Breast Implants and Connective Tissue Disease

Charles Hennekens, I-Min Lee, Nancy R. Cook, Patricia Hebert, Elizabeth Karlson, Fran LaMotte, JoAnn Manson, Julie Buring; JAMA.1996;275:616-621

These researchers reported a relative risk of 1.24 (95% C.I. 1.08-1.41) for women with breast implants and any connective tissue disease. The researchers claim that the major contribution of this study is to exclude large risks of connective tissue diseases following breast implants. In the press coverage of the study, the researchers stated that the risk was so small that the 24 percent increase in risk amounts to only one extra annual case for every 3,000 women.

Let's see, approximately 1 million women have had breast implants, so 1 million divided by 3,000 is about... only 333 new cases of connective tissue disease each year. If each gets a verdict of $1 million... well, you do the math. And this is on an annual basis! So much for only one extra annual case for every 3,000 women!

Of course, I will be kind and not harp on the fact that relative risk HAS NOTHING TO DO WITH "RISK!" It is a only statistical evaluation of what happened in a particular population. It is a measure of the strength of association between an exposure and a health outcome. It is not a measure of risk. IT IS 100 PERCENT UNFOUNDED TO CALCULATE POPULATION RISK FROM RELATIVE RISK. Sorry for harping... now back to our regularly scheduled program...

How much confidence can we have about these results? After all, a relative risk of 1.24 is certainly weak enough that it warrants at least some scrutiny. Particularly given that the relative risk is barely statistically significant.

Self-reported cases. Misclassification may play a role given that the results are based on self-reported (and maybe even self-diagnosed) cases of connective tissue disease. For example, it has been reported that as much as 30 percent of self-diagnosed rheumatoid arthritis is erroneous.


No distinction of type of implant. The researchers did not ascertain the type of implant received (i.e., silicone gel-filled, saline-filled, etc.). Even if the crude association is true, it may have nothing to do with silicone implants.


Failure to confirm cases. My biggest gripe is that there was no validation of the self-reported cases of connective tissue disease. Imagine publishing a major study without even verifying that the individuals who reported health effects actually had the reported health effects. Hennekens et al. are saving validation for the "next phase of our study." What a gravy train! How do I get on?

This study probably does convincingly show that there is no large risk from silicone breast implants. Hennekens et al. got that much right. But they should have waited until the cases were verified before anything was published about a barely significant 1.24 relative risk based on unverified self-reported data.

I'm sure the plaintiff lawyers are cheering this study, but I give it one long loud

BOOOOOOOOOOO...

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.