The objective of the study was to identify factors which predispose women to smoking relapse postpartum and the factors that prevent the relapse, in order to design effective interventions to reduce the rate of smoking relapse after delivery. A prospective cohort study was conducted in 2004 and 2005 in Lodz, Poland. One hundred thirty-eight women who quit smoking for the pregnancy period were enrolled into the study between 32 and 36 weeks of pregnancy and were followed up at 3 months after delivery. Self-reported non-smoking status at enrollment and postpartum was verified using saliva cotinine level analyzed by liquid chromatography with tandem mass spectrometry (LC-MS/MS). The cut-off point of 10 ng/ml was adopted for the saliva cotinine level. About half of the study population relapsed into smoking within 3 months after delivery. Most (49%) relapses occurred in the first 4 weeks postpartum. Three factors were identified as directly influencing smoking relapse after delivery: need to cope with stressful situations (OR = 4.7; 95% CI 1.7-14.3), type of quitting attempt (OR = 4.0; 95% CI 1.1-16.1), and smoking environment at home (OR = 7.2; 95% CI 2.2-28.3). These three factors accounted for 84.0% (95% CI 52.7-96.1) of relapses. The profile of women who relapse to smoking after delivery can be described as quitting smoking only for pregnancy and postpartum period, resorting to smoking in response to stressful situations, and living with smokers. Those factors should be taken into account while conducting antismoking intervention for pregnant and postpartum women.