Rethinking Breast Cancer Screening: Why Mammograms Alone Aren't Enough

Keren Landman is a senior reporter covering public health, emerging infectious diseases, the health workforce, and health justice at Vox. Keren is trained as a physician, researcher, and epidemiologist and has served as a disease detective at the US Centers for Disease Control and Prevention.
On May 9, the US Preventive Services Task Force (USPSTF) released a draft of their new guidelines on who should be screened for breast cancer. The most significant change: recommending that women with average breast cancer risk start getting mammograms every two years beginning at 40, instead of starting at age 50.

According to the recommendation’s authors, the reasons for the change included a noticeable increase in breast cancer cases among women in their 40s between 2

015 and 2019, as well as a higher likelihood of late cancer diagnoses — and of dying due to breast cancer — among Black women in particular.

“The most important thing for women to know is to begin screening at the age of 40 because it just might save your life,” said USPSTF chair Carol Mangione in a video on the USPSTF website. (Before the new recommendations, USPSTF encouraged women to start screening between 40 and 50 based on their choice.)

But several experts say the new guidance is not that simple. For a test so often touted as lifesaving, there’s a surprising lack of consensus about how much mammography actually prolongs lives, and whether the unnecessary medical care mammograms sometimes lead to is worth their benefits.

Breast cancer causes more than 42,000 deaths in the US each year, and about 40 percent more Black women die due to breast cancer than White women. The causes of this disparity are complex — and recommending more mammograms to all Americans as the solution comes with increased physical, emotional, and economic costs.

Will they be worth it? Let’s get realistic about what these recommendations can and can’t do.

The new recommendations can increase rates of breast cancer diagnosis and treatment — which is both good and bad
The USPSTF’s recommendations offer guidance on how to use mammograms for screening average-risk people. That’s worth keeping in mind for a couple of reasons. First, higher-risk people — like those with family histories of breast cancer — may need to be screened differently.

Additionally, not every mammogram that gets done is for screening: These recommendations do not apply to, for example, someone who detects a lump in their breast. (In that scenario, because the person has a symptom before they get the test, the mammogram is a diagnostic test — and the procedure and what follows is a little different.)

When screening tests work, they find medical problems early, before people have any indication that they might have that problem. Mammograms can do this, flagging tiny changes associated with many types of breast cancer. Some breast cancers are slow-moving while others grow quickly; some are treatment-responsive and others are resistant to treatment; some readily invade other tissues and others resolve on their own. Mammograms can help spot them all.

But they can also detect lots of breast changes that aren’t abnormal, like benign texture differences in the breast tissue called fibrocystic changes. The trouble is, they’re not great at distinguishing what’s harmful from what’s benign. Therefore, abnormal mammograms are generally followed by a cascade of other evaluations to help determine what’s going on, depending on what they’ve found.

Increasing the number of mammograms people get can increase the number of abnormalities that get diagnosed and will lead to more follow-up ultrasounds, biopsies, and other procedures. It can also increase the number of people getting treated for conditions judged to be cancerous, with a range of surgeries, chemotherapies, and radiation treatments.

But how much more premature breast cancer death can all this added care prevent? That’s not a simple question to answer.
The thing about breast cancer is that most types are so slow-moving and so responsive to therapy that even if people don’t get breast cancer treatment until they first notice a lump, they’re still highly likely to survive cancer.

Yes, diagnosis in the early stages of these cancers leads to better outcomes than late-stage diagnoses — but those early stages last a while, with the median time around 10 years altogether.

(About one in 10 breast cancers diagnosed in the US are very aggressive and resistant to treatment, and they usually move too quickly to be caught by a mammogram done only once every two years.)

That means that for people who are vigilant about changes in their bodies and have decent access to health care, the value-add of mammograms over simply being aware of your body is pretty small.

“The better we get able to treat a disease, the less important it becomes to find it,” said Gilbert Welch, a senior researcher at Brigham and Women’s Hospital in Boston who studies cancer overdiagnosis. In other words, when we find treatments that work at many different stages of the disease, the benefits of early diagnosis shrink dramatically. For breast cancer, those treatments are the anti-hormonal therapies that emerged in the 1990s and dramatically changed breast cancer survival rates.

A recent analysis by the Lown Institute, a nonprofit healthcare think tank, highlighted some key USPSTF figures that helpfully show the limits of mammograms.

The analysis imagines a world without screening mammograms, in which women seek evaluation for breast cancer only when they notice a breast lump or other concerning symptoms. According to the USPSTF’s models, about 28 out of every 1,000 women in this world would die from breast cancer at some point in their lives.

Then, let’s look at what the previous guidelines — recommending regular mammograms starting at age 50 — can accomplish. If all women adhered to those guidelines, then seven of those breast cancer deaths would be prevented.

However, the USPSTF estimates those mammograms would also result in 1,021 false-positive mammograms and 148 biopsies that turn out to be benign. It would also lead to 10 cases of overdiagnosis — that is, treatment of cancer that never would have harmed them, to begin with.

That’s bad. False-positive mammograms are associated with psychological consequences, such as anxiety and sleep issues, that persist for years afterward. And the unnecessary surgeries, medications, and radiotherapies that follow breast cancer overdiagnosis cost the US an estimated $4 billion a year.

Now, what can we expect under the new guidelines, with the mammogram age starting at 40? The USPSTF projects that would save an additional two of those 1,000 women from breast cancer death (technically, one and a half; the authors rounded up). But significant added harms would come with all of those extra positive mammograms: This scenario leads to an additional 62 benign biopsies and two additional overdiagnosed cases.

What all of this means is that the new recommendations can potentially lead to a lot of unnecessary, and potentially harmful, care. To some people, those trade-offs, even if they harm hundreds of people, are worth saving even one extra life.

What complicates things even further is that these are calculations set in the ideal world of a statistical model. The complexity of the real world leads to some changes in math, especially for Black women.