This is for emergency medicine trainees preparing the MRCEM SBA (Intermediate) around clinical shifts. The honest limitation comes first: at the last check on 19 July 2026, PassMedicine (passmedicine.com) does not publish an MRCEM SBA or MRCEM Intermediate bank at all — its own site directs emergency medicine candidates elsewhere. So what follows is a transferable method, not a product endorsement: how to run a disciplined first-pass, review and exit loop on whichever RCEM-blueprinted bank you actually use, keeping unseen questions in reserve so your final readiness signal stays honest.
What PassMedicine offers for MRCEM SBA right now
Lead with the fact that reshapes the plan. PassMedicine is a strong, well-known UK bank, but for MRCEM it is not the tool: on its own site it states that for MRCEM Primary and Intermediate it recommends the specialist provider FRCEMtutor rather than offering its own bank.
| Item | Status (last checked 19 July 2026) |
|---|---|
| MRCEM SBA / Intermediate bank on passmedicine.com | Not offered. PassMedicine's site directs MRCEM candidates to FRCEMtutor. |
| What PassMedicine does cover (vendor-reported) | MRCP Part 1 and 2, MRCGP AKT, UKMLA / finals, PLAB 1, MSRA, the Prescribing Safety Assessment, the SJT, DRCOG, USMLE Step 1 and UCAT. |
| Separate "Passmed UK" site (passmed.uk) | A different website advertising an MRCEM Intermediate bank (vendor-reported ~2,400+ MCQs, priced in US dollars). Not the same product; verify independently. |
| Adaptive engine | None on either site — self-configured filters and a dashboard, no machine-learning feed. |
| Price and access for MRCEM | Not applicable to PassMedicine; verify on whichever MRCEM bank you choose. |
The consequence is simple: do not buy PassMedicine for the SBA. What its users value — a self-directed loop over a large bank with clean analytics — is a method, and the method transfers. The rest of this article is that method, written for FRCEMtutor or any RCEM-blueprinted bank, with iatroX used only for fresh, unseen measurement.
The exam you are actually preparing for
The MRCEM SBA is the written single-best-answer component of the Intermediate examination, sat after MRCEM Primary and before the OSCE. It is 180 SBA items across two 120-minute papers of 90 questions each — a pace near 80 seconds per item — blueprinted to the 2021 RCEM curriculum and sampling Specialty Learning Outcomes 1 and 3–7. Recent pass rates sit around 45–50%. The blueprint expresses the curriculum as six clinical domains, and the weighting is the single most important planning fact:
| RCEM SBA domain | Indicative items (of 180) |
|---|---|
| Care of the complex, stable patient | 55 |
| Resuscitation and the unstable patient | 40 |
| The injured patient | 30 |
| Paediatric emergency medicine | 25 |
| Procedural skills | 20 |
| Complex and challenging situations | 10 |
Two things follow. The complex stable patient is by a distance the largest domain, yet the least glamorous and routinely under-revised. Resuscitation is the domain trainees over-practise out of enthusiasm, while paediatric emergency medicine and procedural skills are quietly under-sampled. Confirm the current blueprint and official sample questions on the RCEM website before relying on any figure.
Baseline week: measure before you personalise
Before any bank's filters shape your feed, establish a baseline you can trust. Run two or three fresh, timed, mixed blocks in iatroX that sample across all six domains at 80 seconds per item, unassisted and without immediate answer-checking. Record accuracy per domain, not one headline number. As the Q-bank percentage article sets out, a bank average over questions you have already met measures recall, not readiness. You will likely find complex-stable and paediatric accuracy below your resuscitation accuracy — and that asymmetry is the reason for the plan.
