P407 Validation of a newly developed patient-reported experience measure for people with Inflammatory Bowel Disease

Abstract

Patient-Reported Experience Measures (PREMs) assess a patient’s healthcare experiences and are important to improve the quality of healthcare. A PREM specific to inflammatory bowel disease (IBD) was co-produced with service users. This study aimed to validate the PREM. The IBD PREM contains 38-items and comprises three domains: ‘the care team’, ‘what matters to me’, and ‘living with Crohn’s or Colitis’. Scores can be calculated for each domain and summated to give a total score. Validation was completed on data collected from participants who completed the PREM every 3 months alongside the IBD-Control measure of disease activity at a single UK tertiary IBD service. The PREM was assessed for acceptability; reliability (internal consistency and repeatability); known groups validity; criterion validity and responsiveness. Face and content validity were completed before finalisation using ‘Think Aloud’ interviews. The dataset used for validation was extracted in September 2023 and contained responses from 287 participants (63% female, 89% White British, median age: 52) comprising 1,699 PREM responses between November 2021 and August 2023. Acceptability was demonstrated with very low rates of missing data (four items with >5% missing) and median submission duration of 4.8 minutes (IQR: 3.6 – 6.6). High Cronbach’s alpha scores (0.85, 0.97, 0.88 and 0.97 for each of the domains and total PREM score respectively) show internal consistency of the measure. However, the particularly high value on one domain and the total score suggest some overlapping items may remain. An intraclass corelation coefficient of 0.88 (95% CI: 0.85 to 0.91) suggests good repeat reliability over a two-week timeframe. Good comprehension of the measure by service users satisfied face and content validity. Known groups validity was supported with lower score on ‘living with Crohn’s or Colitis’ domain the for those with an IBD Control score of less than 13 by 0.38 (95% CI: 0.23 to 0.54), p<0.001). A low to moderate correlation of 0.37 (95% CI: 0.32 to 0.42) was found between the total PREM score and the IBD Control score, as expected. Ceiling and floor effects had a maximum of 6% demonstrating potential responsiveness of the PREM. The change over a year detected by those that stated that had experienced a change was 0.53 (95% CI: 0.41 to 0.66, n=38) on a five-point scale.Conclusion The co-produced PREM is an acceptable, valid, reliable, and responsive tool to measure experience of IBD care for quality improvement. Further work should assess the potential for item reduction as well as correlation to other related measures.