Only One in 2000 Women Benefit from Mammography
Posted by medconsumers on December 1, 2006
Over the 31 years of publishing our newsletter, we have reported frequently about the lack of honesty in the mammography-screening information directed at women. Whether it comes from the American Cancer Society or a family physician, the information tends to be heavy on the lifesaving benefits and light on potential harms.
Here’s something a woman is unlikely to hear: “For every 2000 women* who go for mammography screening throughout a ten-year period, one woman will have her life prolonged. BUT ten additional healthy women, who would not have received a breast cancer diagnosis had they avoided mammography screening, will be treated unnecessarily for a breast cancer that would never have become life-threatening. And more than 200 women will experience significant psychological distress for many months because of false-alarm findings on a mammogram.” click here for the source of this quote
This important information, so crucial to a woman’s informed decision-making, comes from a recent update of a 2001 Cochrane Review of the world’s major mammography screening clinical trials with a combined total of a half a million women. In all seven trials, which form the basis of the U.S. screening recommendations, women were randomly assigned to receive screening mammograms or not. These trials, conducted in North America and Europe, were published many years ago, but researchers continue to argue over their interpretation and conclusions.
* These statistics apply to women, aged 50-69 years. Odds of lifesaving benefit are even smaller for women in their forties.
Original Review in 2001
The 2001 Cochrane Review of these trials was co-authored by Peter Gotzsche, MD, of the Nordic Cochrane Centre in Copenhagen, as is the 2006 update published recently in the Cochrane Library. The original review triggered a prolonged debate in the American media in 2001-2002 after The Lancet published a summary of its results. The lifesaving benefit of mammography screening was shown to be far more modest than previously thought and the harms had been greatly underestimated.
The Cochrane Review also found that the mammography-screened women in the trials had more mastectomies, more lumpectomies, and many more cases of ductal carcinoma in situ (DCIS) than the women who were not screened. Although most cases of DCIS, a tiny non-invasive cancer within the milk duct, do not progress, they were—and in some parts of the country, still are—always treated with breast removal. Now radiation therapy is the more common treatment for DCIS.
In the two trials judged to be of the highest quality by the Cochrane reviewers, there were 30% more cancers in the mammography-screened groups than in the unscreened groups and 31% more mastectomies and lumpectomies. Yet surprisingly, in both of these trials, conducted in Canada and Malmo, Sweden, respectively, mammography screening did not reduce the rate of breast cancer deaths. The other five trials, judged to be of inferior quality by the Cochrane reviewers, did show a modest reduction in breast cancer deaths.
Old Trials, New Data
You might wonder why we are still learning new information from these seven trials, which were conducted between the 1960s and 1980s. The answer lies in the Cochrane reviewers’ thorough assessment of the enormous amount of data that was originally withheld and, in some cases, continues to be generated by these trials. The data, it should be noted, were requested by the Cochrane reviewers.
The long-term data provided by the Canadian trial are a case in point. Mammography’s ability to detect many more cases of DCIS has long been known. Not until this trial published 11- to 16-year follow-up results for women in their forties did it become evident that mammograms also detect non-palpable invasive breast cancers that do not progress. This means that some breast cancers can begin to invade the surrounding tissue, but go no further.
This was demonstrated by the Canadian trial results: There were 43 more invasive breast cancers and 42 more cases of DCIS in the women who underwent mammography screening, compared to those who did not. Yet 16 years later, the breast cancer death rate of both groups was exactly the same. (It is likely that there were just as many breast cancers of both types in the women not given mammograms, but these women benefited from being unaware of them.) Currently, doctors are not able to determine which cases of DCIS and invasive cancer will not progress to become life-threatening. All are treated as if they will be lethal.
Bottom Line: Mammography screening was introduced 35 years ago to reduce the rate of breast cancer deaths by finding and treating breast cancer before it can be felt by a woman or her doctor. even trials have produced contradictory results about the reduction in deaths. And some of these trials found that mammography can detect non-palpable breast cancers—both invasive and noninvasive—that do not progress. All are treated as if they will. For the first time, the benefit and harms have been quantified for women who want to make an informed decision. Based on the combined results of these trials, the Cochrane reviewers produced the one in 2000 women will have her life-prolonged estimate stated at the beginning of this article.
“Women, clinicians and policy makers should consider the trade-offs carefully when they decide whether or not to attend or support screening programs,” concludes the updated Cochrane Review, which was originally commissioned and funded in 1999 by the Danish Institute for Health Technology Assessment and the (Danish) National Board of Health.
Maryann Napoli, Center for Medical Consumers ©