Development and evaluation of the Women's Sexual Interest Diagnostic Interview (WSID): a structured interview to diagnose hypoactive sexual desire disorder (HSDD) in standardized patients.

Introduction: Female sexual dysfunction (FSD) is a common disorder in postmenopausal women. Currently, there is no clear "gold standard" for the diagnosis of FSD.

Aim: The aim of this study was to evaluate the interrater reliability of the Women's Sexual Interest Diagnostic Interview (WSID), a new structured clinical interview designed to diagnose hypoactive sexual desire disorder (HSDD). The reliability of additional interview questions focused on the diagnosis of other types of FSD was also evaluated.

Main Outcome Measures: The main outcome measure was the level of agreement in the diagnosis of FSD among clinical experts, between clinical experts and study coordinators, and between clinical experts and patients' self-reported interactive voice response system (IVRS) version of the WSID.

Methods: Two versions of WSID were developed based on current diagnostic criteria: a clinician-administered version using a structured interview guide, and a patient self-report version using an IVRS. Three sexual medicine experts developed 20 clinical scenarios portraying cases and noncases of HSDD and other FSD diagnostic subtypes. Ten actresses with experience in standardized patient interviewing rehearsed these scenarios and performed the scripted patient roles in a standardized clinical interview with clinical experts (not the author of the script) and study coordinators, on a one-on-one basis, using the WSID interview format. In addition, all actresses completed the IVRS version of the WSID. Interviews were videotaped and viewed by the expert panel. In each instance, the diagnosis that the interview was scripted to portray was considered as the "gold standard." Kappa (kappa) coefficients were utilized to assess the level of agreement among experts, between study coordinators and the "gold standard", and between the IVRS version of the WSID and the "gold standard".

Results: All experts agreed with the gold standard diagnosis provided by the author of the script (kappa=1.0). Similarly, there was perfect agreement among the experts on the presence of depressive symptomatology (kappa=1.0). On the related diagnoses of arousal disorder, orgasmic disorder, and sexual pain disorder, kappas of 0.894, 0.966, and 0.946 were observed (P<0.0001 for all comparisons). When study coordinator's WSID diagnoses were compared with the "gold standard," kappa for HSDD was 0.851; sensitivity was 0.864, and specificity and positive predictive value (PPV) were 1.00 (P<0.001 for all comparisons). When diagnoses obtained via IVRS interviews were compared with the "gold standard", kappa for HSDD was 0.802, sensitivity was 0.818, and specificity and PPV were 1.00 (P<0.001 for all comparisons).

Conclusion: Agreement as estimated by kappa coefficients was consistently high in both clinician-administered and patient self-reported IVRS versions in the diagnosis of HSDD.