Peer simulated patients: what a program gains, and what it hands to faculty
A 2020 review of 12 studies found no skill difference between peer and standardized patients. A 2026 review says the saving can shift work to faculty.
ClinicalSim Team
ClinicalSim
A peer simulated patient program looks like the cheap way out of a standardized patient shortage. Senior students play the patient, junior students get more practice, and the actor budget stays where it was. The evidence says the learning holds up for a good share of what a medical school teaches, and it also says the work doesn't disappear. It moves off the SP payroll and onto the educators who recruit, train, calibrate, and look after the students doing the portraying.
What is a peer simulated patient?
A peer simulated patient is a medical student, usually a senior one, trained to portray a patient so that other students can practice clinical encounters. A 2020 systematic review of 12 studies, five of them randomized trials comparing peers with simulated patients, found no difference in communication skills between the two groups and equal or greater gains in empathy with peers (Dalwood et al., Medical Education, 2020). A 2026 narrative review by Pollak and Feinstein in Postgraduate Medical Journal concludes that "Institutions should view PSP programs as complementary to, rather than a replacement for, occupational SPs."
So the useful question for a clerkship or simulation director is which encounters a peer can carry, and what the school has to build so peers carry them safely.
What the trials show about learning
The Dalwood review is the best summary of the comparative evidence, and its authors are candid about its limits. The 12 studies covered entry-level health care students broadly, not only medical students, and varied so much in design, outcome measures, and how the peer patients were trained that the authors couldn't say which design features make a peer program work. They called for more rigorous research across a wider range of outcomes.
One randomized trial in a Heidelberg pediatrics rotation shows what the comparison looks like up close, and its authors reported it in three papers. Sixty-nine fifth-year students trained on the same nine cases, each pairing a common pediatric problem with a communication problem involving parents, with either peer role-play (34 students) or standardized patients (35). Students rated the two methods equally realistic (5.60 against 5.53 on a six-point scale), but rated standardized patients as more worthwhile, more useful for communication training, and better preparation for real parent encounters, at 5.00 against 4.29 (p<.009) on that last measure (Bosse et al., BMC Medical Education, 2010). Those are student perceptions, not performance.
The same trial measured performance in a six-station OSCE with standardized patients, scored with the Calgary-Cambridge guide. The peer role-play group scored 81.6 percent against 78.0 percent for the standardized patient group (p=.021), mostly because of higher scores in one domain, understanding the parents' perspective (Bosse et al., Patient Education and Counseling, 2012). The authors later noted that the training ran over a few weeks and that they couldn't judge whether the skills lasted (Bosse et al., BMC Medical Education, 2015).
So the honest reading is that peers can teach communication about as well as SPs for students at this level, in cases like these, over a short course. That's useful, and it's also a smaller finding than "peers work as well as SPs."
Where the savings come from, and what the count leaves out
The 2015 paper on that trial is the clearest published accounting of what a peer program saves. Training with standardized patients took 172 man-hours against 112 for peer role-play, 53.6 percent more, counting SP time and tutor time (Bosse et al., BMC Medical Education, 2015).
The detail matters more than the total. Tutor hours were identical in both arms, and the entire 60-hour gap was SP time in the training sessions, because the two groups used the same cases and tutors and the only change was who played the patient. Both arms still used standardized patients for the OSCE.
The count covered only SP and tutor hours, so the student role-players' own time wasn't in it, and the authors also left out contingency costs and the general running of their SP training center. They write that they "cannot draw conclusions regarding cost-effectiveness" of the methods over the long run.
A school reading that study should take away one number and one warning. The number is that swapping actors for students in practice sessions saved actor hours in one well-run program. The warning is that nobody counted the hours it takes to turn a student into a reliable patient.
What a peer program hands to faculty instead
Pollak and Feinstein are direct about where that cost goes. Peer programs "require upfront investment in case design, training, supervision, and QA," and they "may shift workload from hiring actors to building educator and governance capacity." Their implementation framework lists the jobs that land on educators:
- Recruiting and selecting senior students for reliability, communication skills, and motivation to teach.
- Training them in portrayal, consistency, and script fidelity, and separately in giving feedback.
- Quality assurance through calibration, observation, and audits.
- Psychological safety through prebriefing, role clarity, confidentiality norms, stop rules, and debriefing.
- Choosing which scenarios peers portray and which go to professional SPs.
- Evaluating outcomes for junior learners, peer patients, and the institution.
The educators who train those portrayals are doing simulation education, and the Certified Healthcare Simulation Educator exam from the Society for Simulation in Healthcare is the credential built for that role.
