October 11, 2005

Fear can be useful...

A doctor argues in the LAT that getting people frightened over the threat of Avian Flu is a bad idea. [Thanks to Betsy Newmark] I think he doesn't have a grasp of how to match danger against probability...and of course that's always a tricky problem, because there are many low-probability high-danger scenarios in life, most of which will not happen. Dr Siegal writes:

...Fear is a warning system intended to alert us to impending danger. The bird flu, though a potential large-scale danger, is not impending.

This is very misleading. It's like taking the high likelihood that a big quake will hit San Francisco in the next 50 years and saying: "an earthquake is not impending." It's not something we can say is or is not "impending" (at our present state of knowledge)--it could happen tomorrow or not at all

The facts are these: The current H5N1 avian influenza virus has not mutated into a form that can easily infect humans, and the 60 people in the world who have died of this bird flu have done so not because this bug is on the road to mutation but because millions of birds throughout Asia have been infected, and the more birds that have it, the more likely that an occasional human bird handler will be infected.

The actual mutation feared is something that could take less than a minute. There isn't any such thing as "being on the road to mutation." This is similar to the mistake of flipping a coin and getting "heads" ten times in a row, and then thinking your next flip just has to be "tails." Each flip of the coin has the same probability, no matter what happened in the past.

Most human influenzas begin as bird flus, but many bird flus never change to a form that can harm us. Though flu pandemics occur on the average of three times per century, and we are clearly overdue (the last was in 1968), there is absolutely no indication that the transformation to mass human killer is about to happen. The threat is theoretical. Unfortunately, the attention it has received makes it feel like something terrible is inevitable.

EVERY threat that hasn't yet happened is "theoretical." It's silly to imply that this makes the threat less. And again, the mutation isn't something that "gives an indication that it is about to happen."

Why the overreaction? For one thing, direct comparisons to the Spanish flu of 1918, a scourge that killed more than 50 million people worldwide, has alarmed the public unnecessarily. In fact, there are many scenarios in which the current bird flu won't mutate into a form as deadly as the 1918 virus.

Of course there are. But the number of scenarios doesn't change or help us gauge the probability of any one scenario.

And even if we accept the Spanish flu scenario, health conditions in 1918 were far worse in most of the world than they are now. Many people lived in squalor; 17 million influenza deaths occurred in India, versus about half a million deaths in the U.S. There were no flu vaccinations, no antiviral drugs, and containment by isolating infected individuals wasn't effective, largely because of poor information and poor compliance.

Many people still live in squalor, and population density is now much greater. Flu vaccine won't be available for at least 6 months after the new virus appears, and there are questions about the efficacy of anti-viral drugs. Containment is unlikely to work, because Flu is more infectious than almost any disease.

Today's media reach could be a useful tool to aid compliance. Of course, the concern that air travel can spread viral infections faster may be valid, but infected migratory birds were sufficient in 1918.

Siegal has an anti-fear ax to grind, (he's pushing his book on the subject) and here he's talking nonsense. There is NO QUESTION that air travel can spread flu faster, and the 1918 flu was a Human Flu (of Avian origin) and spread from human to human.

Unfortunately, public health alarms are sounded too often and too soon. SARS was broadcast as a new global killer to which we had zero immunity, and yet it petered out long before it killed a single person in the United States.

That SARS petered out has no relevance to the POTENTIAL danger it posed. A Flu might fizzle out, or mutate into a less dangerous form. Or might not. That doesn't change its POTENTIAL danger, which is what we need to deal with. Siegal is basicaly saying that, because we didn't have an earthquake this year, the likelihood of an earthquake has decreased. That's idiocy. And the fear of SARS galvanized the efforts to contain it, and possible saved us from disaster.

SARS was something to be taken seriously, but the real lessons of SARS, smallpox, West Nile virus, anthrax and mad cow disease weren't learned by our leaders — that potential health threats are more effectively examined in the laboratory than at a news conference.

Bullshit. The political impetus that puts money in the laboratory stems from the scary news conference. And in many flu scenarios the laboratory work is irrelevant, and the preparations that ordinary people should make are vital. And will be stimulated by fear.

With bird flu, scientists have been working on the structure of the viruses in an attempt to protect us. Studies published in the journals Nature and Science over the last six years have given scientists a road map with which to track the current bird flu and alert health officials if it mutates further.

Again, a mutation could change things instantly.

It is reasonable to try to control the bird flu while it remains in the bird population. There is great value in improving our emergency health response system and upgrading our vaccine-making capacity. Government subsidies in these areas could make the public safer.

And they may not. But government has been very slow to prepare, and now that fear is growing, it is starting to move. Fear is USEFUL. It would be better if we all acted like wise philosophers, but we don't. And anyway, here we have a supposedly wise professor, who seems to know nothing of the rules of probability.

But, right now, there is no value in scaring the public with Hitchcockian bird flu scenarios. The public must be kept in the loop, but potential threats should be put into context. The worst case is not the only case.

What "context?" We don't really know. There are many many scenarios where I play golf in the rain, and have a good time. And one scenario where I get struck by lightning. So, not to worry, right? The worst case is not the only case, right?

Posted by John Weidner at October 11, 2005 9:09 AM
Weblog by John Weidner