A110 CLINICAL DECISION SUPPORT SYSTEM FOR IBD FLARE MANAGEMENT AND CORTICOSTEROID ADMINISTRATION: PRELIMINARY RESULTS FROM AN INTERRUPED TIME SERIES

Induction and maintenance of remission is a primary treatment goal for inflammatory bowel disease (IBD). Known gaps exist in adherence of practitioners to international guidelines for IBD care and, in other specialties, it has been shown to take up to 17 years for evidence to be translated into clinical practice. One problem; guidelines themselves are not actionable because they largely describe ‘what’ but not ‘how’. As a solution, our group developed a clinical decision support (CDS) tool for IBD in line with current guidelines. This was embedded into the Clinical Information System (CIS), EpicCare Ambulatory, as automated alerts, call-to-actions, and electronic order sets. Although similar tools have been built in other Epic systems and used for disease management, there is limited data on impact. We aim to assess the effect of our CDS tool on IBD specialists’ adherence to guidelines, as well as system satisfaction. A pragmatic, pre- and post-implementation design with a two-phase interrupted time-series. The interruption is activation of the CDS tool, accompanied by user training and a memo with instructions for use. Each data point will correspond to one month of individual patient visits, with a total of 18 months of data, 9 pre- and 9 post-interruption. A questionnaire including the Workflow Integration Survey (WIS) was administered to all users pre-interruption to collect demographic and baseline use data. The STARE-HI guidelines for health informatics evaluations were used in study design and assessment of appropriateness. The CDS tool was activated in October 2018. Ten IBD specialist users have completed the baseline questionnaire. Demographics; median of 7.0 (IQR: 6.0 to 7.0) years in practice, 5.0 (IQR: 4.0 to 6.0) years using any EMR, and 3.0 (IQR: 3.0 to 4.0) years using EpicCare. Users were asked to what extent they felt EpicCare was being fully utilized (Likert scale); 5 (50%) selected ‘neutral’, 5 (50%) selected ‘to a good extent’ (none selected ‘to the full extent’). Results from the WIS; mean of 4.0 (SD: 0.69) for navigation, 3.37 (SD: 0.93) for functionality, 3.50 (SD: 0.86) for ease of use, 3.30 (SD: 0.75) for workload. The Likert scores from individual WIS items are shown in Figure 1. At baseline, users are relatively satisfied with the different domains of the pre-existing system. About half of users feel that duplicate information is entered during patient encounters, and some feel that the system is not fully utilized. We will finish collecting adherence and application data and re-administer the WIS score again post-study. We hope to improve or at least do no harm to the current user experience, while improving guideline adherence and quality of care. CCC, CIHRUniversity of Alberta

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.