Quantifying Potential Bias Resulting From Child Age on Screening for Hyperactive/Impulsive Presentations of ADHD

Abstract

Objective: This study aims to quantify the potential age bias in screening of hyperactive/impulsive presentations of ADHD in children ages 5 to 12 through comparison of age-based and overall percentiles in screening. Method: A referred clinical sample of 307 children ages 5 to 12 with behavioral concerns completed the Vanderbilt Attention Deficit-Hyperactivity Disorder Diagnostic Parent Rating Scale (VADPRS) and were formally evaluated for ADHD with a diagnostic interview. Analysis utilizing logistic regression and receiver operating characteristic (ROC) curves was performed to compare the screening performance of agebased and overall percentiles. Results: The age-based percentiles demonstrated no improvement in the analyzed models compared to overall percentiles in hyperactive presentation ADHD screening. This finding was present in the overall sample and in the sub analysis of the 5 to 6 year old children. Conclusions: This study identifies no improvement in modeling of hyperactive/impulsive ADHD screening when considering a child’s age using age-based percentiles.