Part B is not an examination of what you know. It is an examination of what you can deliver — under time pressure, in front of an examiner, in a format that rewards structure and penalises hesitation. Most candidates preparing for the MRCS Part B OSCE revise exhaustively for the former and almost not at all for the latter.
The revision hours go in. The knowledge accumulates. And then, inside the surgical viva station, something goes wrong that no amount of reading could have prevented — because reading was never the problem.
Say Your Answers Out Loud
There is a version of every answer that exists only in your head. It is always coherent, always complete, always well-sequenced. It is also an entirely unreliable predictor of OSCE performance.
The spoken version — produced in real time, under observation, without the safety net of internal revision — is a different thing altogether. It is where structure collapses mid-sentence, where clinical vocabulary becomes elusive, where the probe “go on” produces a silence rather than a continuation. Ericsson’s foundational work on deliberate practice established that expert performance is not a product of passive accumulation but of structured, effortful rehearsal under conditions that replicate the target task [1]. Moulaert and colleagues confirmed that how one practises matters as much as how much one practises — and that repetition alone, without the deliberate conditions of performance, produces diminishing returns [2].
Duvivier et al. (2011) tested 875 medical students across three years and identified four components of deliberate practice. Which ones actually predicted clinical skill performance?
Source: Duvivier RJ et al. BMC Med Educ. 2011;11:101. N=875 students (90% participation).
Begin from the first week of revision. Every image. Every station type. Every clinical scenario. The discomfort is productive. It fades. What remains is a candidate who has already done, several hundred times, the thing they are being asked to do on exam day.
Structure Is the Examiner’s First Question
An examiner assessing a nine-minute MRCS OSCE station is not simply collecting correct facts. They are evaluating whether this candidate thinks like a clinician — whether, given a set of information, they can impose an appropriate order on it and lead the examiner through it. Structure is therefore not a stylistic preference. It is the framework within which all other marks are awarded, and it is assessed before a single item of content is heard.
Anatomy and surgical pathology stations demand orientation before identification, identification before function, function before clinical application. An examiner presented with a prosection wants to be led — “This is a posterior view of the right upper limb at the level of the elbow, demonstrating the contents of the cubital fossa” — before a single structure is named.
Clinical examination stations require a presentation, not a recitation. Diagnosis or differential first, relevant positive and negative findings second, investigative and management pathway third. Presenting findings in the order they were elicited mirrors a student ward round.
History-taking stations test logical prioritisation under conversational pressure. A candidate who reaches the safety-critical history in the final minute of a nine-minute viva station has failed structurally regardless of the quality of their interpersonal technique.
Applied surgical science and data interpretation stations require interpretation, not description. Presenting an ABG as “pH 7.28, pCO² 55, bicarbonate 28” scores nothing. Interpreting it as a compensated respiratory acidosis in a post-operative thoracotomy patient, and framing the clinical implication immediately, demonstrates the reasoning the mark scheme is awarding.
Communication stations are assessed on structure as much as content. Communication skills in surgical residency have been shown to require deliberate, structured practice across all levels of training — it is not a trait but a learned competency [3].
Across all station formats: decide your opening sentence before you begin, and practise it until it is automatic.
To Probe or Not to Probe: The Examiner’s Dilemma
The probe — “go on”, “and what else?”, a deliberate pause — is the most misread signal in a surgical viva or OSCE station. The prevailing candidate interpretation is that the probe signals error. What follows is a retraction of a correct answer, an apology, and a collapse of the structure that had been carrying the station.
The examiner is managing a genuine tension. Each question carries a maximum of two or four marks. The decision to probe — investing fifteen or twenty seconds of station time — versus moving on is a real-time calculation about whether the candidate has more to give. A candidate who responds to “go on” with a fluent continuation gets probed generously. A candidate who freezes signals the ceiling has been reached; the examiner moves on.
Research consistently demonstrates that OSCE-related anxiety does not itself cause failure — six out of eight studies in one systematic review found no significant association between anxiety levels and OSCE performance [4]. The freeze is not anxiety. It is an undertrained response to an unfamiliar stimulus.
The probe is not a correction. It is a real-time calculation — and candidate behaviour determines the outcome.
