The safety of mammography screening and its ability to reduce mortality from breast cancer has become a highly controversial subject. The outcomes and clinical applicability of several major trials have been challenged by reanalysis of data. The most comprehensive data have been generated by Swedish studies that are population-based and focused on the role of mammography alone, without the use of breast examination. Studies from other countries have not always been open to the entire population, and some have incorporated breast examination or focused on comparison of different screening methods. Nyström and colleagues reviewed all available Swedish data on community mammography screening up to 1996 to determine the effects on breast cancer mortality.

The four major trials (Malmö, Two-County, Stockholm, and Göteborg) invited 129,750 women to participate in mammogra-phy screening and matched these women with 117,260 controls. The women ranged from 40 to 74 years of age, and the median follow-up ranged from 4.4 to 18.8 years. Collectively, 511 deaths from breast cancer occurred in women invited for screening compared with 584 in control women. These deaths were calculated to be equivalent to a reduction of 21 percent in breast cancer mortality. The greatest reduction (33 percent) was observed in women who were 60 to 69 years of age on entry to screening trials, but significant benefit was found in all groups older than 55 years. The benefit in mortality continued for several years after the end of the studies.

The authors conclude that mammography screening provides a significant, but age-dependent, reduction in mortality from breast cancer. The recent criticisms of Swedish randomized mammography screening for breast cancer are misleading and scientifically unfounded.