Abstract
Background We tested the utility of the SWAN (Strengths and Weaknesses of ADHD Symptoms and Normal Behavior Rating Scale) for measuring attention-deficit/hyperactivity disorder (ADHD) traits in population-based genetics. We examined the convergent, predictive, and discriminant validity of the SWAN parent‐ and new youth-report scale while creating norms and clinical cut-offs. We tested if high ADHD traits were associated with ADHD diagnosis and polygenic risk and if low ADHD traits pointed to another psychopathological phenotype or genetic risk.
Methods We collected parent‐ and youth-reported SWAN scores in a community sample (n=15,560; 6-18 years of age). Sensitivity-specificity analyses determined SWAN scores that discriminated a community ADHD diagnosis (n=972). Cut-offs were validated in a clinic sample (266 ADHD patients; 36 controls). We examined the relationship of SWAN scores with anxiety, obsessive-compulsive (OC) traits as well as ADHD, obsessive-compulsive disorder (OCD), and anxiety disorder using polygenic risk scores.
Results SWAN scores showed high convergent validity and distinguished ADHD participants with high sensitivity and specificity in the community sample. The community-based threshold discriminated ADHD clinic cases from controls with a sensitivity of 86% and specificity of 94%. High ADHD traits were associated with high anxiety, but not OC, traits. High SWAN scores and those above the community-based cut-off were only associated with ADHD polygenic risk.
Conclusions The SWAN is useful for genetic research because it predicts ADHD diagnoses with high sensitivity and specificity and is associated with ADHD polygenic risk. Cut-off values and norms are presented. Low ADHD traits were not associated with other psychopathology.