More than half of older women who get annual mammograms over the course of 10 years can expect to be called back at least once because of a false alarm, while one in 12 will be referred for a surgical biopsy that will turn out to be negative, a study released today shows.
The study, published in the Oct. 18 issue of the Annals of Internal Medicine, looked at the records of nearly 170,000 women from seven regions around the United States who had their first mammogram between the ages of 40 and 59. They found that a woman who starts getting annual screenings at age 50 has a 61 percent chance over the next 10 years of getting a mammogram that turns out to be a false positive — meaning the radiologist says the mammogram is abnormal when there is no cancer present.
However, when the researchers analyzed the records of roughly 4,500 women who actually had breast cancer, they found that those screened every two years were slightly more likely to be diagnosed with late-stage cancer than those screened every year, suggesting a benefit of annual screenings. But the increase was so small that it was not statistically significant, meaning the increase could be due to chance.
According to Karla Kerlikowske, M.D., a breast cancer researcher at University of California in San Francisco and one of the study's authors, other research into the rates of false positives were based on mathematical models, which can never be completely precise. But in this study, she says, "We looked at results from real women who have had four or five or six mammograms, so we know the exact false positive rate. We also know the exact rate of false positive results for biopsy, which is fairly high."
The researchers found that women in their 40s and 50s had similar risks of a false positive during 10 years of screening, but over the course of a lifetime women who started regular screening mammograms at age 40 rather than 50 were likely to have more recalls for suspicious results because of the extra decade of testing. Women who had an earlier mammogram available for comparison halved the probability of being recalled for further testing.