This workflow is for higher specialty trainees (ST4 and above) in acute internal medicine who are using the Society for Acute Medicine's revision resources to prepare for the written SCE Acute Medicine. It addresses the knowledge-recall component — the two-paper, best-of-five exam — and turns a small set of official examples into a diagnostic coverage audit. The principal limitation is honest and structural: the Society curates and signposts; it is not a large adaptive question bank, and the official examples are finite and once-seen.
That constraint is not a criticism. Official and society material is the calibration gold-standard precisely because it defines the exam's own idea of a fair question. The job is to extract maximum diagnostic value from that finite set, then hand daily volume to an unseen bank so you never contaminate your best measuring instrument.
What the Society for Acute Medicine offers for SCE Acute Medicine right now
The current-state box below is what you are actually working with. Treat every figure as last checked on 21 July 2026 and re-verify on the source pages before you rely on it.
| Field | Detail (last checked 21 July 2026) |
|---|---|
| Provider | Society for Acute Medicine (SAM), acutemedicine.org.uk — the UK specialty society, not the examining body |
| What it is | Curated revision signposting: an SCE Revision Resources guide (PDF, updated December 2024), reading lists, curriculum links and pointers to the Federation's official sample questions — not a hosted question bank |
| Live question count | None hosted by SAM. The official practice items are the Federation's sample "Best of Five" questions on the SCE Acute Medicine page — a small, finite set |
| AI / adaptive features | None |
| Price / access | Free to access; some material is member-facing |
| Component supported | The written 200-question knowledge exam only. SAM does not deliver a timed mock under exam conditions |
Read that box carefully, because a great deal of online advice quietly assumes the Society hands you a bank. It does not. What it gives you is arguably more valuable at the calibration stage — an authoritative reading map and a route to the examiner's own sample questions — but it will not generate the hundreds of unseen items you need to build stamina and expose blind spots. Distinguishing the Society's genuine offer from third-party claims is the first act of an honest revision plan.
The exam you are actually sitting
The SCE in Acute Medicine follows the standard MRCP(UK) Specialty Certificate Examination structure. It is two papers of 100 best-of-five questions each — 200 questions in total — with each paper lasting three hours, sat in one day, computer-based on the Surpass platform at a test centre. There is one mark per correct answer and no negative marking, so every question deserves an answer. What changes between specialties is the blueprint, not the machinery.
The Acute Medicine blueprint is a broad general-medicine map, weighted toward the systems that dominate the acute take. The indicative distribution across the 200 questions is below; the examiners note the actual count in any diet may vary.
| Blueprint domain | Indicative questions (of 200) |
|---|---|
| Cardiovascular medicine | 20 |
| Gastroenterology and hepatology | 20 |
| Neurology and ophthalmology | 20 |
| Respiratory medicine | 20 |
| Medicine in the elderly | 18 |
| Diabetes and endocrine medicine | 14 |
| Infectious diseases | 14 |
| Musculoskeletal system | 12 |
| Cancer and palliative care, haematology | 10 |
| Clinical pharmacology and poisoning | 10 |
| Critical care medicine | 10 |
| Renal medicine | 10 |
| Other (allergy, clinical genetics, dermatology, immunology, patient safety, psychiatry, public health) | 22 |
Two features of that table should shape your audit. First, no single system dominates: four domains sit at 20 questions and the long tail of smaller domains adds up to real marks. Second, critical care and clinical pharmacology and poisoning carry only about ten questions each despite defining the day job — a reminder that the SCE samples breadth of internal-medicine knowledge, not the intensity of your resuscitation practice. The official Federation page and the JRCPTB Acute Internal Medicine curriculum are the authoritative references; anything a revision provider tells you about weighting should be checked against them.
Inventory your official material: unseen, attempted once, or contaminated
Before you sit anything, list every piece of official or society material you hold and label its diagnostic state. An item is only a clean measurement once. The moment you have worked through it and read the answer, it measures recognition, not reasoning.
| Material | Source | Diagnostic label |
|---|---|---|
| Federation sample best-of-five questions | thefederation.uk SCE Acute Medicine page | Unseen (until first attempt) |
| SAM SCE Revision Resources guide (Dec 2024) | acutemedicine.org.uk | Reference, not a test — a content map |
| JRCPTB Acute Internal Medicine curriculum | jrcptb.org.uk | Syllabus / blueprint anchor |
| Any released past examples you have already reviewed | Various | Contaminated by review — do not use for calibration |
The point of the labels is discipline. Your unseen official items are scarce and irreplaceable, so they are spent deliberately, once, as a calibration probe — never dribbled away as casual daily practice.
