In recent days, the White House and Congress have been distancing themselves from the U.S. Preventive Services Task Force's controversial new breast-cancer screening guidelines. But after all the blaring headlines and government double talk, plus this week's news that cancer death rates continue to fall (all deaths were down by 1.6 percent per year between 2001 and 2006; breast cancer deaths, by almost 2 percent per year), how in the world can women figure out if regular mammograms are indeed good for their health? The new guidelines, which the task force stands by, come out against routine screening of women in their 40s and call for cutting in half screening of those over 50 and stopping altogether at 74. This schedule, which would bring the typical woman down to about 12 mammograms total from two to three times that many, is in contrast to most current guidelines (and the task force's previous position), which advise screening every year starting at age 40. The downgrade of mammograms has left women with some heavy-duty misconceptions: Early detection of cancer does not save lives. Forty-year-olds see trivial or no benefits from cancer screening. And the new guidelines are better because they're based on science, rather than sloppy emotion.
Welcome to the future, where the battle of dueling medical directives on how to care for patients effectively but at the lowest possible cost is just beginning. The breast-cancer screening saga points to the kind of struggles that will be rampant as comparative-effectiveness studies—whether of mammograms, testing for prostate cancer, or a new drug or procedure for heart failure—are translated into medical decisions by Uncle Sam, who will also be determining what everyone's health insurance will and won't cover. Most of these decisions will be clear, based on good science and a consensus of opinion, but some will struggle in the gray zone where legitimate disagreements can run hot. Mammograms are a case in point. Women have been told that the new guidelines are based on the latest science. What needs to be made clear is that the same latest science also shows that the current generous screening approach saves many more lives—although it consumes more resources.
[Slide Show: 11 Screening Tests You Should (or Shouldn't) Get.]
Skeptics of mammograms are right that the test is imperfect, and it needs to get better. And indeed, medical costs are unsustainable and must be contained in some fashion. But U.S. breast-cancer screening policies are a big part of why America has among the lowest breast-cancer death rates of developed countries, and in my view, it's highly questionable to start rationing there, particularly without full transparency. The champions of mammogram-lite are misleading the public when they claim that cutting women's mammograms by more than half over their lifetime will have trivial effects. Finding cancers early that lurk in solid organs saves lives overall and makes treatment easier. With breast cancer, studies of large populations have shown this time and again; one just released from Harvard Medical School reported mortality rates in Massachusetts that are roughly 10 times as high in women who do not get regular mammograms as in those who do.
Even the task force's own exhaustive analysis of clinical trials, including those involving women in their 40s, demonstrated a fairly consistent 15 percent or more reduction in mortality because of mammograms. Powerful computer modeling techniques that projected mortality based on numerous screening schedules—varying by the age that screening began, when it stopped, and whether it was yearly or every other year—demonstrated an impact on women's lives that was two to three times that great. The more frequent the mammograms, the modeling showed, the better.
When "public good" means more deaths. Among the several screening options deemed by the USPSTF research to be "efficient" in terms of health benefit (survival) balanced against resource use (cost), the one that saved the most lives is virtually the same as the schedules advised by most doctors today and endorsed by the American Cancer Society and the American College of Obstetricians and Gynecologists. It was the most intensive program—amounting to an average of 37 screenings for women who live happily to age 84. The task force chose instead just one-third that many screenings, accepting roughly 80 percent of the survival benefits of the intensive program in exchange for an estimated 50 percent of the costs.
This cost-benefit trade-off surprises me, even in the context of health reform. The general understanding of comparative effectiveness decision making is that treatments that reduce costs are preferred—assuming they provide the same or a better benefit. The new guidelines would change that equation, upsetting existing practice and accepting a scientifically predictable loss of lives. We're talking about human lives here, not pork bellies.
What's the evidence that lives are at stake? Consider some of the task force's computer-generated comparisons of the mammogram-lite and mammogram-rich approaches. The most extreme model, produced jointly by the University of Wisconsin and Harvard, shows that the new guidelines would bring a mortality reduction (compared with women who have no screening) of 28 percent; the one that starts screening yearly at age 40 would amount to a 54 percent reduction. If one looks at the number of years of life gained—which is greater when younger women's lives are saved—the task force's approach brings 95 extra years per 1,000 women, far below the 202 years gained with the more intense screening schedule. Boiled down to its essence, curbing women's appetite for, or access to, mammograms could easily mean that more than 100,000 women would be absent from the ranks of the 2.5 million breast-cancer survivors. That's roughly two Vietnams or more than 30 9/11s.