OBJECTIVES: To assess the prevalence and determinants of smoking prior to and during pregnancy in Lebanon. METHODS: A cross-sectional study using two structured instruments. One instrument included information on demographic characteristics, smoking patterns in the index pregnancy and previous pregnancies, use of prenatal health services, stressful life events, and social support during pregnancy. The second was the Arabic General Health Questionnaire (GHQ-12). Women who delivered in 11 randomly selected hospitals in Beirut and its suburbs within 24 hours were asked to consent to participate in the study. The total sample interviewed was 576 women. RESULTS: The prevalence of pre-pregnancy smoking was 32% and 20% for smoking in pregnancy. Considering argileh smoking, the prevalence of tobacco use in pregnancy increased to 27% in Beirut and 25% in the suburbs. Pre-pregnancy smoking was associated with older maternal age [OR = 1.08, 95% CI (1.03, 1.14)], low and medium education [OR = 2.22, 95% CI (1.22,4.04)], increased psychiatric distress [OR = 3.11, 95% CI (1.77,5.46)], and a husband who smoked [OR = 5.00, 95% CI (2.98,8.39)]. Continued smoking during pregnancy was associated with low and medium education [OR = 3.77, 95% CI (1.31, 10.8)], younger age [OR = 1.11, 95% CI (1.02-1.20)], and a heavy pre-pregnancy smoking pattern [OR = 13.9, 95% CI (1.40,137.4)]. CONCLUSION: Policies and programs to eliminate or reduce smoking during pregnancy should be targeted toward young and less educated females and involving the spouse. Obstetricians should promote smoking cessation during pregnancy using evidence-based methods.