Session: Genetic, Biological, and Social Dimensions of Child and Adolescent Development
A Genetically Adjusted Propensity Score (GAPS) Analysis of Adverse Childhood Experiences and Adolescent Psychopathology (Athena Chow, UCL)
Adverse childhood experiences (ACEs) are associated with psychopathology, but the extent to which these associations are confounded by genetic and environmental vulnerabilities remains unclear. Using data from 6,431 children in the UK Millennium Cohort Study, we investigated (1) whether children exposed to ACEs have pre-existing genetic and environmental vulnerabilities, and (2) whether ACEs remain associated with psychopathology after adjusting for these vulnerabilities using genetically adjusted propensity score (GAPS) matching. Genetic liabilities for psychopathology and lower educational attainment, alongside socioeconomic disadvantage, prenatal risks, and socioemotional difficulties were associated with increased odds of ACE exposure. Following adjustment for genetic and environmental vulnerabilities, parental psychopathology, emotional neglect, and interpersonal victimisation remained significantly associated with adolescent psychopathology; conversely, household-level ACEs (e.g., parent substance use, divorce, domestic violence) were largely attenuated. Our findings suggest that parental psychopathology, neglect, and victimisation impact adolescent psychopathology independently of genetic and environmental confounding, highlighting them as critical intervention targets.
Nature and Nurture in the Production of Childhood BMI (Nancy Daza Baez, UCL)
We estimate a dynamic production model for childhood BMI and body fat percentage that integrates trio genotypes (mother–father–child), parental investments in diet and exercise, and local environmental factors. Using the UK MCS at ages 3–17 with polygenic scores, regional food prices, and spatial access measures, we apply a control-function approach exploiting random genetic segregation and exogenous cost shifters to identify causal investment effects. Three findings stand out. First, BMI and body fat exhibit strong state dependence, implying early divergences compound through adolescence. Second, the child’s own genetic risk rises in importance with age, whereas parental genetic effects attenuate, consistent with direct genetic influence dominating environmental transmission. Third, returns to investment are age-specific.
Youngest in the school year: Exploring how trajectories of socio-emotional development differ by month-of-birth in England (Maxime Perrott, Education Policy Institute and University of Bristol)
This paper visualises trajectories of socio-emotional development by month-of-birth between ages 3 and 17, using multi-level growth curve models applied to Millennium Cohort Study (MCS) data. The research explores when month-of-birth gaps emerge and how they evolve in parent-measured Strength and Difficulties Questionnaire (SDQ) outcomes. Findings suggest gaps emerge during early childhood (between ages 5 and 7), peak during late-childhood and adolescence (between ages 11 and 14) then close when young people leave compulsory schooling (at age 17). While, month-of-birth differences emerge earliest for Emotional Symptoms at age 5, they are widest for Hyperactivity/Inattention at age 11. Finally, the analysis contrasts parent and teacher-measured SDQ outcomes at ages 7 and 11, revealing particularly large month-of-birth gaps in teacher-measured Hyperactivity/Inattention symptoms. Overall, these findings suggest that being the youngest in the school year has a cumulative effect on socio-emotional difficulties, an effect that is compounded by the transition into secondary school.
Do early internalising and externalising problems mediate the effects of the educational attainment polygenic index (Chloe Austerberry, University of Cambridge)
The educational attainment polygenic index (EA-PGI) is widely used, yet the pathways linking it to educational outcomes remain unclear. Internalising and externalising problems plausibly mediate this link, but very few studies have directly tested this mediation. We addressed this gap by analysing data from 8,014 parent–offspring trios in the Millennium Cohort Study, testing whether parent-child EA-PGI associations with educational performance at ages 7 and 17 were mediated by parent-reported child internalising and externalising problems at ages 3 and 5. We found that externalising—but not internalising—problems significantly mediated these associations, suggesting that early child externalising problems may represent one pathway linking education-associated genetic propensity (of children and their parents) to children’s later educational performance.
