November 18, 2026 · Journal of the American College of Surgeons · DOI: 10.1097/XCS.0000000000002296

Role of Surgeon-Model Interaction in the Real-World Performance of Surgical Risk Prediction Tools.

Listen to this summary

This study examined how surgeons' interpretations of surgical risk model predictions, and their disagreements with those predictions, affect real-world performance of three deployed tools (NSQIP-SRC, POTTER, and MySurgeryRisk) across two academic centers. Across 1,702 predictions from 31 surgeons, surgeons disagreed with the model 53% of the time, but adding their responses to model predictions did not significantly improve discrimination (AUCs rose to 0.68, 0.83, and 0.86, respectively, none statistically significant). The authors conclude that surgeon-model disagreement is not uniformly helpful or harmful, so real-world performance should be evaluated by considering how clinicians interpret and modify predictions, not only by conventional model metrics.

Sofia Volpi, Tyler Loftus, Jayson Marwaha, Margaret Berrigan, Zeynep Bilgili, Anupamaa Seshadri, Daniel Bent, Stephen Odom, Charles Parsons, Alok Gupta, Charles Cook, Azra Bihorac, Gabriel Brat

This is one of 33,000+ journals available on OSLR. Try it free for 14 days.

Free 14-day trial. 33,000+ journals. Cancel anytime.

14-day free trial. No commitment.

“

"Oslr has become part of my weekly routine on my day off. The clinical relevance of the summaries is outstanding — I'd rate it 9/10. Being able to consume research hands-free is a huge advantage for busy physicians."

Dr. Jennifer Thompson

Dr. Jennifer Thompson

Portland, OR

Stay current without falling behind

33,000+ journals. 3-minute audio summaries. Free for 14 days.

Download on the App StoreGet it on Google Play