Bloomberg News: The dangers of excessive mammograms

A 25-year study of 90,000 women found that mammograms do nothing to lower the death rate from breast cancer. That's pretty strong evidence, to say the least. And it's only the latest from many studies over the past several years indicating that mammography often doesn't help prevent advanced cancer.

Yet women — and, even more crucially, their doctors — remain unwilling to give up annual mammograms. Five years ago, when a panel of experts convened by the U.S. government looked at the available evidence and concluded that women in their 40s should stop being screened for breast cancer, and that those aged 50 to 75 should be screened only every other year, 3 in 4 women said they simply disagreed. The secretary of Health and Human Services felt obliged to speak out against the experts' recommendation.

What's going on here? A fear of breast cancer, to begin with, which is understandable. But there's also an abiding belief that the best way to fight cancer is to find any sign of it early and root it out — despite evidence demonstrating that's not entirely true, and that overscreening can lead to overdiagnosis and overtreatment. What's needed is a strategy to align public perception with scientific consensus.

One easy way to do so is through language. It's neither correct nor helpful to use the word "cancer" to describe the tiny lesions that mammograms are so good at detecting but that usually don't turn into lethal cancer. Autopsy studies suggest that 4 in 10 women from ages 40 to 70 have these lesions without knowing it and die of something else.

Last summer, a National Cancer Institute working group endorsed the term "idle." This has the benefit of being both scientifically precise (it stands for indolent lesions of epithelial origin) and conversationally apt (a growth can be described not as "cancerous" but as "idle").

The group also recommends setting up observational registries for such lesions to make it easier to study exactly how they progress, and it says doctors should raise the threshold for deciding which of these lesions need to be biopsied.

Those last two recommendations show that medical practice can also lag the scientific consensus. In fact, the American College of Obstetrics and Gynecology stubbornly maintains its recommendation that women get a mammogram every year once they turn 40.

The doctor-patient relationship is by definition personal, of course. But a woman has a right to expect her doctor to be aware of the latest research and apply it to her case. In the case of breast cancer, the evidence is clear: More tests do not reduce mortality, but fewer tests may reduce anxiety.