59. LEVERAGING THE ATTENTION-DEFICIT/ HYPERACTIVITY DISORDER POLYGENIC SCORE TO INVESTIGATE ATTRITION IN THE NORWEGIAN MOTHER, FATHER AND CHILD COHORT STUDY

Background Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder where the core symptoms are associated with increased attrition in longitudinal studies, potentially leading to systematic bias and reduced external validity. The ADHD Polygenic Score (PGS) could provide a stable, phenotype-independent approach to studying attrition. We examined whether maternal and child ADHD PGS were associated with attrition in a large population-based longitudinal birth cohort when adjusting for sociodemographic factors. Methods Data were drawn from the Norwegian Mother, Father, and Child Cohort Study. ADHD PGS was calculated for mothers and children of European ancestry. Logistic regression models examined associations between ADHD PGS and missingness from 6 months to 8 years of follow-up. Models were adjusted for maternal age, education, and income, and compared with the base model by Likelihood Ratio tests and Nagelkerke R². Results The sample included 70,755 participants, of whom 79% had missing data at one or more time points. Attrition increased from 23.2% at 6 months to 65.0% at 5 years and remained high at 8 years (62.5%). Participants with missing data had younger mothers and lower maternal education and income. Both maternal and child ADHD PGS were consistently associated with missingness across follow-up, with maternal PGS explaining more variance (maternal PGS ΔR²=0.0039 at 0.5 years; ΔR²=0.0036 at 1.5 years; ΔR²=0.0036 at 3 years; ΔR²=0.0031 at 5 years; ΔR²=0.0037 at 8 years). Associations remained statistically significant when adjusting for socioeconomic covariates, although effect sizes were reduced. Discussion ADHD PGS in both mothers and children were consistently associated with missingness across eight years of follow-up, independent of sociodemographic factors. Although effect sizes were small, the findings indicate systematic, non-random attrition associated with neurodevelopmental genetic susceptibility. As polygenic scores are fixed across the lifespan and obtained independently of questionnaire completion, attrition or participant engagement, they are not themselves subject to attrition-related bias. As such, they may be promising tools for identifying missingness mechanisms, addressing bias, and improving imputation strategies in longitudinal psychiatric research.