The scarcest resource in MSRA preparation is not time or questions — it is unseen questions, the only material that can honestly measure readiness. This Medibuddy workflow is built to protect that resource: a first-pass, review and exit plan that uses the adaptive engine for learning while preserving a reserve of never-seen questions for calibration. It is the implementation companion to our Medibuddy MSRA analytics audit, and it treats the Professional Dilemmas paper as the separate project it is.
What you are working with
Medibuddy's MSRA bank holds 4,000+ questions including 190+ Professional Dilemmas scenarios, covers both papers, and runs an adaptive engine with Topic Mastery and spaced repetition; access is £30–£60 across one to twelve months. The exam is Professional Dilemmas (95 minutes, ~50 consensus-scored scenarios) plus Clinical Problem Solving (75 minutes, ~86 items at sub-minute pace).
Why preserving unseen questions matters
Once you have answered a question, it can never again give you a clean first-attempt reading — it measures recognition thereafter. An adaptive engine, left unmanaged, will cycle you through the bank until almost nothing is unseen, at which point you have no honest readiness signal left inside the platform. The discipline is to ring-fence a reserve from the start: never let the adaptive feed touch a designated block of questions (or use a second, free source such as iatroX for measurement), so that in the final fortnight you still have cold questions to calibrate against.
First pass: learn broadly, preserve deliberately
In the first-pass phase, run the adaptive engine on the bulk of the CPS bank to target weaknesses, but hold back two reserves. Reserve one: a designated set of Medibuddy questions you never open in adaptive mode, kept for timed self-tests. Reserve two — simpler and cleaner — use iatroX's free MSRA practice as your unseen-measurement source throughout, so your entire Medibuddy bank can be spent on learning while your readiness signal comes from questions the Medibuddy engine never selected. Either way, first-attempt measurement stays uncontaminated.
Review: match the fix to the miss, not the mode
Review each miss by error type — knowledge gap, misread stem, premature closure, guideline error, calculation error, time-pressure error — and let the fix follow. Medibuddy's spaced repetition handles scheduled re-tests of genuine gaps; pace errors need timed blocks, not more content; guideline errors need a dated UK source. Do not let the adaptive engine's re-tests become your only review, because they train recognition of specific items rather than transfer.
The Professional Dilemmas boundary
Medibuddy's 190+ PD scenarios are useful for format rehearsal, but PD is scored against expert consensus, so bank PD accuracy is a familiarity signal, not a validated readiness measure. Treat PD as a separate project: two timed, consensus-reviewed sessions a week from the start, using official-style material for the scoring standard. No adaptive engine has a view of the paper that carries half your marks.
Exit criteria
You are ready to move to full simulation when every CPS domain is covered above its floor, first-attempt accuracy on your preserved unseen questions is stable for two weeks, CPS pace is under the paper's per-item budget, and your PD ranking logic is measurably improving against consensus material — not when the Medibuddy bank is exhausted.
A seven-day pattern for applicants
Monday: 50 Medibuddy CPS questions in adaptive mode, misses reviewed by type. Tuesday: a timed PD session — Medibuddy scenarios for format, then consensus-reviewed against official-style material. Wednesday: a timed, unseen 50-question CPS block in iatroX (free for MSRA) — your preserved, uncontaminated readiness signal. Thursday: 30 forced-pace CPS questions plus spaced review. Friday: second PD session. Saturday: a full timed both-paper simulation, same-day review. Sunday: rest. Medibuddy learns adaptively and rehearses PD format; official material anchors PD scoring; iatroX preserves the unseen signal.
A worked example: how the reserve saves your final fortnight
The value of preserving unseen questions is easiest to see by imagining not doing it. A candidate runs Medibuddy's adaptive engine hard for eight weeks, and by the final fortnight has answered nearly the whole bank. Now they want an honest readiness reading — and there is none available, because every remaining question is one they have seen, so any percentage measures recognition. They walk into the exam with a dashboard full of green and no idea what their cold first-attempt performance actually is. The preserved-reserve version is different: from week one, either a designated block of Medibuddy questions is ring-fenced from the adaptive feed, or a separate free source such as iatroX carries all the measurement, so in the final fortnight there is still a supply of genuinely cold questions to calibrate against. The reserve is what makes the last two weeks informative rather than reassuring.
