The New England Journal of Medicine
©Copyright, 1998, by the Massachusetts Medical Society
VOLUME 338 A PRIL 16, 1998 NUMBER 16
J OANN G. E LMORE , M.D., M.P.H., M ARY B. B ARTON , M.D., M.P.P., V ICTORIA M. M OCERI , P H .C., S ARAH P OLK , B.A., P HILIP J. A RENA , M.D., AND S UZANNE W. F LETCHER , M.D.
A BSTRACT Background The cumulative risk of a false positiveresult of a breast-cancer screening test is unknown. Methods We performed a 10-year retrospective co-hort study of breast-cancer screening and diagnosticevaluations among 2400 women who were 40 to 69years old at study entry. Mammograms or clinicalbreast examinations that were interpreted as indeter-minate, aroused a suspicion of cancer, or promptedrecommendations for additional workup in women inwhom breast cancer was not diagnosed within thenext year were considered to be false positive tests. Results A total of 9762 screening mammogramsand 10,905 screening clinical breast examinationswere performed, for a median of 4 mammograms and5 clinical breast examinations per woman over the 10-year period. Of the women who were screened, 23.8percent had at least one false positive mammogram,13.4 percent had at least one false positive breast ex-amination, and 31.7 percent had at least one falsepositive result for either test. The estimated cumula-tive risk of a false positive result was 49.1 percent (95percent confidence interval, 40.3 to 64.1 percent) after10 mammograms and 22.3 percent (95 percent confi-dence interval, 19.2 to 27.5 percent) after 10 clinicalbreast examinations. The false positive tests led to870 outpatient appointments, 539 diagnostic mam-mograms, 186 ultrasound examinations, 188 biopsies,and 1 hospitalization. We estimate that among wom-en who do not have breast cancer, 18.6 percent (95percent confidence interval, 9.8 to 41.2 percent) willundergo a biopsy after 10 mammograms, and 6.2 per-cent (95 percent confidence interval, 3.7 to 11.2 per-cent) after 10 clinical breast examinations. For every$100 spent for screening, an additional $33 was spentto evaluate the false positive results.
To read the rest of the study, click here to download the PDF.