A new structured interview for children with autism spectrum disorder based on the DSM-IV.

BACKGROUND: Autism spectrum disorder (ASD) is a common neurodevelopmental disorder in children. The clinical spectrum of ASD includes autism, childhood disintegrative disorder Asperger syndrome and pervasive developmental disorder not otherwise specified (PDD-NOS). Although the DSM-IVcriteria are well acceptedforASD diagnosis, there are some known limitations for clinicians. The most important issue is lack'ofspecific age-appropriate items in each domain. Thus, the DSM-IVneeds some modifications in order to be appropriate for clinical use.

OBJECTIVE: To develop a structured interview for children based on the DSM-IVdiagnostic criteria ofautism and PDD-NOS. MATERIAL ANDMETHOD: From June 2006 to December 2008, 140 Thai children, 121 boys and 19 girls, already diagnosed with ASD, were recruited through the child development clinics of Ramathibodi and Thammasat University Hospitals in Thailand. A 26-item structured interview was developed with scoring according to the DSM-IVdiagnostic criteria for autism andPDD- NOS. To test the accuracy of the structured interview and its reliability, 32 children with ASD were selected and interviewed by four clinicians using the new instrument. One clinician interviewed the parents or caregivers, while three others independently took notes and observed the play behavior of the children. All items from the structured interview as scored by each clinician were compared using inter-rater agreement statistics (Kappa). All of the original 140 patients were then clinically diagnosed again using the structured interview and the results were compared with the initial diagnoses.

RESULTS: Ofthe 140patients originally diagnosed with ASD, 110 and 30patients were finally diagnosed with the new interview as having autism and PDD-NOS, respectively. The initial diagnoses from 15 cases (10.7%) were changed according to the structured interview Inter-rater reliability among the four clinicians showed a good level ofagreement (Kappa = 0.897) with statistical significance (p<0.001). The authors only compared the items in the structured interview between the autism and PDD-NOSgroups from 105 cases aged 2-5 years (79 cases with autism and 26 cases with PDD-NOS) because there were only 4 cases with PDD-NOS in the other age groups. Highly significant differences (p<0.001) in clinical items between patients with autism and patients with PDD-NOS from the final diagnoses were noted in 6 of 8 items in the category of restricted, repetitive and stereotyped patterns ofbehavior, interests and activities, which were more common in the autism group than the PDD-NOS group. In addition, the autism group had higher frequencies of using finger-pointing to indicate interest rather than verbalization, and idiosyncratic language, than the PDD-NOS group.

CONCLUSION: The newly developed structured interview for Thai children with ASD had a high level ofinterrater reliability between four clinicians. However, most children tested using this structured interview were 2-5years ofage, and the study did not include non-autistic groups. The application ofthis structured interview needs further study with a wider variety ofcases, such as ASD cases from different age groups, children with delayed development and normal children.