Evidence-based decision support for pediatric rheumatology reduces diagnostic errors.

Area of Science:

• Medical Informatics
• Clinical Decision Support
• Pediatric Rheumatology

Background:

• Global shortage of pediatric rheumatology specialists limits access to care.
• Non-specialists often face challenges in diagnosing complex pediatric rheumatologic disorders.
• Diagnostic decision support software (DDSS) was evaluated to improve non-specialist diagnostic accuracy.

Purpose of the Study:

• To assess the efficacy of DDSS in improving diagnostic accuracy for pediatric rheumatologic conditions among non-specialists.
• To determine if DDSS can help alleviate the shortage of pediatric rheumatology experts.

Main Methods:

• Clinicians used vignettes of actual clinical cases to generate differential diagnoses before and after using the SimulConsult® DDSS.
• The DDSS tool utilized Bayesian pattern matching, incorporating disease specifics, findings, and temporal information.
• Efficacy was measured by comparing diagnostic accuracy with and without DDSS assistance.

Main Results:

• Diagnostic errors decreased significantly from 28% (unaided) to 15% (aided) (p < 0.0001).
• Improvement was most pronounced in emergency medicine physicians and clinicians with less than 10 years of practice.
• Error reduction was achieved even with clinicians using additional resources before DDSS use.

Conclusions:

• DDSS can effectively reduce diagnostic errors made by generalists.
• The software improves the utilization of relevant clinical information by non-specialists.
• DDSS shows potential to enhance generalists' diagnostic capabilities in underserved specialties like pediatric rheumatology.