Interobserver variation in the assessment of pelvic organ prolapse

Abstract

The aim of this study was to determine the interobserver agreement of two grading systems for pelvic organ prolapse: the vaginal profile and the International Continence Society (ICS) draft proposal. Forty-nine consecutive women referred for evaluation of urinary incontinence and/or pelvic organ prolapse were studied. Patients were first examined by a physician and a nurse clinician using the vaginal profile, followed by an examination according to the technique described in the ICS draft proposal for standardization of terminology (1994). kappa statistic and Pearson's correlation coefficient were used to determine interobserver variability for the ICS system by overall stage, by stage-specific comparison, and by specific anatomic location. The vaginal profile was evaluated by obtaining a kappa for overall degree of prolapse, stage-specific comparison and by anatomic area. The kappa for the ICS stage was 0.79 (P < 0.001), and the kappa for the vaginal profile by area of greatest prolapse was 0.68 (P < 0.001), indicating substantial interobserver agreement for both systems. The ICS system was noted to have substantial interobserver agreement by a stage-specific comparison. All anatomic locations of the ICS staging system were found to correlate significantly, and a high degree of interobserver precision was found. The vaginal profile also showed significant interobserver agreement by overall degree of prolapse, by specific degree of prolapse, and by anatomic area. It was concluded that both the proposed ICS staging system and the traditional vaginal profile show significant interobserver agreement both by overall stage, stage-specific analysis and specific location. The registered nurse examination correlated well with the physician examination, indicating that the most important factor in obtaining reproducible results may be definition and close attention to examination technique.