RESUMO
Guidelines exist for the assessment of attention deficit/hyperactivity disorder (ADHD), but they are often unclear as to how a clinician should consider multiple informants, methods, and co-occurring symptoms to reach an overall diagnostic probability for an individual patient. The current study used receiver operating characteristic analyses and evidence-based medicine methods to evaluate the Achenbach System of Empirically Based Assessment measures and the Conners' Continuous Performance Test for ADHD diagnosis in youth. Children (n = 379) and their parent(s) presented at an outpatient clinic for a psychoeducational assessment. Analyses examined the diagnostic efficiency and utility of study measures for predicting a best-estimate ADHD diagnosis. The Child Behavior Checklist Attention Problems construct, Teacher Report Form Attention Problems construct, and Hit Reaction Time Standard Error showed adequate diagnostic efficiency and unique contributions to the prediction of ADHD, Combined Type diagnosis. None of these measures showed good diagnostic efficiency or utility for the prediction of ADHD, Predominantly Inattentive Type. Child anxiety did not moderate the relations between predictors and ADHD diagnosis. Both the Child Behavior Checklist and Teacher Report Form Attention Problems constructs can discriminate youth with ADHD, Combined Type from other clinic-referred youth. Although Hit Reaction Time Standard Error also showed diagnostic utility, the decision to include a computerized measure should consider time and expense and be utilized in cases where diagnostic probability is unclear. Finally, anxiety may be associated with elevated attention problems, but it does not appear that anxiety affects diagnostic cutoffs for ADHD.
