A Critical Appraisal ofthe Canadian National Breast Cancer Screening Study ’ 661

661 B REAST cancer is the leading cause of death due to cancer in most countries of the Western world, and therapy has to date had little impact on mortality from the disease. The development of mammography offered the prospect of detecting breast cancer at an earlier stage than had previously been possible; hence, this suggested that the systematic use of mammography might lead to detection of a substantial proportion of breast cancers when they are still curable and to a reduction in mortality. Several clinical trials have been conducted to test this hypothesis, and, while their results vary somewhat, they do in general indicate a reduction in breast cancer mortality as a result of screening with mammography. The magnitude of this reduction in women aged 50 years and older is about 30%; we are aware of no randomized controlled trial performed to date in younger women that showed a statistically signfficant reduction in mortality, except the Health Insurance Plan of New York. However, a recent metaanalysis of trials conducted in Sweden shows a 13% reduction in mortality for younger women (not statistically significant) (1). A 30% reduction in mortality for a disease as common as breast cancer is of