Mammography Still Saves Lives

Daniel B. Kopans, MD


February 20, 2014

In This Article

Editor's Note:
In a recent article published in JAMA Internal Medicine, H. Gilbert Welch, MD, MPH, and Honor J. Passow, PhD, present their estimates of the benefits and harms of screening mammography by calculating reductions in 3 outcomes (breast cancer death, false-positive results, and overdiagnoses) in a hypothetical population of 1000 women aged 40-60 years who were screened for 10 years. Using data from previous studies, they estimate a mortality reduction between 0.3 and 3.2 lives per 1000 women. They estimate that 490-670 recalls and 70-100 false-positive biopsies would take place per 1000 women over a 10-year course of screening mammography, with 14 cases of overdiagnosis of breast cancer. Finally, they estimate that reducing mammography to every 2 years would cut overdiagnosis of breast cancer in half.[1]

Medscape invited Dr. Daniel Kopans, Professor, Department of Radiology at Harvard Medical School, and Senior Radiologist in the Breast Imaging Division at Massachusetts General Hospital, to respond to Drs. Welch and Passow's conclusions. Dr. Kopans has conducted many studies on breast imaging in breast cancer and screening mammography.

An Old Argument Resurfaces

The recent study by Welch and Passow[1] published in JAMA Internal Medicine is another step in the ongoing effort to discourage women from having mammography and their physicians from recommending it. Dr. Welch, who has published similar analyses before,[2,3] has once again made the argument that most people being screened for a cancer will not benefit from the screening because most people do not develop the cancer each year or, in fact, over many years.

This argument first dates to the 1990s, when it was used to discourage the use of breast cancer screening. At that time, I gave it the name "dilutional nihilism,"[4] to reflect the effect of taking an important benefit for individuals and minimizing it by dividing the benefit over the entire population. For example, assuming seatbelts save the lives of 12,000 of us each year, this amounts to a benefit for "only" 0.003% of the population. To "dilutional nihilists," this would hardly seem worth the expense of installing seatbelts and the inconvenience of wearing them.

Would proponents of this perspective argue that because only 3% of women who die each year in the United States die of breast cancer, then a reduction of 30% would "only" reduce all deaths by 1%? Welch and Passow's use of "absolute" risks is reasonable if you want to compare the financial costs of various interventions.

"Quality-adjusted life-years saved (QALYS)" is a measure used routinely to compare interventions. If 1 woman's life is saved and she lives for 30 years, and 1000 women are needed to be screened to save that 1 life, then it is like saying that everyone got an additional 0.03 years of life (30/1000 = 0.03). Saving 1 life out of 1000 is like giving each of 1000 women an additional 11 days (0.03 × 365 = 11) of life. These types of analyses have been used to confuse the uninformed who think that the intervention only added 11 total days of life.

Unless an infection is prevented from spreading, or we as individuals have need of healthcare, none of us benefits directly from the hundreds of billions of dollars spent each year on the healthcare provided to others. Does this mean that we should abandon all healthcare that benefits only the individual? Those who argue against breast cancer screening would have us ignore the fact that when a woman is diagnosed with breast cancer, it is 100% for that woman. A 30% reduction in breast cancer deaths each year means that 10,000-15,000 fewer women die each year from breast cancer in the United States.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.