The PISQ-IR: considerations in scale scoring and development.

INTRODUCTION: This paper provides a detailed discussion of the psychometric analysis and scoring of a revised measure of sexual function in women with pelvic floor disorders (PFD): the Pelvic Organ Prolapse Incontinence Sexual Questionnaire, IUGA-Revised (PISQ-IR).
METHODS: Standard tools for evaluating item distributions, relationships, and psychometric properties were used to identify sub-scales and determine how the sub-scales should be scored. The evaluation of items included a nonresponse analysis, the nature of missingness, and imputation methods. The minimum number of items required to be answered and three different scoring methods were evaluated: simple summation, mean calculation, and transformed summation.
RESULTS: Item nonresponse levels are low in women who are sexually active and the psychometric properties of the scales are robust. Moderate levels of item nonresponse are present for women who are not sexually active, which presents some concerns relative to the robustness of the scales. Single imputation for missing items is not advisable and multiple imputation methods, while plausible, are not recommended owing to the complexity of their application in clinical research. The sub-scales can be scored using either mean calculation or transformed summation. Calculation of a summary score is not recommended.
CONCLUSION: The PISQ-IR demonstrates strong psychometric properties in women who are sexually active and acceptable properties in those who are not sexually active. To score the PISQ-IR sub-scales, half of the items must be answered, imputation is not recommended, and either mean calculation or transformed sum methods are recommended. A summary score should not be calculated.