Wednesday, December 5, 2007

Science Behind the Seasonal Flu: Why Winter?

In October of 2007, Anice Lowen, Samira Mubareka, John Steel, and Peter Palese published a paper reporting that the reason we suffer influenza in the winter is “low relative humidity produced by indoor heating and cold temperatures … [which] favor influenza virus spread.”

Is it just me?  Or does that seem self-evident, obvious even?  I’m quite sure my grandmother told me that!

But it turns out that understanding how the flu is passed from host to host and why outbreaks happen in the winter has remained a mystery for many years.  This October I heard Dr. Marc Lipsitch, professor of Epidemiology from Harvard School of Public Health, talk about the great epidemic of 1918 and how many questions we still have about the virus.  Scientists and science writers love to talk about what we know rather than the unknown, the answers rather than the questions.  Although the great questions are obvious to the scientific community, they are often implicit and not part of the public dialogue.  And that’s why, when we learn something we thought we always knew, it can come as a surprise.

Isn’t that amazing:  that as much as we DO know about health and disease, there are still such basic things to learn about everyday phenomena like how the flu is spread? And we can be certain that, at some point in the not-too-far-distant future we’ll take this too for granted.  This little bit of scientific knowledge will be completely integrated into what we teach and learn about human health, biology and medicine as if we had always known it. 

But that’s a pity. Although it’s unlikely that anyone then will remember such a classic study, it’s worth trying.  Even then it will be important to remember:  that this knowledge was not always known; that this new theory was one of a variety of controversial, plausible explanations;  and that someone who cared needed to ask very specific questions, conduct experiments, and finally to collect evidence which supported or contradicted these alternative theories.  And we should remember that it all began with an idea in the imagination of the investigator and ended when a community of autonomous and highly independent thinkers have reproduced his results and challenged every aspect of his procedures and logic.

So why do we get the flu in the winter?  An article in the New York Times provided some interesting context for their scientific paper:
As long as flu has been recognized, people have asked, Why winter? The very name, “influenza,” is an Italian word that some historians proposed, originated in the mid-18th century as influenza di freddo or “influence of the cold.”
There was no shortage of hypotheses. Some said flu came in winter because people are indoors; and children are in school, crowded together, getting the flu and passing it on to their families.  Others proposed a diminished immune response because people make less vitamin D or melatonin when days are shorter. Others pointed to the direction of air currents in the upper atmosphere.
Although scientists have been speculating about this for centuries, it has been surprisingly difficult to study: first because it was unethical for scientists to expose people to the virus under controlled conditions; and second, until now, scientists did not have laboratory animals they could use to study the transmission of the human flu virus.

As is often the case, this science began with an accident and an observant, curious mind.  Dr. Palese, one of the investigators in the study, was reading a paper published in 1919, soon after the influenza epidemic.  From the article he learned that, in addition to 20 million human beings, the laboratory guinea pigs at Camp Cody in New Mexico apparently also succumbed to the virus.  Could these animals help him discover how the flu virus was actually transmitted? 

Is it possible that nobody else who read this article actually asked this question?  Apparently, yes.

In 2006, 87 years after the article was published, Dr. Palese acquired several of the guinea pigs and discovered that unlike laboratory mice, these animals can infect one another much the way people do.  Equipped with an animal model, he enlisted the help of three colleagues and began testing their hypothesis “that ambient air temperature and RH [relative humidity] impact the efficiency with which influenza virus is spread.” [1]

They constructed an apparatus to keep infected animals in isolated chambers under controlled conditions. Air was forced to flow through chambers with the infected animals into chambers with healthy animals, allowing only the temperature and humidity to vary. 

