Screening Mammograms Overdiagnosed More Than 1 Million Women

This paper is going to create discussion and concern, and get people thinking again about mammography. But from my point of view it isn’t the final answer.

Dr. J. Leonard Lichtenfeld

There’s been a substantial amount of research and commentary on the role of mammography – discussion that has split people into two political camps: one saying it’s the major reason for the reduced mortality and one saying it has no value whatsoever.

I think the truth lies somewhere in between.

Something has clearly affected breast cancer mortality in the past few decades. Before the 1990s, the rate of breast cancer death in this country was a flat line that had not changed for decades. Then suddenly it began to drop, and it has continued to do so. We are clearly doing something right. The question is: What is it? Probably a combination of mammography, improvements in adjuvant chemotherapy, and a general increase in breast awareness.

In the 1970s, when I was beginning practice – and before that, as a family member – breast cancer was not something anyone spoke about. It was never mentioned publicly. Now we are much more aware. Women are tuned in to the topic and aware of the need to do self-exams. It’s a national discussion.

Both our surgeries and our chemotherapy are much improved. But even now, if a woman presents with a palpable breast lesion, the odds are that it’s going to be fairly sizable and have lymph node involvement. Can we assume that we are able to effectively treat every woman in this group? I’m not sure we can.

This leads us to mammography. There is no question that it identifies subclinical lesions. But we have to recognize the problem of overdiagnosis and overtreatment.

Mammography and other new imaging modalities are driving the point of detection to much earlier in the history of a woman’s breast cancer. We have recognized from autopsy studies that certain cancers can exist in the body for long periods of time without ever causing any problems. We know that most women with breast cancers don’t have any readily identifiable risk factors. And we don’t have a test that allows us to tell most women whether or not their particular cancer is likely to be aggressive.

The current study doesn’t really help us with these questions. There are severe limitations on these data. How often did the women actually have a mammogram? What we consider "regional disease" today is not the same as it was in 1975. Advanced disease today isn’t the same as it was in 1975. There are cultural and insurance barriers that affect access to care and, thus, affect mortality. All of these issues must be weighed into the equation.

Right now, we at the American Cancer Society are still confident in our recommendation for women older than 40 to have an annual screening mammogram and clinical breast exam. There are other recommendations from other groups, which may be a better fit for other women. The important thing is for a woman and her physician to pick a program and stay with it.

Everyone wants clear-cut answers in a world that is not clear-cut and is unlikely to become so. So women and doctors are left in the difficult position of weighing what is best for them.

I think we would all be reluctant to completely give this up. A breast cancer diagnosis in the 40s can kill a woman in her 50s. I’m concerned that we will begin missing breast cancers now that will kill in 10-15 years, and that we will lose the gains we’ve seen in mortality.

Dr. J. Leonard Lichtenfeld is the deputy chief medical officer for the national office of the American Cancer Society.