Choose the calibration date
Pick a single date to sit your official probe cold. Too early and you measure only what you have not yet revised; too late and you cannot act on the findings. For most trainees revising around clinical work, a probe four to six weeks before the exam is the sweet spot — late enough that a first content pass is complete, early enough to redirect several weeks of study. Diarise it as a fixed appointment, because the temptation to "just read a few more topics first" quietly turns a measurement into a rehearsal.
Reproduce exam conditions exactly
The finite official set is too small to fill a three-hour paper, so pad it to a realistic block using unseen transfer questions, and then reproduce the conditions that matter: roughly 1.8 minutes per question, single sitting, no reference material, no phone, best-of-five response only, and a genuine break between blocks as you would get between papers. The goal is not a score you can boast about. It is a clean read on pace, on where your certainty and accuracy diverge, and on which domains collapse under time pressure. Interpreting that read is where most candidates go wrong — see Your Q-Bank Percentage Is Not Your Exam Score for why a headline percentage is the least useful number the sitting produces.
Code every error by domain, cognitive process and format
After the probe, resist the urge to file each miss under a subject and move on. Code it three ways, because the fix is different for each.
| Error code | Blueprint domain | Cognitive process | Format | Corrective action |
|---|---|---|---|---|
| E1 | Cardiovascular | Knowledge gap (did not know the fact) | Text stem | Learn: targeted reading from the SAM map, then test on unseen items |
| E2 | Respiratory | Data interpretation (ABG/spirometry misread) | Data interpretation | Drill interpretation sets, not more prose |
| E3 | Elderly medicine | Prioritisation/judgement (knew facts, chose wrong "best") | Management sequencing | Practise ranking the safest next step under time pressure |
| E4 | Endocrine | Recognition-only (right answer, wrong reasoning) | Text stem | Re-test blind later; a lucky hit is a hidden gap |
| E5 | Any | Careless/pacing error | Any | Fix process: flag-and-return discipline, not knowledge |
Coding this way turns a vague "I got 62%" into an action list. Ten E1 errors say learn; ten E2 errors say drill interpretation; ten E3 errors say your knowledge is fine but your judgement under time is not — a completely different intervention.
When you read the coded probe, look at the shape of the errors, not the total. A cluster of E1 knowledge gaps in one or two systems is the easiest problem to fix and the least worrying — targeted reading closes it. A scatter of E3 judgement errors across many domains is more serious, because it means your facts are sound but your selection of the safest option under time is not, and that is trained by more timed best-of-five practice, not by more reading. A run of E5 pacing errors says the problem is process — flag-and-return discipline — and has nothing to do with knowledge at all.
Map errors to fresh practice without burning official questions
Each coded error now points to a specific transfer task on unseen material, keeping the official questions locked away. An E1 cardiovascular gap becomes a short focused read followed by a mixed unseen block that includes cardiovascular items among distractor topics, so you cannot pattern-match your way to the answer. An E2 interpretation error becomes a run of data-only items. Crucially, you never re-do the official question that exposed the gap; you rebuild the capability on new stems and let the official set stay pristine for a later, genuinely unseen re-probe if a new sample is released.
This is exactly the discipline behind the two-Q-bank rule: one resource is your daily working bank; a separate, unseen resource is your measuring instrument, and you never let them bleed into each other.
When to repeat, and when to switch to transfer questions
Only re-run a formal official probe when you have genuinely unseen official material — a newly released sample, or items you have provably never touched. Everything else is transfer practice on a specialist bank and on iatroX. If you find yourself "revising" by re-reading questions you have already seen, you have stopped measuring and started memorising a specific set of stems, which the real exam will not reuse. Track a rolling first-attempt accuracy on unseen blocks instead; that trend, not a re-sat familiar paper, tells you whether you are improving.
A seven-day plan for a busy acute take
Consider Priya, an ST5 in acute internal medicine, roughly six weeks out, working a mix of days and on-calls. She uses the Society's resources for one defined job — mapping content and running her single official calibration probe — and uses iatroX for adaptive transfer practice, without any claim that iatroX runs a proprietary algorithm; it simply serves unseen, mixed-topic, timed blocks and re-surfaces her misses on a spaced schedule.
- Day 1 (post-nights, light): Read the SAM revision map for one weak system identified last week. No testing. Twenty minutes.
- Day 2: One 30-item unseen iatroX block, mixed topics, timed. Code every error E1–E5.
- Day 3 (on call): Nothing formal. Read one NICE or SIGN summary relevant to a Day 2 miss on the commute.