Gina Kolata reported in the Times:
They discovered that transmission was excellent at 41 degrees. It declined as the temperature rose until, by 86 degrees, the virus was not transmitted at all.  The virus was transmitted best at a low humidity, 20 percent, and not transmitted at all when the humidity reached 80 percent.
From this, they confirmed that:
Flu viruses spread through the air, unlike cold viruses, Dr. Palese said, which primarily spread by direct contact when people touch surfaces that had been touched by someone with a cold or shake hands with someone who is infected, for example.
But they also demonstrated that:
Flu viruses are more stable in cold air, and low humidity also helps the virus particles remain in the air. That is because the viruses float in the air in little respiratory droplets, Dr. Palese said. When the air is humid, those droplets pick up water, grow larger and fall to the ground.
The coverage in the is excellent with regard to what they learned about the flu virus, how hard it has to been to work on transmission specifically in the past, and how Dr. Palese discovered the animal model in the paper from 1919. 

The technical paper published in PloS Pathogens, on the other hand omitted this rich context, but nevertheless offered other interesting insights to the real work of the scientist which might be rewarding to the persistent, lay reader. In it I discovered, for example, that they had to rinse virus particles from the noses of the healthy guinea pigs to determine when and to what extent their inoculated neighbors infected them.  I also learned that they observed conventions that regulate how these animals are treated and that they were anesthetized before they were tested for infection. I could see that there were precautions – including the use of a ‘sentinel animal’ – that demonstrated that the researchers handling the animals were not inadvertently spreading the virus themselves, invalidating their careful experiment. And finally, underlying this simple experiment, the collaborative, rich web connecting many laboratories and technology providers is also quite apparent, all predicated on openness and trust.

This is all part of that texture, the tangible world of the researcher, which is transparent and obvious to other researchers but quite opaque to the lay reader and difficult to imagine.  How often do we refer to an ‘animal model’ without appreciating what it really is, how fortunate we are that we can study human disease in animals and how we can extend what we learn about these animals to ourselves?

And with regard to the really broad question I posed at the beginning of this essay:  what can this study teach us about the scientific community as a whole?  First, when scientists ask ‘why’ the flu is transmitted more readily in winter they really mean ‘how’. Although science can tell us a lot about how things work, it’s unable to help us understand what things mean.  I don’t want to suggest that these larger questions are not important to scientists – of course they are.  In fact, the human dimensions of health and illness, life and death motivate lots of scientists and doctors.  But you won’t find that in their own writing:  it is assumed.

And second, also rarely discussed, is the role of curiosity and relentless inquiry as the engine that drives science, not glory or riches (although they help too in some cases).  How many people would have read that paper and thought nothing of the guinea pigs death from human influenza in the great epidemic of 1918?   Lots of them, apparently!  Gina Kolata of the Times was right to include this crucial part of the story.  But she neglected to consider how Dr. Peter Palese might have been different, how he was able to see something new and ask an original question. 

Let's make the implicit, explicit, enumerating a list of what Dr. Palese had to believe, first of all, and then what he had to DO to make this science thing work: 

  1. He had to assume the mechanism of transmission was knowable and observable.
  2. He was optimistic about our ability as humans and his own capacity as a scientist to imagine new and untested ideas, to devise strategies to test their validity in the lab and then to use reason, extending their application to the real world.
  3. He had a precise question  -- he knew what he wanted to know.
  4. He was on the lookout for an animal model he could use to test the transmission of human flu.
  5. He was willing and able to do the test.
  6. He was committed to sharing his results with a community of peers, allowing them to decide if and to what extent they contribute to our understanding of the natural world.
Will we remember all this when we teach how the flu virus is transmitted ten years from now?  Is it enough to remember what we know about the flu?  Or is it just as important to recall how we know what we know, including some of the less obvious assumptions on this list?

If we succeed in communicating how we know what we know clearly embedded in specific examples of scientific research -- something real and true and human about the culture and practice of science – perhaps then we will become wiser, more mature as we puzzle over the limits of scientific knowledge, uncertainty, risk and controversies about global warming or stem cell research or mercury in tuna fish or the demise of the polar bear, for example. 

How can simple stories like this one help us answer another, larger question:  What is science, anyway?


[1]  Influenza Virus Transmission Is Dependent on Relative Humidity and Temperature by Lowen AC, Mubareka S, Steel J, Palese P PLoS Pathogens Vol. 3, No. 10

[2] Study Shows Why the Flu Likes Winter by Gina Kolata, New York Times, 12/5/2007