A 75 g oral glucose tolerance test was carried out on 953 pregnant women who were identified on the basis of clinical risk factors. The tests were analysed by the WHO criteria: 826 were normal, 120 showed impaired glucose tolerance, and 7 identified diabetes. A number of obstetric and perinatal outcome measures were compared between the groups with normal and impaired glucose tolerance, and also with 135 women who had pre-existing Type 1 diabetes and delivered during the study period. There was no significant difference in the incidence of antenatal complications between mothers with normal and impaired glucose tolerance. There was a higher rate of induced labour (p < 0.05) and caesarean section (p < 0.01) in the impaired glucose tolerance group compared to the normal group, but no difference in fetal outcome or neonatal morbidity. All of these outcome measures were increased in the Type 1 diabetic pregnancies.