- Day 4: The official calibration probe — Federation sample items padded to a 50-question timed block under strict conditions. Code, do not grade.
- Day 5: Convert the probe's coded errors into targeted reading plus one 20-item interpretation drill (ECGs, ABGs, sodium trends).
- Day 6: One 40-item unseen mixed block. Compare first-attempt accuracy with Day 2.
- Day 7: Review only. Re-test the five items you most wanted to get right, blind, from Day 2's misses now re-surfaced by spaced retrieval.
The Society material does one thing supremely well here — it anchors the week to the examiner's own standard — while the unseen bank supplies the volume, the timing stress and the spaced re-exposure that a finite official set physically cannot.
Decision checklist: continue, supplement, switch or stop
Base the next move on measured gaps, not novelty or sunk cost.
- Continue with the Society-anchored workflow if your unseen first-attempt accuracy is rising and your blueprint coverage is even.
- Supplement with a dedicated SCE bank (StudyPRN and BMJ OnExamination both publish SCE Acute Medicine banks — vendor-reported) if you are running out of unseen questions in a weak domain.
- Switch your primary bank only if a stratified sample is out of step with current UK guidance or over-represents one system.
- Stop adding new questions when coverage is complete, high-confidence errors are near zero and your pacing is comfortable — then consolidate rather than chase a higher percentage.
Three mistakes this workflow is designed to stop
The first is spending the finite official questions casually until none remain for a clean calibration; they are your best measuring instrument and deserve to be used once, deliberately. The second is treating the Society's reading map as the whole plan — it is an excellent content anchor, but without a separate source of unseen volume you will arrive under-practised on pacing and breadth. The third is re-sitting questions you have already seen and calling it revision; a familiar paper measures memory of those items, not readiness for the fresh ones the exam will actually use. Track unseen first-attempt accuracy instead.
Frequently asked questions
Is the Society for Acute Medicine's material enough for SCE Acute Medicine on its own? No, and it is not designed to be. The Society provides a revision map, reading lists and a route to the Federation's finite official sample questions; it does not host the large volume of unseen practice items you need to build stamina and expose blind spots across a 200-question blueprint. Use it as your calibration and content-mapping layer, and pair it with at least one full question bank for daily volume.
Which SCE Acute Medicine component does the Society's material not reproduce well? It does not reproduce a timed, full-length mock under exam conditions, and it cannot supply the repeated unseen exposure that trains pacing across two three-hour papers. Because the official sample set is small, it is a single-use calibration probe rather than a source of daily practice, so timed stamina and breadth have to come from a separate bank.
How many Society questions should I complete per day for SCE Acute Medicine? Effectively none on a daily basis, because the finite official set should be reserved for one deliberate calibration sitting rather than drip-fed. For daily volume, a sustainable target for a working trainee is roughly 30 to 50 unseen questions a day from a bank, always reviewed and coded, rather than a larger number skimmed without analysis.
When should I stop using the Society's material and move to mixed mocks? Move to mixed, timed, full-length mocks once you have completed a content pass and run your single official probe, typically in the final three to four weeks. At that point your marginal gains come from integration and pacing under time pressure, which only mixed unseen blocks and full mocks can train — keep the Society's reading map for closing specific gaps the mocks reveal.
How should I combine the Society's material with iatroX without duplicating practice? Give each a single, non-overlapping job: the Society's resources are your content map and official-calibration source, and iatroX is your unseen measurement and spaced-retrieval layer for cross-specialty internal-medicine knowledge. iatroX is not a specialty-specific SCE bank, so it will not duplicate the Federation's sample items; it supplies fresh, mixed-topic timed blocks and re-surfaces your coded misses, which is exactly the volume the official set cannot provide.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 21 July 2026. Vendor-reported figures (for example question counts and prices at StudyPRN and BMJ OnExamination) are labelled as such and were correct on the pages cited at the last-checked date; verify current numbers on the product pages before relying on them. Disclosure: iatroX operates a UK question bank and clinical-knowledge platform and therefore competes with some resources mentioned here; its role in this workflow is confined to cross-specialty unseen measurement and spaced retrieval — jobs the Society's material does not claim to perform — and it is not a specialty-specific SCE Acute Medicine bank. Corrections are welcome via the feedback route on iatrox.com. References: the Federation of Royal Colleges of Physicians SCE Acute Medicine page; the Society for Acute Medicine SCE resources; the JRCPTB Acute Internal Medicine curriculum; the iatroX comparison hub; and the exam-level hub, What MCQ Banks Cannot Prepare You for in SCE Acute Medicine.