Reliability of the DSM-III-R childhood anxiety disorders using structured interview: interrater and parent-child agreement.

OBJECTIVE: The aim of the study was to examine the interrater and parent-child agreement for the major child anxiety disorders.
METHOD: One hundred sixty-one children and their parents underwent a semistructured interview (Anxiety Disorders Interview Schedule for Children). To increase external validity, clinicians did not receive specific, extensive training in diagnosing anxiety disorders apart from their standard qualifications. The design of the study allowed for calculation of agreement between raters based on information obtained from the parents alone, from the child alone, or through combined information from both the parents and child, and for calculation of agreement between information obtained from the parents and information obtained from the child.
RESULTS: Levels of interrater agreement either as principal or additional diagnoses were moderate to strong for all of the major childhood anxiety disorders (kappa values .59 to .82). In contrast, parent-child agreement was poor for most diagnostic categories (kappa values .11 to .44).
CONCLUSIONS: The data indicate that, despite the fact that parents and their children do not demonstrate strong agreement, the DSM-III-R childhood anxiety disorders can be reliably diagnosed by pairs of general clinicians using structured interviews.