We carried out a randomized trial of invitation to screening mammography in the city of Gothenburg, Sweden, to estimate the effect of screening on breast cancer mortality in women under age 50 years. A total of 11,724 women aged 39-49 were randomized to the study group, which was invited to mammographic screening every 18 months; 14,217 women in the same age range were randomized to a control group, which was not invited to screening until the fifth screen of the study group. Breast cancers diagnosed in both groups between randomization and immediately after the first screen of the control group were followed up for death from breast cancer to the end of December 1994. There was a significant 44% reduction in mortality from breast cancer in the study group compared to the control group (relative risk [RR] = 0.56, P = 0.042, 95% confidence interval [CI]: 0.32-0.98). A conservative estimate based on removal of the cancers detected at the first screen of the control group gave an RR = 0.59 (P = 0.069, 95% CI: 0.33-1.05). The true answer is likely to lie between the two estimates. These data suggest that mammographic screening can reduce breast cancer mortality in women under age 50, particularly if high-quality mammography is used and a short interscreening interval is adhered to.