<div style="background:var(--card);border-radius:8px;padding:14px 18px;text-align:center;margin-bottom:12px;">
<div style="font-size:14px;font-weight:500;color:var(--midnight);">Candidate gives a partial answer</div>
<div style="font-size:12px;color:var(--muted);margin-top:4px;">1 mark earned · 1 mark still available</div>
</div>
<div style="text-align:center;margin-bottom:12px;font-size:20px;color:var(--magenta);">↓</div>
<div style="background:var(--midnight);border-radius:8px;padding:14px 18px;text-align:center;margin-bottom:16px;">
<div style="font-size:14px;font-weight:500;color:#fff;">Examiner calculates</div>
<div style="font-size:12px;color:var(--lilac);margin-top:4px;">Probe (20s cost) vs Move on (preserve time for remaining questions)</div>
</div>
<div style="display:grid;grid-template-columns:1fr 1fr;gap:12px;margin-bottom:16px;">
<div>
<div style="background:var(--gb);border-radius:8px;padding:12px 14px;text-align:center;margin-bottom:8px;">
<div style="font-size:12px;color:var(--gd);font-weight:500;margin-bottom:4px;">Candidate seems to have more</div>
<div style="font-size:22px;color:var(--gm);">↓</div>
<div style="font-size:13px;color:var(--gd);font-weight:500;">“Go on.”</div>
</div>
<div style="display:grid;grid-template-columns:1fr 1fr;gap:6px;">
<div style="background:var(--gb);border-radius:6px;padding:10px;text-align:center;">
<div style="font-size:11px;color:var(--gd);font-weight:500;margin-bottom:4px;">Fluent continuation</div>
<div style="font-size:18px;font-weight:500;color:var(--gm);">2 marks</div>
</div>
<div style="background:var(--rb);border-radius:6px;padding:10px;text-align:center;">
<div style="font-size:11px;color:var(--rd);font-weight:500;margin-bottom:4px;">Freeze or retraction</div>
<div style="font-size:12px;color:var(--rm);margin-top:2px;">Examiner moves on</div>
</div>
</div>
</div>
<div>
<div style="background:#F1EFE8;border-radius:8px;padding:12px 14px;text-align:center;margin-bottom:8px;">
<div style="font-size:12px;color:#5F5E5A;font-weight:500;margin-bottom:4px;">Ceiling reached</div>
<div style="font-size:22px;color:#888780;">↓</div>
<div style="font-size:13px;color:#5F5E5A;font-weight:500;">Examiner moves on</div>
</div>
<div style="background:#F1EFE8;border-radius:6px;padding:10px;text-align:center;">
<div style="font-size:11px;color:#5F5E5A;font-weight:500;margin-bottom:4px;">1 mark awarded</div>
<div style="font-size:12px;color:#888780;margin-top:2px;">Next question</div>
</div>
</div>
</div>
<div class="finding-box">The probe is an invitation, not a correction. The examiner didn’t move on — that is information. The next correct sentence is the only appropriate response.</div>
In a borderline result, the difference between thirteen and fourteen marks — between amber and green — is often decided by how a candidate responds to a probe in the closing minutes of a station. The examiner did not move on. That is information.
Nine Minutes Is a Design Decision, Not a Constraint
The structure of a Part B station — twenty questions, nine minutes — is calibrated to produce time pressure that distinguishes candidates who have practised from those who have not. MRCS Part B carries a pass rate of approximately 50%, with candidates permitted up to four attempts.
Researchers have investigated every measurable predictor of Part B performance — knowledge score, training grade, deanery, ethnicity, first language, and exam date. The results are striking.
Researchers combined the most important measurable factors in surgical training — knowledge score, training grade, deanery, ethnicity, language, and exam date.
Source: Cleland J et al. Ann R Coll Surg Engl. 2018;100(6):424–427. Full multivariate model R²=0.22, n=2,612 candidates.
High-mark and safety-critical items must be addressed first — not because they are more interesting, but because they carry the marks that cannot be recovered if the station ends before they are reached. Set a stopwatch. Run the station. Stop at nine minutes regardless of where you are.
The Mark Scheme Is the Starting Point, Not the Afterthought
Most candidates find that when they score a practice station against the mark scheme rather than their own impression of it, the gap between the two is instructive. This is not surprising — the literature on self-assessment in health professions education consistently finds that students overestimate their performance, with clinical teachers and examiners assigning substantially lower scores than candidates give themselves [5, 6].
Among 426 medical students assessed on clinical performance — how accurately did they judge their own results?
