The XX Factor

Breast Implants, Cancer, and the Problem With “Significant” Results

Just a few years after debunking the myths about links between breast implants and connective tissue disease, last week the FDA announced more potential problems with implants: Women with saline and silicone gel-filled implants may have an increased risk of anaplastic large cell lymphoma (ALCL). ALCL, a rare malignant tumor, can occur in different areas of the body, including the breast, lymph nodes, and skin. You’ve never heard of it because it’s so rare: In the United States, it’s diagnosed in only one in 500,000 women annually. It affects the breasts even less often: in women without implants, it’s diagnosed in just 3 out of 100 million women annually. The reason for the FDA’s report? About 60 cases have been identified among the 5 million-10 million women worldwide who have breast implants.

So what’s the big fuss? Why did Well, the New York Times’ health blog, give its story the scary and overblown title “Breast Implants Linked to Rare Cancer” ? Why are these truly tiny numbers making headlines that are probably freaking out women all over the world?  Part of the problem is poorly translated scientific jargon. In one of the FDA reports, the risk of ALCL is described as “very small but significant.” In science, “statistical significance” refers to something that is unlikely to have occurred by chance. According to the report, the appearance of ALCL in women with implants “may not be coincidence.”  In other words, ALCL may occur more frequently among women with implants than among those without implants. But here’s the problem: While the numbers may be different-say, one in 500,000 vs. six in 500,000, to take a crude example-in either case the risk is vanishingly small. (Compare that with the one in 400,000 annual risk of dying in an airplane crash.)

The dictionary definition of “significant” implies something quite different-something important, something big, something that can make a noticeable difference. I understand that the FDA is publicizing this possible association so as to encourage women and their doctors to report ALCL cases to a new registry. That is a good thing. But it’s not good that the word “significant” is unnecessarily stressing out many of the millions of women with breast implants. FDA scientist Dr. Binita Ashar commented, in one news report, on a “very small but increased risk.” That sounds better than “a small but significant risk.” Health and science reporters, and especially the people who write the FDA press releases, should think twice before using the s-word.