First pass: set domain floors the score cannot hide
On your first pass, do not work question-to-end or let "topics I enjoy" drive selection. Set explicit domain floors: a minimum number of first-attempt questions per domain before the first pass counts as finished. Weight the floors to the blueprint, then deliberately protect the under-sampled domains.
| Domain | Blueprint share | First-pass floor (of ~900) |
|---|---|---|
| Complex, stable patient | ~31% | 280 |
| Resuscitation / unstable | ~22% | 200 (a cap too — resist over-practising) |
| Injured patient | ~17% | 150 |
| Paediatric EM | ~14% | 125 (actively protect) |
| Procedural skills | ~11% | 100 (actively protect) |
| Complex / challenging situations | ~6% | 50 |
Floors stop the algorithm — or your own preference — hiding an unattempted area behind a rising percentage, and force the complex-stable domain to get the attention its 55-item weight deserves. Track attempted counts weekly and top up whichever floor lags.
An error taxonomy that tells you what to do next
Reviewing every wrong answer the same way wastes time. Sort each error into one of six types, because the type dictates the fix:
- Knowledge gap — you did not know it. Fix: a short source read, then a fresh transfer question.
- Misread stem — you missed a discriminating detail. Fix: a stem-marking habit, not more content.
- Premature closure — you anchored and stopped. Fix: a forced differential before committing.
- Guideline error — your answer was out of date or non-UK. Fix: verify against a current UK source and date it.
- Calculation error — a dose, rate, weight-based paediatric or unit slip. Fix: drill the calculation type under time pressure.
- Time-pressure error — right answer, too slow. Fix: pacing practice.
Log the type beside each miss. After a fortnight the distribution is diagnostic: a wall of paediatric knowledge gaps means content work; a wall of time-pressure errors late in blocks means pacing, not knowledge.
Review intervals: repeat, transfer, space or read
Not every error deserves an immediate re-attempt, which mostly trains recognition of that item. Decide from the error type: a new transfer question (ideally unseen, in iatroX) for knowledge gaps and premature closure; spaced review at three and ten days for newly learned facts; a short source read for guideline errors; and an immediate light repeat only for genuine misreads. This keeps seen-question exposure low, preserving unseen questions for honest measurement.
The mixed-block switch
Topic-filtered practice is right early, while you build coverage. Switch to timed random mixed blocks when three criteria are met: every domain floor is cleared; first-attempt accuracy is stable within a few points across at least four domains; and pacing sits comfortably inside 80 seconds per item. After that, most practice should be full mixed blocks at exam pace — the skill the SBA actually tests.
Exit criteria that mean something
You are ready to ease off new questions when you can evidence five things — none of which is bank completion: coverage (every domain sampled in roughly blueprint proportion); stable first-attempt performance across two or more fresh, timed, mixed blocks; pacing that holds across a 90-item paper without the last 20 collapsing; retention on delayed recall; and calibration against the RCEM official sample questions. A percentage can be high while coverage and retention quietly fail.
Worked example: seven days around clinical work
A realistic week for a trainee on full shifts, using one MRCEM bank for structured practice and iatroX only for unseen measurement — no proprietary-algorithm claims involved.
| Day | Primary MRCEM bank (one job) | iatroX (unseen measurement) |
|---|---|---|
| Mon | 40 questions, complex-stable floor | — |
| Tue | Review Monday's errors by taxonomy | 1 fresh 20-item mixed block, timed |
| Wed | 40 questions, paediatric + procedural floors | — |
| Thu | Spaced recall of the week's knowledge gaps | 1 fresh 20-item mixed block, timed |
| Fri | 30 questions, injured patient + challenging situations | — |
| Sat | Full timed 90-item mixed paper on your bank | — |
| Sun | Rest, or read two guideline sources you got wrong | Compare unseen accuracy vs baseline |
The design keeps roughly two-thirds of effort on your primary bank and one-third on unseen measurement — an application of the two-Q-bank rule.
Three mistakes this plan is designed to stop
First, mistaking a rising percentage for readiness — the number climbs as you re-see questions while unattempted domains stay unattempted; the floors and a coverage matrix fix this. Second, over-practising resuscitation at the expense of the 55-item complex-stable domain. Third, burning your only unseen questions early by doing everything on one bank. For the full matrix method, see the completion-is-not-coverage pillar.
