RSNA 2016: Study Confirms Benefits of Screening Mammography After 75


Should women continue to get screening mammograms after age 75? A study1 presented November, 28, 2016, at the annual meeting of the Radiological Society of North America in Chicago, demonstrated unequivocal ongoing benefits of mammography after age 75.

Working with data from the ACR’s National Mammography Database, Cindy S. Lee, MD, and co-authors found increased rates of breast cancer detection among women aged 75 to 90, as well as lower recall rates compared to younger women, providing much-needed evidence for older women, referring physicians, and breast imagers on which to base screening decisions and recommendations.

The study was prompted in part by the 2009 United States Preventative Task Force guidelines for mammography, which stated that there was not enough evidence to assess whether mammography provides more benefit than harm in women older than 75. Seven years of controversy and uncertainty among patients and referrers ensued.

Lee, a breast imager at the University of California, San Francisco, frequently fields questions from women who hit age 75 and wonder whether they should return the next year for screening. The authors sought to provide the missing data that would help women aged 75 and older decide whether or not to continue breast cancer screening.

“All prior randomized, controlled trials intentionally excluded women older than 75, limiting available data to only small observational studies,” explains Lee, assistant professor in residence, University of California, San Francisco. “That’s why the task force said it didn’t have the evidence. We wanted to study and address this knowledge gap by looking at the relationship between patient age and screening performance, and how well screening mammography works as women get older.”

Four Performance Metrics

Working with the ACR’s National Mammography Database, the authors analyzed data from more than 5.6 million screening mammograms performed in 150 facilities across 31 states in the U.S. between January 2008 and December 2014. Lee et al sorted the data— representing 2.5 million women older than 40—into age cohorts in five-year increments.

The performance of screening mammography was measured by four metrics: cancer detection rate, recall rate, positive predictive value for biopsy recommended (PPV2), and biopsy performed (PPV3). Overall, the mean cancer detection rate found was 3.74 per 1,000 exams, with a recall rate of 10%, a PPV2 of 20%, and a PPV3 of 29%.

When sorting patients by age, however, Lee et al found a corresponding decline in recall rates as women grow older, as well as corresponding increases in the number of cancers detected, PPV2 rates, and PPV3 rates.

For instance, women aged 40-44 (635,202 exams) had a mean recall rate of 14%, a cancer detection rate of 1.72 per 1,000, a PPV2 rate of 8%, and a PPV3 rate of 11%. In stark contrast, women aged 75-79 (304,908 exams) had a recall rate of 7%, a cancer detection rate of 5.6 per 1,000, a PPV2 rate of 33%, and a PPV3 rate of 47%.

“As a woman gets older, we are catching more cancers and calling fewer people back,” Lee says. “We were expecting screening performance to be a little better [over time] because breast cancer is a disease of older women, but this is significantly better. Why are we stopping screening so early if benefits continue into the 90s?”

Other Considerations

From a health policy perspective, other considerations are important, Lee notes. “Per the American Cancer Society and medical specialty societies, everyone agrees that to get maximal benefit from screening you must have at least 5 to 7 years of life expectancy,” she says. “If you don’t think your patient will live for another 10 years, then screening is probably not a priority now. This really is a patient-centered decision.”

Lee recommends that physicians have a discussion with their older patients to assess the patient’s health condition, comorbidities, interest in being screened, and willingness to be treated if cancer is found.

Lee, who chairs the research subcommittee for the National Mammography Database, says further studies utilizing the NMD are planned. To encourage practices to join the NMD, the ACR has waived the fee for radiology practices meeting the following criteria:

  • all existing ACR-accredited Breast Imaging Centers of Excellence; and
  • all practices currently participating in the ACR Dose Registry Index.

Benefits of participation include bi-annual reports on physician and facility performance, with side-by-side comparison with national and regional benchmarks. It also fulfills the Physician Quality Reporting System (PQRS) requirement for CMS.


  1. Lee,C, Sengupta,D, Burleson,J, Bhargavan-Chatfield,M, Sickles,E, Burnside,E, Zuley,M, Current Era Screening Mammography Outcomes from the National Mammography Database, Involving Nearly 7 Million Examinations. Radiological Society of North America 2016 Scientific Assembly and Annual Meeting, November 27–December 2, 2016, Chicago Accessed December 8, 2016