Ten years ago I opened my driver’s license renewal notice and out dropped a pink slip reminding me to have an annual mammogram. While it was touching to know that the New York State Department of Motor Vehicles cared about my breast health, I saw the pink slip as yet another example of over-the-top mammography promotion. (My husband has yet to receive any health instructions from the DMV.) When a test’s lifesaving benefit has been oversold to the public for over three decades—and the harms downplayed—any cutback in recommendations will be met with a firestorm of anger.
That’s exactly what happened yesterday when a highly respected organization recently broke ranks with others that issue screening guidelines and advised women at average risk for breast cancer to begin having regular mammograms at age 50, not age 40. The U.S. Preventive Services Task Force also changed two longstanding recommendations: women age 50 to 74 should have a mammogram every two years rather than annually (cuts down on the biopsies without altering the number of lives saved), and doctors should stop teaching women how to do breast self-examination (no benefit, more biopsies). Shock, horror, and condemnations followed the announcement in the media on Monday. Not a surprising reaction considering the advice to the public preceded the science.
The recommendation to start mammography screening at age 40 was premature to begin with, dating back to the early 1970s. Incredibly, it was based on one clinical trial conducted in the 1960s that found a 30% reduction in breast cancer deaths in the women over age 50 but no benefit for women in their forties. That clinical trial, known as the Health Insurance Plan of Greater New York study, had followed the participants for three to five years. Not unreasonably, public health officials at the time thought that the younger women in this study would eventually show a reduction in breast cancer mortality with longer followup. They didn’t bother to wait and the aggressive promotion of mammography took off after President Nixon declared the “war on cancer.” Thus began a nationwide campaign, sponsored by the American Cancer Society and the National Cancer Institute, urging women over the age of 35 to seek yearly mammograms complete with the overly optimistic message that early detection will save their lives.
Radiation exposure was the only acknowledged harm in this era, and it was usually dismissed as unimportant in comparison to the benefit of finding breast cancer early. By the 1980s, mammographic techniques had improved and the radiation exposure greatly reduced. A modest reduction in breast cancer deaths in younger women was found in some, but not all, clinical trials done in other countries. The largest trial intended to answer the question of mammography’s value to women in their forties found no lifesaving benefit. Basing its estimate on the trials that did find a benefit, the U.S Preventive Services Task Force concluded that one in every 1,900 women who undergo an annual mammography over a ten-year period will avoid a breast cancer death.
In 2001 the Lancet published the first evidence indicating that mammography screening leads to overtreatment without reducing mortality. The authors, both researchers at the Nordic Cochrane Centre, did an in-depth assessment of the data generated by all the randomized mammography screening trials. To this day the Cochrane researchers continue to reassess the research (read a 2009 update of their findings). What they have found is entirely counterintuitive and not confined to women in the forties. Mammograms can detect tiny cancers, but not all of them would become deadly or even produce symptoms if left untreated. Yet virtually all are treated aggressively because no test can accurately sort out the potentially lethal cancers from those that do not progress. Overdiagnosis and overtreatment are the terms for this problem, which was quantified in a study published recently in the British Medical Journal, showing that one in three cancers found on a mammogram would not become life-threatening. This should have been the major reason for the change in recommendations, but the media coverage frequently presented mammography’s major harm as the anxiety from false alarms and breast biopsies. As for the other major harm—radiation exposure—the USPSTF cites this estimate, “annual mammography of 100,000 women for 10 consecutive years would result in up to 8 radiation-induced breast cancer deaths.”
The USPSTF based its revised recommendations on the findings of a panel of experts made up of research physicians and most crucially, biostatisticians who quantified some of the harms as well as the benefit. Their task was to assess the evidence generated by the world’s gold standard trials, as well as observational studies. While we all hope that our physicians are familiar with the evidence supporting their advice, the sad reality is that many are not. Most simply follow the advice of organizations like the American Cancer Society and the American College of Obstetricians and Gynecologists. What’s more, it’s as hard for them to change their thinking as it is for us. Here’s just one example. From the professional advice given my daughter and her friends, I know that the long-discredited American Cancer Society recommendation for women to have a baseline mammogram in their late thirties continues to live on. (The ACS quietly withdrew this recommendation in 1992, but it takes time for the word to get out.)
Other sources of information:
If you want to make an informed decision about mammography, go directly to the scientific evidence instead of your doctor. Here’s a timeline for the key studies and previous failed attempts to raise the starting age for mammography which was published recently in the New York Times. This is a “summary of the evidence” from the randomized trials and observational or population studies (i.e., mammography done in the real world), which formed the basis for the USPSTF’s revised recommendations. And this is the National Cancer Institute’s summary of pretty much the same clinical trials (note that the first trial that justified the advice to start mammography at age 40 is described as being of “poor quality”), as well as population studies. The NCI summary is difficult to read, but this sentence from the summary is easy to understand, “Screening for breast cancer does not affect overall mortality, and the absolute benefit for breast cancer mortality appears to be small.” This applies to women who start mammgraphy at age 50 as well as those who started at 40.
Using the same data, researchers at the Nordic Cochrane Centre have produced a more understandable summary for women that is available on the Web (see link within this article). To increase your chances of having a mammography performed skillfully, read How to Select a Mammography Facility . And to reduce your chances of an inaccurate diagnosis, read “When to get a 2nd pathology opinion.”
Maryann Napoli, Center for Medical Consumers(c)
Two days after the USPSTF issued its recommendations, HHS Secretary Kathleen Sebelius issued a statement distancing the Obama Administration from the new guidelines. In short, she told women to ignore them and claimed disingenuously that more research is needed. This is very disappointing to those of us who had high hopes that health care reform would mean that medical treatment and screening decisions, as well as cost-cutting measures, would be evidence-based. Once again, mammography’s evidence is clouded by politics. This time around, however, the politics make sense. No doubt, Secretary Sebelius did not want to derail the current health care reform efforts.