OET Speaking for doctors

Doctors usually walk into OET Speaking expecting it to be the easy sub-test. You consult with patients every day; a four-minute role-play with a simulated patient sounds like a Tuesday morning. Then the feedback report comes back Band C on Relationship-Building and Patient-Centred Approach, and it makes no sense — you have been communicating with patients competently for years. The report feels wrong.

It is not wrong. It is measuring something specific that medical training actively trains out of you. The communication style that makes you efficient and trusted on a real ward — concise, authoritative, decisive — is close to the opposite of what the OET rubric rewards. This post is about the traps that catch doctors specifically, as distinct from the nurse-focused failure patterns covered in why clinical skill hurts your OET score. The underlying principle overlaps, but the doctor-specific version has its own flavour, and the fixes are worth spelling out on their own.

Before the traps, the frame that matters: the OET examiner is not assessing whether you are a good doctor. They are assessing whether you perform a particular communicative ritual — collaborative, unhurried, explicitly empathetic, structurally redundant. If you have not internalised the nine criteria, read how OET Speaking is actually scored first; everything below assumes you know what the rubric is watching for.

Trap 1: diagnostic efficiency reads as not listening

Doctors are trained to reach a working diagnosis fast. You take a history with a purpose, filter for the relevant, and discard the noise — that efficiency is a clinical virtue and, on a busy clinic day, a necessity. In the OET role-play it costs you marks.

The rubric's "finding out and understanding the patient's needs" cluster rewards visible listening: open questions, follow-ups that show you absorbed the answer, acknowledgement before moving on. A doctor's instinct is to ask a tight, closed, diagnostic sequence — "Any chest pain? Shortness of breath? When did it start?" — that gets to the answer in three questions. Efficient, clinically sound, and rubric-poor, because it gives the examiner no evidence of the broad, patient-led elicitation they are scoring.

The fix is to deliberately slow the history. Ask one open question — "Tell me what's been happening" — and let the patient run before you narrow. Reflect back what you heard. It will feel inefficient and slightly redundant. That feeling is the sign you are now doing it right for the exam.

Trap 2: the explaining reflex becomes a monologue

Doctors explain for a living, and many are good at it. But "good at explaining" in a clinic often means delivering a fluent, complete, uninterrupted account of a condition and its management. The OET rubric does not reward the complete monologue — it rewards chunked information delivered in pieces, with comprehension checks between each piece.

This is a structural criterion, not a content one. You can explain a diagnosis flawlessly and still lose the mark because you delivered it as one continuous block. The examiner wants to hear you stop after each idea and check: "Does that make sense so far?" "Have you come across this before?" The chunk-and-check rhythm is exactly what the annotated role-play transcripts mark line by line — the marks attach to the pauses, not just the explanation.

For doctors the discipline is counterintuitive: you must interrupt your own fluent explanation to check understanding, even when you are certain the patient is following. The check is the scored behaviour.

Trap 3: clinical authority overrides shared decision-making

This is the deepest trap and the one doctors find hardest to hear. Medical training builds a decision-making voice — you weigh the evidence and recommend a course, because that is your job and patients come to you for exactly that judgement. "I'd recommend we start you on this medication and review in two weeks" is appropriate, competent, and what most patients want from their doctor.

The OET Patient-Centred Approach criterion scores whether you negotiate the decision rather than deliver it. The same clinical content, framed as a shared choice — "There are a couple of options here; one would be to start a medication now, another is to try lifestyle changes first and review. What are your thoughts?" — earns the mark that the confident recommendation does not. The examiner is not testing whether your recommendation is correct. They are testing whether the surface form of your speech treats the patient as a co-decider.

Doctors often resist this because it can feel like abdicating clinical responsibility — and in real practice, over-offering options to a frightened patient who wants direction can be poor care. The OET booth is not real practice. For the four minutes of the role-play, over-offer options on purpose. Demonstrate the negotiation the rubric is scoring, then return to your normal clinical judgement the moment you leave the exam centre.

Trap 4: composure reads as coldness

Doctors are trained to stay calm in the face of distress. Equanimity under pressure is a professional asset — a panicking doctor helps no one. But the OET rubric scores explicit, visible empathy under Relationship-Building, and a doctor's trained composure can register to the examiner as emotional flatness.

When the simulated patient expresses fear or distress, the clinical instinct is to absorb it quietly and move to reassurance through competence. The rubric wants the empathy spoken aloud: "That sounds really frightening — I can understand why you're worried." It feels almost performative to a doctor who is used to conveying reassurance through steady presence rather than emotional language. The exam rewards the explicit verbal acknowledgement, so make the empathy audible even when your instinct is to demonstrate it through calm.

Trap 5: jargon you no longer notice

Years in medicine make clinical language invisible to you. "We'll need to monitor your renal function", "this is a benign presentation", "we'll titrate the dose" — these feel like plain English after a decade on the wards. To the OET examiner scoring your appropriateness of language for a layperson, unexplained jargon is a clear deduction.

The trap is specifically that you cannot hear your own jargon any more. The fix is a habit: every time a clinical term leaves your mouth, append a plain-language gloss — "your renal function, that's how well your kidneys are working." Doctors who practise this find an uncomfortable number of terms need glossing, precisely because so much of their vocabulary has become invisible to them. Building back that awareness is most of the work.

A worked example: the same content, two ways

Take a single clinical moment and watch the marks move. The simulated patient has been prescribed a statin and says, worried, "I've heard these can cause memory problems — my neighbour stopped taking them for that."

The clinically efficient response — accurate, confident, the kind you would give on a real ward — sounds like: "The evidence doesn't support a link between statins and memory problems. The benefits for your heart far outweigh any theoretical risk, so I'd recommend you continue." Every word is true. It scores poorly, because it dismisses the concern, delivers a verdict rather than negotiating, and offers no acknowledgement of the fear underneath the question.

