DEVELOPMENT AND VALIDATION OF A PATIENT-REPORTED OUTCOME MEASURE FOR CROHN’S DISEASE: THE CROHN’S DISEASE-HEALTH INDEX (CD-HI), A FULLY VALIDATED TOOL TO BOLSTER CLINICAL TRIAL AND RESEARCH INFRASTRUCTURE

Abstract

Abstract STUDY AIM To bolster Crohn’s disease (CD) clinical trial and research infrastructure through the development and validation of a sensitive, multifactorial, disease-specific, patient-reported outcome measure (PRO) for patient monitoring in therapeutic trials of CD. BACKGROUND As therapeutic advancements are made in the field of CD, it is vital to have a fully validated, sensitive, and reliable, disease-specific PRO to accurately measure changes in disease burden. Importantly, the U.S. Food and Drug Administration has encouraged the use of PROs to measure therapeutic benefit of new treatments on patient populations. This research describes the creation, testing, and validation of a new, multifactorial PRO: The Crohn’s Disease-Health Index (CD-HI). METHODS Questions were selected for the CD-HI based on their importance to patients, potential to respond to therapeutic benefit, reliability, internal consistency, and generalizability (Figure 1). We beta tested the CD-HI to evaluate its clarity, ease of use, and applicability to the CD patient population. We performed test-retest reliability, known groups validity, internal consistency, and area under the curve (AUC) analyses. RESULTS The final version of the CD-HI contains 12 subscales that measure disease burden in the following areas: 1) fatigue, 2) dietary restrictions, 3) gastrointestinal health, 4) sleep and daytime sleepiness, 5) bowel and bladder function, 6) emotional wellbeing, 7) joint health, 8) pain, 9) neck and back health, 10) activity participation, 11) social wellbeing, and 12) skin health. The CD-HI also provides a composite measure of total disease burden. Fifteen participants participated in beta testing and identified the CD-HI as easy to use, clear, relevant, and comprehensive. Twenty-three individuals with CD participated in test-retest reliability testing of the CD-HI, which indicated a high reliability of the individual questions, subscales, and full instrument (intraclass correlation coefficient = 0.84 for the full instrument). The CD-HI and its subscales demonstrated a high internal consistency (Cronbach’s α = 0.98 for the full instrument). The CD-HI total and subscale scores successfully distinguished between patient cohorts with differing disease severity. AUC analysis demonstrated a high sensitivity and specificity of the CD-HI in distinguishing between those with CD on disability vs. not on disability, with an AUC value of 0.9087 (Figure 2). CONCLUSIONS The CD-HI is a fully validated, sensitive, and accurate marker of disease burden in CD. This disease-specific PRO is available for use by researchers, clinicians, patient organizations, and companies to monitor patients and measure disease burden during therapeutic trials