Oxford Medicine Interview
Two panel interviews, no written test, and a UCAT that has already done its work
Two panel interviews, no written test, and a UCAT that has already done its work
Oxford Medicine interviews are unlike any other medical school interview in the UK. They are not competency-based panels designed to test communication skills or empathy through structured scenarios. Instead, they are academic tutorials — conducted by the very tutors who would teach you if you were offered a place — designed to assess how you think under pressure, how you engage with unfamiliar ideas, and whether you have the intellectual curiosity that Oxford's tutorial system demands. Standard interview preparation, or even strong A-level revision, will not be enough on its own. Understanding exactly what Oxford is looking for, and practising the right kind of thinking, is what separates candidates who receive offers from those who do not. The second half of this page takes one of Oxford's own question types — a blood pressure that falls after surgery — and does the arithmetic out loud twice, because the obvious way through it predicts the opposite of what the patient is doing.
Oxford medicine interviewers move fast between disciplines, and the Medicine pack is built around that range: mole and concentration calculations rooted in blood chemistry, pedigree-based inheritance problems, the physiology of the kidney and fetal circulation, and ethics questions on euthanasia and confidentiality. No free sample exists for this one; the pack is sold on the strength of what's inside.
The Medicine pack — £180
Ten questions fill 22 pages, laid out in three passes each: the question alone, then the hints for when you stall, then the answer worked through in full. One PDF, one payment, instant download.
Get the Medicine pack — £180Most Oxford Medicine applicants are interviewed at two colleges: their chosen college and a second college assigned by the admissions pool. Each interview typically lasts between 20 and 30 minutes, and you will usually face two interviewers — often a biomedical scientist and a clinician, or two academics with different specialisms. The questions are rarely straightforward, and they are not designed to have a single correct answer. Tutors want to see how you respond when pushed beyond what you already know.
The format varies slightly between colleges, but the underlying approach is consistent: interviewers will introduce a problem, listen to your initial response, and then probe further. They may offer new information mid-question, challenge your reasoning, or ask you to reconsider a conclusion you have just reached. This is not hostility — it is the tutorial method in action. The ability to update your thinking in response to new evidence, rather than defending a position out of anxiety, is one of the most important qualities Oxford tutors are looking for.
Interviews typically take place in December, and shortlisting is based on your UCAS application, your personal statement, and your UCAT score. By the time you reach the interview stage, your academic potential has already been assessed on paper. The interview exists to test something different: your capacity to think scientifically in real time.
Oxford Medicine does not currently require a written admissions test at the interview stage. The UCAT is used earlier in the process, as part of the shortlisting criteria, alongside your predicted grades and personal statement. If you have been invited to interview, your UCAT score has already done its work — it will not be revisited or discussed during the interview itself.
What this means in practice is that your interview preparation should not be focused on test technique or data interpretation drills. The skills that matter now are different: scientific reasoning from first principles, the ability to articulate uncertainty, and genuine engagement with biological and ethical complexity. Candidates who arrive at interview still in a UCAT mindset — looking for the most efficient answer rather than the most intellectually honest one — often struggle with Oxford's open-ended questioning style.
Two colleges, one of them assigned, and no written test to fall back on
Oxford Medicine applicants face a second set of interviewers they did not choose, and by that stage the UCAT has done its work. Everything remaining rests on scientific reasoning in conversation.
That is an awkward thing to revise for, because no syllabus is attached to it. What can be practised is the move these questions all turn on: taking a sentence about a patient, converting it into a relationship you can manipulate, and then testing your own answer before the interviewer has to. One question of that kind is taken apart below. Drilling that move is what the £180 Medicine pack is for.
The most effective preparation combines three things: building a strong foundation of scientific understanding beyond A-level, practising thinking aloud with a knowledgeable interlocutor, and reading widely in areas that Oxford tutors find genuinely interesting.
On the scientific side, you should be comfortable with core concepts in physiology, biochemistry, genetics, and cell biology at a level that goes beyond your A-level syllabus. Oxford tutors frequently introduce problems that require you to apply familiar principles in unfamiliar contexts — for example, using your understanding of osmosis to reason about a clinical scenario you have never encountered. You do not need to have memorised the answer; you need to be able to construct a plausible explanation from what you do know.
Thinking aloud is a skill that requires deliberate practice. Many strong candidates lose marks not because their reasoning is wrong, but because they go silent when uncertain, or jump to a conclusion without showing their working. Oxford tutors want to follow your thought process. If you do not know something, say so — and then reason towards an answer anyway. A response such as "I'm not certain, but if I think about the underlying mechanism, I would expect..." is far more impressive than silence or a guess presented as fact.
