The Canadian National Breast Screening Study: A Perspective on Criticisms

15 February 1994 | Volume 120 Issue 4 | Pages 326-334

Recently published 7-year results from the Canadian National Breast Screening Study (NBSS) generated much controversy and criticism. In women aged 40 to 49 years at entry, no reduction in breast cancer mortality was observed when screened women were compared with virtually unscreened women. In women aged 50 to 59 years, breast cancer mortality was similar when annual screening with mammography and physical examination was compared with annual screening with physical examination alone. Although NBSS results in 40- to 49-year-old women are similar to those from previously published screening studies, critics have attacked the study's design, randomization, execution, mammography, follow-up procedures, contamination of controls, and analysis. The absence of benefit observed in mammographically screened women who were 50 to 59 years old has been used to support criticism of mammography. Important facts have been ignored. The NBSS controls, aged 50 to 59 years, unlike in other studies, received thorough annual physical examinations. Cancer detection rates in both age groups were higher in the mammography than the comparison groups. Screen and interval cancer detection rates, sensitivity and specificity estimates, and prevalence to incidence ratios at first screen met or exceeded standards established by other screening studies. Claims that randomization was flawed, in particular, that more symptomatic women were assigned to mammography, are not supported by the distribution of descriptive variables collected before randomization was done. As for the "contamination" of 26% of controls aged 40 to 49 years, who reported receiving mammography, it is improbable that single or occasional diagnostic mammograms in one quarter of the control group could obliterate the benefit of four or five annual mammograms in almost 100% of the mammography group. Much remains unknown about the efficacy of breast screening.

From the Central Office, National Breast Screening Study at the University of Toronto, Ontario.
Requests for Reprints: Cornelia J. Baines, MD, 12 Queen's Park Crescent West, 3rd Floor, Toronto, Ontario, Canada M5S 1A8.
Acknowledgments: The author thanks Dr. Howard Seiden, who identified the origin of the claims that cancer detection was delayed 2 to 5 years, Dr. Andrew D. Baines, who designed Table 7 and who, along with Professor Gail McKeown-Eyssen, read and improved many drafts.
Grant Support: By the Canadian Cancer Society, Health and Welfare Canada, Heritage Fund Alberta, Manitoba Health Services Commission, Medical Research Council of Canada, Le Minister des Affaires Sociales du Quebec, National Cancer Institute of Canada, Nova Scotia Department of Health, and the Ontario Ministry of Health. The author received partial salary support from the National Cancer Institute of Canada.