Decision checklist: continue, supplement, switch or stop
- Continue your current MRCEM bank where domain floors are being met and first-attempt accuracy is rising on fresh questions.
- Supplement with a second, unseen bank once the first's percentage no longer moves you — measure on the second, learn on the first.
- Switch primary banks only for a measurable coverage gap (for example, thin paediatric or procedural coverage), not for novelty.
- Stop expanding into new questions when all five exit criteria are met; past that, more volume adds fatigue.
Bottom line
PassMedicine is a capable bank, but not for MRCEM — it does not offer this exam and says so itself. What transfers is the discipline: a self-configured loop with domain floors, an error taxonomy, sensible review intervals and honest exit criteria, run on a genuine RCEM-blueprinted bank, with unseen questions preserved for measurement.
Frequently asked questions
Is PassMedicine enough for MRCEM SBA on its own? No — not because it is weak, but because it does not cover this exam. As checked on 19 July 2026, PassMedicine does not publish an MRCEM bank and directs candidates to FRCEMtutor. For MRCEM you need a genuinely RCEM-blueprinted bank, and no single bank answers the readiness question alone, because once you have seen its questions your percentage measures recall rather than mastery.
Which MRCEM SBA component does PassMedicine not reproduce well? It reproduces none of it, because it does not offer the bank. More usefully: whichever written SBA bank you choose, none reproduces the OSCE that follows, and no multiple-choice bank rehearses live resuscitation, team working or hands-on procedural performance — those are trained on the shop floor and in the OSCE.
How many PassMedicine questions should I complete per day for MRCEM SBA? The question does not apply, since PassMedicine has no MRCEM bank. On whichever bank you use, roughly 30–50 first-attempt questions a day is sustainable around clinical work, always with at least as much review as answering, weighted toward under-sampled domains rather than favourites, ramping toward full timed papers near the exam.
When should I stop using PassMedicine and move to mixed mocks? For MRCEM you never start with PassMedicine, because it is not an MRCEM resource. On your actual bank, move from topic-filtered practice to timed mixed mocks once every domain floor is cleared, first-attempt accuracy is stable across several domains, and pacing sits comfortably inside 80 seconds per item.
How should I combine PassMedicine with iatroX without duplicating practice? Because PassMedicine has no MRCEM bank there is nothing to duplicate for this exam. The principle still holds: learn on one bank and measure on another, so you never test yourself on questions you have already seen. Use iatroX for fresh, timed, blueprint-sampled blocks that show where you actually stand.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026. The finding that PassMedicine does not offer an MRCEM bank reflects passmedicine.com at that date, where MRCEM candidates are directed to FRCEMtutor; exam format and blueprint figures reflect the published RCEM MRCEM SBA structure at that date, so verify the current blueprint and official sample questions on the RCEM website before relying on any number. Any third-party question counts, prices or features here are vendor-reported and should be confirmed on the relevant product page.
Disclosure: iatroX operates its own question bank and clinical-knowledge tools, so it competes with the resources discussed here. This is written as an exam-level method rather than a product pitch, and iatroX is positioned only for the job it suits — fresh, unseen questions for baseline and readiness measurement — not as a replacement for your primary MRCEM bank, for teaching or for the OSCE. Corrections are welcome through the feedback route on iatrox.com. Compare resources on the iatroX comparison hub.
References: Royal College of Emergency Medicine — MRCEM SBA format, 2021 curriculum, blueprint and official sample questions (rcem.ac.uk); PassMedicine exam list (passmedicine.com); NICE, CKS, SIGN, SmPC/eMC, Resuscitation Council UK and NHS content for guideline currency; iatroX, why your Q-bank percentage is not your exam score and the completion-is-not-coverage blueprint-matrix method.