As Many as 25% of Cancers Detected by Mammograms May Be Overdiagnosed, Study Finds
By Brenda Goodman, MA
WebMD Health News
Latest Cancer News
Reviewed by Laura J. Martin, MD
April 2, 2012 -- A new study suggests that routine mammograms, long pitched to women as lifesaving tests, may also be causing substantial harm.
The study estimates that as many as 1 in 4 cancers detected over a decade by routine mammograms are cancers that won't grow or spread, cause symptoms, or lead to death.
Instead, these "overdiagnosed" cancers are treated with surgery, powerful drugs, and radiation, all when the cancer wouldn't have made a woman sick in the first place.
"We are curing people who don't need to be cured," says Otis W. Brawley, MD, chief medical officer of the American Cancer Society.
Brawley reviewed the study for WebMD but was not involved in the research.
He says doctors have known for some time that not all breast cancers are dangerous. But he says it's been difficult to figure out how many breast cancers are being treated when they probably don't need to be.
In part, that's because there aren't any tests that can distinguish between cancers that are harmful and those that may not otherwise affect a person's health. And when doctors find cancer, they treat it, of course. Not doing so would be unethical. So there's never been a population of women who were diagnosed with cancer but then left untreated so doctors could see what would naturally happen to those cancers over time.
"This is one of the best studies ever designed to try to figure that out," Brawley says.
Estimating Overdiagnosis in Breast Cancer
The study included nearly 40,000 women diagnosed with invasive breast cancers in Norway before and after 1996, when the country began offering regular mammograms to residents in some regions.
By comparing the cancers diagnosed in women who were screened vs. those who were not, researchers estimated that 15% to 25% of breast cancers found by routine mammograms were overdiagnosed.
Put another way, researchers say that for every 2,500 women aged 50 to 69 offered mammograms for 10 years, just one additional life would be saved, 20 women would be diagnosed with cancers that needed treatment, and six to 10 women would be overdiagnosed.
"What women have been told before is, 'You look for cancer and we'll save you,'" says researcher Mette Kalager, MD, a breast cancer surgeon and an epidemiologist with the Harvard School of Public Health. "That's not the whole story. You will be saved without screening as well."
"I think we have to inform women about the downside or harm of mammography screening," she says.
The study is published in the Annals of Internal Medicine.
The number of women who are overdiagnosed in the U.S. is likely to be higher, experts say, because women in this country often start getting mammograms in their 40s, rather than in their 50s as women in Norway do, and Americans are generally screened more often, every year instead of every two years.
"If you're starting in the U.S. at a younger age, and you're doing screening more frequently, that means you have more chances to be screened, and every time you're screened you're at risk of overdiagnosis," says Joann G. Elmore, MD, MPH, a professor of medicine at the University of Washington in Seattle.
Elmore wrote an editorial on the study, but she was not involved in the research.
She says previous studies have shown that radiologists in the U.S. have about a 10% recall rate; that is, they call about 10% of women back for more testing because they're worried about something they see on the X-ray. In other countries, radiologists have about a 2% recall rate. That higher index of suspicion also raises the likelihood of overdiagnosis.
Weighing the Benefits and Harms of Mammograms
Overdiagnosis, it should be noted, is not the same thing as another risk of cancer screening tests: false-positive results. A false-positive is a result that looks suspicious and may lead to more testing, but the patient is ultimately cleared of cancer.
"That test is eventually shown not to be cancer, but in the time it takes to figure that out, women are made anxious, many are never totally relieved," says H. Gilbert Welch, MD, MPH, a professor of family and community medicine at Dartmouth Medical School in Hanover, N.H.
Arguably, Welch says, overdiagnosis is even more insidious.
"You're treated for a breast cancer that was never going to bother you. Now you're talking about real intervention. Mastectomy, lumpectomy, surgery, radiation," he says.
Not to mention the mental toll of a cancer diagnosis.
Welch points to the fact that for every 2,500 women screened for a decade with regular mammograms, one life will be saved thanks to early detection.
But, he says, "It's very relevant what happens to the other 2,499 women. In this country, somewhere around 1,000 will have at least one false-positive result and about half will have a biopsy, so that's a lot of angst there."
Based on this study and several others, Welch says it looks like "somewhere between five and 15 will be overdiagnosed to receive unnecessary treatment for cancer."
"We need to work to make the deal better for patients," Welch says. "There might be reasons to screen less often, look less hard, stop screening earlier."
To Screen or Not to Screen?
To try to reduce some of the harms associated with screening, in 2009, the U.S. Preventive Services Task Force changed its recommendations for routine mammograms, saying women should get one every two years starting at age 50, rather than every year beginning at age 40.
Those recommendations, however, are at odds with guidelines from other groups, including the American Cancer Society and the American College of Radiology, who say screening should start at age 40 or even younger for high-risk women.
For women who are weighing the risks and benefits of getting regular mammograms, Welch says he advises patients to remember that cancer screening is a choice made after consultation with your doctor regarding risks and benefits of screening.
Health Solutions From Our Sponsors
©2012 WebMD, LLC. All Rights Reserved.