182 Japanese and 161 Caucasian breast cancer patients participated in an epidemiologic case-control study from 1975-1980. They were subsequently followed until the end of 1987 to determine their survival status. Among the Japanese, patients with regional or distant disease had a relative risk (RR) of death of 13.0 (95% Confidence Interval (CI), 4.3-39.1) compared to those with in situ or localized disease, and obese patients had a RR of death of 3.5 (95% CI, 1.3-10.0) compared to non-obese subjects. Among the Caucasians, patients with advanced disease had a RR of death of 4.3 (95% CI, 1.8-10.5) compared to those with in situ or localized disease, and patients with a high fat intake had a RR of 3.2 (95% CI, 1.2-8.6) compared to subjects with a low fat intake. Menopausal status (pre- or postmenopausal) and replacement estrogen use were not related to survival from breast cancer in either ethnic group. When Japanese and Caucasian patients were compared with each other, there was no significant difference in survival between them.