Start with the finding that governs this whole article: as of 20 July 2026, BMJ OnExamination does not sell an SCE Palliative Medicine question bank. Its published SCE range covers Acute Medicine, Endocrinology & Diabetes, Geriatric Medicine, Respiratory Medicine and Neurology — not Palliative Medicine. So there are no BMJ OnExamination palliative analytics to audit. This audit instead does two useful things: it teaches you how to read Q-bank analytics against the official palliative blueprint whatever bank you use, and it names the option that genuinely does cover this exam.
What BMJ OnExamination offers for SCE Palliative Medicine right now
| Item | Detail (vendor-reported, checked 20 July 2026) |
|---|---|
| SCE Palliative Medicine bank | Not offered |
| BMJ OnExamination SCE range | Acute Medicine, Endocrinology & Diabetes, Geriatric Medicine, Respiratory Medicine, Neurology |
| Genuine SCE Palliative Medicine bank | StudyPRN (599 questions including a 100-question mock, vendor-reported) |
| iatroX role | Cross-specialty knowledge and unseen-MCQ measurement layer — not a palliative-specific bank |
If you searched for "BMJ OnExamination SCE Palliative Medicine analytics" expecting a dashboard to interrogate, the honest answer is that there is nothing to interrogate, because the product does not exist. The generic analytics model BMJ OnExamination runs on its other SCE banks — first-attempt accuracy, peer percentile, themed mock scores and difficulty tags — is worth understanding, because a bank that does cover palliative medicine (StudyPRN) shows you similar numbers, and the reasoning below transfers directly. Confirm all figures on the respective product pages; vendor claims move.
Exam anchor: the SCE format and the Palliative Medicine blueprint
The SCE is identical in structure across specialties: two papers of 100 best-of-five questions, 200 in total, three hours per paper, one day, computer-based on Surpass, one mark per correct answer, no negative marking. Only the blueprint changes. The Federation's SCE Palliative Medicine blueprint (updated 2023) distributes the 200 marks as follows:
| Blueprint domain | Questions | Share |
|---|---|---|
| Other symptoms/clinical problems secondary to life-limiting disease | 48 | 24% |
| Pharmacology and therapeutics | 35 | 17.5% |
| Management of life-limiting disease | 20 | 10% |
| Clinical problems unrelated to cancer | 20 | 10% |
| Management of emergencies | 20 | 10% |
| Care of the dying patient and their family | 20 | 10% |
| Pain (assessment and management) | 15 | 7.5% |
| Legal issues | 15 | 7.5% |
| Practical procedures | 5 | 2.5% |
| Rehabilitation | 2 | 1% |
That distribution is the official requirement. Any vendor's coverage is a third-party approximation of it, and the two most heavily weighted domains — symptom control secondary to life-limiting disease, and pharmacology and therapeutics — are where the exam lives.
Define every metric before you trust it
Whatever bank you use, the home screen shows a handful of numbers. Define each one precisely, because a vague sense of "doing well" is how candidates walk into an under-prepared domain:
- First-attempt accuracy — the only score that reflects genuine recall; everything else is contaminated by exposure.
- Repeat accuracy — how you do on items you have seen before; flatters you and measures recognition.
- Percentile — your position relative to other users, whose readiness and timing you cannot see, so treat it as weak context, not a verdict.
- Predicted score — a modelled projection; never mistake it for a pass guarantee.
- Coverage — proportion of the bank attempted, which is exposure, not blueprint coverage.
- Difficulty — a vendor tag, not a standardised difficulty index; use it to sequence, not to judge readiness.
- Time per item — the most under-used metric; pace discipline protects marks on two three-hour papers.
