The Mammogram Melee: How Much Screening Is Best?

Women’s breasts are not the usual topic of public discourse in Washington, at least not outside the context of a scandal. But for the past few weeks, the question of when women should be screened for breast cancer has become the subject of intense medical debate, partisan congressional bickering and a whole lot of confusion among mothers, daughters, sisters and friends, not only inside the Beltway but throughout the rest of the country.

On Nov. 16, the 16 members of the independent U.S. Preventive Services Task Force recommended that most women delay routine mammograms until age 50 (instead of 40, as the panel advised in 2002). The task force cited enhanced analysis of the risks and benefits of screening as the reason for the new guidelines. But the recommendations went straight to the heart of the emotionally charged debate over the Democratic-sponsored health care reform legislation that is working its way through Congress. Republicans like Representative Marsha Blackburn charged that “this is how rationing begins. This is the little toe in the edge of the water.” No one was more surprised, or less prepared, for the uproar over the new guidelines than the advisory panel itself. As a result, the merits of what the group is now recommending have been obscured by all the political smoke. Dr. Diana Petitti, a professor at Arizona State University and vice chair of the task force, says, “Our attempt to communicate [the risks and benefits of] routine screening was definitely lost.”

The rumblings over the mammography message provide a useful window into why U.S. health policy does not always dovetail with the best available medical evidence, and certainly not with the best available data on costs. By and large, American patients (not to mention politicians and cancer advocacy groups) still subscribe to the view that every life is worth saving, no matter the cost, and that when it comes to prevention, screening is always good and more is always better. For decades, patients have been steeped in the notion that frequent screening is not just beneficial but also essential to the early detection of cancer. But such personal calculations do not apply in the same way to an entire population, where the benefit to some must be weighed against the harm to others.

Judging from the reaction to the new breast-cancer recommendations, that is proving to be a difficult reality for Americans to accept. But the task force’s mammography advice probably marks the leading edge of similar recommendations to come. Already, new guidelines on Pap smear screening for cervical cancer have delayed the age at which young women should begin regular testing and reduced the frequency of testing in older women. Doctors are also questioning the usefulness of prostate-cancer screening among otherwise healthy middle-aged men, as studies begin to show that the test, which has many risks, may not necessarily lead to fewer deaths from the usually slow-growing cancer. The Senate health reform bill currently being debated would also rely on the task-force guidelines to determine what preventive medical services private insurers would be required to cover at no cost to patients. In a sign of how contentious evidence-based approaches may become, Secretary of Health and Human Services (HHS) Kathleen Sebelius quickly distanced the Obama Administration from the new mammography advice. She said in a statement, “Our policies remain unchanged,” and cast doubt on whether private insurance–required by most states to cover routine mammograms beginning at 40–would be affected. “Keep doing what you’re doing,” Sebelius advised women.

Gail Wilensky, a former high-level HHS official under George W. Bush, called the Obama Administration’s response “appalling and outrageous and a very good reminder of how hard it is going to be to move to evidence-based medicine whenever it goes against a sacred cow.”

The Limits of Screening

A decade ago, doctors saw breast cancer as a monolithic disease that always progressed the same way, beginning with a single mutant cell that continued to divide and spread to the rest of the body. At the time, screening all women made sense, especially since annual mammograms had reduced deaths from breast cancer 3% each year since 1990. But as Dr. Russell Harris, a professor of medicine at the University of North Carolina at Chapel Hill and a member of the task force, points out, breast cancer occurs less frequently in younger women, and not every cancer is the same. Some tumors are indolent and slow-growing; others are aggressively malignant and blanket a body within months. Mammography is best at spotting the slowest-growing tumors, which are most common and generally do not spread beyond the breast or require treatment. Although these tumors are malignant, they rarely go on to cause clinical symptoms. But when detected, they are still treated as if they were potentially faster-growing–with a combination of chemotherapy, radiation, surgery and hormone therapy. “We can’t figure out which is which,” says Harris. “So we end up having to treat them all.”

