Session Type: ACR Poster Session C
Session Time: 9:00AM-11:00AM
Background/Purpose: This projects seeks to respond to the critical shortage of pediatric rheumatologists encapsulating the diagnostic information of the field in an advanced diagnostic decision support software (DDSS) tool. We present here an assessment of the benefits of such rheumatology assistance in the evaluation of actual case vignettes.
Methods: The evaluation used the SimulConsult DDSS tool, based on Bayesian pattern matching with temporal onset of each finding in each disease. The software covered 5,405 diseases, taking into account the incidence, treatability, relative frequency, age of onset and age of disappearance for each finding in each disease. The output includes a rank-ordered differential diagnosis and a display of clinical and lab findings ranked by pertinence (measured by the ability of a finding to modify the differential diagnosis) and cost-effectiveness in distinguishing among diagnoses. Rheumatology information was entered by junior clinician curators primarily from textbook material and edited by senior clinicians. A checklist of 46 common rheumatologic findings was developed to assist in both curation of information and entry of findings in clinical use.
Results: Twenty-six testers were asked to evaluate 8 case vignettes of real patients with laboratory-established diagnoses (6 had pediatric rheumatologic diagnoses; 2 had other conditions with some rheumatologic findings). Of the 26 testers, 13 were “junior” clinicians in the final year of residency or first post-residency year. The remaining 13 had been practicing for at least 10 years (“senior”). Ten of the 26 testers were pediatric rheumatologists, 9 were pediatric emergency medicine physicians and 7 were general pediatricians. Clinician testers generated a differential diagnosis before and after using diagnostic decision support software. Overall, testers demonstrated a significant reduction in diagnostic errors following introduction of the software, from 28% unaided to 15% using decision support (p<0.001). Error reduction was significantly larger for emergency medicine physicians compared to generalists and rheumatologist (p=0.013). This error reduction occurred despite the fact that testers employed an “open book” approach to generating their initial lists of potential diagnoses, spending an average of 8.6 minutes using sources of medical information generally available on the internet before using the diagnostic software.
Conclusion: Use of DDSS can reduce diagnostic error, cutting a greater amount of error for generalists, but a greater percentage of error for rheumatologists. Junior rheumatologists equipped with decision support were able to rival the diagnostic accuracy of senior colleagues at baseline, an improvement that could further extend the availability of pediatric rheumatology consultations. These findings suggest that decision support can reduce diagnostic errors and improve use of relevant information by practitioners, potentially helping to address the shortage of experts in pediatric rheumatology and similarly underserved specialties.
To cite this abstract in AMA style:Athreya B, Son MB, Hausmann JS, Ang E, Zurakowski D, Segal M, Sundel R. Evidence-Based Decision Support for Pediatric Rheumatology Reduces Diagnostic Errors, with the Potential to Reduce Capacity Shortage [abstract]. Arthritis Rheumatol. 2015; 67 (suppl 10). https://acrabstracts.org/abstract/evidence-based-decision-support-for-pediatric-rheumatology-reduces-diagnostic-errors-with-the-potential-to-reduce-capacity-shortage/. Accessed November 28, 2020.
« Back to 2015 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/evidence-based-decision-support-for-pediatric-rheumatology-reduces-diagnostic-errors-with-the-potential-to-reduce-capacity-shortage/