OSCE feedback examples: what examiners really mean
You get your OSCE feedback and it reads, "Good. Some disorganisation." You're staring at a single line that could mean five different things, and you've got another circuit in three weeks. What now.
This page takes common-sounding feedback lines and decodes them into plain meaning you can act on. Every example line here is an illustrative composite written for this post, not a quote from a real examiner or a real mark sheet.
Then we show how to turn each into practice before the next OSCE, with concrete fixes you can rehearse and recognise under time pressure.
What OSCE feedback usually looks like, and why it's so vague
Most OSCE feedback lands as a few words per station, a pat then a poke, and it often reads like this. Illustrative composites for this post, not real quotes or a school's mark sheet: Good rapport, missed safety net questions; thorough history, no differential, plan unclear. Examiners write between candidates, so brevity beats precision.
In examiner comments from Galway, a text-mined analysis of written comments for 1,034 undergraduates across five cohorts, the most frequent word over five years was "good", which the authors call "indicative of potentially non-specific feedback". Generic terms give some reinforcement but may not drive meaningful improvement, so they propose structured, station-specific statements that describe the learning gap and observed behaviours.
One UK national assessment uses its own marking approach, see how CPSA stations are scored, but comments are still often brief because examiners are moving fast between candidates.
What does 'good' or 'satisfactory' actually mean on OSCE feedback?
It usually means you met the expected standard for that station or domain, with no standout errors or standout strengths recorded. It's reassurance that you were safe and competent on the day, but it's not a map for what to fix next or how to improve.
The useful bit sits in the domains and any written line. If you see good or satisfactory across the board, you'll get further by hunting for the one concrete criticism and the missed ticks in history, data gathering, explanation or management, and how those were judged by checklists versus global ratings.
In Year 4 at Monash University Malaysia, 116 students compared face-to-face with enhanced written feedback, and 75% preferred the latter, while the marking rubric, framed as 'what examiners look for', ranked highest in a Monash Malaysia comparison. It measured perceptions at one site, not later performance, and schools differ.
OSCE feedback examples for history stations, decoded
These feedback lines are illustrative composites written for this post. They are not quotes from any examiner, school or real mark sheet.
"Missed red flags"
What it usually means: you closed without safety questions for a time sensitive diagnosis. In the Julian Underhill AI case, you didn't ask about facial weakness, double vision, limb weakness or slurred speech, and missed that he'd go home for two weeks unless told it's urgent. What to practise: before closing, run a fixed red flag bundle for sudden hearing loss.
"Too many closed questions"
What it usually means: too many yes-no questions hid the signal. With Julian, "any pain?" went nowhere because he plays it down when it doesn't hurt. What to practise: start with two open prompts, then funnel, for example "talk me through the morning you noticed it".
"Didn't explore ICE"
What it usually means: you didn't explore ideas, concerns and expectations (eliciting the patient's ICE). With Julian, you missed that he thinks it's wax or fluid, expects syringing or drops, and worries about ringing and looking foolish at work. What to practise: ask "what were you worried it might be?" and "what were you hoping we'd do today?"
"Summary didn't reflect the history"
What it usually means: your recap contradicted what you heard. In Julian's case, you said "blocked ear, probably wax" despite sudden unilateral loss with constant ringing, no pain or discharge, and a stated plan to wait unless you explain urgency. What to practise: a 20 second template, age and job, onset and laterality, key positives and negatives, and the patient's main concern.
"Disorganised history"
What it usually means: messy flow and an unclear timeline. With Julian, you bounced between tinnitus, past colds, then back to onset, so the picture never settled. What to practise: a fixed micro-structure, one liner, chronology, associated otological and neurological symptoms, impact, ICE, and red flags check.
"Didn't check for pertinent negatives"
What it usually means: you skipped discriminators that separate diagnoses. In Julian, you had to be told there was no ear pain or discharge and no facial weakness, double vision, limb weakness or slurred speech, rather than asking them. What to practise: carry a list of which negatives to ask for sudden hearing loss and tick them off before closing.
OSCE feedback examples for communication and the global rating
Illustrative composite lines, not quotes from any real examiner or mark sheet.
"Poor rapport"
What it usually means: The patient didn't feel heard or safe, sparse empathy, little eye contact, no use of their name or concerns. What to practise: Open with name, role, permission, reflect and summarise concerns, add one empathy line; see building rapport in stations.
"No signposting"
What it usually means: You jumped between topics without saying what was next, the patient got lost and your organisation slipped. What to practise: Use brief headers, for example "I'll ask about medicines, then today's symptoms", close with a one-line summary.
"Used jargon"
What it usually means: You used technical terms without checking understanding, the patient looked confused, and marks in communication suffered. What to practise: Swap to plain English, define unavoidable terms, and use teach-back for key points.
