Evidence
Start with a complete conversation and the feedback a participant receives. Then review how we build cases, score practice, and define the limits of those scores.
4 complete pediatric examples include the recording, full transcript, and feedback report, with no sign-in required. They let you inspect the words behind a score and judge whether the feedback would support practice in your program. They don't establish training effectiveness or patient outcomes.
Open the complete examplesOur methodology explains case development, the scoring process, and how transcript evidence supports feedback. It also sets out what a score can and can't establish.
Read the methodologyThe framework library links to the sources behind the named standards. Their published reliability comes from the settings and raters those studies tested. It doesn't establish the reliability of AI scoring in ClinicalSim.
Review the framework sourcesOur research page lists conference presentations, including records from the International Pediatric Simulation Symposium and Workshops, the Pediatric Academic Societies Meeting, and the Critical Care Congress. A presentation record isn't evidence of a confirmed product outcome.
View presentations and research collaborationThe communication frameworks we apply were built for trained human raters. Applying them through AI in simulated encounters goes beyond the settings in which their published reliability was established.
Consistency between model runs isn't agreement with expert human raters. We haven't completed faculty-rater validation on a customer's own rubric, and we don't claim our scores are more accurate than faculty judgment.
Read the scoring method and limitsClinicalSim output is formative. It is evidence that informs program judgment, and it is not intended for employment, credentialing, privileging, licensure, or other high-stakes decisions.
ClinicalSim does not predict patient experience scores, readmissions, safety events, claims, or other clinical or business outcomes. Any comparison with institution-held outcome data requires a separate study plan.
Studies of ClinicalSim are underway at partner institutions. We won't publish their outcomes until the study owners confirm the record in writing. The public conference list records presentations, and it doesn't imply a peer-reviewed outcome publication.
Discuss a research questionThe evaluation guide covers intended use, rollout needs, and who takes part in the review. The trust page describes data handling and the questions IT and privacy teams should examine.
We'll walk through the product, the standard you want participants to practice against, and the evidence your institution needs to review.
Talk with us