BMJ OnExamination is a large, BMJ-edited MRCP Part 1 bank with genuine editorial pedigree — and, like every bank, a dashboard that measures how you are performing inside the app rather than whether you are ready for the exam. This audit is for candidates deciding how far to trust its numbers. The principal limitation: features built for engagement (a daily leaderboard, peer comparison) and for self-directed difficulty selection produce statistics shaped by your choices, not by the exam's blueprint.
What BMJ OnExamination offers for MRCP Part 1 right now
Checked against the vendor's public product page on 19 July 2026: 3,190 questions published by BMJ Publishing Group, with subscriptions from £53.99 (one month) to £159.99 (twelve months). Features include selectable difficulty, "revision plans to turn weaknesses into strengths", peer comparison, a daily leaderboard, mock tests curated by BMJ's medical editorial team from recent exam themes, an offline app, a 10-question group mode and a free "10 questions a day" trial; the platform reports over 200,000 registered users and alignment to Royal College of Physicians standards. Two honest framings: the BMJ editorial provenance is a real quality signal, and the "adaptivity" here is user-selected difficulty plus revision plans, not a machine-learning engine — so read the "engine" as the practice loop you configure.
The exam the numbers must answer to
MRCP(UK) Part 1 is two three-hour papers, 100 best-of-five questions each, no negative marking. The Federation publishes the per-paper specialty distribution: clinical sciences 25, clinical pharmacology and therapeutics 15, then the major organ specialties at 14 each, down to oncology 5, medical ophthalmology 4 and palliative medicine 4. Clinical sciences plus pharmacology are nearly a fifth of the paper, and the small domains still contribute marks candidates fail margins smaller than.
What each metric actually measures
First-attempt accuracy on unseen items is the only number that behaves like the exam. Repeat accuracy rises with exposure regardless of learning. The peer-comparison and leaderboard figures tell you where you sit relative to other users — motivating, but the comparison population is self-selected and at every stage of preparation, so a good position is not a pass probability. Difficulty is user-selected here, which is a double-edged metric: practising at your chosen difficulty flatters or frightens you depending on the setting, and neither matches the exam's fixed mix. Coverage counts attempts, the raw material for the blueprint audit. Time-per-item, read per domain, separates accurate-but-slow from fast-but-wrong.
Selection and difficulty bias
Any loop where you choose difficulty and topic biases every statistic. Set difficulty high and your accuracy deflates; set it comfortable and it inflates; and because you chose, neither is comparable with a mixed, unseen, timed block at exam difficulty. The daily leaderboard adds a subtler pull: it rewards volume and streaks, which is motivating but optimises for engagement rather than blueprint coverage. The general rule holds — any percentage produced under conditions you chose is not comparable with the exam, and only unseen mixed timed blocks resemble it. This is the argument of Your Q-Bank Percentage Is Not Your Exam Score.
Audit your distribution against the blueprint
Once a fortnight, set your per-specialty attempted counts against the official distribution. BMJ OnExamination's revision plans steer you toward your weak areas, which is helpful, but they optimise for your improvement curve, not the exam's weighting — so check that clinical sciences and pharmacology (40 marks combined) and the micro-domains are attempted in blueprint-like proportion, not left thin because the plan or the leaderboard pulled you elsewhere.
What a credible readiness signal requires
Five conditions, all of them: unseen questions, timed at exam pace (about 1.8 minutes per item), mixed across specialties, no assistance, and a sample of 100+ questions across multiple sittings. BMJ OnExamination's editorial mocks, run under timed conditions, are the most exam-like signal inside the product; its difficulty-selected revision blocks, by design, are not, because you chose the conditions.
When to override the selector
Force full-pace mixed blocks at exam difficulty weekly, regardless of what the revision plan or leaderboard rewards. Force clinical sciences and pharmacology on a rota. Force the micro-domains — ophthalmology, palliative medicine, oncology — that any self-directed loop under-samples. And resist optimising for the leaderboard: a streak is not a blueprint.
A worked dashboard example
Suppose after four weeks: overall 70% (mostly at "medium" difficulty); cardiology and gastroenterology strong and heavily practised; clinical sciences 56% over 90 attempts; pharmacology 58% over 100; ophthalmology 12 attempts; palliative care none; a strong leaderboard position. Read properly: the strong systems can be maintained with one small block; clinical sciences and pharmacology are under-performing on nearly 20% of the exam and take the largest quota; ophthalmology's 12 attempts are noise, not a measurement; palliative care is a blank spot; and the leaderboard position is motivating but irrelevant to readiness. Next week is quotas, not a pass prediction.
A seven-day pattern for busy trainees
Monday: 40 BMJ OnExamination questions in a weak blueprint domain at exam difficulty, explanations read. Tuesday: 30 more plus review of the week's errors. Wednesday: a timed, unseen 50-question mixed block in iatroX's free MRCP Part 1 bank — unseen items, adaptive selection across topic boundaries, and a first-attempt signal the leaderboard cannot give you. Thursday: rest or 20 light questions. Friday: 40 questions in the blueprint domains your audit flagged, timed. Saturday: a BMJ OnExamination editorial mock or a second iatroX mixed block, reviewed same day by error type. Sunday: spaced review of the week's errors. BMJ OnExamination does edited-content drilling and mocks; iatroX does unseen transfer measurement; neither is asked to do the other's job.
