Abstract
Aim Paediatric inflammatory bowel disease (IBD) is a chronic relapsing condition requiring adherence to complex treatment regimens to achieve best outcomes. Adherence is frequently low in this population but can be improved by increasing disease- and treatment-related knowledge. The IBD-knowledge inventory device (IBD-KID) is a knowledge assessment tool specifically developed and validated for children with IBD. To analyse IBD-KID participant response patterns in order to review the strength of the tool. Methods A cohort of children with IBD completed IBD-KID, and their responses were used to assess the tool's validity and feasibility. Item response analysis assessed the item difficulty and the ability of items to discriminate between high/low scorers. The analysis considered item structure, readability and the effectiveness of multiple choice items. Results A total of 105 completed IBD-KID assessments showed that 12 items (52%) had an acceptable difficulty level, and 17 (74%) were effective at discriminating between high/low scorers. Nine (61%) had good readability, but comprehension levels ranged from 5 to 18 years. Seven (30%) had elevated 'don't know' responses, highlighting the need for content and construction review. Of the 10 multiple choice items, 9 were complex and not functioning efficiently. Internal consistency was acceptable but could be improved by removing two items. Conclusions The response analysis metrics were reviewed by an expert panel and provided a framework for IBD-KID improvements with the aim of increasing discrimination and reducing difficulty without adversely affecting reliability. The proposed revisions will address components that may have caused children to answer incorrectly due to confusion rather than lack of knowledge.