Finnish Indoor Radon Study: Right Answer, Wrong Reason

Anssi Auvinen, Hona Makelainen, Matti Hakama, Olli Castren, and Eero Pukklala, Heikki Reisbacka, Tapio Rytomaa J Natl Cancer Inst 1996;88:966-972

Finnish researchers investigating the effect of indoor radon exposure on the risk of lung cancer have reported that indoor radon exposure did not increase lung cancer risk. The key reported result was a statistically insignificant increase in risk on the order of 1 percent (i.e., relative risk = 1.01, 95 percent confidence interval (0.94-1.08).

Many have been saying for years that no science adequately supports the proposition that indoor radon causes lung cancer. Does this study help?

The answer is a resounding "NO!" Why?

Because, practically speaking, case-control epidemiology is inherently incapable of identifying whether OR NOT small risks exist. And radon is certainly a small risk, if it is a risk at all. Using case-control epidemiology to identify and quantify small risks is like trying to see an atom with the naked eye.

What is case-control epidemiology? As I explained in Science Without Sense: The Risky Business of Public Health Research, case-control epidemiology is basically a quick-n-dirty epidemiologic technique involving a look back in time at a population and trying to figure out its past exposure.

In a nutshell, it's like recreating history -- a technique that may be great for politicians, but makes for lousy science.

Recreating exposure history is not a very exact science. Think about it. How does an epidemiologist "know" with sufficient precision how much radon someone was exposed to 30 years ago? How does an epidemiologist confidently rule out exposure to other lung cancer risk factors? It's simply impossible.

Case-control epidemiology, particularly when looking at environmental risk factors, is generally not exact enough to detect small risks.

Finally, there is the problem of proving a negative. Let's say that there was no radon-induced lung cancer in this particular population. That certainly doesn't preclude a future study of a different population from finding such a risk.

For those of you who are willing to accept this result, be careful. You may be signing on to a methodology (i.e., case-control epidemiology) that will come back to haunt us all.

To the Finnish epidemiologists: Nice try, but no cigar!

Material presented on this home page constitutes opinion of the author.

Copyright © 1996 Steven J. Milloy. All rights reserved. Site developed and hosted by WestLake Solutions, Inc.