TEST-RETEST RELIABILITY AND VALIDITY OF THE STRUCTURED INTERVIEW FOR SLEEP DISORDERS ACCORDING TO DSM-III-R

Objective: The purpose of this study was to evaluate the reliability o f sleep disorder diagnoses in DSM-III-R by using a newly developed int erview, the Structured Interview for Sleep Disorders According to DSM- III-R (SIS-D) and to evaluate the concordance between these diagnoses and sleep laboratory data. In addition, the sources of disagreements b etween two interviewers in the diagnoses given to the same patient wer e determined. Method: Two different interviewers used the SIS-D to dia gnose 68 patients with complaints of sleep disorders. The concordance between these interviewers' diagnoses and polysomnographic findings wa s investigated by using kappa statistics. Results: There were excellen t reliabilities for almost all current main diagnostic categories and good concordance between diagnoses made on the basis of the structured interview and polysomnographic data. The main source of disagreement between interviewers was found in the symptom information given by the patient. Conclusions: These findings provide support for the utility of DSM-III-R sleep disorder diagnoses and for their retention in DSM-I V. These findings also accord well with a recent literature review of the DSM-III-R diagnosis of primary insomnia by the DSM-IV Work Group o n Sleep Disorders. The good concordance between interview diagnoses an d polysomnographic data suggests that a structured interview such as t he SIS-D may be a useful screening instrument. The authors discuss the implications of these findings for the polysomnographic evaluation of chronic insomnia.