The aim of this study was to assess the relationship between preterm/early preterm delivery and active smoking as well as environmental tobacco smoke (ETS) exposure in a sample of pregnant Italian women. A case-control study was conducted in nine cities in Italy between October 1999 and September 2000. Cases of preterm birth were singleton babies born before the 37th gestational week; babies born before the 35th gestational week were considered early preterm births. Controls were babies with gestational ages >or= 37th week. A total of 299 preterm cases (including 105 early preterm) and 855 controls were analysed. A self-administered questionnaire was used to assess active smoking and ETS exposure, as well as potential confounders. Multivariable logistic regression analysis showed a relationship between active smoking during pregnancy and preterm/early preterm delivery [adjusted ORs: 1.53; 95% CI 1.05, 2.21 and 2.00; 95% CI 1.16, 3.45, respectively]. A dose-response relationship was found for the number of cigarettes smoked daily. The adjusted ORs were 1.54 and 1.69 for preterm babies and 1.90 and 2.46 for early preterm babies for 1-10 and >10 cigarettes/day respectively. ETS exposure was associated with early preterm delivery [adjusted OR 1.56; 95% CI 0.99, 2.46] with a dose-response relationship with the number of smokers in the home. Smoking during pregnancy was strongly associated with preterm delivery with a dose-response effect. ETS exposure in non-smoking women was associated only with early preterm delivery.