« Advertising markets in everything | Main | Revisiting Irish growth »
How much should we worry about pandemics?
Brett Stephens isn't that worried:
In other words, despite all the processes of globalization that are said to be leading us toward nature's great comeuppance, trend lines indicate we are better equipped than ever to minimize the effects of a pandemic.
Why? Because wealthier people tend to be healthier people, and because wealthier societies have more to invest in medicine and research, and because a higher standard of living tends to correlate with more personal space. Also, because globalization means information sharing across boundaries, and rapid adoption of best practices, and greater transparency.
I say think probabilistically, a concept not prominent in his piece. A one percent chance of one hundred million deaths is, in expected value terms, one million deaths and that is a big deal. Probably the United States is less vulnerable than it was in 1918, but how many people would die in China, India and many other locales? How much disruption to trade, travel, and the world economy would take place? Even in the United States, our public health systems would break down quickly and render many modern medical advances useless (e.g., when would the Tamiflu run out?). Having lots of living space is wonderful, but it pays off only if people stay home from work and that means dealing with massive absenteeism. Not pretty. Better safe than sorry.
Oddly Stephens never mentions that we are living in a raging epidemic now, namely AIDS, which has run for several decades. For all the virtues of retrovirals, the modern world was quite slow in combating or even checking the disease and still many people, including U.S. citizens, engage in very risky behavior. Our collective response was not terribly impressive. Greater wealth does help, but greater wealth also means we should spend more to limit the problem.
Going back to the flu, I was struck by this passage:
In each of the four major pandemics since 1889, a spring wave of relatively mild illness was followed by a second wave, a few months later, of a much more virulent disease. This was true in 1889, 1957, 1968 and in the catastrophic flu outbreak of 1918, which sickened an estimated third of the world's population and killed, conservatively, 50 million people.
I should add that we're not yet "out of the woods" on this wave, since there is a reasonable probability of sustained human-to-human transmission starting in at least one country. And a virus which lives in many people is a virus which can mutate.
The main thing we should do -- invest in public health infrastructure -- is in any case a good idea with many possible payoffs, whether a pandemic comes or not. It is a better investment of money than pursuing the ideal of universal health insurance coverage. I might add that one of the better arguments for universal coverage is simply that it could lead to better monitoring of some public health issues.
Posted by Tyler Cowen on May 5, 2009 at 07:23 AM in History, Medicine | Permalink
Comments
So four of the last flu epidemics and 116 of the last non-epidemics were preceded by a mild spring flu season. In other words, a mild spring flu season predicted 120 of the last four flu epidemics.
Posted by: Asher at May 5, 2009 7:39:58 AM
Keep washing your hands after being in public spaces, touching public surfaces, shaking hands, etc. And stop touching your face!
Also, eat lots of fruits and vegetables.
Posted by: anon at May 5, 2009 7:41:24 AM
Asher, this is a quite new strain, different from most other years. Besides, it happening 4 times out of 120 is still quite something to worry about.
Posted by: Tyler Cowen at May 5, 2009 7:49:28 AM
4 different ways of saying the same thing.
1) How is the diarrhea / wealth interaction not like the flu / wealth interaction as far as outcomes are concerned?
2) Even with the AIDS example, increasing wealth decreases death (see first world vs third world).
3) That a reasonably large fraction of world is likely to become infected by the flu (making it a pandemic) seems likely. But it's potential impact is decreasing over time due to increasing wealth.
4) Even if *everyone* caught it, no one would worry about a global hangnail pandemic. That deaths are likely to be way down due to increasing wealth is a good reason to worry less about pandemics.
Posted by: Jody at May 5, 2009 8:13:07 AM
Ahhh... one more thing...
[Given a large number of deaths of poor people around the world] "How much disruption to trade, travel, and the *world economy* would take place?"
Morals aside and focusing purely on the economics, that they are poorer is indicative that their marginal contribution to the global economy is proportionally smaller so the disruption will be less.
Lacking actual numbers, it seems not unreasonable to me to believe that interruptions to trade among the wealthier to combat the possibility of infecting the poorer (e.g., by encouraging everyone to not ride a plane even though the chance of death in a wealthy society is quite low) could easily have a greater impact on the global economy than the impact from the deaths of the poorer people around the world.
Thus to maximize the growth in the world economy, it may well be preferable for the wealthier societies to worry less instead of worry more.
