Wynder Validates Science Without Sense

Ernst Wynder, Am J Epidemiol 1996; 143;747- 749

In my recent book Science Without Sense: The Risky Business of Public Health Research, I wrote that

Finding the right risk to "discover" is the critical first step. If you [the researcher] pick the right risk, its intrinsic characteristics will make most of the risk assessment process a mere formality. Pick the wrong risk, and the only thing at risk is your career.

In this invited commentary, famed epidemiologist Ernst Wynder writes that

...if, in the late 1940s, my idea had been to study the relevance of electromagnetic fields to brain cancer rather than whether cigarette smoking caused lung cancer, my career might well have taken a very different turn.

Case closed.

Actually, Wynder goes on a little more about weak association epidemiology. I'll let him take it away...

...[T]oday we find numerous publications with relative risks of less than 2 that do not discuss the extent to which [causal criteria for epidemiology] fit with their conclusions. Studies reporting on the relation of alcohol to breast cancer, for example, generally do not consider that the global distribution of alcohol consumption does not correlate with incidence of breast cancer. Epidemiologic studies on diesel exhaust exposure and its relation to lung cancer do not report that positive animal studies were based on a major overload of the rat's pulmonary system. Investigations suggesting that cigarette smoking relates to cancer of the cervix do not report that the marked increase in smoking in women is inconsistent with the steep reduction in cervical cancer rates. Usually, inconsistencies between studies, which often exist in reports of weak associations, are not fully presented. In short, as we determine whether a relation is causative, the criteria of judgment including consistency, time trends, dose response, and biologic plausibility are often neglected. We should more diligently consider these criteria. We should not rush to judgment about a causative implication when in fact the word "association" ought to be used. Here again, we recognize that the wish bias tends to lead the investigator to conclude that a reported association is causative or to give that implication.

It almost makes me wonder whether Wynder, in fact, wrote Science Without Sense.

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.