Continue, supplement, switch or stop
Continue while unseen mixed performance is rising and your blueprint audit is roughly level. Supplement (with unseen mixed blocks and transfer practice) if BMJ OnExamination's numbers rise while unseen performance stalls. Switch only for a measurable gap — a domain covered thinly for you, or analytics you cannot get; the leaderboard is not a reason to stay and novelty is not a reason to leave. Stop question-grinding and move to mock conditions in the final fortnight when coverage is complete, first-attempt accuracy is stable and pacing is on target — completion of a bank, or a leaderboard streak, is not an exit criterion.
The leaderboard trap, spelled out
BMJ OnExamination's daily leaderboard is the feature most likely to distort a preparation, precisely because it is motivating. Leaderboards reward volume and streaks, which nudges you toward fast, comfortable questions that lift your position — and none of that maps onto blueprint coverage or transfer. A candidate optimising for the leaderboard can post an impressive streak while leaving clinical sciences, pharmacology and the micro-domains under-practised, because those are slower and less satisfying to grind. Treat the leaderboard as you would a fitness-app streak: pleasant motivation, no evidence of readiness. If you find yourself choosing what to practise by what protects your streak rather than what your blueprint audit demands, the feature has captured your revision, and the fix is to set your quotas from the audit and let the leaderboard fall where it may.
The difficulty-selection subtlety
The second BMJ-specific trap is user-selected difficulty. Because you choose the level, your accuracy figure says as much about your difficulty setting as about your knowledge — a comfortable 80% at "easy" and an alarming 55% at "hard" can reflect identical underlying ability. The exam has a fixed difficulty you do not control, so the only accuracy figure that matters is on unseen, mixed, exam-difficulty material. Use difficulty selection deliberately — harder blocks to stress-test a strong domain, easier ones to build confidence in a new one — but never read a difficulty-selected percentage as a readiness signal, and always calibrate against exam-difficulty unseen blocks. The BMJ editorial mocks, run at realistic difficulty, are the honest gauge inside the product; the difficulty-selected revision blocks are for learning, not measuring.
Frequently asked questions
Is BMJ OnExamination enough for MRCP Part 1 on its own? Its BMJ-edited volume and mocks can carry much of the preparation, but it cannot provide the one measurement that predicts the exam — performance on unseen, mixed, timed material at exam difficulty — so add independent mock-condition practice and treat the leaderboard as motivation, not readiness.
Which MRCP Part 1 component does BMJ OnExamination not reproduce well? The exam's fixed blueprint weighting and difficulty under unseen timed conditions: user-selected difficulty and revision-plan steering systematically under-rehearse whichever domains and difficulty levels you do not choose.
How many BMJ OnExamination questions should I complete per day for MRCP Part 1? 40–60 on weekday study days and 80–100 on a weekend day; distribute by your blueprint audit rather than by the revision plan or leaderboard, and protect two timed mixed blocks per week.
When should I stop using BMJ OnExamination and move to mixed mocks? When every blueprint domain is attempted above your coverage floor, first-attempt accuracy is stable across two weeks, and pacing sits at or under 1.8 minutes per item — usually the final three to four weeks.
How should I combine BMJ OnExamination with iatroX without duplicating practice? BMJ OnExamination for edited-content domain drilling and its mocks; iatroX for unseen timed blocks, transfer measurement and adaptive selection that surfaces related weaknesses you did not choose — so your readiness signal comes from questions neither you nor a revision plan selected.
The bottom line for busy trainees
The honest one-line verdict on BMJ OnExamination for MRCP Part 1: a large, BMJ-edited bank with genuine editorial pedigree and two engagement features — a daily leaderboard and user-selected difficulty — that quietly distort a preparation if you let them. The editorial quality is a real strength; the leaderboard is motivation, not readiness, and difficulty selection means your accuracy figure says as much about your setting as your knowledge. So the discipline is to set quotas from a blueprint audit rather than a streak, practise mostly at exam difficulty, and measure readiness only on unseen, mixed, exam-difficulty blocks — with the BMJ editorial mocks as the honest gauge inside the product. Buy it for the edited content and the mocks, ignore the leaderboard, and let an unseen source carry the measurement, and it is a strong core; optimise for the streak and the comfortable difficulty, and its healthy-looking dashboard becomes exactly the false comfort this series warns about.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026; BMJ OnExamination figures (3,190 questions; £53.99–£159.99; 200,000+ users) are vendor-published and were checked on that date — confirm current counts and prices on the product page. Disclosure: iatroX operates a free competing MRCP Part 1 bank, so this audit names BMJ OnExamination's strengths plainly and confines the iatroX role to jobs the audited product does not claim. Corrections via the feedback route on iatrox.com. References: MRCP(UK) Part 1 format and blueprint (thefederation.uk/examinations/part-1/format); BMJ OnExamination product page (onexamination.com); related reading: PassMedicine vs Pastest vs BMJ OnExamination for MRCP and why your Q-bank percentage is not your exam score.