Pediatric Musculoskeletal Radiograph Interpretation Among Orthopaedic Surgery Postgraduate Trainees: A Competency-Based Education Intervention
Carl A. Starvaggi, Maryse Bouchard, Mark Camp, Sasha Carsen, Alexandra J. Mortimer, Ryan Katchky, Ron El-Hawary, Lori A. Archer, Neil Saran, Jennifer Stimec, Kai Ziebarth, Kristina Keitel, Kathy BoutisBackground:
Pediatric musculoskeletal radiographs (pMSK) are the primary imaging modality for evaluating injured children, yet there is no widely adopted, standardized approach to teach and assess this specific skill among orthopaedic surgery postgraduate trainees (OS-PGT). We evaluated an online educational and assessment intervention using pMSK radiographs to improve fracture/dislocation detection and risk stratification among OS-PGT.
Methods:
We conducted a multicenter prospective cohort study using a web-based educational intervention, recruiting a convenience sample of OS-PGT. The intervention included 1609 pMSK radiograph cases organized into 6 body regions. OS-PGT judged whether a fracture/dislocation was present or absent, and when present, localized the abnormality and assigned a risk category. Participants received immediate feedback after every case, continuing until reaching a predefined 90% performance standard. Following the intervention period, participants were invited to complete a post-intervention survey.
Results:
A total of 64,489 unique case interpretations were generated by 115 OS-PGT. From initial to final, 25 cases were completed, and diagnostic accuracy increased by 10.5% (95% CI: 9.4%, 11.6%), localization accuracy by 11.9% (95% CI: 10.7%, 13.0%), and risk stratification accuracy by 20.5% (95% CI: 18.9%, 22.1%), with the greatest gains observed for ankle/foot radiographs. Across 246 attempts to reach the performance standard, 192 (78.0%) were successful. The median number of cases required per body region was 141 (IQR: 96, 313), completed in a median of 34.9 minutes (IQR: 23.8, 77.6). Error rates were lower when participants were certain of their answers (7.8% incorrect “definite” interpretations vs. 23.7% “probable”). Forty-five (39.1%) participants completed the post-intervention survey, and 38 (84.6%) supported integrating the intervention into orthopaedic residency training.
Conclusion:
This study shows that a structured, web-based intervention combining performance-based assessment with case-based practice led to substantial and efficient learning gains in pMSK radiograph interpretation among OS-PGT and helped identify diagnostic challenges. These findings support the use of such educational interventions as a practical strategy to supplement orthopaedic residency training in targeted diagnostic skills.
Levels of Evidence:
Level II—prospective cohort study.