It’s impossible to say how frequently such “overdiagnosis” occurs, according to the task force, but the data did conclusively show that in order to save the life of one woman in her 40s from breast cancer, 1,904 women would have to be screened every year for up to 20 years. Because it judged that the risks of harm from annual screening outweighed the benefits, the panel issued its controversial recommendation that most women ages 40 to 49 need not get routine mammograms. “We felt that women would be better served if they understood the trade-off between the benefits, harms and risks of starting at 40 or waiting a few years into their 50s,” says Petitti.

That calculus is precisely what drives comparative-effectiveness research, a strategy embraced by both the House and Senate health care reform bills: figuring out which tests and treatments work best–instead of using every available treatment just because it’s there–while saving money without adversely affecting health. Using magnetic resonance imaging (MRI) to screen for breast cancer, for example, isn’t necessary for the vast majority of women who are at low risk of the disease; because most tumors are not aggressive, most women will not benefit from finding the first signs of tiny tumors that an MRI can detect.

Both bills in Congress would set up new institutes to organize and fund more comparative-effectiveness research, ostensibly to help guide health care policy. (The $787 billion American Recovery and Reinvestment Act of 2009 has already authorized $1.1 billion for the field.) And yet as Diana Buist, a researcher at Group Health in Seattle who received some of the stimulus funding, says, “[Comparative-effectiveness research is] a hard sell. It always has been.” According to a 2007 Congressional Budget Office (CBO) report on the topic, “Some experts believed that less than half of all medical care is based on or supported by adequate evidence about its effectiveness.” Instead, said the CBO, health care in the U.S. is often motivated by factors like “enthusiasm for the newest technology” and a fee-for-service payment system that rewards doctors based not on outcomes but the number and price of treatments they prescribe and perform.

The status quo won’t be easy to change, largely because evidence-based medicine often runs counter to our personal understanding of risk. It’s intuitively difficult for a woman in her 40s to stop getting annual mammograms when she is fully aware that they could save her life. Feeding this instinct is the relentless effort on the part of doctors and disease advocacy groups to promote preventive-health behaviors. Many feel the push may have done the public a disservice by instilling the belief that screenings are purely beneficial. “We have not rounded out that discussion with the American public about the harms,” says Dr. Therese Bevers, a professor of clinical cancer prevention at M.D. Anderson Cancer Center in Houston.

Even doctors are not always mindful of the risks of overscreening. Busy physicians often prescribe routine screening as a substitute for in-depth discussions with patients about their individual risk of developing cancer and the relative benefit a yearly mammography would offer. “These are hard conversations,” says Dr. Eric Winer, director of breast oncology at the Dana-Farber Cancer Institute and chief scientific adviser to Susan G. Komen for the Cure. “So it’s easier to tell women to get a mammogram beginning at age 40. But simple isn’t always what’s best.”

The Calculus of Risk

So is there any way to reconcile the weight of evidence about screening with patients’ natural impulse to take every available precaution? Science may ultimately provide a way out. A better understanding of what triggers abnormal growth of breast tissue, for example, would lead to more effective and targeted therapies, rather than the treat-one-treat-all approach in place today. Current screening can pick up only suspicious growths; smarter technology could help doctors know exactly what to do with them. “The goal is not finding cancer; the goal is reducing death and suffering,” says Dr. Lisa Schwartz, a professor of medicine at the Dartmouth Institute for Health Policy and Clinical Practice.

In the future, that may help reduce the risks to women from screening, says Dr. Stephen Taplin, program director for the Breast Cancer Surveillance Consortium at the National Cancer Institute, which is reviewing its guidelines that now advise biennial mammography for women in their 40s. But that will also require women to come to grips with the full range of risks associated with being proactive. Taplin says the oft-cited statistic that 1 in 8 women will eventually develop breast cancer requires doctors and patients to be vigilant but judicious about using mammography. Says Taplin: “We need to think about the seven who don’t get breast cancer as well as the one who will.”