Editorial: Value of Screening; New Study Reassesses Mammograms

Women seeking definitive guidance about breast cancer screening — when and how often to get mammograms — received little of it from the Nov. 22 Associated Press story about a recent study of the test. The analysis, led in part by Dartmouth’s Dr. H. Gilbert Welch and published in the current New England Journal of Medicine, reviewed data from between 1976 and 2008 and found that the advent of widespread screening had little impact on breast cancer mortality. But while the study certainly raised questions about the value of mammograms and suggested that they result in some unnecessary treatment for cases that would not have caused death, medical experts were hardly prepared to advise women to make any alterations in their testing regimen.

Clearly, regular screening doesn’t always make sense as a general practice because at a young age not enough cancers will be detected to justify the cost, not to mention whatever harm might result from the testing itself, including the anxiety resulting from false-positives. This study, however, went beyond a rudimentary cost-benefit analysis by focusing on outcomes. Looking at data gathered over more than three decades when mammograms had become a regular preventive measure, the study found that the number of early-stage cancers detected had more than doubled, but that late-stage cancers decreased by only 8 percent. If early detection was as effective as hoped, the considerable increase in detection (and, often, treatment) of early-stage cancer should have produced a much more significant drop in late-stage disease. The study estimated that more than a million women were treated for non-lethal breast cancer during the period under review.

“We’re diagnosing a lot of something else — not cancer,” said Dr. Archie Bleyer, one of the study’s co-authors. “And the worst cancer is still going on, just like it always was.”

Similar questions resulted from comparing the mortality rate of women age 40 and over — those who are advised to get regular mammograms — with that of younger women. Mortality dropped significantly for the older group, 28 percent, but even more, 41 percent, for younger women, who presumably were not getting mammograms. That would suggest that progress in reducing breast cancer mortality came by way of improved treatment, not earlier detection.

Judging by those quoted in the AP story, not all cancer experts are prepared to accept the suggestion that breast cancer is being overdiagnosed, or that mammography should be regarded as anything but an indispensable tool. And no one is suggesting that women should alter their approach to getting regular mammograms to protect themselves from what remains the leading cause of cancer deaths in women. The American Cancer Society recommends yearly screening after age 40, while a government appointed task-force recommends every-other-year mammograms between the ages of 50 and 75.

But even if the study offers little practical advice for women, it does highlight how much more work remains to be done to improve doctors’ ability to distinguish among the types of breast cancer and determine the best course of treatment. It also provides additional evidence for the case that Welch and The Dartmouth Institute have been increasingly persuasive in making: More medicine is not necessarily better medicine.

“Our technology has brought us to the place where we can find a lot of cancer,” said Dr. Len Lichtenfeld, the American Cancer Society’s deputy chief medical officer. “Our science has to bring us to the point where we can define what treatment people really need.”