The mortality costs of extremely cold weather

by on January 7, 2014 at 9:22 am in Medicine, Science | Permalink

Here is an NBER paper from 2007 — perhaps it is timely today — by Olivier Deschenes and Enrico Moretti:

We estimate the effect of extreme weather on life expectancy in the US. Using high frequency mortality data, we find that both extreme heat and extreme cold result in immediate increases in mortality. However, the increase in mortality following extreme heat appears entirely driven by temporal displacement, while the increase in mortality following extreme cold is long lasting. The aggregate effect of cold on mortality is quantitatively large. We estimate that the number of annual deaths attributable to cold temperature is 27,940 or 1.3% of total deaths in the US. This effect is even larger in low income areas. Because the U.S. population has been moving from cold Northeastern states to the warmer Southwestern states, our findings have implications for understanding the causes of long-term increases in life expectancy. We calculate that every year, 5,400 deaths are delayed by changes in exposure to cold temperature induced by mobility. These longevity gains associated with long term trends in geographical mobility account for 8%-15% of the total gains in life expectancy experienced by the US population over the past 30 years. Thus mobility is an important but previously overlooked determinant of increased longevity in the United States. We also find that the probability of moving to a state that has fewer days of extreme cold is higher for the age groups that are predicted to benefit more in terms of lower mortality compared to the age groups that are predicted to benefit less.

Ungated versions of the paper can be found hereAddendum: The published version is here, with slightly different numbers.

ummm January 7, 2014 at 9:31 am

but the left insists that not only is the world getting warmer, but cold weather is beneficial . The truth of the matter is, humans suck at cold weather.

IVV January 7, 2014 at 9:36 am

I’ve never heard the argument that cold weather is beneficial. However, I have noticed the paucity of data about which places experience warmer winters.

Gordon Mohr January 7, 2014 at 1:31 pm

One study suggesting cold weather improves per capita output appeared here less than 3 weeks ago:

http://marginalrevolution.com/marginalrevolution/2013/12/does-a-warm-climate-discourage-economic-output.html

Given this mortality info, though, maybe the effect comes more from reducing the ‘capita’ than increasing the ‘output’?

Gil January 7, 2014 at 10:24 pm

Because no one ever dies from heat waves.

Alex Godofsky January 7, 2014 at 9:38 am

However, the increase in mortality following extreme heat appears entirely driven by temporal displacement, while the increase in mortality following extreme cold is long lasting.

Aren’t they all temporal displacement, just in different degrees? After all, everyone does eventually die.

Dan Weber January 7, 2014 at 9:48 am

I’m trying to point out how you are technically wrong but cannot. :P

My understanding, though, is that heat waves *specifically* take out those who have months left to live, while the deaths from cold snaps are generally spread over the entire population.

Does “cold deaths” mean people dying from exposure, or does it include accidents from icy/snowy roads?

Alexei Sadeski January 7, 2014 at 10:17 am

We all have months to live, when ya think about it.

PD Shaw January 7, 2014 at 10:18 am

The paper seems to focus on ages 65 and up as those most affected by cold deaths, particularly it appears through cardiovascular disease.

Accidents seem to be a wash: “Eisenberg and Warner (2005) found that on
snow days there were more nonfatal accidents than on dry days, but less fatal crashes.
They also found evidence of behavioral adjustment in the sense that the first snowy day
of year was associated with substantially higher accident risk than subsequent snow days.”

Footnote: “cold temperature reduces mortality for young adults (aged 20-34) through a marked reduction in motor-vehicle accidents fatalities.”

Steve-O January 7, 2014 at 12:06 pm

It always bugged me that traffic deaths during a storm were attributed to the storm. What about all the lives saved because people weren’t out driving around? Or at least be consistent and do stories about all the traffic deaths in San Diego due to the 70 degree weather.

JWatts January 7, 2014 at 1:00 pm

” all the traffic deaths in San Diego due to the 70 degree weather.”

Particularly motor cycle deaths.

