In the past decade the prevention and management of prematurity have begun to be addressed with more appropriate designs. A few strategies-very few-can now be recommended. A few, some widely implemented, can be abandoned. The risks and benefits of most interventions still require clarification.It is commonly agreed among clinicians, researchers, and health policy makers in industrialized countries that the crucial perinatal problem is prematurity. One response to this recognition has been an emphasis on the need for a better understanding of the pathophysiology of prematurity as the key to prevention. Eastman's (19) classic comments-"only when the factors causing prematurity are clearly understood can any intelligent attempt at prevention be made"-is still quoted by those who hold this view, and it could still be regarded as largely true.In contrast, several recent articles call for a greater sense of urgency, exhorting clinicians to stop worrying about causes and to implement preventive programs instead (32;33;36). It is tempting, in the face of stable prematurity rates and the insatiable demand for more neonatal intensive care facilities, to respond enthusiastically to the call. The only trouble with this advice is its lack of clarity about preventive programs: What is the scope of prevention and what components of which programs actually work?First, the scope for prevention is limited by the fact that more than 60% of preterm deaths occur before labor or in infants with lethal malformations (86;87). Even when the infant is alive and not malformed at the onset of labor, some preterm deliveries are elective and others are complicated by factors that contraindicate delaying the delivery (e.g., bleeding). The remaining cases of preterm birth involving spontaneous uncomplicated preterm labor or preterm premature rupture of the membranes (PPROM) with a healthy fetus offer the widest scope for prevention; in some cases, however, delivery is already inevitable at the time of hospital admission. Potential preventive efforts in the uncomplicated group include true primary prevention, early identification of preterm labor or its premonitory signs, and inhibition of preterm labor.Though it is often blurred in practice, the distinction between prevention and inhibition of preterm labor is an important one because the potential for net negative outcomes is even greater with prophylactic interventions. These interventions are used in women at risk, rather than in ones with an established problem, and are used for a greater proportion of the pregnancy.