Estimating Risks

The April 24, 2010 issue of Science News has an interesting article entitled "Gambling on Experience" which reports on research into how we (people) guestimate risk in real life. I call it interesting both because it is an interesting social question, but also because of a recent dialogue about the topic of cell phones and brain cancer risk reported in a popular magazine. I’ll end up rehashing a little of that conversation, but the interesting part is rethinking the discussion in light of the article. Go ahead, have a look at the article, I won’t even be offended if you don’t come back to read the rest of my comments….

The main point of the article is that our experience significantly affects how we estimate risk even in the face of explicit warnings. Think the Challenger accident. Morton-Thiokol repeatedly warned about launching in cold weather due to the O-rings, but past experience said the O-rings would hold, so NASA launched with deadly results. That’s an institutional example, but….

The researchers also looked at compliance to the FDA warning (October 2007) warning that went out not to use certain adult cough suppressants for small children because of a small but significant risk of death. In the face of limited information (mostly advertisments) about the warning, parents with children over the age of 2 compliance was only about 15%, apparently due to experience like that of Morton-Thiokol: well, it never happened before. As a parent of two, I can relate to that hand-waving argument for safety.

They make similar comparisons to other risk assessments, where sometimes people underestimate the risk (like above) and sometimes overestimate the risk. In some cases (the study looked at precautionary actions by students in Isreal during 2-year period of September 30, 2000 through August 31, 2002 when there were terrorist bombings on 71 days), the extra caution makes survival sense. Being extra careful and overestimating the risk really is the safe thing to do.

On the other side, expertise can lead to underestimating the risk. Writing on the issue of the debunked link between the MMR vacine and autism,  "M.D.s tend to underestimate the possibility of patients developing real but infrequent vaccination side effects and are befuddled by parent’s unfounded autism concerns." Experienced doctors may have never seen any vaccination side-effect, even ones that are well-established (but rare). As a result, they may underestimate the true risk.

Again, I can relate to that, having been a "radiation worker" for about 6 years during graduate school. We were trained in radiation safety and the goal, of course, is to make sure that accidents truly are rare (as in so rare the just don’t happen). But sometimes a little knowledge isn’t so good. I only once ever had my dosimeter show anything and that was only at the lowest level measurable. But one one or two occasions I took a shortcut through an experimental hall while the accelerator beam was one. I knew that the intrinsic level of the radiation was low and that I would only be near the site for a few seconds as I passed through. The risk really was low, but how low? I didn’t calculate it or think about it, I relied on experience.

So what about that cell phone and brain cancer connection? The current concensus is no link but a lot of people are still worried. The problem comes back to how we estimate risk and how we assess anecdotal reports. How many of you have ever actually been in contact with a person with brain cancer? Not many (I haven’t). But we’ve all heard the stories and after a while, they build up to sound like a major risk. My issue with the article we were what I consider journalistic license to sell magazines. Let’s put a picture of a pack of cigarettes next to a cell phone. Guilty by association? Or cite 50 year old research on microwaves from an era when no one had a cell phone and the devices in question pumped out tons of power compared to a moder microwave oven much less a cell phone. Oh, and let’s say "The only honest way to think of our cell phones is that they are tiny,
low-power microwave ovens, without walls, that we hold against the
sides of our heads." Uhm, really, the only honest way? So everyone else is a liar?

You get the point. I have no problem with limiting cell phone use. I resisted getting a pager for work until they made me, then I resisted getting a cell phone, and I still take vacations to places without coverage which I consider a feature. But I think that when it comes to estimating technological risk, the internet has been a horrible influence. It takes little effort to find information (a good thing) and filter only the parts you like/want to assemble them into a reasonable sounding argument. The time has come where every citizen has to be an amateur scientist just to understand the world around them.