Abstract
Aims: To study the prospective relationship between maternal smoking during pregnancy (MSP) and behavioral problems, heavy alcohol use, daily smoking, and ever use of cannabis in the offspring, and to assess the role of confounding and mediating factors in a systematic way. Methods: Population-based cohort study of 2,230 respondents, starting in 2001 when respondents were around the age of 11 years, and two follow-up measurements at intervals of about 2.5 years (response rates of 96.0 and 81.4%). Results: Almost one third of the respondents’ mothers had smoked tobacco during pregnancy. These respondents were at an increased risk for all outcomes except internalizing problems (significant odds ratios ranged from 1.40 to 2.97). The successive models showed that the potential confounding factors reduced the strength of all relationships. In the full model, the strongest relationship was found for mothers who smoked more than 10 cigarettes a day during pregnancy and daily smoking in early adolescence (odds ratio: 1.56), but none of the relationships were statistically significant. Conclusions: MSP is a marker for future behavioral outcomes in the offspring, but reducing the prevalence of MSP is unlikely to make a meaningful contribution to the prevention of these problems in adolescents.