Posted by: Jody at May 5, 2009 8:27:57 AM
Two points:
Safe is not necessarily better than sorry. The creation of "cry wolf" reactions is a genuinely risky outcome.
Great care is needed not to confuse the cost of reaction and over reaction with unavoidable or close to unavoidable cost.
The latter feeds the former over time in a dangerous fashion.
Posted by: Brent Wheeler at May 5, 2009 8:38:38 AM
Wouldn't massive investment in public health infrastructure make health cheaper anyway, moving us towards the goal of universal coverage? (If the infrastructure's in place, universal coverage seems like it would also help prevent deaths in case of an epidemic - people would get treated without worrying about having to pay.) Not to ignore their relative importances, but it seems like health infrastructure investment and universal coverage are complementary goals.
Posted by: Neal at May 5, 2009 8:46:20 AM
but it seems like health infrastructure investment and universal coverage are complementary goals.
What does "universal coverage" actually mean in a world of infinite desires and finite resources?
Everyone can get a band-aid now (heck, come to my house and I'll give you a band-aid if you need one. I might even throw in some Neosporin.), but I don't foresee any system of "universal coverage" in the near future that will cover cryonically freezing my head (a potentially useful medical treatment).
Sure the weasel word is "essential" health care but that's a very elusive target given infinite desires.
Posted by: Jody at May 5, 2009 8:59:13 AM
In each of the four major pandemics since 1889, a spring wave of relatively mild illness was followed by a second wave, a few months later, of a much more virulent disease.
Well that's odd -- because as I understand it, that's not the way pathogens evolve. When they first jump from other species, they may 'accidentally' kill their new human hosts, but from the pathogen's perspective, this is a bad mistake (because dead people can't support or transmit the pathogens). Even seriously ill people aren't very useful to pathogens, since seriously ill people don't go out and interact with other potential hosts. So the evolution is in the direction of viruses becoming more easily transmissible but less lethal. But I suppose 'virulent' can refer to either transmissibility or severity.
Still, I'm less worried that I was because the fatality rate now seems much lower than originally believed and will drop over time as the virus evolves.
Posted by: Slocum at May 5, 2009 9:25:00 AM
The analogy to AIDS has been going around for a while now and I think it's overstated. AIDS took a long time to spread to the point where it required a major public health response. And even at that point, becoming infected required intimate contact that excluded large portions of the population. It took a decade for AIDS to even reach North America. It took, what, a month for swine flu?
Posted by: Mike at May 5, 2009 9:36:56 AM
Tyler, I'd take Asher's point to be that the "mild spring flu season" is not causally connected to "severe epidemic".
I'd take issue with "greater wealth also means we should spend more to limit the problem." Greater wealth means we CAN spend more, but we can also spend that money teaching deaf-mute New Guinea tribesmen about Shakespeare's sonnets, or providing psychotherapy to disturbed wombats. "Can" doesn't mean "should".
Posted by: Laserlight at May 5, 2009 9:45:31 AM
Tyler, I've seen you mention several times the value of investing in public health infrastructure, but what does this mean exactly (community health clinics?), and do you mean in the U.S. specifically or more in the global context? Just curious.
Posted by: Franknfurter at May 5, 2009 9:53:36 AM
There are several simple factors at work here as well. The industrialized world is much more robust in terms of respiratory health, body mass, nutrition, and hygiene. Heavier people with fewer lung issues (less heavy particulate pollution, less smoking, etc.) simply live through diseases at higher rates.
Posted by: The Other Eric at May 5, 2009 10:03:58 AM
"Tyler, I'd take Asher's point to be that the "mild spring flu season" is not causally connected to "severe epidemic"."
I'd bet "mild spring flu season of a new strain" is causally connected to "severe epidemic months later." But what I'd like to see how often a new strain has a mild spring season that doesn't result in a pandemic. Identification is probably tricky here, since technically each flu virus is slightly different, but I'm sure someone would have the data on H1N1-like viruses that died out quickly in the past century.
Posted by: HH at May 5, 2009 11:24:49 AM
Slocum,
Pathogens can evolve either for increased or decreased virulence depending on the method of transmission. The classic case is cholera, which has evolved to extreme virulence in third-world settings because it can most easily infect caregivers that way. In the USA, it has evolved to such low virulance that it was endemic to the Gulf Coast areas of TX and LA for decades before health officials noticed it.
http://jama.ama-assn.org/cgi/content/abstract/247/11/1598
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=202726
In short, pathogen evolution supports the contention that wealth=health.
