A new validated score for detecting patient-reported success on postoperative ICIQ-SF: a novel two-stage analysis from two large RCT cohorts.

The Patient Global Impression of Improvement (PGI-I) and International Consultation of Incontinence Questionnaire - Short Form (ICIQ-SF) are validated instruments for the assessment of patient reported outcome measures (PROM) following treatment of stress urinary incontinence (SUI). However, there is a paucity of evidence as to what represents a successful postintervention ICIQ-SF score. To determine the correlation between the postoperative ICIQ-SF scores with the PGI-I outcomes, the latter was considered one of the standard PROM following surgical treatment for SUI. The aim of this study was to determine, and if appropriate validate, an ICIQ-SF cut-off score that can predict a successful PROM as determined by PGI-I. Four large datasets yielding 674 ICIQ-SF score/PGI-I outcome data pairs were used in this study for (a) determining and (b) validating the cut-off ICIQ-SF score for a successful PGI-I outcome. Two long-term follow-up datasets were used representing follow-up periods of 3 and 8 years of a randomized controlled trial (RCT) performed between April 2005 and April 2007 in a tertiary urogynaecology centre in Scotland, UK. All patients had urodynamic SUI or mixed urinary incontinence (MUI, with predominant SUI) and were randomized to treatment with either an inside-out or an outside-in transobturator tape (TVT-O or TOT, respectively) as a sole procedure. The datasets yielded 432 ICIQ-SF score/PGI-I outcome data pairs. Successful outcome was defined as "very much improved/much improved" on the PGI-I scale. SPSS v. 22.0 (IBM Corp., Armonk, NY) was used for all statistical analyses. The correlations and cut-off scores generated were then validated on two independent datasets representing the 1-year and 4-year follow-up periods of the multicentre RCT in six units in the UK. The datasets yielded 242 ICIQ-SF score/PGI-I outcome data pairs. All patients had urodynamic SUI or MUI (with predominant SUI) and were randomized to either adjustable single incision minisling (SIMS) or TVT-O. Significant correlations at the 0.01 level (two-tailed) were clearly demonstrated between ICIQ-SF scores at follow up and PGI-I outcomes in terms of success/failure in both the generation and validation datasets. Higher ICIQ-SF scores correlated with a 'poorer' PGI-I score. Using ROC analysis, a postoperative ICIQ-SF score of 6 was validated as approximately 90 % sensitive and 85 % specific for success/failure with a high Cohen's kappa coefficient of 0.83 (95 % CI 0.74 - 0.89). This two-stage study provided a robust well-validated postoperative ICIQ-SF cut-off score (of 6/21) that is likely to be associated with a patient-reported successful outcome on the PGI-I following surgical treatment with a midurethral sling in women at different stages of follow-up over 1 - 8 years. Such a cut-off score could enable the comparison of results between various studies and serve as a valuable guide for surgeons to counsel patients before and/or after surgical treatment. Our study fills a research gap in providing a way to compare trial results when baseline ICIQ-SF scores are not available.