Randomized population mammographic screening trials demonstrated statistically significant mortality reduction in screened women. However, in large general populations, it is unclear how screening impacts death from breast cancer. In a previous report, 75% of breast cancer deaths occurred in the small proportion of unscreened women. That conclusion needs confirmation.

6,997 invasive breast cancer diagnoses occurred in a large hospital consortium between 1990 and 1999. Among all subsequent deaths through 2007, breast cancer deaths in Massachusetts women were documented by actual review of hospital and out-patient records. Regular screening was defined as two or more screening mammograms at intervals of two years or less in asymptomatic women.

After 12.5 (8-17) years median follow-up, 461 deaths from breast cancer were confirmed. 72 deaths (15.6%) resulted from non-palpable  screen detected cancers; 44 deaths (9.6%) resulted from palpable interval cancers, a total of 116 deaths (25.2%) were in regularly screened women. 322 deaths (69.9%) occurred in women who never had screening mammography, and 23 deaths (5%) occurred after one or more previous mammograms, none within two years of diagnosis. Thus 345 breast cancer deaths (74.8%) occurred in women not regularly screened.

The most effective method of avoiding death from breast cancer is for women to participate in regular screening mammography.