Selection bias: why adaptive-style feeds distort the average
Any feed that preferentially serves you weak areas — whether an algorithm or your own "practise what I'm bad at" habit — pushes your visible average down and makes it non-comparable with a mixed, unseen block. If two-thirds of your recent questions came from your worst domains, a 60% average is a different, better thing than 60% on a balanced mixed sample. This is why the headline percentage is not your exam score, and why every readiness judgement should rest on a balanced, unseen, timed block rather than the running average the platform shows you. The practical fix is to keep one weekly block deliberately balanced and unseen, and to read that block — never the cumulative dashboard — as your progress marker.
Blueprint audit: compare your attempts with the official weighting
The single most valuable audit you can run costs nothing: tabulate your attempted questions by domain and compare that distribution against the official blueprint above. If you have done 200 questions but only three touched "management of emergencies" and none touched "legal issues", your 10% and 7.5% domains are unaudited no matter how good your overall number looks. Build the comparison as a simple two-column table — official share versus your attempted share — and let the gaps set next week's quotas. This is the completion-is-not-coverage discipline applied to palliative medicine.
Readiness test: the conditions for a credible signal
A number only means "ready" when it comes from the right conditions. Require all five: unseen items you have not worked before; timed at exam pace; mixed across the blueprint rather than topic-filtered; no assistance of any kind; and a sufficiently large sample — a handful of questions is noise. A 70% on a small, seen, untimed, single-topic block tells you almost nothing; the same number on a large, unseen, timed, mixed block is a signal worth acting on.
Algorithm override rules: force the low-volume material
Adaptive-style feeds and self-directed habits both under-serve the domains that are small but examinable. Override deliberately to force in:
- Management of emergencies — malignant spinal cord compression, hypercalcaemia of malignancy, major haemorrhage, superior vena cava obstruction, seizures at the end of life.
- Legal and ethical issues — the Mental Capacity Act, best-interests decisions, advance decisions to refuse treatment, DNACPR and ReSPECT processes, verification of death.
- Calculations — opioid dose conversions, breakthrough dosing and syringe-driver arithmetic, where a single error changes the answer.
- Pharmacology detail — antiemetic choice by mechanism, corticosteroid indications, and the SmPC/eMC facts that underpin safe symptom control.
- Practical procedures and rehabilitation — the smallest blueprint domains (procedures at 2.5%, rehabilitation at 1%) that any weak-area feed almost never surfaces; force a handful in so they are not left unaudited at zero.
These are precisely the areas a comfortable overall percentage hides.
Worked dashboard example: turning analytics into quotas
Suppose a hypothetical palliative dashboard shows 220 questions attempted, 64% first-attempt accuracy, a 58th percentile, and this attempted distribution: symptom control 40%, pharmacology 22%, pain 12%, care of the dying 10%, emergencies 6%, everything else under 4% combined, with legal issues at 0%. The average looks reassuring. The audit does not: emergencies and legal issues are unaudited, and pain is under-weighted relative to its share. Next week's quotas write themselves — a fixed block of legal/ethics items, a block of emergencies, and a pain top-up — with no change to the domains already above their floors. Notice what we did not do: we did not convert 64% into a pass probability, because no honest model supports that leap.
Worked example: a seven-day plan around clinical work
Because BMJ OnExamination has no palliative bank, run this on the bank that does (StudyPRN) plus iatroX for unseen measurement — the analytics reasoning is identical.
- Monday (45 min): StudyPRN timed block weighted to symptom control; code misses.
- Tuesday (40 min): Forced legal/ethics block (the 0% domain from the audit); one source read on the Mental Capacity Act.
- Wednesday (30 min): iatroX small unseen mixed block; log first-attempt accuracy by domain.
- Thursday (45 min): Emergencies block — spinal cord compression, hypercalcaemia, haemorrhage; code misses.
- Friday (20 min): Spaced review of oldest misses; opioid-conversion calculation drill.
- Saturday (90 min): Longer timed mixed block; rebuild the attempted-versus-blueprint table.
- Sunday (40 min): iatroX unseen mixed block as the readiness check; set next week's quotas.
iatroX is the unseen measurement layer here, not a palliative bank; it tells you whether the week's learning transfers, with no proprietary-algorithm claims on either side.
