[Interim results of a prospective randomized study of self-examination for early detection of breast cancer (Russia/St.Petersburg/WHO)].

Training in breast self-examination (BSE) technique involved 57,712 women, aged 40-64, at 14 out of randomly selected out-patient hospitals in St. Petersburg (1985-1989). Another 64,759 women selected at another 14 out-patient hospitals were in control. All patients with detected tumor pathology of the breast were biopsied and treated at the Institute's Clinic. The study focused on breast cancer incidence, survival and mortality. More women in the BSE group sought medical advice for suspected pathology (4,300) than those in control (2,438; p < 0.05). There were 493 cases of breast cancer in the BSE group with 157 fatalities, 446 cases of breast cancer with 167 fatalities in the control group. There was no significant difference in tumor stage. Nine-year survival (after Kaplan-Meyer) from the time of tumor detection was 65% in the study group and 55% in control (log rank 0.774; p > 0.05). There has been no significant difference in death rates in both groups for the past ten years. The study is to continue until the year 2001.