Assessing match and mismatch between practitioner-generated and standardized interview-generated diagnoses for clinic-referred children and adolescents.

Although the Diagnostic and Statistical Manual of Mental Disorders (DSM) is widely used in both clinical and research settings, little is known about agreement between clinician and standardized research diagnoses. Diagnoses generated by the Diagnostic Interview Schedule for Children (DISC-P-2.3) were compared with clinician-generated diagnoses for 245 referred youths. Agreement was poor for all individual disorders and broader diagnostic clusters. Agreement was higher for externalizing categories than for internalizing, but no association was found between agreement and child, family, or clinician characteristics. Clinicians were more likely than the DISC to assign 1 diagnosis and less likely to assign 0 diagnoses, suggesting that clinic policies may play a role. Implications for the use of the DSM across different settings are discussed.