A few days after my 40th birthday, I had a routine appointment with my gynecologist. As she turned to leave the exam room, she handed me a slip of paper without saying a word. It was a prescription for a mammogram.
“Wait — you’re not going to discuss this with me?” I asked. I knew that the U.S. Preventive Services Task Force (USPSTF) had advised that for women in their 40s, the decision to start mammography should be an individual one, based on family history and a patient’s values.
“I am discussing it. I’m telling you to get a mammogram,” she replied. When I asked her why, she told me because it could save my life.
What she neglected to tell me is that a mammogram was, in my case, more likely to hurt than help me. Few doctors take the time to mention the risks of mammography — especially, the danger of overdiagnosis — that a mammogram might lead a patient to get needled, sliced, zapped with radiation and possibly treated with tamoxifen, a drug that increases risk of uterine cancer, for a breast lesion that wasn’t life-threatening in the first place.
Most people believe that breast cancer is inevitably a progressive disease that will kill you if you don’t remove it in time. According to this idea, which I call the relentless progression model, every big cancer is harmful, every small one is less so and every cancer is curable if only you catch it in time. It’s an appealing, intuitive idea — except that a growing body of research suggests that it’s wrong.
Scientists now know that breast cancers can behave in different ways. The disease falls into three general behavioral categories, according to Barnett Kramer, director of the National Cancer Institute’s Division of Cancer Prevention. Kramer refers to the three types as turtles, birds and bears.
Turtles are cancers that progress so slowly that they will never metastasize or harm you. Finding these cancers won’t save your life, because they never endangered it in the first place, Kramer says. Birds, on the other hand, are extremely aggressive cancers that are programmed to become deadly. These cancers aren’t helped by mammography, either, because they spread before they can be detected on a mammogram.
It’s only the bears whose future is altered by mammography, Kramer says. Bears may eventually kill you if they’re not treated, but they spread slowly enough that a mammogram can detect them in time to make a difference.
Here’s the problem: At the moment, we don’t have a reliable way to distinguish turtles, birds and bears. While scientists are working to find genetic markers that predict how a cancer will behave, right now the only way to know which way a cancer will act is to wait and see what it does. This means that if a mammogram finds a breast cancer, the safest course of action is to assume that it’s a bear. This inevitably leads to some women with turtles and birds receiving treatment that doesn’t help.
Those with turtles end up getting treated for cancers that were never destined to cause any harm. Women who have such cancers are grateful because they believe their lives have been saved. Their doctors and radiologists become even more convinced that mammography saves lives. In fact, these women are victims of overdiagnosis.
A study published in the New England Journal of Medicine in 2012 calculated that one-third of newly diagnosed cases of breast cancer are turtles. The study’s author, H. Gilbert Welch, a professor of medicine at the Dartmouth Institute for Health Policy and Clinical Practice, estimates that more than a million American women have been treated for an early-stage breast cancer that would never have hurt them.
For bears, mammograms make a difference, which is why the USPSTF recommends them every two years for women age 50 to 74. According to the National Cancer Institute, having a mammogram every year for a decade reduces the typical woman’s risk of dying from breast cancer by about 1 percent.
But at my age and with my family history (which includes two aunts with the disease), the National Cancer Institute’s risk calculator suggests that my chance of developing invasive breast cancer over the next five years is 0.7 percent, which is the average risk for a woman of my age. The tool also calculates that I have an 89 percent probability of remaining free of breast cancer from now until age 90. Those are odds I can live with.
How would a mammogram improve my already favorable odds?
Last year, I asked Welch to calculate a woman’s risk of dying from breast cancer with and without mammography. His numbers showed that the risk for a 50-year-old woman who had no mammograms over the next 10 years is reassuringly low: between 0.39 and 0.48 percent over the next 10 years. Having an annual mammogram during those 10 years only reduces that number by a sliver: to 0.34 to 0.37 percent. For someone like me — a woman in her 40s — a mammogram’s benefits are even slighter.
“This is a choice, not a public health imperative,” Welch says.
The USPSTF has calculated that 1,900 women age 40 to 49 must be screened with an annual mammogram for 10 years to save a single life. During that time, the other 1,899 will gain no benefit from screening but will be subjected to 1,330 false alarms, 665 breast biopsies and eight breast cancer diagnoses that do not change the course of the disease.
This means that my chances of being harmed by a mammogram are far greater than my likelihood of being helped.
I realize that by opting out of mammography, I’ll miss the very modest chance that such a scan would save my life. But I’m OK with that. If I feel a lump or another symptom, I’ll get checked out. This is how breast cancers were detected before mammography, and even women who undergo yearly mammography may still find a cancer as a lump in between mammograms. (And these so-called interval cancers are most common in women in their 40s.) But until then, I’ll trust that feeling healthy means that I am.