Perinatal/neonatal intensive care can be evaluated in terms of its efficacy, effectiveness and efficiency. There is good experimental evidence from randomized controlled clinical trials that intensive-care interventions in the perinatal period are efficacious in reducing perinatal/neonatal mortality and fetal/neonatal morbidity. The effectiveness of intensive care programs in large populations has not been tested experimentally. However, population surveys show that perinatal/neonatal mortality is declining and it is likely that this is due in large part to improved perinatal care. Moreover, low birth-weight infants born in Level 3 hospitals have a lower neonatal mortality rate than LBW infants born in hospitals that are less well staffed and equipped. The true size of the reduction in fetal/neonatal mortality that is attributable to perinatal interventions is difficult to estimate from the experience of referral hospitals because of selection bias (both postnatal and prenatal). Thus, the size of the reduction in mortality resulting from the intensive care of VLBW infants is quite striking in hospital-based studies, but more modest in area-based studies. The efficiency of perinatal interventions in reducing death and disability takes into consideration both the health outcomes and the costs attributable to perinatal intensive care. Although neonatal intensive care saves lives, it is doubtful that the rate of handicap in very low birth-weight infants has been much affected. The immediate and long-term costs of neonatal intensive care are high, but a systematic economic evaluation has not yet been published.