don’t actually find it surprising that humans prefer an optimum temperature, and that wide variations from it will lead to more deaths. And then, like radioactivity, you can extend that thinking to say that if varying the temperature +/- 100C kills 50% of the people, then varying it +/- 1C will kill 0.5% of the people. Is this what the paper has done? I wouldn’t be surprised…
But where is the evidence, or, indeed, the assertion, that Global Warming will INCREASE temperature variation? I would have thought that if you added extra heat, you would tend to suppress variation….


So basically, mon 25c no problem, tues 25c no problem, weds 25c no problem, thurs 26c, omg they’re all going to die!