A patient-reported outcome measure comprising the stool frequency and abdominal pain items from the Crohn's Disease Activity Index: psychometric evaluation in adults with Crohn's disease

Background The Stool Frequency (SF) and Abdominal Pain (AP) items from the Crohn's Disease Activity Index are together referred to as the "Patient Reported Outcome" (PRO). The SF item measures the number of very soft/liquid stools and the AP item measures abdominal pain severity, which are common Crohn's disease (CD) symptoms that patients consider important to treat. This study evaluated the psychometric properties of both PRO items separately and estimated thresholds for clinical remission in moderately to severely active CD. Methods The measurement properties of the PRO items were analyzed using pooled data from VIVID-1 (NCT03926130), a Phase 3, randomized, placebo- and active-controlled study in adults with moderately to severely active CD. Analyses used weekly average scores of the SF and AP items at Weeks 0 (Baseline), 4, 12, and 52. Remission thresholds were estimated using the Patient Global Rating of Severity (PGRS) and Patient Global Impression of Change (PGIC) as primary anchors as well as qualitative evidence from exit interviews. Results Data from 1065 participants (mean age: 36.2 years [standard deviation: 13 years]) were analyzed. During the trial, scores improved for both PRO items. Both items demonstrated moderate-to-good test-retest reliability for participants defined as stable based on PGRS and PGIC. Most correlations of related assessments were moderate (0.30 <=|rho| <0.70) with SF and moderate-to-large (0.30 <=|rho| <= 0.90) with AP. By contrast, as anticipated, both items had weak correlations (|rho| <0.30) with endoscopic and laboratory assessments. The PRO items could discriminate between groups of participants known to differ based on other assessments. The PRO items were able to detect change, as score changes in both items between Baseline and Weeks 12 and 52 differed significantly between most PGRS and PGIC categories. Anchor-based analyses combined with responses from the exit interviews suggested that an SF score of <= 3 and an AP score of <= 1 could together represent clinical remission. Conclusion These results support the reliability, construct-validity, and responsiveness of both PRO items in moderately to severely active CD and confirm previously suggested scores for both items that could represent clinical remission. Trial registrationClinicaltrials.gov, NCT03926130. Registered 23 April 2019, https://clinicaltrials.gov/study/NCT03926130.