Environment and Disease: Humidity May Be Key to Flu Virus Survival

Jeff Shaman combines physics and biology in his analysis of the factors that explain flu outbreaks. (Photo: Karl Maasdam)

October 28, 2009

More water vapor means less flu virus transmission

By Nick Houtman

In Brief

Understanding why flu season begins and ends could help public health officials in their efforts to manage outbreaks. Jeff Shaman in the College of Oceanic and Atmospheric Sciences has demonstrated that absolute humidity is a good predictor of flu virus survival and transmission.

Winter means coughs, colds and flu as surely as it does longer nights and shorter days. The seasonal cycle brings a regular round of flu vaccinations and public health warnings. Now, research by Oregon State University atmospheric scientist Jeff Shaman suggests another way to attack influenza: turn up the humidity.

His results come as scientists continue to debate exactly how the flu virus makes its way from one person to another (think kissing, sneezing, shaking hands or just plain breathing). And while Shaman and colleagues continue to investigate the relationship between the virus and the amount of water vapor in the air, he urges caution in using humidity as a cure-all. He points out that plenty of other microorganisms, some dangerous to human health, do quite well in high humidity.

Shaman's analysis of flu transmission and humidity hit the headlines (The New York Times, The Oregonian, Science magazine) last spring after he co-authored a report in the Proceedings of the National Academy of Sciences. He and Melvin Kohn, an epidemiologist with the Oregon Department of Health Services, had re-analyzed data from a published study of flu transmission in guinea pigs.

Trigger Point

The original study by researchers at Mt. Sinai School of Medicine in New York looked at the effects of temperature and relative humidity on transmission of influenza using virus-infected guinea pigs in climate-controlled chambers. The researchers used 20 different combinations of temperature and relative humidity in an effort to identify a trigger point for changes in transmission of the virus between infected guinea pigs and adjacent control animals.

In general, the study found that there were more infections when it was colder and drier. However, Shaman and Kohn demonstrated that relative humidity could only explain about 12 percent of the variability of influenza virus transmission from these data.

In addition, numerous other experiments, dating back to the 1940s, have shown that low relative humidity favors increased influenza virus survival. In their PNAS analysis, Shaman and Kohn demonstrated that relative humidity explains only about 36 percent of influenza virus survival. The Oregon researchers then retested the data using absolute humidity and found a dramatic rise in accounting for both transmission (50 percent, up from 12 percent) and survival (90 percent, up from 36 percent).

Encased in Water

The difference between relative and absolute humidity lies at the heart of the issue. Each measures airborne water vapor in a different way. Relative humidity compares the amount of moisture to a maximum level (or saturation) at a given temperature. Absolute humidity is the amount of water vapor amount of water vapor, regardless of temperature. The difference is crucial. The hotter the air, the more water vapor is needed to reach saturation (i.e. 100 percent relative humidity). Absolute humidity is thus a more direct indicator of the amount of water vapor present.

Shaman has presented his results to colleagues at Harvard and Stanford and is scheduled to give a talk to the American Meteorological Society in January. A question that he is often asked focuses on how humidity affects the virus. "Why would a virus, encased inside a droplet of water and glycoproteins and other junk, care about the humidity levels in the ambient air?" he says.

"I don't have the answer. You could understand maybe temperature or something that directly controls rates of evaporation. The PNAS paper didn't find any evidence of that. Maybe humidity is a proxy for a combination of other variables."

Taking Action

Nevertheless, Shaman's results have practical potential. If low humidity favors virus survival and transmission, perhaps weather patterns could signal the likelihood of a flu outbreak. Or humidity adjustments in buildings (theaters, hospitals, nursing homes) could reduce the probability of contributing to one. What's needed before such measures are taken, he says, are laboratory studies to understand exactly how the virus spreads from one person to another.

Shaman is continuing to work on the relationship between flu outbreaks and weather patterns. His expertise is in the physics of large-scale atmospheric circulation, and his papers focus on phenomena such as El Nino in the Pacific and trends in the Atlantic and West African monsoons. However, he also specializes in how atmospheric patterns affect diseases such as malaria and encephalitis. So he finds himself working with researchers in other disciplines: biology, engineering, virology.

It's in this interdisciplinary collaboration where he thinks interesting scientific advances can be made. "We can bring different perspectives to a problem, literally address multiple aspects of the same issue," he says. "Sometimes the ability to move between scientific languages and work with people with different expertise is helpful."

Download article as PDF