Aims To examine whether child and adolescent psychopathology predicts subsequent tobacco use at 14 and 21 years of age.
Design Prospective birth cohort study.
Setting Data are taken from the Mater Misericordiae Hospital and University of Queensland Study of Pregnancy and its outcomes (MUSP), a prospective longitudinal study which recruited women at their first antenatal visit in Brisbane, Australia.
Participants A 5-, 14- and 21-year follow-up of children whose mother's were recruited into the MUSP birth cohort study at their first antenatal visit.
Measurements Psychopathology exposure was measured using the Achenbach's Child Behaviour Checklist (CBCL) at 5 years, the Youth Self Report (YSR) at 14 years and the Young Adult Self Report (YASR) at 21 years. Outcome measures were the children's tobacco smoking status at the 14 and 21 years' follow-up and the Composite International Diagnostic Interview (CIDI) based DSM-IV nicotine dependence at 21 years' follow-up.
Findings Externalizing symptoms had the strongest association with subsequent tobacco use. Children who met the criteria for CBCL aggression at 5 years were more likely to be tobacco smokers at the 14-year follow-up. YSR externalizing behaviours at the 14-year follow-up predicted tobacco smoking, but not DSM-IV nicotine dependence at the 21-year follow-up. Internalizing behaviour (anxiety/depression) was associated with a reduced rate of smoking at the 14- and 21-year follow-ups, but externalizing behaviour and attention problems at 14 and 21 years were associated separately and cumulatively with nicotine dependence at the 21-year follow-up.
Conclusion Childhood and adolescent psychopathology predict tobacco smoking, but some forms of psychopathology predict increased (aggression/delinquency; attention problems) and other forms decreased (anxiety/depression) smoking. There may be some benefits in targeting children with early onset aggressive/delinquent behaviour problems with tobacco smoking prevention initiatives.