Nate January 7, 2014 at 11:15 am

Heart attacks due to shoveling are extremely common.

mpowell January 7, 2014 at 11:26 am

Yeah, but you can measure correlations. In a given population, say people age 65, there will be individuals with shorter and longer expected remaining lifespan depending on their current health circumstance. If excess mortality due to extreme heat is highly correlated with individuals with shorter current remaining lifespan then we call that ‘temporal displacement’. If excess mortality due to extreme cold is distributed randomly through the population, that’s a different result. The concept of ‘temporal displacement’ is perhaps vague but the statistical result they have identified is not and has some useful meaning for sure.

Alex Godofsky January 7, 2014 at 12:11 pm

Yes, I know, see “just in different degrees”.

chuck martel January 7, 2014 at 10:06 am

The concept is obvious. Brazilians greatly outnumber Eskimos.

prior_approval January 7, 2014 at 10:30 am

‘These longevity gains associated with long term trends in geographical mobility account for 8%-15% of the total gains in life expectancy experienced by the US population over the past 30 years.’

So, in all those countries with cheaper health care systems, where life expectancy has been rising, the fact that the population can’t move a thousand miles to the south means that their gains are even more impressive after discounting this 8%-15% American mobility bonus.

Pshrnk January 7, 2014 at 10:58 am

+1

kiwi dave January 7, 2014 at 11:18 am

Alternatively, if you take into account that Europe has *much* milder temperatures than the continental U.S. (in terms of both heat and cold — compare January and July temperatures in, say, Chicago v. Berlin or New York v. Paris), maybe the it’s surprising that the U.S.’s mortality rate is as low as it is. On those occasions in recent memory in which Europe has had extreme temperatures, the performance has not been so good. For example, the 2003 European heat wave killed 70,000 people — including almost 15,000 in France and 9,000 in Germany — from temperatures that would not have been considered exceptional (and certainly not deadly) in the American midwest (only on occasion did the temperature reach 40 Celsius). UK health authorities had predicted 40,000 excess deaths from the winter of 2005-2006 (http://www.nzherald.co.nz/world/news/article.cfm?c_id=2&objectid=10619380&pnum=2) — again, with temperatures that would be absolutely unexceptional in most American cities.

Tom West January 7, 2014 at 11:26 am

Let’s be a little reasonable. Temperatures that are well outside the expected bounds will cause increased mortality in any location. If you get extreme temperatures regularly, you are far better prepared for them. I don’t credit Winnipeg’s universal health-care system for the fact that ten’s of thousands aren’t dying when the temperature drops.

kiwi dave January 7, 2014 at 11:29 am

Tom,

I agree that events outside the usual bounds are going to increase mortality. My point is that you need to take into account the fact that different places have different climate-related burdens. The fact that the U.S. and Canada cope, on a regular basis, with temperatures that would be seen in Europe as supernatural is something that needs to be taken into account.

Tom West January 7, 2014 at 5:10 pm

Agreed. However, outside of some Canada-US hospital studies that were attempting to actual improve patient welfare through best practices, the only use I’ve seen cross-border studies put to is to beat the other side of the universal health-care debate. Such comparisons have sufficient ambiguity that you can explain away the ones that don’t favor you and use the ones that do to prove your opponents want to destroy healthcare, despite the evidence. So attempting to add yet another factor why our scores are inferior/superior is mostly a waste of time.

Color me cynical, but I have yet to see an opinion in the health-care debate (including my own) that isn’t far more about moral principles than actual health outcomes.

kiwi dave January 7, 2014 at 11:27 am

the fact that the population can’t move a thousand miles to the south

Wow, with that level of geographic illiteracy, you must be an American.

The whole of Scandinavia is more than 1,000 miles north of all of Greece and Spain and the entire Italian mezzogiorno. The Costa del Sol is more than 1,000 miles south of Manchester, or Hamburg, or Amsterdam. And if your point is that Europeans can’t escape to somewhere else in Europe where they can avoid cold winters, that is demonstrably false (have you been to the Spanish mediterranean coast?) And, anyway, as I pointed out above, western European winters are not as cold as continental American winters to start with.

prior_approval January 7, 2014 at 12:10 pm

‘The whole of Scandinavia is more than 1,000 miles north of all of Greece and Spain and the entire Italian mezzogiorno.’

