The VHA SARC Decision Support Tool: Septic Arthritis Risk Calculation forthe Native Knee

About This Grant

Significance to VA. Septic arthritis (SA) of a native knee is a surgical emergency and efficient diagnosis and treatment are critical as delay to treatment risks rapidly progressive arthritis, osteomyelitis, limb loss, and death. Diagnosis is complicated by the existence of other pathologies with similar presentations. There are no clear diagnostic criteria and pre-existing arthritis and crystal arthropathy are common among Veterans, particularly complicating the diagnosis. Evidence-based information at the right time and place will fill a large clinical gap and help reshape underlying assumptions currently driving decision making for knee SA in Veterans. Innovation and Impact. There are no established evidence-based guidelines or tools to accurately diagnose native knee SA. The PI developed a novel multivariable model in a non-Veteran population that demonstrated strong diagnostic characteristics. This model has been validated in a pilot Veteran population with similar performance characteristics. In addition, an on-line risk calculator known as the Septic Arthritis Risk Calculator (SARC) decision support tool (DST) has been developed and undergone usability testing. VA providers have expressed an interest in seeing the model validated in a national VA population and the DST integrated into the VA electronic health record (EHR). If successful, this will be the only tool of its kind and will facilitate decision making for an emergent clinical problem with significant risks of over and under treatment. Specific Aims. (1a): Externally validate the current SARC prediction model in a national VHA patient population while developing and externally validating a second model using predictors that do not require provider assessment to reduce clinical burden; (1b): Compare performance characteristics of the existing SARC model with the newly developed model and determine which model will be most appropriate for converting into the SARC DST (risk calculator) to be integrated into the VHA EHR; (2): Perform necessary programming to integrate the SARC DST into the VHA EHR, working with VA Information Technology specialists and extracting predictors from the CDW to ensure the final tool facilitates auto-population all predictors possible and is available in the VA EHR; (3): Evaluate the content, structure, and usability of the EHR-integrated SARC DST and patient educational handout using a mixed-methods approach, finalizing the tools based on the findings; (4): Identify barriers and facilitators to implementation of the SARC DST through a formative evaluation with clinician participants to identify preferred clinician users, timing of use and methods/vehicles for sharing results with patients, so that it is available for use across VHA. Methodology. We will identify a national sample of Veterans through the VA Corporate Data Warehouse (CDW) and validated through chart notes to validate the current prediction model and develop and validate an enhanced model with alternative predictors that are readily available in the EHR. We will develop the programming infrastructure for the EHR integrated SARC DST by collaborating with VA information technology specialists. This will include leveraging VA secure servers that have the capacity to link directly to the CDW to extract patient data in real time to facilitate auto-population. The tool will undergo usability testing from a national sample of orthopedic, infectious disease, emergency, and hospitalist physicians, as well as Veteran patients previously diagnosed with SA. Key stakeholders will provide feedback on implementation barriers and facilitators. Path to Translation/Implementation. This tool will be considered a “best practice” method to facilitate accurate and timely diagnosis of SA in Veterans. Upon completion of this grant, the tool will be made available VA wide with a planned strategy to champion the tool and to apply multiple vehicles for training and education to ensure uptake and sustainability. Future work will include implementation studies to evaluate the tool's effect on treatment and patient outcomes as well as the potential for initiatives evaluating provider and facility quality of care, the need for which is supported by literature on overuse of surgery.