Drinking Water and Birth Defects

Andrew T.L. Chen, John A. Reidy, and L owell E. Sever Am J Epidemiol 1996;143:1179-1180

A 1995 study authored by Frank J. Bove et al. tried to link public drinking water contaminated with trihalomethanes and carb on tetrachloride with various birth defects. ( Am J Epidemiol 1995;141:850-862). The results of Bove's study were actually cited in Congressional testimony by EPA Administrator Carol Browner as evidence that hazardous waste sites cause health effec ts.

Of course what Administrator Browner didn't realize (or did she?) was that Bove had decided not to perform any tests to see whether his results were statistically significant -- for all he knew, his results occ urred totally by chance. Bove decided that the estimated risks were real as long as he was able to compute a relative risk greater than 1.0 that was based on at least two cases!

I would say that jettisoning statistical significance is MOST JUNK SCIENTIFIC of Bove et al., but his cockamamie standard for deciding when risks are real easily wins that competition. This study is so bad that it has earned Bove et al. retroactive placement in the Junk Science Hall of Shame!

In addition to not testing for statistical significance, Bove also couldn't develop any biologic plausibility (i.e., no biological basis, theory or support) for his results. Now come other (government-funded) researchers who have actually tried developed a theory for Bove's results -- results which are probably meaningless for more reasons than just statistical significance (it was an ecologic-type study after all).

Isn't trying to f igure out biological plausibility for Bove's study kind of like trying to figure out the aerodynamics of UFOs or the physiology of ghosts?

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.