Identification of high risk is only moderately successful at the booking visit; most risk factors only give a relative risk of around three, so that most of the high-risk group do not experience the adverse outcome, and most adverse outcomes occur in low-risk women. Risk factors are useful in planning for confinement and extra care, but since new problems can arise at any time and most antenatal admissions are for conditions arising in spite of antenatal care (Chng et al, 1980), some care should be offered to all women. Traditional schedules of care, however, have no scientific justification. Identification of low risk is also fallible and there seems to be an irreducible minimum of unpredictable problems which will arise even in low-risk women. Methods need to be found to reduce the lack of continuity which often results from unscheduled transfers of care.