Source: PMC11515314. N=426 first-semester medical students. ρ=−0.590 (p<0.001) between actual score and self-assessment.
It is worth considering the mark scheme not as an afterthought but as the starting point — the map from which the revision territory is drawn. Structured, written feedback against a mark scheme has been shown to be both desired by candidates and effective at identifying gaps that self-assessment routinely misses [7].
After their OSCE, students received three types of feedback. How many reviewed each?
Source: Pecaric MR et al. PubMed 28521646. N=92 first-year, 86 second-year students.
The Borderline Performance: A Diagnostic, Not a Verdict
The borderline result is the most instructive outcome the MRCS Part B produces, and the most misunderstood. It is not a near-miss. It is a precise diagnosis. The candidate who scores amber across multiple domains has not fallen short through ignorance — in almost every case, the knowledge was present. What failed was the delivery.
MRCS performance has been shown to vary significantly by CST deanery, suggesting that the gap between passing and borderline candidates is substantially determined by the quality and structure of preparation — not by innate ability. This is a trainable gap.
Part B pass rates vary significantly by training deanery (p=0.013). Odds ratios for passing at first attempt, relative to the lowest-performing reference group.
Sources: Rashid P et al. Med Educ 2021; Ellis R et al. BJS 2021.
What a borderline OSCE performance looks like in practice:
- Answers that are broadly correct but unstructured — the right content, in the wrong order, without a clinical frame
- Safety-critical points mentioned incidentally rather than led with — present in the answer, absent from the mark scheme score
- Freezing on the probe — interpreting “go on” as correction rather than invitation, retracting a correct answer and losing both the mark and composure
- Running out of time before the final questions — front-loaded detail on early questions at the expense of the marks-dense later ones
- Describing data rather than interpreting it — reading the numbers rather than telling the examiner what they mean for this patient, now
- Communication that is clinically accurate but patient-uncentred — technically complete, experientially inadequate
Each of these is recoverable. None requires more knowledge. All require deliberate, repeated practice of a specific skill. Moon and colleagues demonstrated that deliberate practice, structured around mastery learning with rigorous outcome measurement, is an effective remediation strategy specifically for candidates performing below expectation in clinical skills assessments [8].
Borderline and failing candidates were invited to a structured deliberate practice workshop before their end-of-year OSCE. Both groups had the same prior performance profile.
Source: Times Higher Education, 2025. Reflective deliberate practice remediation workshop. Both groups had failed or were borderline on mid-year OSCE assessments.
The borderline result, in almost every case, is not discovered until the feedback letter arrives. By then, the sitting is over. The work is to surface the borderline before the exam does.
References
- Ericsson KA, Krampe RT, Tesch-Römer C. The role of deliberate practice in the acquisition of expert performance. Psychol Rev. 1993;100(3):363–406.
- Moulaert V et al. The effects of deliberate practice in undergraduate medical education. Med Educ. 2004;38(10):1044–52.
- Griffen FD, et al. Communication skills training in surgical residency. J Surg Educ. 2014.
- Sturpe DA. Systematic review of student anxiety and performance during OSCEs. Curr Pharm Teach Learn. 2020.
- Hawkins SC et al. The accuracy of health professions students’ self-assessments. Am J Pharm Educ. 2021;85(4):8405.
- Eva KW, Regehr G. Self-assessment in the health professions. Acad Med. 2005;80(10 Suppl):S46–54.
- Sterz J et al. Implementation of written structured feedback into a surgical OSCE. BMC Med Educ. 2021;21:210.
- Moon SH et al. Deliberate practice as an effective remediation strategy. J Korean Med Sci. 2019;34(11):e84.
- Ali et al. Performance under pressure: OSCE Anxiety Scale. ResearchSquare. 2025.
- Duvivier RJ et al. The role of deliberate practice in clinical skills. BMC Med Educ. 2011;11:101.
- Rashid P et al. MRCS performance by CST location. Med Educ. 2021.
- Cleland J et al. Predictors of success in MRCS. Ann R Coll Surg Engl. 2018;100(6):424–427.
- Ellis R et al. MRCS performance predicts surgical career outcomes. BJS. 2021;108(Suppl 6).
- Pecaric MR et al. Medical students review of formative OSCE scores. PubMed 28521646. 2017.
- Intercollegiate Surgical Curriculum Programme. MRCS examination structure. rcseng.ac.uk. Accessed May 2026.