Family physicians delivering obstetrical care in Nova Scotia were invited to participate in a fetal risk project, conducted from 1971 to 1975. Thirty percent of the physicians participated, and data were collected on 17,270 patients, using the Goodwin, Dunne and Thomas(4) fetal risk scoring system. Patients scoring four to ten on the entire system accounted for 60% of stillbirths and 68% of neonatal deaths; when the 68% of neonatal deaths; when the non-gestational portion alone was used, patients scoring three to six accounted for 52% of stillbirths and 57% of neonatal deaths. When the scoring system was altered, patients scoring three to six accounted for 61% of stillbirths and 62% of neonatal deaths. Patients of participants did not appear to fare better than those of non-participants.