Session: Adolescent neurodevelopment, behaviour and mental health
Influence of pubertal development on behavioral disorders in adolescents, mediated by body mass index (BMI) – evidence from a large British cohort (Luciana Maestri Karoleski, Federal University of Sao Paulo / University of Pisa)
Puberty involves cognitive and biological maturation, often coinciding with an increased incidence of behavioral disorders. Objective: To investigate whether BMI mediates the association between pubertal development and behavioral outcomes. Methodology: Using Millennium Cohort Study data (n = 5,838), counterfactual mediation analyses examined the direct and indirect effects of the pubertal latent trait (ages 11 and 14) on behavioral disorders at age 17, with BMI as the mediator. Results: At age 11, puberty showed direct effects on emotional symptoms and peer problems. By age 14, a full mediation effect via BMI was observed for peer relationship problems, while hyperactivity showed partial mediation. Conclusions: These findings provide empirical evidence for the association between pubertal timing and adolescent behavioral health and highlight the significant mediating role of BMI.
Characterising the longitudinal relationship between social media use and psychiatric diagnoses in secondary care in adolescents in England (Tom Metherell, UCL Centre for Longitudinal Studies)
Concerns that excessive use of social media platforms may be driving increased incidence of mental health conditions in adolescents have attracted significant media and research attention. Using the linkage of the NHS Healthcare Episode Statistics (HES) dataset to the Millennium Cohort Study, we aimed to relate measures of social media use with health service use related to diagnosed mental health disorders. Our modelling strategy was informed by co-production with advisors with lived experience of mental distress and social media use.
We did not find any significant relationship between social media use and instances of mental health service use between ages 11 and 14 or between ages 14 and 17. However, post hoc exploratory analyses suggested a possible non-linear relationship. We conclude that a more nuanced consideration of online risks and harms is needed.
Disentangling late-onset ADHD in adolescence: Informant-specific evidence from polygenic risk, neurodevelopmental CNVs, and childhood resources (Egle Padaigaite-Gulbiniene, Cardiff University)
Using data from the Millennium Cohort Study, we aimed to understand differences between childhood-onset and late/adolescent-onset ADHD. We compared late-onset ADHD to childhood-onset persistent ADHD in their associations with genetic liability and childhood resources. We also examined whether these associations differ by informant. ADHD symptoms were assessed at ages 7, 11, and 17 using parent- and self-rated scores on the Strengths and Difficulties Questionnaire. Compared to childhood-onset persistent ADHD, parent- and self-rated late-onset ADHD was characterised by higher childhood internal and external resources, while self-rated late-onset ADHD by lower genetic liability to ADHD. Our results suggest that late-onset ADHD may onset later because of childhood resources, and that self-rated late-onset ADHD may represent a partly distinct phenotype than childhood onset ADHD.
Investigating the associations between attention-deficit hyperactivity disorder (ADHD) symptoms and physical activity across childhood and adolescence in a representative UK birth cohort (Amandine Senequier, Queen Mary University of London)
Young people with attention-deficit hyperactivity disorder (ADHD) tend to have higher physical activity (PA) levels than their peers, yet how this relationship varies across development and by sex is unclear. Using Millennium Cohort Study data from ages 7-17 (n=10,757-14,043), we examined associations between ADHD symptoms and PA measured via accelerometery and questionnaires. Higher accelerometer-measured moderate-to-vigorous PA (MVPA) was associated with higher ADHD symptoms at ages 7 (b=0.009, p<0.001) and 14 (b=0.004, p=0.004), particularly in males. Mixed-effects models revealed no change in associations over time. MVPA at age 7 predicted higher ADHD symptoms in later adolescence. Higher reported PA at ages 7, 11 and 14 was associated with lower ADHD symptoms. Findings highlight the importance of developmental stage and sex in understanding the links between ADHD and PA.
Social Media Use in Early Adolescence and Mental Health in Young Adulthood: The Roles of Distrust, Social Isolation, and Polygenic Risk for Psychopathology (Dimitris Tsomokos, UCL Institute of Education)
Using representative Millennium Cohort Study data from the UK (N = 8,403; multiply imputed), we tested whether social media use (SMU) at age 11 predicts psychological distress at age 23 via distrust and social isolation at age 14. Higher SMU predicted greater distrust (a = 0.035, p = 0.043) but not social isolation or distress directly. Both mediators predicted distress (distrust: b = 0.267; social isolation: b = 0.557; both p < .001), with a significant indirect effect through distrust only (ab = 0.009, p = 0.045). In a genotyped subsample (N = 3,960), polygenic risk scores (MDD, ADHD, neuroticism, externalising) attenuated the distrust pathway; a significant negative direct effect of SMU on distress also emerged (c = ?0.135, p = 0.016). Distrust may link early SMU to young-adult psychological distress, though genetic liability for psychopathology confounds this association.
