The results of a randomized clinical trial of a prenatal self-help smoking cessation program are reported in terms of the pregnancy and cost outcomes. The study population were the socioeconomically and ethnically diverse members of a large health maintenance organization (HMO) who reported that they were smoking at the time of their first prenatal visit. The intervention consisted predominantly of printed materials received through the mail. Compared with the usual care control group, women assigned to the self-help program were more likely to achieve cessation for the majority of their pregnancy (22.2 percent versus 8.6 percent), gave birth to infants weighing on average 57 grams more, and were 45 percent less likely to deliver a low birth weight infant. An economic evaluation of the self-help program was conducted from the perspective of the sponsoring HMO. Based upon the expenditures associated with the neonates' initial hospital episode, the intervention had a benefit-cost ratio of 2.8:1. These findings provide strong evidence to support widespread incorporation of smoking cessation interventions as a standard component of prenatal care.