RESUMO
OBJECTIVE: To study the prospective association between birth weight and attention problems and to explore the role of maternal body mass index (BMI) in this association. STUDY DESIGN: In 6015 children of a population-based cohort (Rotterdam, The Netherlands, 2001-2005), information on birth weight was collected and gestational age-adjusted SDS were calculated. At age 6 years, parents assessed attention problems with the Child Behavior Checklist. We used linear regression to study the association of birth weight with attention problem score and examined the modification of this association by maternal early pregnancy BMI. RESULTS: The observed association between birth weight and attention problem score was curvilinear (adjusted ß per birth weight SDS(2): 0.02, 95% CI 0.00; 0.03, P = .008); the turning point equals 3.6 kg at term. In analyses of the extreme tails of the birth weight distribution, the associations with attention problem score disappeared after adjustment for socioeconomic confounders. Maternal early pregnancy BMI moderated the association of child birth weight with attention problem score (P interaction = .007, with curvilinear term in model). CONCLUSIONS: Higher birth weight was related to less attention problems but from a birth weight of about 3.6 kg or more, a higher birth weight did not reduce the risk of attention problems any further. However, in children of obese mothers (BMI >30 kg/m(2)), high birth weight may increase the risk of attention problems.