Psychiatric symptoms at age 8 as predictors of specialized health service use for psychiatric disorders in late adolescence and early adulthood: findings from the Finnish Nationwide 1981 Birth Cohort Study.

Background: A majority of adult psychiatric patients have suffered of psychiatric symptoms already during childhood or adolescence. Previous studies on continuity of mental health problems have included only children with diagnoses and thus missed symptomatic children without mental health service contact, or as in longitudinal cohort studies, have suffered from considerable losses of subjects during follow-up. The aim of the present study was to examine the association between multi-informant ratings of childhood psychiatric symptoms and cumulative incidences of diagnosed mental disorders in early adulthood in a large population-based cohort. Methods: The original sample consisted of 6,017 Finnish children born in 1981. Parents and teachers completed the Rutter questionnaire, while the children completed the Children's Depression Inventory (CDI) at the age of 8. The follow-up information was obtained from the Finnish Care Register for Health Care. Analyses were done with the Cox regression model. Results: A total of 2,717 males (12.4%) and 2,683 females (12.8%) were diagnosed with a mental disorder in early adulthood. Parent-/teacher-reported high levels (above the 90th percentile) of conduct problems (males, Hazard Ratio (HR)=2.5, 95% CI=1.6-3.7; females, HR=1.5, 95% CI=1.01-2.1) and anxiety (males, HR=1.5, 95% CI=1.1-2.1, females, HR=1.7, 95% CI=1.2-2.4), and child self-reported depressive problems (males, HR=1.7, 95% CI=1.2-2.3; females, HR=1.6, 95% CI=1.1-2.2) were key predictors of any mental disorder in early adulthood. Conduct problems predicted psychotic disorders and substance-related disorders among males. Anxiety problems predicted psychotic disorders, anxiety, and depression. Child self-reported depressive problems predicted male depression. Attention-deficit hyperactivity disorder (ADHD) symptoms and low school performance did not predict any outcome in multivariate analyses. A non-nuclear family living situation at age 8 predicted most outcomes. Conclusions: Adult-reported conduct and emotional problems in children, but not ADHD symptoms, independently predicted mental health service use and any psychiatric diagnosis in late adolescence and early adulthood, emphasizing the need for early identification of childhood mental health problems. Similarly, child self-reports of depressive problems already at age 8 predicted adult outcomes. Our findings emphasize the importance of multi-informant assessment and early targeted interventions for conduct and emotional problems in early school years.