We appreciate the thoughtful review of our article by Jellinek and colleagues and the opportunity to respond to the points they raised.
Jellinek et al argue that the performance of the screening tests should be tested against levels of impairment rather than tested by comparing scores with diagnostic classification. Regardless of the intent of instruments such as the PSC and the Child Behavior Checklist, they are being used as screening tools for mental disorders by pediatricians (survey of Pediatric Research in Office Settings Network, 1993 [R. Wasserman, MD, oral communication, 1993]) and to classify chronically ill children as having a disorder.1,2 Therefore, their appropriateness in this context seems to be an important avenue for exploration.
Moreover, we agree that impairment is critical in the categorization of disorder.3 However, the recommendation to compare the screening instruments (the Child Behavior Checklist, the PSC, and the Child Depression Inventory)