AIMS: To study the predictive associations between maternal smoking and the impact of quitting smoking during pregnancy and offspring daily smoking at age 15-16 years. DESIGN: The Northern Finland Birth Cohort 1986 (NFBC86) includes 99% of all births in the region and has an ongoing follow-up. Data were collected using questionnaires at 24th gestational week during pregnancy and after delivery, and at follow-up in 2001-02, when the offspring were aged 15-16 years. SETTING: Northern Finland. PARTICIPANTS: NFBC86 included 9432 live born children. Data regarding maternal smoking during pregnancy and offspring smoking at age 15-16 years were available for 4462 subjects (47.3% of the original sample). MEASUREMENTS: The outcome was offspring's self-reported daily smoking. Maternal smoking during pregnancy was considered using a four-class variable: (1) no smoking, (2) mother had smoked, but had quit smoking before becoming pregnant, (3) mother quit smoking during the 1st trimester and (4) mother quit smoking after the 1st trimester or continued smoking throughout the pregnancy. Information regarding paternal smoking during pregnancy, maternal and paternal smoking and education level, family structure and dwelling at offspring's age 15-16 years were considered potential confounding variables. FINDINGS: Continuing smoking after the 1st trimester increased the odds of daily smoking among offspring, independently of confounding factors [odds ratio (OR) = 1.8, 95% confidence interval (CI) = 1.3-2.5]. Continuing to smoke after the 1st trimester was associated with higher odds compared with quitting smoking during the 1st trimester. Also, parental smoking at offspring age 15-16 years increased the odds of offspring daily smoking, independently of prenatal smoking exposure. CONCLUSIONS: Prenatal smoking exposure increases the risk for offspring adolescent daily smoking. Quitting smoking during the early stages of pregnancy may decrease the odds for offspring smoking.