Shared Genetic Risk in the Association of Screen Time With Psychiatric Problems in Children

IMPORTANCE Children's exposure to screen time has been associated with poor mental health outcomes, yet the role of genetic factors remains largely unknown. OBJECTIVE To assess the extent of genetic confounding in the associations between screen time and attention problems or internalizing problems in preadolescent children. DESIGN, SETTING, AND PARTICIPANTS This cohort study analyzed data obtained between 2016 and 2019 from the Adolescent Brain Cognitive Development Study at 21 sites in the US. The sample included children aged 9 to 11 years of genetically assigned European ancestry with self-reported screen time. Data were analyzed between November 2021 and September 2023. EXPOSURE Child-reported daily screen time (in hours) was ascertained from questionnaires completed by the children at baseline. MAIN OUTCOMES AND MEASURES Child psychiatric problems, specifically attention and internalizing problems, were measured with the parent-completed Achenbach Child Behavior Checklist at the 1-year follow-up. Genetic sensitivity analyses model (Gsens) was used, which incorporated polygenic risk scores (PRSs) of both exposure and outcomes as well as either single-nucleotide variant (SNV; formerly single-nucleotide polymorphism)-based heritability or twin-based heritability to estimate genetic confounding. RESULTS The 4262 children in the sample included 2269 males (53.2%) with a mean (SD) age of 9.9 (0.6) years. Child screen time was associated with attention problems (beta = 0.10 SD; 95% CI, 0.07-0.13 SD) and internalizing problems (beta = 0.03 SD; 95% CI, 0.003-0.06 SD). The television time PRS was associated with child screen time (beta = 0.18 SD; 95% CI, 0.14-0.23 SD), the attention-deficit/ hyperactivity disorder PRS was associated with attention problems (beta = 0.13 SD; 95% CI, 0.10-0.16 SD), and the depression PRS was associated with internalizing problems (beta = 0.10 SD; 95% CI, 0.07-0.13 SD). These PRSs were associated with cross-traits, suggesting genetic confounding. Estimates using PRSs and SNV-based heritability showed that genetic confounding accounted for most of the association between child screen time and attention problems and for 42.7% of the association between child screen time and internalizing problems. When PRSs and twin-based heritability estimates were used, genetic confounding fully explained both associations. CONCLUSIONS AND RELEVANCE Results of this study suggest that genetic confounding may explain a substantial part of the associations between child screen time and psychiatric problems. Genetic confounding should be considered in sociobehavioral studies of modifiable factors for youth mental health.