P234 Development of a patient reported disease activity score to screen for mucosal inflammation in inflammatory bowel disease.
M. de Jong*1, J. Degens1, T. van den Heuvel1, M. Romberg-Camps2, B. Winkens3, T. Markus4, A. Masclee1, A. van Tubergen5, D. Jonkers1, M. Pierik1
1MUMC, division Gastroenterology-Hepatology, Maastricht, Netherlands, 2ORBIS Medical Centre, Gastroenterology and Hepatology, Sittard, Netherlands, 3Maastricht University, Department of Methodology and Statistics, Maastricht, Netherlands, 4CCUVN, CCUVN, Woerden, Netherlands, 5MUMC, division Rheumatology, Maastricht, Netherlands
Integrated care and patient empowerment improve the outcome of chronic diseases. Telemedicine programmes are of interest for Inflammatory Bowel Diseases (IBD), but should include adequate monitoring of mucosal inflammation to prevent longterm complications. Different clinical activity questionnaires have been developed for systematic follow-up of disease activity in Crohn's disease (CD) and ulcerative colitis (UC). However, none has been validated against endoscopy, which is the golden standard for assessing mucosal inflammation. Recently published validated clinical activity scores include laboratory parameters and are therefore not suitable for telemedicine programmes. The objective of this study was to develop the first patient reported disease activity score for IBD patients to predict endoscopic disease activity, which can be used in telemedicine programmes.
Twenty-three questions regarding disease activity in IBD were selected based on literature review and expert opinion. Consecutive patients undergoing a colonoscopy for clinical evaluation between March 2013 and April 2014 were invited to fill out this 23 item questionnaire 24 hours before endoscopy (i.e. prior to bowel cleansing). Mucosal inflammation was assessed during endoscopy with the simplified endoscopic activity score for CD (SES - CD) and the Mayo endoscopic subscore (MES) for UC. Questions were reduced to a total of 6, based on individual correlation coefficients with endoscopic inflammation and expert opinion. Then, logistic regression was used to find the best fitting model. ROC curves were used to find the most sensitive cut-off value.
Ninety-eight CD patients (41.8% male, mean±SD age 44.7 ± 14.2 years, 55.1% active disease) and 80 UC patients (58.8% male, mean±SD age 52.2 ± 15.3 years, 63.8% active disease) were included.The multivariable logistic regression model for CD with a sensitivity of 90.4% (specificity 40.9%) included questions on blood loss, number of stools, urgency, fatigue and IBD symptoms in general. The multivariable logistic regression model for UC with a sensitivity of 88.3% (specificity 65.5%) contained items on blood loss, number of stools, urgency, abdominal pain and IBD symptoms in general.
We developed a patient reported disease activity score with a high sensitivity for detecting endoscopic disease activity in IBD patients. Such a tool is warranted in telemedicine programmes for screening patients who need further assessment of disease activity with biochemical markers and/or endoscopy. At present we are validating the score in an independent patient cohort.