Super-curricular preparation matters significantly at Oxford. Reading around your subject — through journals such as The Lancet, BMJ, or New Scientist, or through books like The Gene by Siddhartha Mukherjee or Do No Harm by Henry Marsh — gives you the intellectual vocabulary to engage with questions that go beyond the syllabus. Tutors are not expecting you to have read specific texts, but they can tell immediately whether a candidate has spent time genuinely thinking about medicine as a discipline, rather than simply preparing answers.
One habit is worth building before December: say what you are about to do before you do it. “I am going to write down what blood pressure is made of, then change one term and see what happens” costs four seconds and tells the interviewer where you are heading, so that when you stall they can nudge the step you are actually on rather than guess at it. Candidates who narrate their route are easier to help, and the people asking spend the whole conversation deciding whether you would be worth helping for six years.
For further reading and worked examples, our Oxford Medicine interview questions with ethics and scientific reasoning model answers blog post walks through real question types with detailed commentary on what strong answers look like. You may also find our Oxford Medicine interview questions with model answers resource page useful for structured self-study. If you are also considering the other leading university for Medicine, our Cambridge Medicine Interview preparation page covers the differences in format and approach.
The following questions are representative of the kind of problems Oxford Medicine interviewers use. They are not trick questions, but they are deliberately open-ended and require you to reason carefully rather than recall facts.
The first question in that list is the one most likely to arrive with numbers attached. Here it is in the packs' three layers, with the wrong turnings left in.
“Ten minutes after an operation this patient’s mean arterial pressure has fallen from 90 to 54. Their heart rate has gone from 70 to 140. Take me through what the body is doing, and tell me whether the faster heart is helping.”
Nothing else is supplied: no age, no blood-loss figure, no drug chart. Any further number you assume out loud.
Follow-ups an interviewer might reach for once the arithmetic above stalls, roughly in this order:
Mean arterial pressure is cardiac output times total peripheral resistance, and cardiac output is heart rate times stroke volume. So the monitor number is three quantities multiplied — rate, stroke volume, resistance — and whatever has gone wrong sits in one of them. After surgery the middle one is the suspect: volume lost is filling lost.
My first pass was to check whether the faster heart had already dealt with it. Take the baseline at 70 beats a minute and 70 mL a beat: 4,900 mL a minute. Suppose theatre has brought stroke volume to 42 mL, six tenths of what it was. At the old rate, 70 × 42 = 2,940 mL a minute, 60% of baseline, and with resistance unchanged the pressure falls in the same proportion — 60% of 90 is 54, the number I was handed. Then I doubled the rate: 140 × 42 = 5,880 mL a minute, 120% of the original output, which puts the pressure at 108 and makes this patient hypertensive. They are not. The third prompt is the interviewer declining to rescue me.
The error sits in the word unchanged. My 42 mL was measured at 70 beats a minute and I carried it to 140 as though rate and volume were independent. They are not: a beat must both eject and fill inside whatever time it is given. At 70 the cycle lasts 60 ÷ 70 = 0.857 seconds. Hold ejection at 0.30 seconds — a stipulation, and the weakest plank in what follows — and filling gets 0.557. At 140 the cycle is 60 ÷ 140 = 0.429, so filling gets 0.129. The rate doubled; the time to fill fell to 0.129 ÷ 0.557 = 23% of what it was.
So try the far end and let filling volume be proportional to filling time. Stroke volume then goes as 60/HR − 0.30, and output as HR × (60/HR − 0.30), which multiplies out to 60 − 0.30 × HR: a straight line falling with rate, 39 units at 70 and 18 at 140, less than half. And at 60 ÷ 0.30 = 200 beats a minute the line reaches zero, so it predicts a patient at 200 has no cardiac output at all. A model that puts a living patient at zero has failed a test it set itself, and cannot be trusted in the middle of its range either.
Both attempts made the same mistake in opposite directions: each held one factor still and moved the other. Output is rate times volume; volume falls as rate rises, because filling is what a faster rate takes its time from; so output rises by less than the rate does, and past some rate it turns over. The linear version overstates the fall because diastole is not filled evenly — most of the volume arrives early, so the time a fast rate removes is the least productive there is. The tachycardia is worth something and is not a solution: the pressure has not come back because the volume lost in theatre is still lost.
Resistance is the third quantity and the reflex raises it too, which should hold the pressure up better than my arithmetic did; that my numbers landed exactly on 54 is luck, not confirmation, and I would rather say so than have it said to me. Ejection time also shortens as rate rises, handing some filling time back and making my 23% too harsh, though not enough to reverse the direction.
None of it is knowledge about surgery. Every number was handed over or stipulated, and the biology needed is one relationship from AS level. What is watched is whether “the heart is beating faster” can become a term in a product you are willing to move; whether you notice that your own second model predicts something absurd at 200; and whether the assumption holding the answer up is volunteered or has to be extracted. They are not picking the candidate who is right, but the one they could teach for six years, and the tell is what you do with a result you did not want.