Sure is – Germany’s 1600 kilometers sits in the middle. Well, in the case of Greece and Spain, that also includes much of France or Austria and the Balkan countries, but why bother with details.

‘Costa del Sol is more than 1,000 miles south of Manchester, or Hamburg, or Amsterdam. And if your point is that Europeans can’t escape to somewhere else in Europe where they can avoid cold winters, that is demonstrably false…’

Nope – the amount of nothern Europeans that move to sourthern Europe is perhaps worth a major real estate bubble, but it is not exactly common.

‘…have you been to the Spanish mediterranean coast?’

Why yes – a family friend we visit, a founder of a German three letter ERP software company, owns a house there. Your point being?

‘And, anyway, as I pointed out above, western European winters are not as cold as continental American winters to start with.’

I’m a northern Virginian that has lived in southern Germany for more than 20 years – I find the weather colder here, in general. And Virginia is not really considered all that southern by those fleeing such northern states as Mayland, Pennsylvania, or New York. Admittedly, New Englanders find Virginia’s summer’s brutal – none of them, from my experience, consider the winters all that mild (on a longer term basis, that is). Neither Virginia nor Germany nor New England have continental weather – even if this region of Germany actually has higher record temperatures than where I grew up (with old fashioned New English levels of air conditioning). However, only someone who hasn’t lived in both places would look at such statistics and think Germany is warmer (hint – orbital mechanics).

Careless January 8, 2014 at 4:06 pm

Unless it did in the past few days, it’s been over 3 years since DC has gotten 2″+ of snow.

mike January 7, 2014 at 11:41 am

Yes, let’s start talking about differences between the US and those countries other than the health care system

prior_approval January 7, 2014 at 12:15 pm

When up to 15% of the increase in American increase in life expectancy can be explained by nothing more than moving south, an option not really available to other countries, it is interesting to use such data corrections when talking about health care.

Unlike the spurious statistics previously posted here, where American life expectancy rose if violent/accidental death was excluded, while everybody else’s life expectancy decreased.

This study seems fairly well based actually (the reduction in accident deaths of a certain cohort being an interesting point, for example), and for those so interested in pointing out just how exceptional America is, perfectly suited.

Well, except for the fact that it just adds to the evidence that the American health care system is even more expensive and less effective than its defenders realize, of course.

interested January 7, 2014 at 12:23 pm

Hey prior_approval,

Just curious what brought you to Germany and kept you there. I’m wondering because I too am American and secretly dream of relocating to Germany to work while I’m still unburdened by a wife and kids. Right now I work in finance in NYC.

TMC January 7, 2014 at 12:33 pm

I always had the impression prior was kicked out.

JWatts January 7, 2014 at 1:07 pm

No, I think he had to go on the run to avoid the bounty the Koch brothers placed on his head for exposing the right wing/libertarian/Mercatus Center conspiracy.

prior_approval January 7, 2014 at 1:07 pm

‘.. what brought you to Germany .. ‘

My wife.

prior_approval January 7, 2014 at 1:13 pm

‘I always had the impression prior was kicked out.’

Well except for the parents that worked for 3 letter agencies, as do a couple of my siblings. I had to turn down an unsolicited job offer for one of them. I preferred GMU, actually. But then, I was naive back then about how propaganda worked, even if I accepted the paycheck for a while.

prior_approval January 7, 2014 at 1:54 pm

‘the Koch brothers placed on his head for exposing the right wing/libertarian/Mercatus Center ‘

Why? They are extremely proud of what they have done. It is just their lower level staff that seem to not understand this pride, fearing that if others knew what the Kochs’ supported, it would be rejected.

Cowards, but ones the Koch brothers have seemingly rewarded well.

