Background: Methodological challenges such as confounding have made the study of the early determinants of mental health morbidity problematic. This study aims to address these challenges in investigating antenatal, perinatal and postnatal risk factors for the development of mental health problems in pre-school children in a cohort of Western Australian children. Methods: The Raine Study is a prospective cohort study of 2,868 live born children involving 2,979 pregnant women recruited at 18 weeks gestation. Children were followed up at age two and five years. The Child Behaviour Checklist (CBCL) was used to measure child mental health with clinical cut-points, including internalising (withdrawn/depressed) and externalising (aggressive/destructive) behaviours (n = 1707). Results: Multinomial logistic regression analysis showed that the significant risk factors for behaviour problems at age two were the maternal experience of multiple stress events in pregnancy (OR = 1.20, 95% CI = 1.06, 1.37), smoking during pregnancy (OR = 1.30, 95% CI = 1.06, 1.59) and maternal ethnicity (OR = 3.34, 95% CI = 1.61, 6.96). At age five the experience of multiple stress events (OR = 1.17, 95% CI = 1.08, 1.27), cigarette smoking (OR = 1.19, 95% CI = 1.03, 1.37), male gender (OR = 1.43, 95% CI = 1.02, 2.00), breastfeeding for a shorter time (OR = 0.97, 95% CI = 0.94, 0.99) and multiple baby blues symptoms (OR = 1.08, 95% CI = 1.02, 1.14) were significant predictors of mental health problems. Conclusions: Early childhood mental health is significantly affected by prenatal events in addition to the child's later environment. Interventions targeting adverse prenatal, perinatal and postnatal influences can be expected to improve mental health outcomes for children in the early years.