In this investigation, 45 women with twin gestations in four centers were randomly assigned to either daily home uterine activity monitoring and perinatal nursing support (19) or an education group (26). Patients in the home uterine activity monitoring and perinatal nursing support group monitored uterine activity twice daily, transmitted the data each day, and had 24 hour-a-day access to nursing support. Education patients were counseled regarding the common signs and symptoms of preterm labor. Both groups had scheduled prenatal visits at least every 2 weeks. Sixteen (62%) of the education group and 14 (74%) of the home uterine activity monitoring and perinatal nursing support group developed preterm labor, values that were not significantly different. Of the monitored group who experienced preterm labor, all 14 were dilated 3 cm or less at diagnosis of the first preterm labor, compared with ten of 16 in the education group (P = .01; one-tailed Fisher test). The mean cervical dilatation at the first preterm labor episode in the group receiving daily monitoring and contact (1.6 cm) was significantly less (P = .01) than that in the education group (2.9 cm). Thus, fewer preterm births were recorded in the home uterine activity monitoring and perinatal nursing support group and significantly fewer patients delivered because of failed tocolysis (P = .03).