Ho-ho-ho!
No, that's not Santa Claus. That's me doing a seasonal "Ha-ha-ha!" after reading a new study in the January 1, 1999 issue of the American Journal of Epidemiology reporting that secondhand smoke is associated with breast cancer.
The authors, Timothy L. Lash and Ann Aschengrau of Boston University, studied female residents of five Massachusetts towns between 1983 and 1986, including 265 breast cancer cases and 765 controls. They report two primary findings:
- 1. Subjects exposed to secondhand smoke had a 100 percent increase in breast cancer risk.
- 2. The earlier a female is exposed to secondhand smoke, the greater her risk of breast cancer.
Laughable claims, indeed.
Putting aside the baseline criticisms, including: (1) small study size (most comparisons involve far fewer than 100 breast cancer cases); (2) a likely biased study sample (334 breast cancer from five Massachusetts towns during the period 1983-1986 were nonrandomly whittled down to 265); (3) lack of verified exposure data; and (4) lack of consideration of confounding factors ("known" risk factors for breast cancer account for only a small portion of total breast cancer cases), the claims don't hold water.
In addition to the 100 percent increase in risk for secondhand smokers, the study reported that active smokers had a 100 percent increase in breast cancer risk. Assuming for the sake of argument that tobacco smoke has a causal role in breast cancer, how can active smokers and secondhand smokers have the same level of risk? Aren't active smokers also exposed to their own secondhand smoke?
Worse, when active smokers were compared to all never-smokers, instead of just never-smokers and never-secondhand smokers, the increase in breast cancer risk for active smokers was a statistically nonsignificant 20 percent. Assuming for the sake of argument the tobacco smoke-breast cancer link was true, how likely would it be that secondhand smokers, but not active smokers, would be at risk?
The novel claim that the younger a female is exposed to secondhand smoke, the greater her breast cancer risk, falls apart for similar reasons. Such a relationship did not hold true for active smokers.
A review of co-author Aschengrau's prior work is revealing. She has previously reported that exposure to magnetic fields, estrogenic chemicals in the workplace and radioactive iodine therapy were associated with breast cancer risk -- even though none of the associations were statistically significant!
My favorite Aschengrau study examined cancer risk and residential proximity to cranberry cultivation in Massachusetts. In that study Aschengrau studied eight types of cancer and reported that living near areas of cranberry cultivation was associated with a 570 percent increase in risk of astrocytoma, a type of brain cancer. Of course, the confidence interval (1.6 to 27.8) is wide enough to drive a Mack truck through!
One can only wonder what Aschengrau, sitting in her Ivory tower at Boston University, will dream up next. I'll bet journal editors all over America are making sure their bedroom curtains are closed tight!
Comments on this posting?
Click here to post a public comment on the Trash Talk Bulletin Board.
Click here to send a private comment to the Junkman.
Copyright © 1998 Steven J. Milloy. All rights reserved on original material. Material copyrighted by others is used either with permission or under a claim of "fair use." Site developed and hosted by WestLake Solutions, Inc.
Material presented on this home page constitutes opinion of Steven J. Milloy.