But that fairy tale about youtube and a 4 dollar app being the basis for MRU – hopefully no marginal revolution commenter is naive enough to believe that 1200+ inqbation project hours (www.inqbation.com) are just provided to a couple of GMU professors out of some sense of charity. Someone approved that likely over $100,000 check, quite possibly including Mercatus Center General Director Cowen.

Who just happens to the main figure in MRU – such a coincidence, right?

Ted Craig January 7, 2014 at 12:23 pm

Life expectancy has been rising in every western nation for decades. I’m not sure you have a point. However, if you look at the nations ranked by life expectancy worldwide, moderate-climate countries have an edge in most cetagories. This is especially true when considering how much more the cold countries spend on healthcare. So you could make the case for a temperature premium. A good comparison is Canada and Australia, two very similar countries with identical life expectancies. Canada spends 11.1 percent of its GDP on healthcare while Australia spends 9.1 percent.

prior_approval January 7, 2014 at 1:35 pm

So, more support that this finding is accurate?

Ted Craig January 7, 2014 at 1:50 pm

It might be, but you’re original comment remains weak.

prior_approval January 7, 2014 at 1:57 pm

And yet, it actually rests on empirical numbers, with additional empirical support provided by comparing two Commonwealth countries, with many similarities. Apart from the striking difference in geography, of course.

Floccina January 9, 2014 at 12:55 pm

@prior_approval are you saying that if the USA had the healthcare system of Italy that the USA would have the life expectancy of Italy. Are you also saying that if the Denmark had the healthcare system of Italy that the Denmark would have the life expectancy of Italy. BTW Italy does much better on LE than USA and Denmark.

The Other Jim January 7, 2014 at 10:51 am

So by demanding that we cool the Earth, Al Gore ranks as one of the most wanton murderers of all time.

Sounds like we all need to surround our houses with moats of burning coal. It’s for the children, after all.

We live in interesting times January 7, 2014 at 11:56 am

21st century ice floes?

Ted Craig January 7, 2014 at 11:59 am

This does raise a question. For years, the longevity of Okinawans and Greeks has been attributed to diet. Could climate play as large a role?

prior_approval January 7, 2014 at 1:06 pm

OK, that this hasn’t been removed as spam is a disappointment for a deletion policy that seems so rigorous when applied to whatever our hosts actually care about. (No insulting Fed employees, for example.)

We live in interesting times January 7, 2014 at 1:15 pm

Well, that’s 1 way to solve intergenerational warfare.

Jimbino January 7, 2014 at 1:47 pm

Deaths due to cold or hot weather in Amerika can be almost entirely eliminated. When I lived in Munich, Germany, the government had a policy of subsidizing the owner of a Gasthaus in maintaining heat and allowing pensioners and others in need of warmth to sit inside all day.

Besides providing refuge from the cold, the place offered conversation and commiseration. It would cost very little to do that here.

Where I live in Austin, the city maintains Senior Citizen Centers, where folks can go for a hot meal, read, converse, play pool and bridge. It’s a good idea, but who the hell wants to take a bus halfway across the city only to hang out in a sterile place where you can’t have a beer or a glass of wine, and where you meet only religious folks?

jon January 7, 2014 at 2:54 pm

We need to put a stop to his. Global climate change will only worsen extremes like these. The winter in Europe is exceptionally warm at the moment.

Steve Sailer January 7, 2014 at 3:27 pm

Slipping on ice and breaking your hip and never coming out of the hospital alive is a quite reasonable fear for the elderly in the North.

Seth January 10, 2014 at 5:28 pm

Malaria, yellow fever, typhoid, brain eating amoebas, ect., all more common in warm climates. We should question how deaths from the cold impact our future human evolution, for example, killing out the mentally ill homeless and the sick from our breeding population. Alternatively, since we no longer live as much in a state of nature, these deaths may not be as much of a result of natural selection but rather a result of artificial selection due to unfair opportunities, and thus unfair housing and protection against the cold.

Comments on this entry are closed.

Previous post:

Next post: