Hypothesis / aims of study
Assessment of sexual function is typically based on self-administered questionnaires, with the Female Sexual Function Index (FSFI) among the most popular tools. Still, the disease-specific – Pelvic Organ Prolapse/Incontinence Sexual Questionnaire – IUGA Revised (PISQ-IR) has no fixed cutoff values until today. The aim of the study was to find the cutoff scores of PISQ-IR for women with pelvic floor disorders (PFD).
Study design, materials and methods
A cross-sectional study was conducted in 457 consecutive women referred to a gynecology clinic due to PFD. Medical history was taken and urogynecological examination was performed according to the standards of International Continence Society. The degree of prolapse was assessed using the Pelvic Organ Prolapse Quantification (POP-Q) system. The subjects provided information on their sexual activity and completed two questionnaires: general (FSFI) and disease-specific (PISQ-IR). Sexually active women with PFD were enrolled for further analysis and categorized according to their FSFI scores < 26.55 - sexual dysfunction and > 26.55 - no sexual dysfunction . Medical history, urogynecological examinations and questionnaire scores were compared. Receiver operating characteristics-curve (ROC) was used to test how well PISQ-IR allowed for discrimination between patients with and without sexual disorders. Area under the curve (AUC) statistic was calculated in order to measure the PISQ-IR Summary Score efficiency in the aforementioned prediction. The cutoff values which minimalize (1-specifity) and maximize sensitivity were selected. A subject-to-item ratio of 10:1 has been recommended for psychometric work, so a sample size of at least 120 sexually active women was required to fulfill the abovementioned criterion. The analysis included only the respondents who had complete information in all domains of the FSFI and PISQ-IR questionnaires.
Out of 457 women with PFD, 227 (49.7%) were sexually active. The response rate was 88.1%. Therefore, a total of 200 sexually active women were recruited for the study: 129 (66.5%) with FSFI scores < 26.55 and 71 (35.5%) with FSFI scores > 26.55. The POP assessment revealed stage 0 POP-Q in 6 (3%) women, I - 13 (6.5%), II - 69 (34.5%), III - 96 (48%), IV - 16 (8%). Women form both groups did not differ in terms of age, education, parity, marital status, POP-Q stage, history of previous surgeries and the diagnosis of various PFD. Women with FSFI scores indicating sexual dysfunction had higher BMI (27.9 ± 4.7 vs. 26.5 ± 4.1, p<0.05) and were more often postmenopausal (72.9% vs. 57.7%, p=0.03). After dichotomizing patients into two groups according to their FSFI results, the differences in PISQ-IR scores were statistically significant in all domains (Table 1). On the basis of ROC curve analysis, the PISQ-IR Summary Score of 2.68 was determined to be the optimal cutoff for distinguishing between dysfunctional and non-dysfunctional women. The assumption that women with scores above the cutoff values have good sexual function resulted in the true positive rate (sensitivity) of 90%, and represented the proportion of sexually active women which were correctly classified by the PISQ-IR Summary Score as healthy. The false-positive rate of 27% (1-specificity) represented the proportion of women with sexual dysfunction which were incorrectly classified as healthy (Figure 1).
Interpretation of results
Sexual function was impaired in 66.5% of women with PFD. The statistical description of the effectiveness of classifying the subjects into two FSFI groups on the basis of PISQ-IR, AUC=0.85 deemed the scale suitable for that purpose. The cutoff score of 2.68 for PISQ-IR Summary Score allowed to diagnose sexual dysfunction in sexually active women with PFD with 90% sensitivity and 73% specificity; 73% of the cases were correctly classified as dysfunctional and 90% as non-dysfunctional.