November 22, 2005

Overstating Health Impacts of Global Warming

Filed under: Health Effects

The best way to garner headlines in the global warming game is to generate scary scenarios about how many people will die in its wake. While many people view global warming as some esoteric concern of environmentalists, it does at least raise one’s eyebrows when you hear a phrase like “global warming deaths.”

It’s little surprise then that a “Review” article that just appeared in Nature magazine has caught so much attention. The review by Jonathan Patz of the University of Wisconsin and three colleagues essentially is a selective culling of the scientific literature—some recent, some not—on climate change and possible health impacts across the planet. And it should also be little surprise to the readers of this column that prospects are bad.

In an effort to provide balance, we’ll briefly review key portions of the paper and provide a much-needed perspective that was unfortunately missing.

Patz begins with the 2003 heat wave in Europe. First, it is not possible to say this or any heat wave was caused by global warming, despite some climate modeling efforts that, according to Patz, “[demonstrates] a causal link.” It is impossible, and in fact is irresponsible, for any climatologist to claim that any given weather event could not have happened if not for increasing atmospheric greenhouse gases. Yes, 2003 was a very warm summer in Europe, but the fact the similar conditions occurred there in the very distant past essentially pretty much debunks this “global warming” hypothesis.

The more relevant question is why so many Europeans died in August, 2003. Here is where culture conspired with climate to raise the death toll. The month-long August vacation is a cherished European tradition. It’s not unusual for many countries to effectively shut down while the epicenter of the population shifts southward to Mediterranean beaches. This includes a reduction in medical staffing, less oversight of one’s elderly parents, etc. The French government caught flak for the high death toll, and rightly so. Undoubtedly the same weather conditions in July would have produced substantially fewer deaths.

But the cultural factor is never mentioned when you have a global warming hook.

The “theory” that leads to such sloppy thinking about heat waves is that with global warming, climate will be more variable. While the jury is still out on this, there is plenty of evidence for the United States that the opposite is true. In a series of studies by Indiana University’s Scott Robeson, for a large set of U.S. cities, he found that in places that have warmed, most exhibited less temperature variability, not more. Regrettably, these and other key papers were not part of Patz’ review.

Patz continues by talking about impacts that urban “heat islands”—the heat trapping effects of buildings and paved surfaces combined with less vegetation—that result in most large cities being significantly warmer than the surrounding countryside. He is correct. In fact, the urbanization effect exceeds the background rate of global warming significantly, in some cases by an order of magnitude or more. So if this is such a problem, we should expect people living in cities to be dying from heat exposure with reckless abandon.

The graph below shows the aggregate heat-related death rate toll for 28 of the largest U.S. cities from 1964–1998. There is a statistically significant decline in heat-related over the period. During the same time, temperature and humidity conditions increased by an average of almost 1°C, partly and probably mostly, because of heat island effects. Why aren’t more people, instead of less people, dying from heat exposure, as postulated by Patz?

U.S. heat-related mortality

Figure 1. Annual population-adjusted heat-related mortality averaged across 28 major U.S. cities. Each bar represents a different decade, beginning in the mid-1960s and ending in the late 1980s. Heat-related mortality has been steadily declining (adapted from Davis et al., 2003).

It’s simple. People, by and large, are not stupid. If it’s too hot, they go into air conditioning. If it’s too cold, they turn up the heat, go into the sun, put on a jacket, etc. The fact that Phoenix has a thriving population in a valley that is essentially inhospitable to human life speaks volumes for the adaptability of humans to overcome the limitation imposed by Nature. And the last time we checked, most elderly people move to Phoenix or Miami thinking they might prolong their lives by living away from harsh winter weather, not so they could die sooner. But global warming scaremongers depend upon the stupid people hypothesis to generate high mortality figures.

Later in the review, the authors wax poetically about the potential health impacts of El Niño across the globe. Epidemics of malaria and Rift Valley fever. Dengue hemorrhagic fever in Thailand. Hantavirus pulmonary syndrome in the Desert Southwest. Waterborne diseases in Peru. Cholera in Bangladesh. Etc.

One teensy problem…El Niño (or ENSO) is not related to global warming. The author admits this (sort of): “Although it is not clear whether and how ENSO dynamics will change in a warmer world, regions that are currently strongly affected by ENSO…could experience heightened risks if ENSO variability, or the strength of events intensifies.” Sure. An equally likely scenario is that the impact of all of these diseases will be reduced if global warming generates fewer and weaker El Niños. But this probability was not discussed. It is not scientifically rigorous to write a paper about global warming impacts and to spend pages talking about impacts from something that is unrelated to global warming. Somehow the folks who reviewed this paper for Nature magazine missed this minor point.

Finally, the review paper pulls out a three-year old World Health Organization study and suggests that climate changes that have occurred in the last 30 years could have caused 150,000 deaths per year worldwide. Based on a back-of-the-envelope estimates using current global population and mortality rate estimates, we determined that “global warming” is responsible for zero point two percent of all deaths. This is a remarkably small number based upon WHO estimates that are undoubtedly an exaggeration in the first place. Another way to look at this is that during the last 30 years, primarily as a result of technologies developed in a world powered by fossil fuels (the emissions from burning them are the presumed culprit behind the 150,000 annual deaths), average human life expectancy has increased significantly, especially in places such as China and India. This increase has resulted in roughly 1 billion fewer deaths during the past 30 years. While one could quibble about the specifics, it is clear that fossil fuels have been responsible for averting about 20 to 40 times the number of deaths that they purportedly haves caused.

The most important and interesting aspect of the Nature review article is that Patz, whose primary expertise is in vector-borne diseases like malaria, has the least confidence about the global warming-malaria link. His discussions and review of the vector borne disease literature is fairly balanced and contains many of the key caveats. Unfortunately, this balanced tone does not permeate most other aspects of the report.

There is no doubt that climate change will have some impacts, both positive and negative, on global health. One could just as easily write a review about how a warming planet is producing myriad health benefits. It would not be published in Nature magazine, however, and rightly so, because it is not represent a fair, accurate, and thorough overview of the scientific literature. After reading the Patz review, it’s clear that this standard of objectivity is selectively applied.


Davis, R.E., et al., 2003. Changing heat-related mortality in the United States. Environmental Health Perspectives, 14, 1712-1718.

Patz, J., et al., 2005. Impact of regional climate change on human health. Nature, 438, 310-317.

Robeson, S., 2002. Relationships between mean and standard deviation of air temperature: implication for global warming. Climate Research, 22, 205-213.

No Comments

No comments yet.

RSS feed for comments on this post. TrackBack URI

Sorry, the comment form is closed at this time.

Powered by WordPress