Frequently asked questions

ClinicalSim is a voice-based AI clinical simulation built for communication practice in medical education. Learners talk through high-stakes conversations with AI patients and receive rubric-scored feedback tied to the transcript. Named physicians write and review each case, and the report names the competency and communication frameworks it uses.

No. USMLE discontinued Step 2 CS in 2021, and no national successor now assesses spoken clinical communication. ClinicalSim adds repeatable practice and formative evidence within a program. It is not a national exam or a replacement for human judgment.

ClinicalSim extends standardized patient (SP) programs and does not replace them. SP encounters remain important for live coaching and high-stakes assessment. ClinicalSim adds repeatable practice between those encounters, without requiring another actor or room for each attempt.

A general AI assistant can improvise a conversation. ClinicalSim uses physician-authored cases with fixed objectives, named frameworks, repeatable scoring logic, and transcript evidence a faculty member can review across learners.

Communication remediation is the structured practice and feedback a program uses to help a learner improve interpersonal and communication skills. In a CERA survey of 267 family medicine program directors, 93% reported at least one resident in remediation during the prior three years, and 50% selected an accessible remediation toolkit as the most important tool for improving the process (Frazier et al., Family Medicine, 2021). ClinicalSim adds repeatable practice and transcript evidence for faculty and CCC review.

Learners practice privately, without a faculty observer in the encounter. They can repeat a case before sharing the report with a coach or program.

Voice practice can surface pacing, silence, word choice, and real-time responses to emotion. It cannot assess eye contact, body language, or physical presence, so those skills remain part of live encounters and human review.

Learners can practice on any device, a phone, a tablet, or a desktop, wherever they are. ClinicalSim runs entirely in the browser, so there is nothing to download and no app to install.

A typical ClinicalSim encounter takes between 3 and 10 minutes, short enough to fit into a clinical day and repeat as often as a learner needs.

ClinicalSim currently supports English.

Named physicians write and review each case for clinical accuracy, standards alignment, and fit with its learning objectives. The full process is documented on the methodology page.

Faculty can work with ClinicalSim staff to develop a case for a specific learner group and communication objective. Case development support depends on the pilot or study scope.

Before release, each case is run repeatedly to check the AI patient's role, the scoring and feedback, and whether the case gives learners a fair chance to show each assessed behavior. Clinical and technical reviewers revise the case before publication.

Each completed simulation produces scored fields from the case rubric and a narrative report. My Progress uses those scores to show assignment completion and competency trends, based on the program's configuration. A case that cannot assess performance above a certain level is excluded from the overall rollup when that ceiling would make strong performance look low. Learners do not see peer comparisons.

Each encounter is a voice conversation between the learner and an AI patient designed for the case, captured as a timestamped transcript. For every scored competency and framework step, the platform draws one or two verbatim excerpts that demonstrate the behavior, or documents its absence, so each score is traceable to the moment that supports it, rather than serving as an unexplained rating.

GME cases use the specialty-specific ACGME Milestones 2.0 and quote the relevant descriptors from each specialty's document. The report scores only the subcompetencies the case gives the learner a fair chance to show. Faculty can use the report as one source of evidence in CCC review.

UME cases align to the Foundational Competencies for Undergraduate Medical Education (AAMC, AACOM, and ACGME) and the AAMC Core Entrustable Professional Activities (EPAs) for Entering Residency. Because the Foundational Competencies are not published with the milestones' five-level scale, ClinicalSim does not assign a numeric level for UME; it records whether each competency was demonstrated and scores performance through the applied communication or skill rubric. Entrustment remains a program decision that this evidence informs.

No. The ACGME Milestones are formative and were not designed for high-stakes external decisions, and ClinicalSim treats milestone-aligned output accordingly, as evidence that informs program judgment, not as a stand-alone basis for high-stakes decisions.

A competency framework is the governing-body standard a case uses, such as the ACGME Milestones 2.0 in graduate medical education or the Foundational Competencies in undergraduate medical education. A communication framework is a published model of communication behavior, such as SPIKES or Calgary-Cambridge. A rubric turns the relevant framework into scored items for that case.

Yes. A case may use more than one communication framework when each one measures a different part of the task. Every framework names its published source, and programs may add their own rubrics. The report scores each framework separately.

ClinicalSim uses representative frameworks including SPIKES, the Kalamazoo Essential Elements Communication Checklist (KEECC-A), SEGUE, NURSE, REMAP, SBAR, I-PASS, TeamSTEPPS, CANDOR, and Calgary-Cambridge. Full citations are maintained in the ClinicalSim Frameworks Bibliography; see the full methodology page for representative sources.

Each encounter produces a report with rubric scores, strengths, priority gaps, and suggested next steps. Every scored item cites evidence from the transcript. The report is formative evidence for the learner and faculty reviewer, not a verdict or a substitute for human judgment.

Yes. Faculty can rehearse corrective feedback, professionalism concerns, bedside teaching, and peer conversations. They receive the same framework-based report and transcript evidence used in learner cases.

ClinicalSim currently supports graduate medical education, undergraduate medical education, communication remediation, and faculty development. Cases and frameworks change with the learner, specialty, and conversation.

Faculty with program access can review assigned learners' progress, recordings, transcripts, and feedback without attending the practice session. Program permissions determine which learner records each faculty member can see.

Yes. Each practice report maps observed behavior to the relevant milestone and cites the learner's words. A CCC can review it alongside faculty observation and the other evidence it already uses. The report does not replace faculty judgment or the committee's decision.

ClinicalSim stores recordings, transcripts, account data, and program data to provide practice history and faculty review. Access depends on the learner's institutional arrangement. The trust and data handling page describes current collection, access, retention, and deletion practices.

Yes. Depending on the study, ClinicalSim can provide platform access, case development support, participant onboarding, structured exports of study data, and technical documentation for an IRB submission. The research page explains how to propose a study.

Every score cites one or two excerpts from the transcript, so a reviewer can check the rating against what the learner said. ClinicalSim is testing score performance in pilots and does not claim that its ratings are more accurate or fairer than faculty judgment. The report is formative evidence that a faculty member or committee can accept, question, or override.