Feedback is a weak point in at least one published program. In a study of 50 second-year students at one Turkish medical school, peer patients scored above 90 percent on seven of eight observed portrayal items but only 32 percent on giving effective feedback to the student playing the physician (Tengiz et al., Medical Education Online, 2022). The students had been trained in giving feedback, and the authors still found it biased. That school had no standardized patient program to compare against, so the study shows where peer feedback struggles rather than how it compares with SP feedback.
The role-players carry risk of their own. Pollak and Feinstein warn that peer patients face "emotional labor, performance anxiety, and role strain, particularly when portraying emotionally charged scenarios," and they list familiarity effects, reduced professional distance, and confidentiality breaches among the program's risks. Their governance recommendations start with adopting SP best practice standards, then add an opt-out for peer patients, an escalation pathway for distress, and confidentiality agreements. ClinicalSim covered the SP version of those protections in what the ASPE standards ask of a standardized patient program, and a school running a peer program owes its student actors the same ones.
Which encounters suit a peer, and which belong to an SP
The most practical part of the 2026 review is a suitability matrix. It rates foundational history taking and rapport building, and counseling and health behavior change, as highly suitable for peers. Breaking bad news, disclosing an error, and handling conflict are rated moderate, with professional SPs often preferred. Severe mental health crises are rated low to moderate, intimate physical exams low, and summative high-stakes OSCE stations low unless the program is mature, with a professional SP preferred in all three.
For assessment, the review says to "Avoid summative scoring by PSPs" and calls for "explicit separation of teaching and grading roles to minimize bias and conflicts of interest." The review doesn't say how the ratings were set beyond citing studies for each row, and its table notes that suitability varies with local context, learner level, and program maturity. A program should treat the matrix that way: a reasoned starting point, checked against its own cases.
Four decisions to write down before the first session
A school that wants the extra practice a peer program gives can set it up without pretending the work is free. Four decisions cover most of it:
- Which cases go to peers. Start with history taking, rapport, and counseling, where the review rates peers highest, and keep high-affect cases and summative stations with SPs until the program has a track record.
- Who trains and calibrates the peers, and for how many hours. Budget those educator hours explicitly, ideally from the actor hours the program expects to save, so the saving and the new cost sit on the same line.
- How a student declines a role or stops a session. Write the opt-out and the stop rule before anyone is cast, and name who responds.
- Peers never score summatively. Peer patients give formative feedback. Grades come from faculty or trained SPs.
The literature doesn't answer any of these for a given school, so each one gets set locally and written where a new coordinator can find it, the same way a program turns the ASPE standards into local limits.
Where AI patients fit
ClinicalSim is a voice-based AI patient simulation platform for practicing and measuring clinical communication in medical education. Students can practice with AI patients between scheduled encounters, which is the same gap peer programs are built to fill, and that practice complements a standardized patient program. AI patients don't do what a trained SP does: respond as a person in the room, give feedback from having been on the receiving end of the conversation, and carry a summative OSCE station. ClinicalSim's comparison of AI clinical simulation and standardized patients sets out where each is stronger, and why standardized patient programs run out of capacity covers the constraint all three approaches are working against.
Extend your SP program, don't replace it.
References
- Pollak U, Feinstein Y. Training medical students as peer simulated patients in undergraduate medical education: a narrative review and implementation framework. Postgraduate Medical Journal. 2026. doi:10.1093/postmj/qgag100
- Dalwood N, Bowles KA, Williams C, Morgan P, Pritchard S, Blackstock F. Students as patients: a systematic review of peer simulation in health care professional education. Medical Education. 2020. doi:10.1111/medu.14058
- Bosse HM, Nickel M, Huwendiek S, Jünger J, Schultz JH, Nikendei C. Peer role-play and standardised patients in communication training: a comparative study on the student perspective on acceptability, realism, and perceived effect. BMC Medical Education. 2010. doi:10.1186/1472-6920-10-27
- Bosse HM, Schultz JH, Nickel M, Lutz T, Möltner A, Jünger J, Huwendiek S, Nikendei C. The effect of using standardized patients or peer role play on ratings of undergraduate communication training: a randomized controlled trial. Patient Education and Counseling. 2012. doi:10.1016/j.pec.2011.10.007
- Bosse HM, Nickel M, Huwendiek S, Schultz JH, Nikendei C. Cost-effectiveness of peer role play and standardized patients in undergraduate communication training. BMC Medical Education. 2015. doi:10.1186/s12909-015-0468-1
- Tengiz FI, Sezer H, Başer A, Şahin H. Can patient-physician interview skills be implemented with peer simulated patients?. Medical Education Online. 2022. doi:10.1080/10872981.2022.2045670