More women should talk with their doctors about when to start and stop routine mammograms or whether the tests are right for them at all, say two new reports calling for more individualized decision-making around the breast cancer screening tests.
The reports, published Tuesday in JAMA, the Journal of the American Medical Association, include one that focuses on women over age 75 and one that focuses on younger women.
In both cases, the issues are complex, because mammograms confer benefits and risks: Most research suggests they modestly decrease the chance of dying from breast cancer but also lead to many repeat tests and biopsies and to treatment for some slow-growing cancers that would never cause harm. The balance of benefit and harm varies by age but also woman to woman, based on family history and other factors. Attitudes about accepting the risks also vary.
"Different patients are going to make different decisions based on the same information," says Lydia Pace, a specialist in women's health at Brigham and Women's Hospital, Boston.
Pace is co-author of the paper on younger women, which summarizes more than 50 years of research. For example, it says that for every 10,000 women, mammograms probably save five lives of women in their 40s, 10 lives of women in their 50s and 42 lives of women in their 60s. But breast cancer still kills 31 of those screened in their 40s, 62 screened in their 50s and 88 screened in their 60s.
Meanwhile, half of women screened for 10 years have a "false positive" – a suspicious mammogram that leads to a repeat test or a biopsy on a healthy breast – and 19% of the cancers found and treated are not life-threatening tumors, the researchers say.
Such numbers may be especially useful for women in their 40s, who face conflicting official guidelines: The American Cancer Society says they should have yearly mammograms; the U.S. Preventive Services Task Force says women should wait until 50 and then have mammograms every two years.
But the numbers don't help women over age 75. That's because most studies have not included them, even though breast cancer rates peak in old age. Guidelines say older women should make individualized decisions, "but don't tell them how to do that," says Louise Walter, a professor of medicine at the University of California-San Francisco and co-author of the paper on older women.
One big issue to consider is life expectancy, she says. Studies suggest any benefit from mammography is lost in women with less than 10 years to live. Those women face only the harms, such as unneeded biopsies and treatments, Walter says.
"This is not about rationing mammograms," she says. "This is about avoiding harm in women mammograms are unlikely to help."
Doctors have ways to estimate life expectancy (such as the tools at ePrognosis developed by researchers at UCSF). In general, mammography is unlikely to do much good in elderly women already battling any serious illness, such as dementia or heart failure, Walter says.
"If you are a healthy woman in your late 70s or early 80s, there's a good chance you could benefit," she says. But she says even some healthy women may choose to stop. "Some will say 'I just don't want to be poked and prodded.' "
The cancer society recommends mammograms "as long as a woman is in good health."
"There is a point at which screening for cancer should stop," says Richard Wender, chief cancer control officer for the society. Some women with conditions that will end their lives before breast cancer would are getting unneeded mammograms, he says. The same is true for colon cancer tests, he says.
For healthier and younger women, the new reports provide food for thought, he says. But he says the numbers are based on flawed, decades-old studies that might underestimate the benefits of mammogram and subsequent breast cancer treatment today – a limitation researchers acknowledge.
Wender says the downsides of screening are real but "I continue to believe that most people in the United States place a high value on preventing a cancer death and will put up with a lot of downsides."