|| Checking for direct PDF access through Ovid
This study evaluated diagnostic utility of parent and teacher ratings in the attention-deficit hyperactivity disorder (ADHD) assessment and differential diagnosis of a clinical sample of children referred for suspected ADHD. Participants were 184 5- to 12-year-old children for whom the following were available: multimethod multi-informant assessment, firm decision regarding presence or absence of ADHD, and parent-completed Child Behavior Checklist and revised 48-item Conners Parent Rating Scale, and teacher-completed Teacher Report Form and 39-item Conners Teacher Rating Scale. Parent ratings of children diagnosed with and without ADHD were generally similar. In contrast, teachers rated students diagnosed with ADHD as displaying higher levels of behavioral difficulties. Discriminant function analyses indicated parent ratings of narrowband measures of ADHD and broadband measures of externalizing symptoms displayed high sensitivity. Teacher ratings outperformed parent ratings when considering sensitivity, specificity, and overall classification accuracy. For clinically recommended cut scores, teacher measures displayed good specificity and positive predictive power. Combining rating scales within informants. did not improve classification accuracy. Combining across parent and teacher measures produced results consistent with teacher ratings. Results support recommendations to include parent and teacher rating scales in ADHD assessment. Scales contributing most to classification accuracy were those designed to assess ADHD. Imperfect performance of rating scales supports recommendations to include other methodologies in diagnosis and differential diagnosis of ADHD.