Wednesday, June 20, 2012

Do Heat Waves Cause Deaths?

All scientists seek to make "causal statements" rather than simply uncovering correlations.  The U.S northeast is in the midst of a heat wave and the NY Times has made a strong causal statement about last year's August heat wave.  "The last time the temperature broke 90 in New York was in August, as part of a heat wave that claimed at least 31 lives in the city."

How does the NY Times know that 31 people would still be alive had there not been that heat wave?  This is the "causality" challenge.  I acknowledge that it is certainly possible that 31 people did die during the dates of the heat wave but note the causality that the NY Times attributes to the heat wave.   Using Google, I found this Ph.D study of Ireland  .    It does a better job investigating the statistical patterns between heat and death.

Economists have made an important point about studying these correlations.   Read this paper by Enrico and Olivier.   Heat waves cause harvesting as they tend to lead to deaths of people who were vulnerable and would have died soon after the heat wave had the heat wave not taken place.   

Here is  a quote from the authors

"While it is clear that mortality spikes in days of extreme hot or cold temperature,
the significance of those deaths in terms of reduction in life expectancy is much less
clear. The number of deaths caused by extreme temperatures on a given day could be
compensated for by a temporary fall in mortality in the subsequent days or weeks, if
extreme temperature principally affects individuals whose health is already compromised.
This could happen if extreme temperature precipitates the health condition of individuals
who are already weak and would have died even in the absence of the shock. In this case
the only effect of the weather shock is to change the timing of mortality by a few days or
weeks, but not the number of deaths in the longer run. Such temporal displacement is
sometimes referred to as the “harvesting” effect.  Thus, the excess mortality observed on
cold and hot days does not necessarily imply significant permanent reductions in life

UPDATE:  My UCLA colleague Alex Hall's work on predicting climate change's localized impacts for Los Angeles received a nice write up in the LA Times today.   Note that he predicts that West LA won't heat up that much as measured by the count of days exceeding 95 degrees.  So, the adaptation strategy is to allow for high rise buildings in this "temperate" part of the city.  We need the land use zoning laws to be flexible to meet changes in demand.  Right now, the liberal city of Santa Monica makes it quite difficult to build new housing there or to build multi-family housing there.  Both, need to change in our hotter future.

Alex's work is linked to this column about heat waves deaths because he is predicting more heat in East L.A.  An adaptive population needs to have the opportunity to migrate and build high rises in "safer" more pleasant areas.  Such areas exist and self interested people in the future will find them.  Let's see if government induced supply side constraints limit the gains to trade between land owners, real estate developers and households seeking to live well in the hotter future.   The free market can handle these trades if government intervention doesn't inhibit such trades.   The net effect is easier adaptation.