Genetic Associations Between Childhood Psychopathology and Adult Depression and Associated Traits in 42 998 Individuals: A Meta-Analysis

Importance: Adult mood disorders are often preceded by behavioural and emotional problems in childhood. It is yet unclear what explains the associations between childhood psychopathology and adult traits.
Objective: To investigate whether genetic risk for adult mood disorders and related traits influence childhood disorders.
Design: This cohort study meta-analysed data from seven ongoing longitudinal birth and childhood cohorts. The starts of data collection varied, ranging from July 1985 to April 2002. Data analysis was performed from September 2017 to May 2019.
Setting: Seven population-based cohorts from the UK, the Netherlands, Sweden, Norway and Finland.
Participants: 42,998 participants were repeatedly assessed for childhood psychopathology from age 6 to 17 years.
Exposures: Individual polygenic scores (PGS) were constructed in children based on well-powered genome-wide association studies (GWAS) of adult major depression (MD) and bipolar disorder (BD), as well as subjective well-being (SWB), neuroticism, insomnia, educational attainment (EA), and body mass index (BMI).
Main Outcomes and Measures: We used regression meta-analyses to test the associations between the PGS and symptoms of attention-deficit/hyperactivity disorder (ADHD), internalizing problems, and social problems, measured repeatedly across childhood and adolescence, and whether these associations were dependent on childhood phenotype, age, and rater.
Results: We obtained a sample size of 42,998 unique participants aged 6 to 17. The percentage of male participants ranged from 43% to 52% across all cohorts. PGS of adult MD, neuroticism, BMI, and insomnia were positively associated with childhood psychopathology (beta estimates ranged from 0.023 – 0.042, C.I = 0.017 – 0.049) while associations with PGS of SWB and EA were negative (-0.026 – -0.046 (-0.020 – -0.057)). BD PGS did not show any association. We found no effect of age, type of childhood phenotype or rater on the associations except for BMI PGS and EA PGS. We report stronger associations between EA PGS and ADHD compared to internalizing and social problems, and between BMI PGS and social problems and ADHD compared to internalizing. Furthermore, the association between EA PGS and ADHD increased with age.
Conclusions and Relevance: Results from this study suggest the existence of a set of genetic factors influencing a range of traits across the lifespan with stable effects present throughout childhood. Knowledge of underlying mechanisms may impact treatment and long-term outcomes of individuals with psychopathology.