MATERNAL SMOKING DURING PREGNANCY AND OFFSPRINGʼS ATTENTION DEFICIT HYPERACTIVITY DISORDER SYMPTOMS AT AGE 15 YEARS

Background Exposure to tobacco smoke in utero is suspected to be associated with attention deficit hyperactivity disorder (ADHD) and ADHD symptoms in the offspring. Other lifestyle factors during pregnancy may also associate with these disorders. Many studies have suffered from methodological shortcomings, such as recall bias, inaccurate exposure assessment, low statistical power and lack of or insufficient control of confounders. Objective We investigated the association between maternal smoking during pregnancy and ADHD symptoms in adolescence. Method The geographically defined general population based study consisted of 6888 15-year-old adolescents with exposure and outcome data (74% of eligible) from the Northern Finland Birth Cohort born in 1985–86 (NFBC1986). Mothers provided information during pregnancy on their smoking and social standing (SES). At child's age 15 years, parent's assessed adolescent's behaviour by a SWAN questionnaire developed in USA. Scores on the three SWAN subscales-Inattentive (INN), Hyperactive-Impulsive (HI) or Combined (C) are defined from 18 items on the questionnaire. A subject was defined to have potentially ADHD (“ADHD symptoms”), if he/she falls in the upper 5%ile of INN, HI, or C subscales. Unadjusted analyses and stratification were used to study associations and confounding of a wide variety of theoretically associated factors. A multivariate logistic regression model was fitted to assess the independent association between smoking and outcome. Results In unadjusted analyses maternal smoking (yes vs. no), family structure (single parent or reconstructed family vs. always two-parent family), young maternal age at child's birth (< 19y) but not alcohol use during pregnancy (yes vs. no) were significantly associated by about 2-fold risk of ADHD symptoms. Maternal smoking was independently associated with increased risk of ADHD (OR 1.4, 95%CI 1.03–1.8) when adjusted for sex and family structure. After additional adjustment for SES, maternal age and alcohol use during pregnancy, the risk for ADHD symptoms was 1.3 times that of those not exposed to maternal smoking (95%CI 1.0–1.7) Conclusion This study, the largest population-based prospective follow-up on fetal tobacco smoke exposure and later behavioural disorders by age 15 years, supports suggested link between maternal smoking during gestation and ADHD symptoms. The known correlation between family structure, SES and maternal age and the fact that ADHD symptoms fade with age making them more difficult to detect in screening may explain that the association in the “fully adjusted” model was only marginally significant. Prenatal care should encourage pregnant mothers to abstain from smoking during pregnancy.