Decision checklist: continue, supplement, switch or stop
- Continue your current bank if the blueprint audit is filling and unseen accuracy is rising.
- Supplement — for palliative medicine specifically, if your primary source is a general medical bank, add a dedicated palliative bank because BMJ OnExamination does not cover this specialty.
- Switch only for a measurable coverage or currency gap.
- Stop first-pass work and move to pure mixed mocks when the readiness conditions are consistently met.
Bottom line
There is no BMJ OnExamination SCE Palliative Medicine analytics dashboard to audit, because the bank does not exist — a fact worth stating plainly rather than papering over. For this specialty, a dedicated bank such as StudyPRN is the sensible primary, the official blueprint is your coverage yardstick, and iatroX is the unseen measurement layer that keeps your readiness signal honest. Audit your attempts against the blueprint, not your average against your own optimism, and compare options on the iatroX comparison hub.
FAQ
Is BMJ OnExamination enough for SCE Palliative Medicine on its own? No — and not because it is weak, but because it does not offer an SCE Palliative Medicine bank at all; as of 20 July 2026 its SCE range covers Acute Medicine, Endocrinology & Diabetes, Geriatric Medicine, Respiratory Medicine and Neurology. For this exam you need a bank that actually covers the palliative blueprint, such as StudyPRN, so BMJ OnExamination cannot be sufficient on its own for palliative candidates.
Which SCE Palliative Medicine component does BMJ OnExamination not reproduce well? Since there is no palliative product, the honest answer is that it reproduces none of it — but the components hardest for any standard multiple-choice bank to test well are the symptom-control calculations (opioid conversions, breakthrough and syringe-driver dosing), the legal and ethical reasoning around capacity and advance decisions, and the nuanced pharmacology of the two heaviest blueprint domains. Those need deliberate, forced practice wherever you source your questions.
How many BMJ OnExamination questions should I complete per day for SCE Palliative Medicine? None are available, so the practical question is how many palliative questions to do daily on a bank that exists — for most trainees working around clinical duties, roughly 20–30 fully reviewed questions on a weekday with a longer weekend block is sustainable. The limiting factor is review depth on the heavily weighted symptom-control and pharmacology domains, not raw volume.
When should I stop using BMJ OnExamination and move to mixed mocks? For palliative medicine the question is moot because there is no bank to stop using; on whichever bank you do use, move to mixed timed mocks once every blueprint domain is above its floor and your unseen first-attempt accuracy is stable across at least two checks. Topic-filtered practice cues the subject, so the closing phase should be random, timed and mixed.
How should I combine BMJ OnExamination with iatroX without duplicating practice? With no BMJ palliative bank to combine, pair a dedicated palliative bank with iatroX instead: the palliative bank for first-pass learning and review, iatroX for timed, mixed, unseen measurement. iatroX is not a palliative-specific SCE bank — it is the cross-specialty knowledge and unseen-MCQ layer — so use it to check transfer, never to re-test items you have already worked, which is how the two-Q-bank rule protects your calibration.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 20 July 2026. Coverage, counts and prices are vendor-reported and change; the finding that BMJ OnExamination does not offer an SCE Palliative Medicine bank reflects its published SCE range on that date — reconfirm on onexamination.com. Disclosure: iatroX operates a competing question bank; its role here is confined to unseen cross-specialty measurement and spaced retrieval, not palliative-specific content. Corrections are welcome via the feedback route on iatrox.com.
References: The Federation of the Royal Colleges of Physicians — SCE in Palliative Medicine specialty page and 2023 blueprint (thefederation.uk); BMJ OnExamination SCE Resources collection (onexamination.com); StudyPRN Palliative Medicine SCE product page (studyprn.com); NICE (including care of dying adults guidance), SIGN and CKS, and the SmPC/eMC for medicines detail; iatroX internal references — Your Q-Bank Percentage Is Not Your Exam Score and question-bank completion is not coverage.