Ten questions is not what makes it £180
Anyone can print ten questions. Two of the expert packs come out at 8,416 and 13,048 words for their ten — roughly 840 and 1,300 words a question once the prompts and the worked answer are counted in. That is the argument, and the question above is the demonstration: it took nearly nine hundred words to answer honestly, and there is no shorter version that is still true. Written by specialist subject tutors.
Get the Medicine pack — £180The most common mistake Oxford Medicine candidates make is treating the interview as a test of knowledge rather than a test of thinking. Candidates who have memorised impressive-sounding facts but cannot reason flexibly from them are quickly identified. Tutors are not looking for a medical encyclopaedia — they are looking for a future tutorial student.
A second common error is failing to engage with the question as it develops. Oxford interviewers frequently add new information or constraints mid-question. Candidates who ignore this and continue with their original answer, rather than incorporating the new detail, signal that they are not truly listening — a serious concern for a future clinician.
A third mistake is over-preparing scripted answers to ethical questions. Questions about medical ethics at Oxford are not invitations to recite the four principles of bioethics. They are opportunities to reason through a genuine dilemma. Tutors will probe any position you take, so the ability to hold a view tentatively and revise it under scrutiny is far more valuable than a polished but rigid answer.
Finally, many candidates underestimate the importance of scientific depth. Oxford Medicine is a six-year course with a compulsory pre-clinical science component. Tutors want to see that you are genuinely excited by the science of medicine, not just its clinical application.
Knowing more biology does not make the interview easier
Oxford Medicine questions are built to run past whatever a candidate knows, so additional content simply moves the point at which reasoning has to take over. That transition is what is being assessed.
It is also why more content is the last thing to revise. The blood-pressure question above needs one relationship from AS Biology and no clinical knowledge whatever; what it needs is the willingness to double a term and read the consequence honestly, including when the consequence is that your own model has just failed.
Most Oxford Medicine interviews last between 20 and 30 minutes. You will usually have two interviews — one at your chosen college and one at a second college as part of the pool process — so the total time across both interviews is typically 40 to 60 minutes. Each interview will involve two academics, and the pace is often intense, with follow-up questions coming quickly after your initial response.
Oxford interviewers do not expect you to have knowledge beyond A-level Biology and Chemistry, but they will push you to apply those concepts in unfamiliar ways. The questions are designed to be accessible to a strong A-level student who thinks carefully, not to catch you out with university-level content. What matters is not what you know, but how you use what you know.
The most effective practice involves working through unseen scientific problems aloud with someone who can challenge your reasoning — ideally a tutor familiar with Oxford's approach. Reading around your subject and discussing ideas with others also builds the kind of flexible thinking Oxford rewards. Practising with a friend who simply listens is less useful than working with someone who will probe your answers and introduce new information mid-discussion.
Say so, clearly and without panic — and then reason towards an answer anyway. Oxford tutors are not looking for instant recall; they are looking for intellectual honesty and the ability to think under uncertainty. A response that acknowledges the limits of your knowledge and then works through the problem from first principles will always be more impressive than a confident guess or an uncomfortable silence. The worst thing you can do is pretend to know something you do not.
Ten questions, each one carried under three headings in a fixed order: the question by itself, then Prompts, then Suggested answers — so the answer is never sitting beside the question you are meant to attempt first. Panel questions throughout, never MMI stations. Written by specialist subject tutors.
Get the Medicine pack — £180Medicine has no free sample PDF. Nothing on this subject downloads without payment, which is why the question above runs all the way to an answer instead of stopping at an illustration.
Ten questions, each one carried under three headings in a fixed order: the question by itself; then Prompts, follow-up questions an interviewer adds once you fall silent; then Suggested answers, one worked route through the problem, written in the first person. Coverage: scientific reasoning and data interpretation, medical ethics set against the four principles, and the personal-statement follow-ups, in Oxford's panel format, never MMI stations. A PDF, £180 paid once, written by specialist subject tutors.
Yes, and it is the ordinary one: reading the Suggested answer before making a real attempt at its Question. The pack keeps the three parts on separate pages so that takes a deliberate choice, but nothing in a PDF can enforce the choice itself. Do that once and the question stops being useful for practice — you already have a route through it, so meeting it again is revision, not a fresh attempt. The other nine questions are not affected.
No. The UCAT sits at shortlisting, before any of this, and by the time you are in the room it has already done its work; the pack is interview questions only. If you want more of the same reasoning, Physiology and Vet Med 1 are the two packs nearest to Medicine on the pack page, at £180 each, and three packs is where the 20% bundle discount starts — it is applied at checkout rather than asked for.
The Medicine pack is a PDF you can print, £180 paid once. Physiology and Vet Med 1 are the two nearest to it on the pack page, and any three packs take 20% off at checkout.
Get the Medicine pack — £180