Join the 200th Anniversary Celebration

Correspondence

Computer-Aided Mammography Screening

N Engl J Med 2009; 360:836February 19, 2009

Article

To the Editor:

The article on screening mammography by Gilbert et al. (Oct. 16 issue)1 favorably compares single reading plus computer-aided detection with double reading in the United Kingdom. This finding will have no direct effect in the United States, where double reading is rarely performed. However, it is an indirect endorsement of computer-aided detection, because evidence supporting it is less robust than that supporting double reading.2

Major differences exist between European and U.S. screening programs, despite approximately equal cancer-detection rates. European readers are much more experienced, interpreting more than 10 times as many mammographic examinations as U.S readers. European recall rates are 3 to 5%, as compared with U.S. rates of 8 to 12%, and there are fewer biopsies and higher positive predictive values with European readers than with U.S. readers. Digital screening and computer-aided detection are uncommon in Europe. Most European programs offer screening every 2 to 3 years from the ages of 50 to 70 years, whereas U.S. programs provide annual screening beginning at 40 years of age, with no upper age limit.

These widely varying screening methods and guidelines underscore the point that screening is not simply a matter of improving detection. Equally important is determining the cost per life-year saved with comparative programs, allowing countries to balance their individual health care priorities, financial resources, and specific medicolegal and other societal issues.

Ferris M. Hall, M.D.
Beth Israel Deaconess Medical Center, Boston, MA 02215

2 References
  1. 1

    Gilbert FJ, Astley SM, Gillan MGC, et al. Single reading with computer-aided detection for screening mammography. N Engl J Med 2008;359:1675-1684
    Full Text | Web of Science | Medline

  2. 2

    Fenton JJ, Taplin SH, Carney PA, et al. Influence of computer-aided detection on performance of screening mammography. N Engl J Med 2007;356:1399-1409
    Full Text | Web of Science | Medline