Muna J. Tahir, PhD1, Saeid Shahraz, MD, PhD2, Song Wang, PhD3, Peter Nagy, MD4, Pravin Kamble, PhD3 1ICON Plc., South San Francisco, CA; 2ICON Plc., South San Francisco, CA, USA; Tufts Medical Center, Boston, MA; 3Takeda, Cambridge, MA; 4Takeda, Zurich, Zurich, Switzerland
Introduction: The Ulcerative Colitis Electronic Diary (UC e-diary) is a 5-item Likert-based, unidimensional instrument developed to evaluate patient-reported symptoms of UC in adults, including bowel movements, rectal bleeding, diarrhea, urgency, and abdominal pain (range: 0-10, where 10 = worst symptoms). This study aimed to assess the psychometric properties of the UC e-diary.
Methods: We employed data from 2 Phase 3, randomized, double-blind, placebo-controlled studies of ontamalimab vs placebo in patients with moderate-to-severe UC (NCT03259334 and NCT03259308) (N=657). Psychometric properties assessed included: item-level analyses; factor structure; reliability; construct validity; and responsiveness, with clinically meaningful change thresholds derived. An item response theory (IRT) model was fit and item/scale behaviour evaluated.
Results: While rectal bleeding and urgency items exhibited ceiling effects at baseline, ceiling effects were not evident at the total score level. At week 12, four items (rectal bleeding, urgency, bowel movements, and diarrhea) demonstrated ceiling effects, with 30% reaching a score of 0. The factor analysis supported a unidimensional domain structure. Moderate-to-strong item-total correlations and Cronbach’s alpha values (range: 0.88-0.91) were observed at baseline and week 12, indicating high item homogeneity and good internal consistency. The results revealed strong test-retest reliability at the item and total score level (intraclass correlation coefficients: 0.87-0.97). Moderate-to-strong correlations (r=0.47-0.66) with criterion measures (including the Inflammatory Bowel Disease Questionnaire and Mayo Scores) supported convergent validity. Significant differences in UC e-diary total scores were observed between groups defined by clinical remission and improvement at week 12, supporting known-groups validity. The UC e-diary was responsive in patients who indicated a change in other instruments. A 2-point decrease in the UC e-diary reflected meaningful within-patient improvement. Except for the abdominal pain item, the IRT model showed reasonable item and scale level behaviour.
Discussion: Findings indicate that the UC e-diary is reliable, valid and responsive, with a 2-point improvement serving as a reasonable estimate of clinically meaningful within-patient change. This suggests that the tool could be useful to measure a clinical trial endpoint capturing UC-related symptom severity from the patient’s perspective.
Muna J. Tahir, PhD1, Saeid Shahraz, MD, PhD2, Song Wang, PhD3, Peter Nagy, MD4, Pravin Kamble, PhD3. P2633 - Psychometric Properties of the Ulcerative Colitis Electronic Diary (UC e-Diary) in Adult Patients with Moderate-to-Severe Ulcerative Colitis, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.
