Clinicians Comment on Controversial Mammography Study

Edith A. Perez, MD

Our recommendations for screening mammography are based on the totality of the data, and are not going to be currently modified based on the results of this one study. Of particular interest in this Canadian study is the fact that those in the screening group were actually found to have tumors of smaller size and smaller number of involved axillary nodes; these could be important facts beyond the reported overall survival data.

Carol H. Lee, MD

The Canadian National Breast Screening Study (CNBSS) recently reported in the BMJ was criticized when its results were first reported 25 years ago and largely discredited at that time. In fact, the Canadian government, which administers health care in that country, offers screening mammography to its population because the preponderance of evidence is in favor of screening.

The CNBSS is an outlier among the other seven large prospective randomized clinical trials of screening mammography, all of which showed a decrease in breast cancer mortality among the screening group. The criticisms of the CNBSS include very poor image quality, even for the 1980s, and questions about the randomization procedure with substantially more women with advanced cancers included at the start of the study in the screening arm compared to the control group.

Current digital mammography is vastly improved compared to the mammograms produced in the CNBSS. We have been screening in this country for more than 25 years and have seen a concomitant 30% decrease in breast cancer mortality. It has been argued that this is due to advances in treatment rather than in early detection through screening, but those of us who actually take care of patients with breast cancer know that small, localized breast cancers are more often successfully treated than large cancers and/or those that have metastasized.

Reports such as this in the BMJ can raise doubts among the general public about efficacy of screening mammography, which is a public health measure that has been proven to be of benefit. In my opinion, paying so much attention to an already discounted study does not serve a useful purpose.