The cleaner of the two options is usually the second: spend the entire Medibuddy bank on learning, and let an outside unseen source do all the measuring. That removes the discipline of policing which Medibuddy questions you are "allowed" to see, and it guarantees the measurement questions were never selected by the same engine that taught you — which is exactly the independence a readiness signal needs.
The Professional Dilemmas boundary, restated
One thing the reserve cannot fix is Professional Dilemmas, because PD is not a knowledge test with unseen items to preserve — it is judgement scored against consensus. Medibuddy's 190-plus PD scenarios rehearse the format, but a bank PD score is a familiarity signal, not a validated readiness measure, so PD needs its own project: two timed, consensus-reviewed sessions a week using official-style material for the scoring standard. Preserving unseen CPS questions and running a separate consensus-scored PD track are the two disciplines that keep an otherwise strong Medibuddy-centred preparation honest about both papers.
Continue, supplement, switch or stop
Continue while unseen CPS performance climbs and PD logic improves. Supplement PD immediately if your practice is untimed or unreviewed. Switch only for a measurable CPS gap (our MSRA app comparison helps). Stop accumulating CPS volume in the final fortnight; simulate both papers on preserved unseen material.
Three mistakes this workflow is designed to stop
First, letting the adaptive engine consume the whole bank before you need a readiness reading. Once every question has been seen, no honest first-attempt measurement remains inside the platform, so ring-fence a reserve — or, more cleanly, let a separate free source carry all the measurement while the Medibuddy bank is spent entirely on learning. Second, treating bank PD scores as readiness. The 190-plus PD scenarios rehearse format, but PD is scored against consensus, so a separate timed, consensus-reviewed PD track is non-negotiable from week one. Third, reviewing every miss the same way. Match the fix to the error type — a pace error needs timed blocks, a guideline error needs a dated source, a genuine knowledge gap needs a spaced re-test then a fresh transfer question — rather than letting the engine's re-tests become your only review, which trains recognition of specific items. Preserve the unseen signal, run PD as its own project, and match fixes to errors, and an otherwise strong Medibuddy-centred plan stays honest about both papers.
Frequently asked questions
Is Medibuddy enough for MSRA on its own? It covers both papers and drills CPS well, but its PD scenarios rehearse format rather than validate judgement, and readiness needs unseen questions preserved from the adaptive engine — so pair it with consensus-scored PD material and an unseen measurement source.
Which MSRA component does Medibuddy not reproduce well? The Professional Dilemmas paper's consensus scoring, and — within CPS — the sub-minute pace that untimed adaptive practice does not build.
How many Medibuddy questions should I complete per day for MSRA? 40–60 CPS questions on study days by your blueprint audit, with PD sessions and timed blocks protected separately and a reserve of unseen questions preserved for calibration.
When should I stop using Medibuddy and move to mixed mocks? When CPS coverage is complete, unseen first-attempt accuracy is stable, pace is under budget and PD logic is improving — the final two to three weeks, given to both-paper simulation.
How should I combine Medibuddy with iatroX without duplicating practice? Spend the Medibuddy bank on adaptive learning and PD format rehearsal, and use iatroX (free for MSRA) as your preserved unseen-measurement source and Socratic repair — so your readiness signal comes from questions the Medibuddy engine never selected.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026; Medibuddy figures (4,000+ questions incl. 190+ PD; both papers; £30–£60) are vendor-published; MSRA timings are NHS England's. Disclosure: iatroX offers a free competing MSRA bank. Corrections via the feedback route on iatrox.com. References: NHS England MSRA structure (medical.hee.nhs.uk); Medibuddy MSRA pages (medibuddy.co.uk); related reading: the Medibuddy MSRA analytics audit and why your Q-bank percentage is not your exam score.