The rubric-optimised response covers the identical content differently: "That's a really common worry, and I can see why your neighbour's experience would put it on your mind — let me talk you through what the research actually shows. (pause, check) The large studies haven't found a real link between statins and memory. (pause, check) How does that sit with you? If you'd like, we could start them and keep a close eye on how you feel, and you can tell me straight away if anything changes." Same facts. But now there is empathy voiced aloud, information chunked with checks, and a shared plan offered rather than imposed. That is the gap between Band C and Band B, and it is entirely a matter of surface form — the medicine underneath is identical.

GPs and specialists trip differently

The traps land differently depending on your background. GPs and family physicians often fare slightly better on the rapport criteria because their daily work already involves explaining to laypeople and managing the whole person — but they can be the worst offenders on diagnostic efficiency, because high patient throughput trains a very fast, closed history-taking style.

Hospital specialists tend to have the opposite profile. Years of communicating primarily with colleagues, in dense clinical shorthand, makes trap five (invisible jargon) and trap two (the explaining monologue) especially sharp — the specialist's natural register is peer-to-peer, not clinician-to-patient. If you are a specialist, assume your jargon problem is worse than you think and over-correct on plain-language glossing. If you are a GP, assume your history-taking is too fast for the rubric and over-correct on open questions. Knowing your own background's characteristic failure mode lets you target practice instead of spreading it thin.

A four-week retraining plan

For a doctor who has the English but keeps scoring Band C, a focused month usually closes the gap:

• Week 1 — slow the history. Every practice role-play, open with one broad question and let the patient run for thirty seconds before you narrow. The goal is to break the closed-question reflex.
• Week 2 — chunk and check. Add the comprehension checks. After every clinical idea, stop and check understanding before the next. Record yourself and count the checks — aim for one per idea.
• Week 3 — negotiate, don't recommend. Convert every recommendation into an options-and-preference exchange. Over-offer choice deliberately, even when your clinical instinct wants to direct.
• Week 4 — voice empathy and gloss jargon. Layer in the explicit empathy statements and the plain-language glosses, until both are automatic.

Stacking the habits one week at a time stops you trying to fix everything at once and failing at all of it. By week four the earlier habits have started to automate, and you can run full role-plays with all five corrections live.

Retraining without unlearning your medicine

The anxiety doctors voice about all this is understandable: am I being asked to become a worse doctor? No. You are being asked to perform a specific communicative ritual for four minutes, under exam conditions, because that ritual is what the test measures. Your clinical judgement, your diagnostic speed, your decisiveness — keep all of it for the job. Switch into rubric-mode for the role-play the way you would switch registers for any specific audience.

The retraining is mechanical and fast for most doctors, because the underlying English is already strong. The bottleneck is never language — it is habit. A few weeks of practice where you consciously slow the history, chunk-and-check your explanations, over-offer options, voice your empathy, and gloss your jargon is usually enough to move a stalled doctor from Band C to Band B.

The most efficient way to retrain a habit is high-volume, immediate-feedback repetition, which is exactly where practising against an AI patient that talks back earns its keep. You can run a dozen role-plays in an evening and watch whether your Patient-Centred marks improve as you force yourself to negotiate rather than recommend. The feedback loop that takes weeks with a human study partner compresses into a single session, and the specific doctor-traps above are precisely the kind of recurring pattern that shows up clearly when every session is scored against the same nine criteria.

"But isn't this teaching bad medicine?"

This objection comes up in almost every conversation with doctors preparing for OET, and it deserves a straight answer rather than a brush-off. The worry is real: over-offering options to a frightened patient who wants direction can be poor care; spending four minutes negotiating when a patient needs a clear recommendation is not always in their interest. Doctors are right that real clinical communication is more nuanced than the rubric's model.

But the OET role-play is not a simulation of best clinical practice — it is a standardised assessment of specific communicative behaviours, and it has to be standardised to be fair across tens of thousands of candidates worldwide. The rubric rewards visible negotiation and explicit empathy because those are observable and markable; clinical judgement about when to direct and when to collaborate is neither. The test cannot mark your judgement, so it marks the behaviours, and it asks every candidate to demonstrate the collaborative end of the spectrum.

The resolution is to hold both truths at once. For four minutes in the exam booth, perform the collaborative ritual the rubric scores — over-offer options, voice every empathy, negotiate visibly. The moment you leave, return to the full clinical judgement that real patients need, including the judgement to be appropriately directive. Passing OET does not require you to believe the rubric is a complete model of good medicine. It requires you to demonstrate, briefly and on demand, the slice of communication it can measure. Treat it as a register you can switch into, not a philosophy you must adopt.

Exam-day execution

On the day, the doctors who pass comfortably do one thing the strugglers do not: they consciously change register as they walk into the booth. They have a one-line mental cue — something like "slow, collaborative, explicit, plain" — that flips them out of clinic-mode and into rubric-mode for the duration of the role-plays. It sounds trivial; it is the difference between a doctor who reverts to their fast, authoritative default under exam stress and one who holds the trained behaviours when it counts.

The four-minute window is short enough that you can sustain a deliberate register without it feeling unnatural to the examiner — they are not expecting your real consulting style, they are scoring against the rubric. Lean into the ritual, hit each of the five corrections, and trust that your strong underlying English will carry the fluency marks while the deliberate behaviours carry the rapport and structure marks.

You are a good doctor. OET Speaking is not asking you to stop being one. It is asking you to perform, briefly and deliberately, the collaborative ritual that the rubric rewards — and once you can see the five traps above as habits rather than character, retraining them is a matter of weeks, not a verdict on your competence.