"Ran out of time"
What it usually means: You misjudged depth or dwelt on low yield details, then failed to move on, which hits time management. What to practise: Set minute targets per section, cut early when off track, prioritise red flags and decisions, watch the timer.
"Global rating: borderline"
What it usually means: The overall impression sat near the pass line regardless of ticks, with weight on safety and coherence, though schools differ. What to practise: Make structure, safe communication and visible summaries obvious, so the overall read stays safe and organised even if a few boxes are missed.
How do I get more useful feedback from an OSCE examiner or tutor?
Ask for one specific behaviour they observed, when it happened, and what they want to see instead. Ask within the same week while the case is fresh. Open with your self-assessment. Keep it short and concrete so they can answer quickly between cases.
Use Elicit, Support, Plan. Start with your view, ask what to keep, then one change to move your mark next time. That mirrors the elicit, support, plan model on observable behaviours and a clear next step.
At Goethe Frankfurt, a summative third-year surgical OSCE ran. Eight five-minute stations. It involved 351 students and 51 examiners. Students preferred free-text to circled statements. Yet more than half said comments weren't concrete enough. 87.5% of responding students and 91.6% of responding examiners wanted written feedback to continue, and over a quarter of examiners felt writing affected the exam. Student response rate was 38.8%, so read the numbers from a Frankfurt surgical OSCE study cautiously.
- At 3 minutes I hesitated on the differential, what one change would have helped most?
- With 10 minutes to practise one behaviour before next sitting, which would you pick, and what would good look like?
Turning one feedback line into a practice plan
- Pick the single feedback line that cost you the most marks, or that keeps showing up. Any OSCE feedback examples I use here are illustrative composites, not quotes from a real examiner, a named school, or a real mark sheet.
- Translate it into one observable behaviour. "Missed red flags" becomes "before closing, ask the red flag questions for this complaint". With Julian Underhill, 48, sudden one sided hearing loss with tinnitus, use "ask explicitly about neurological symptoms, and what he'll do if nobody explains urgency". That surfaces he'd keep using drops and wait weeks on a time sensitive sudden sensorineural hearing loss.
- Rehearse the behaviour alone in timed reps, across complaints. Ten to twenty seconds per rep, two or three reps per complaint. For structure, use a solo rehearsal routine.
- Re-test with someone watching for that behaviour. Brief them to ignore everything else and mark yes or no, once early and once before you close.
- Log it. Date, station, whether it happened, any friction. Once it is automatic, move to the next feedback line and repeat the five step loop.
Practising with instant feedback between OSCEs
Peers with a checklist. You get real human judgement on rapport, tone and body language, and you can rehearse messy histories that feel like the wards. It takes scheduling, and feedback quality varies, with friends sometimes softening the hard lines you need to hear.
Recorded mocks. Set up your phone, run timed stations, then rewatch. You'll catch hesitations and structure drift, but nobody marks it and clinical gaps can slip by without a second pair of eyes.
AI patient cases. Diagnosica lets you take a history from an AI patient by voice or text, order investigations, commit to a diagnosis and plan, and get a scorecard after every case, on the free tier too, with competency scores and teaching points. Its scoring is competency-based and not mapped to any medical school or exam mark sheet. Every case is written and signed off by a doctor, it runs any hour without booking, and there's a two-minute quick case for daily practice. It has no physical examination practice, and it can't watch your body language or eye contact. It's live and early, so expect rough edges.
Use whichever gives you the most honest reps on your one weak line.
It doesn't give advice about real patients, and it's not a diagnostic system. Treat every output as educational.
Practise history taking Talk or type to an AI patient, then get a scorecard with teaching points after every case, any hour; history taking only, no physical examination. Try a case
What if your feedback says you failed the station?
One failed station rarely decides the whole OSCE on its own. It depends on your school's rules about compensation and minimum station passes, so check your regulations and request a feedback meeting with the assessment lead. Rules differ. Confirm early whether a resit is triggered and what your options are.
Use the failed station's feedback as your plan. Turn each comment into a behaviour to rehearse, like open with a focused safety net. Practise aloud.
For an explainer on resits, compensation and appeals, read what a failed station means. It shows decisions between a single failed station and the final result, then what to check locally.
If your heart rate spikes, it's common. See nerves before the resit for short routines.
Book the feedback meeting, bring three specific questions, and agree next steps. Then schedule short, regular practice targeting the behaviours that tripped you up, not hours of generic revision.
Educational use only, not medical advice. AI-generated; verify clinically against primary sources.
Clinical review pending.
Target weak feedback Pick your one weak feedback line and get scored reps on it; free to start, no card; phone or laptop. Start now
