Weight Gain and Breast Cancer:
Another Nurses Health Study Travesty

Zhiping Huang, Susan E. Hankinson, Graham A. Colditz,
Meir J. Stampfer, David J. Hunter, JoAnn E. Manson,
Charles H. Hennekens, Bernard Rosner, Frank E. Speizer, and
Walter C. Willett
Journal of the American Medical Association 1997;278-1407-1411


Harvard researchers report that avoiding weight gain may be an important factor in reducing breast cancer in post-menopausal women.

But while not gaining too much weight may be a good strategy for staying healthy in general, will it really reduce the risk of breast cancer?

This study certainly can't be considered reliable evidence of that.

In yet another epidemiologic study based on the data tragedy known as the Nurses Health Study, the researchers reported that women who gained more than about 40 pounds since age 18, had a 40 percent increase in breast cancer incidence.

The study isn't convincing for a very basic reason--epidemiologic studies are only good at identifying large risks of rare diseases, like smoking and lung cancer (i.e., an increase in rate of about 2000 percent for a disease that strikes about 1 in 10,000 nonsmokers).

Epidemiologic studies aren't reliable in the case of small risks (i.e., less than 100 percent) of common diseases (like breast cancer, which has as high a rate as 1 in 10 for women in their 80s).

The reason for this is simple. Epidemiologic data are likely to be imprecise. The Nurses Health Study is a great example of this.

The Nurses Health Study is a cohort of more than 95,000 nurses followed since 1976. The nurses filled out an initial health questionnaire in 1976, and follow-up questionnaires have been mailed every two years since.

But the questionnaire responses have not been validated or verified. So who knows how reliable they are? For example, many nurse participants may not be honest about their smoking habits or accurate about their weight gain/loss. A sufficient amount of bad data may give rise to spurious results.

And yet, every two years, we are besieged by a landslide of new studies analyzing the updated Nurses Health Study data set.

It may be a great way for these Harvard researchers to make a living, but it certainly isn't a great way to do science.


Material presented on this home page constitutes opinion of the author.
Copyright © 1997 Steven J. Milloy. All rights reserved. Site developed and hosted by WestLake Solutions, Inc.
 1