Last week, a Danish study published in the Annals of Internal Medicine attracted a lot of attention for claiming that one in three tumors detected during mammograms are harmless, meaning that women are often pressed to needlessly undergo potentially harmful cancer treatments. The study reinforces several other papers supporting the possibility that cancer overdiagnosis is widespread — and that a good portion of mammograms might end up doing more harm than good.
Radiologists and cancer research groups were quick to push back, arguing that the Danish study greatly overestimated false diagnoses and noting that mammograms can catch cancer at its early stages and save lives.
Welcome to one of the most frustrating aspects of health care. There’s so much at stake, and yet so much emotional disagreement. Still, it’s important to note that there’s one thing that just about everyone agrees on: Mammograms are far from perfect, and overdiagnosis is a legitimate problem in cancer treatment. Even the most ardent defenders of mammograms (who are also usually people with some form of vested financial interest in the procedure) are forced to recognize that there is much room for improvement. This tells us we should at least be skeptical of mammography.
If you care at all about health research, this debate is nothing new. Health experts have long thrown doubt on the efficacy of mammograms, pointing to a pile of studies that suggest they aren’t nearly as effective as we’d hope. One massive 2014 study — reviewing almost 90,000 women ages 40 to 59 in the United Kingdom — found no significant difference in cancer deaths between those that received mammograms and those that didn’t. Also, consider that while the United States recommends women receive mammograms way more frequently and at younger ages than women in Europe, we run in the middle of the pack among developed countries when it comes to mortality rates.
Proponents of the procedure usually argue that it’s still good to be vigilant — and that even if mammograms falsely detect dangerous cancers, it should be up to women to decide whether they want to endure the scary rounds of extra tests and biopsies just to be safe. A number of prominent health groups say that women should have regular screenings to detect cancer before age 60 and as early as 40 — and Congress has agreed.
But of course, not all cancerous tumors pose the same health risk. Overdiagnosis can translate to women undergoing unneeded radiation, surgery or even chemotherapy. These treatments, of course, can pose serious threats to women’s health. It’s also extremely costly: One study published in Health Affairs estimated that the United States spends $4 billion annually in unneeded breast cancer treatments.
Part of the problem is that there’s still so much we don’t know about breast cancer. Take, for instance, ductal carcinoma in situ — the most common type of non-invasive breast cancer. Researchers have been studying DCIS for a long time, but they simply can’t agree on what it is — and the debate can get pretty bitter. Some scientists want to consider it a “pre-cancer” or a “risk of a cancer.” Others insist that it’s better to be safe than sorry and recommend treating it through surgery and radiation. For patients, the mixed messaging is confusing and frustrating.
So what should a 40-year-old woman do if she has a mammogram and learns that she has DCIS? There isn’t an easy answer. It’s tempting to get wrapped up in the feel-good, Angelina Jolie narratives, but anecdotes are not good science, and no one should disregard the costs and pains of treatment for an overdiagnosis.
For Aaron Carroll, director of the Center for Health Policy and Professionalism Research at Indiana University School of Medicine, the solution is to recommend that younger women with low risk factors not have mammograms done. His view, he admits, is usually rejected among radiologists — but he stresses that their perspective is informed from a culture that emphasizes preventing diseases regardless of costs.
“Doctors are human beings, and they fall for the same myths and errors that everyone else falls for,” Carroll said in an interview. “I’ll say I have research data behind me, but they say they have clinical experience.”
The most optimistic take on mammography — as described by Otis Brawley of the American Cancer Society in an editorial responding to the Danish study — is to believe that cancer research will be able to tell us more about breast cancers that will let us identify which are dangerous and which are not — possibly at the genetic level.
Hopefully, that’s the solution. In the meantime, the message to women should be to temper their expectations of mammograms. Studies show that patients grossly overestimate the power of mammograms to detect and prevent cancers, but this needs to be corrected. We really shouldn’t be overselling how well we understand this disease.