When Janet Halloran last saw her primary care physician, the doctor asked whether she had undergone her annual mammogram. Yes, she replied, she had.
At 76, Ms. Halloran, a real estate broker in Cambridge, Mass., is past the age that most medical guidelines recommend breast cancer screening for someone with no history of the disease. Even for younger women, the guidelines call for a mammogram every other year, not annually.
So Ms. Halloran could consider stopping mammograms, or at least having them less often. But her doctor has never discussed that prospect. “She says, ‘These are the things you need to do,’” Ms. Halloran said. Besides, she added, it’s an easy test: “Go once a year, hold your breath and you’re done for another year. It’s just routine.”
But for older women, should it be?
“There’s been a lot of uncertainty,” said Dr. Xabier Garcia-Albéniz, an oncologist and epidemiologist at RTI Health Solutions and lead author of a new observational study that tries to answer that question. “This is an area with a complete lack of randomized clinical trials.”
Breast cancer studies, like medical research in general, have often excluded older subjects. So the data on whether mammography improves survival is very limited in women ages 70 to 74, and nonexistent for those 75 and older.
That’s why the independent U.S. Preventive Services Task Force has concluded that while having mammograms every other year improves survival for women ages 50 to 74, there’s “insufficient” evidence to assess their use for those over 75.
The American Geriatrics Society includes screening for breast and other cancers on its Choosing Wisely list of tests that should be questioned. It urges doctors not to recommend it “without considering life expectancy and the risks of testing, overdiagnosis and overtreatment.”
Yet more than half of women over 75 have had a screening mammogram (a test for individuals with no history or symptoms of breast cancer) within the past two years, the Centers for Disease Control and Prevention reported in 2018.
“Whether this investment in breast cancer screening alters survival is a critical question,” said Dr. John Hsu, a health services researcher at Harvard Medical School and senior author of the new study, published in the Annals of Internal Medicine.
The research team used Medicare claims from 2000 to 2008 to follow more than one million women, ages 70 to 84, who had undergone a mammogram.
They had never had breast cancer and had a “high probability,” based on their medical histories, of living at least 10 more years. “That’s the population who will reap the benefit of screening,” Dr. Garcia-Albéniz said, because it takes 10 years for mammography to show reduced mortality.
The researchers divided the subjects into two groups: one that stopped screening, and another that continued having mammograms at least every 15 months. They found that mammograms provided a survival benefit, if a modest one, for women ages 70 to 74. In line with previous research, the study found that annually screening 1,000 women in that age group would result, after 10 years, in one less death from breast cancer.
But among the women who were 75 to 84, annual mammograms did not reduce deaths, although they did, predictably, detect more cancer than in the group that discontinued screening.
“You’re diagnosing more cancer, but that’s not translating to a mortality benefit,” Dr. Garcia-Albéniz said.
Why not? “The cancers themselves might be different at different ages,” Dr. Hsu said. “They might grow faster or slower, or be more likely to spread.”
Treatments may also be less effective at older ages, said Dr. Otis Brawley, an oncologist and epidemiologist at the Johns Hopkins University School of Medicine, who wrote an editorial accompanying the study.
But older people typically are also subject to what researchers call “competing mortality.” Many of the cancers detected by mammography — tiny tumors that earlier technology wouldn’t have spotted — are unlikely to cause any harm if left untreated. But most older people have other diseases that will progress.
“It’s very difficult to tell someone in her 70s or 80s that we’re going to modify your treatment, or not treat you, because of the likelihood that something else will kill you before this cancer will,” Dr. Brawley said.
That reluctance to discuss life expectancy and the limitations of screening also means that many women don’t recognize that, in addition to being inconvenient, expensive and a cause of discomfort or anxiety, mammograms can actually do harm. The tests often prompt unneeded surgery, radiation or drug regimens for cancers that would never have caused symptoms or shortened lives.
Still, because life expectancy varies widely, some very healthy older women may live long enough to benefit and may indeed want screening. Mammograms could lead to treating an aggressive cancer earlier, and with less extensive surgery, for instance.
“I would be very happy if doctors started using our paper to inform the discussion they have with their patients,” Dr. Garcia-Albéniz said.
Yet women remain so committed to regular mammograms that experts doubt they could recruit enough people for a large randomized trial in which half the subjects forgo the tests.
Dr. Mara Schonberg, an internist at Beth Israel Deaconess Hospital in Boston, has worked for years to help women make decisions about breast cancer screening, and has found it tough going.
“These women were told for 40 years to get screened,” Dr. Schonberg said. “They get reassurance from a negative mammogram. And it’s very hard to understand that finding breast cancer early may not help you live longer or better.”
To help explain, Dr. Schonberg developed a decision aid: a brochure, written at a sixth-grade reading level, that uses research findings to explain the pros and cons.
A pilot study showed that, after reading it, women from 75 to 89 were more knowledgeable about mammography, more apt to discuss it with their doctors and less enthusiastic about continuing it.
But they did continue. More than 60 percent, including those with lower life expectancies, had another mammogram within 15 months. A larger study with 546 participants, being readied for publication, will report similar results, Dr. Schonberg said.
Perhaps, as Dr. Brawley said, “the most important thing we can do is get people to understand what the questions are, and to understand that nobody has the exact answers.”
But Dr. Schonberg’s grandmother, who followed her doctor’s recommendation and had a mammogram at age 78, came to a more definitive conclusion.
Ann Schonberg was a Detroit homemaker and a lifelong smoker with mild emphysema. When her mammogram found a small Stage 1 cancer, she underwent a lumpectomy and began endocrine therapy, taking drugs that might lower the risk of recurrence. “She didn’t like how she felt, so she stopped everything after three years,” Dr. Schonberg said.
When Ann Schonberg reached her mid 80s, a mammogram picked up another small cancer, prompting another lumpectomy. At the same time, although she had stopped smoking at 80, her emphysema worsened steadily. That is what caused her death at 88, not breast cancer.
“All the doctors’ appointments, the surgeries, the worry — for her, it was all for naught,” Dr. Schonberg said. Shortly before Ann Schonberg’s death, she told her granddaughter, “I wish I’d never had that mammogram.”