Posted by: Bob Knaus at May 5, 2009 11:30:01 AM
GREAT POST! Should be required readings!!
COMMON CENTS
http://www.commoncts.blogspot.com
Posted by: Steve at May 5, 2009 11:39:39 AM
"Better safe than sorry" is such a lame slogan! True, one might end up sorry for not having taken effective precautions, but how about being sorry for having incurred the cost of useless precautions?
Yglesias's article simply made the point that public health authorities may have acted properly even if the swine flu turns out to be less virulent than they originally expected, since they had to act initially on quite scanty information. But he just assumes, without evidence, that their actions were in fact appropriate, given their initial information.
Posted by: Philo at May 5, 2009 1:18:00 PM
It seems very plausible that this swine flu will percolate around the globe in hiding only to come back strong in the fall. I think the northern hemisphere was lucky that it arrived just as we were coming out of the flu season but that does not mean the southern hemisphere is lucky. The first few cases have started to show up south of Mexico. I think what people miss is that the upcoming flu pandemic we have always talked about was not a big threat because it would be a deadly strain that would kill everyone it infected, it was a big threat because if any flu virus shows up that is novel enough such that virtually nobody has antibodies to it, its spread will be so widespread and in large regions all at the same time such that it will shut the system down or at least slow it down considerably. The threat is not to me, or you, or anyone specific, the threat is to the globe as a whole.
Posted by: CavemanForecaster at May 5, 2009 5:04:04 PM
isn't one major factor missing in your analogy, namely that you can spread the flu by coughing or sneezing near someone, whereas AIDS involves behavior that by-and-large (outside of Africa/lack of blood transfusion testing elsewhere) is voluntary. After the initial realizations about the methods of accidental transmission (eg, blood transfusions) the risk of its spread was largely a matter of education of the public and an exercise in self-control and self-maintenance (via regular testing, full disclosure to partners, etc).
The comparison of the two as diseases of similar "pandemic" potential seems pretty flawed.
Posted by: Neal at May 5, 2009 8:26:29 PM
"A one percent chance of one hundred million deaths is, in expected value terms, one million deaths and that is a big deal."
The "Wow that's a big number" fallacy? Really?!
Disappointing.
Posted by: Vernunft at May 5, 2009 8:30:46 PM
Wouldn't universal health care spending by default improve public health infrastructure (even accounting for waste and rent seeking)?
Posted by: Mike at May 5, 2009 9:46:25 PM
If the main fear is a virus "which can mutate" to become worse, and right now the virus is mild, shouldn't you prefer to catch it sooner, so you have some immunity if/when its worse cousins arrive?
Perhaps this helps explain why global travel and urban density have meant fewer deadly pandemics, rather than more, sine 1918. It's now easier than ever for mild viruses to travel widely, but thanks to wealth/nutrition/healthcare/staying-home-while-sick, still harder for nasty viruses. The selective gradient nudging viruses toward benign evolution is steeper, while 'contagious innoculation' by naturally-occurring mild agents spreads faster.
Posted by: Gordon Mohr at May 6, 2009 12:20:50 AM
"Better safe than sorry" is such a lame slogan! True, one might end up sorry for not having taken effective precautions, but how about being sorry for having incurred the cost of useless precautions?
This is just a framing issue and it's a failure of public education and public relations if the point doesn't get through to enough people. When you renew your car insurance every six months, after having spent the term of the insurance accident-free, do you reflect on how purchasing the insurance was a "useless precaution"?
What counts as a useless precaution is pretty relative, also. In many countries, washing your hands after using the toilet or waiting for traffic to clear up before crossing the street are considered by locals to be useless precautions. In the U.S. and other countries, public health and safety officials have changed attitudes enough to get people to rid their minds of this way of thinking.
Posted by: Ricardo at May 6, 2009 1:51:33 AM
@ricardo: no they haven't 70% still dont wash their hands and thousands die jaywalking every year
Posted by: hESDfeh at May 6, 2009 5:34:23 PM
Some recent research has been conducted on the economic effects of shocks such as stock market crashes and flue on consumption and GDP growth:
http://www.hoover.org/pubaffairs/dailyreport/archive/44358832.html
Now that we've two of those in the mix together, it's certainly some cocktail!
Posted by: Ronan L at May 8, 2009 1:51:37 PM
