Implementation of a Data-Driven Learning Health System in Rheumatology: A Novel Application of Dashboard Quality Reporting to Support Optimal Rheumatoid Arthritis Care

Objectives Learning Health Systems (LHS) leverage healthcare data to drive cycles of knowledge generation for continuous care quality improvement. In Alberta, the recent launch of a provincial health information system (Connect Care, Epic Corporation) that interfaces with analytical and reporting tools, supports the use of routinely collected regional clinical data for care quality improvement efforts. As such, this project aims to implement a provincial LHS to support optimal care for Albertans with rheumatoid arthritis (RA). Methods This was a multiphase project. In Phase 1, we identified a starter set of quality measures. Candidate measures were reviewed for feasibility of operationalization using structured data in Connect Care along with linkage to other administrative datasets (physician billing, pharmacy data, discharge abstract database). In Phase 2, a shortlist of eligible measures that aligned with national measurement priorities were ranked by panelists including providers, field-experts and patient partners. In Phase 3, we operationalized and reported on these measures as dashboards. To do so, a case definition of RA and a list of eligible providers were identified. These definitions were validated by chart reviews completed by 2 rheumatologists and an analyst. The resulting data were reported as dynamic and interactive dashboards on Tableau using near to real-time data from the linked datasets. Dashboards allow providers to access patient-level data on their practice to support quality improvement. Results From a shortlist of eighteen measures, 9 were prioritized by a panel of providers, experts and patient partners (n=7). We developed the following dashboards to display an initial set of measures: 1) number of RA patients followed per practice/site; 2) wait times to first rheumatology consult; 3) RA disease activity assessment (process and outcome measures); 4) gaps in care (lost to follow-up, treatment, or lab monitoring). As of 09/20/2024 there were 8059 individuals with RA under rheumatology care at 4 sites. Only 44% of 6668 encounters had a documented joint count, and 24% had a composite disease activity score calculated. Of the 1600 encounters with a composite score, 17% of patients were in low disease activity/remission. We identified 1,770 individuals with >12 months between rheumatology visits. Only 24% of individuals with RA were seen by a rheumatologist within 6 weeks of referral. Conclusion Using an LHS approach, we aim to shorten and streamline the cycles of transforming data into actionable knowledge for optimized care. We are working with our provincial partners to identify appropriate strategies to address the identified gaps. Practice Reflection Award