Primiparae were randomly assigned to two conditions of routine ultrasound examination at their first antenatal clinic visit: (a) high feedback ultrasound where the monitor screen was visible and the patient was shown the foetal size, shape and movement (N = 67); (b) low feedback ultrasound where the screen was not seen and specific verbal feedback was denied (N = 62). Women were interviewed at 16 weeks gestation. Those receiving high feedback were more likely to report that they had acted on health advice given at the first antenatal visit to reduce their smoking and drinking.