M.Tech. / It is difficult to manage a pregnancy when fetal age, health status or potential pregnancy risks are not known. The aim of this study was to assess the impact of routine ultrasound screening as compared to selective use of ultrasound in the mid-trimester of pregnancy, on women using South African government health services. The three objectives investigated included assessment of the availability of reliable menstrual histories in the study population, the influence of ultrasound dating on obstetric management and the effect of ultrasound on pregnancy outcome due to the early detection of high risk pregnancies. A cluster randomised trial of 962 women was performed to test the hypothesis that midtrimester routine ultrasound screening in low risk pregnancies would result in improved antenatal care and perinatal outcome. Groups of eligible pregnant women were randomly selected to have either a routine scan followed by normal antenatal care or routine antenatal care which only allows for the selective use of ultrasound, in line with South African Antenatal Care Policy. Statistical analysis of the results confirmed that ultrasound dating is a more accurate predictor of the expected date of delivery when compared to other dating methods. Improved pregnancy dating resulted in a significant reduction in induction of labour for post-term pregnancy in the ultrasound screening group, suggesting a positive effect of ultrasound screening on obstetric management. No improvement was demonstrated in perinatal morbidity or mortality. The early detection of anomalous fetuses only led to one therapeutic abortion. The study did not possess the statistical power to demonstrate improved outcomes when multiple pregnancies were detected early in pregnancy. Currently there appears to be no urgent need to implement a routine antenatal screening programme in the Gauteng public health sector. In view of the fact that a third of the participants indicated that they were unsure of menstrual dates, and one third of the participants in the ultrasound screening group presented with an unreliable menstrual history, it is proposed that unsure dates be considered as a valid indication for the selective use of ultrasound in mid-trimester pregnancy.