Development and Validation of the TUMMY-UC: A Patient-Reported Outcome for Pediatric Ulcerative Colitis.

BACKGROUND & AIMS: The TUMMY-UC is a patient-reported outcome measure for pediatric ulcerative colitis (UC) with an observer-reported outcome version for children aged <8 years. It includes eight items selected by concept elicitation interviews. We aimed to finalize the TUMMY-UC by cognitive interviews (stage 2) and to evaluate the index for its psychometric properties (stage 3). METHODS: The TUMMY-UC items were first finalized during 129 cognitive debriefing interviews. Then, in a prospective, multicenter validation study, 84 children who underwent colonoscopy or provided stool for calprotectin completed the TUMMY-UC and various measures of disease activity. Assessments were repeated after 7 and 21 days for evaluating reliability and responsiveness. RESULTS: During stage 2, the items were formatted with identical structure to ensure conceptual equivalence and weighted based on ranking of importance. In stage 3, the TUMMY-UC total score had excellent reliability in repeated assessments (intraclass correlation coefficient, 0.90; 95% confidence interval, 0.84-0.94). It also had moderate to strong correlations with all constructs of disease activity: r = 0.70 with UC endoscopic index of severity, r = 0.63 with the IMPACT-III questionnaire, r = 0.43 with calprotectin, r = 0.80 with the Pediatric Ulcerative Colitis Activity Index, r = 0.75 with global assessment of disease activity, and r = 0.46 with C-reactive protein (all P < .015). The index had excellent discrimination of disease activity, with a score of <9 defining remission (area under the receiver operating characteristic curve, 0.95; 95% confidence interval, 0.93-0.99). The ΔTUMMY-UC showed high responsiveness and differentiated well between children who experienced changed from those with no change. CONCLUSIONS: The TUMMY-UC, constructed from patient-reported outcome and observer-reported outcome versions, is a reliable, valid and responsive index that can be now used in practice and clinical trials.

Agreement on Symptoms Between Children With Ulcerative Colitis and Their Caregivers: Towards Developing the TUMMY-UC.

ABSTRACT: As part of the development of the TUMMY-UC, a patient-reported outcome (PRO) measure for pediatric ulcerative colitis (UC), we aimed to explore agreement on UC symptoms between children and their caregivers. We conducted 44 interviews with children ages 8-12 years, who completed the PRO version of the TUMMY-UC, and their caregivers, who completed the observer-reported outcome (obsRO) version. There was excellent agreement between the total TUMMY-UC PRO and obsRO scores (intra-class correlation coefficient = 0.92 [95% confidence interval 0.74-0.98]). The obsRO scores were always within the same disease-activity category as the corresponding PRO score (ie, remission, mild and moderate-severe disease). There was a strong correlation of the TUMMY-UC PRO and obsRO scores with physician global assessment of disease activity (r = 0.94 and r = 0.90, respectively, P < 0.001) and the pediatric UC activity index (r = 0.95 and r = 0.96; P < 0.001). These data support conceptual equivalence between the PRO and obsRO TUMMY-UC versions, and provide support for their incorporation into one score.

Item Generation and Reduction Toward Developing a Patient-reported Outcome for Pediatric Ulcerative Colitis (TUMMY-UC).

BACKGROUND: The Pediatric Ulcerative Colitis Activity Index (PUCAI) is a noninvasive clinician-based index, which reflects disease severity in pediatric ulcerative colitis (UC) when no endoscopy is performed. Here, we aimed to explore signs and symptoms important to children with UC and their caregivers as the first stage of developing a patient-reported outcome (PRO) measure for pediatric UC (ie, the TUMMY-UC index) to supplement endoscopic assessment. METHODS: Concept elicitation qualitative interviews were performed with children who have UC and their caregivers in 6 centers. Items were rank-ordered by the interviewees according to the frequency of endorsement and importance, graded on a 1 to 5 scale. RESULTS: A total of 46 children (ages 12.5 ±â€Š3.3 years, range 7-18, 48% boys, 83% with pancolitis, 24% with moderate-severe disease) and 33 caregivers were interviewed (ie, 79 interviews). The following items were identified by the children, in decreasing order of weights: abdominal pain (importance × frequency weight 3.9), rectal bleeding (3.6), stool frequency (3.0), stool consistency (3.0), general well-being/fatigue (2.9), urgency (1.9), and nocturnal stools (1.6). Two other items were scored lower: lack of appetite (1.1) and weight loss (0.6). Children 13 to 18 years comprehended adult vocabulary, children 8 to 12 years comprehended simple vocabulary, and younger children had poor understanding in completing the questions. CONCLUSIONS: In this first stage of the TUMMY-UC development, items were generated and ranked by input from patients. These items are now being explored for optimal vocabulary and response options. The TUMMY-UC will supplement the PUCAI in clinical trial outcome assessment.