There is no single best Physician Associate Registration Assessment (PARA) resource, and any article that names one is selling you a ranking rather than a decision. The right stack depends on how many weeks you have, what you can spend, and where your performance actually breaks. This is a decision tree: segment yourself honestly, take a baseline, and follow the branch that matches your profile. It covers the written Knowledge-Based Assessment (KBA) and points you to the right places for the OSCE, which no written bank prepares.
How to use this decision tree
Work through it in order. First segment yourself into a candidate profile. Then take a baseline on unseen, blueprint-stratified questions so your branch is chosen by data, not by anxiety. Then apply the budget and time constraints that are real for you, and read the platform-to-job matrix to fill each role in your stack. The tree assumes one primary question bank, the official RCP material for calibration, and additional tools only where a measured gap justifies them. It does not assume you need everything.
The PARA anchor every branch shares
PARA is RCP-delivered and GMC-regulated (the GMC has regulated physician associates since December 2024). The KBA is 200 single-best-answer questions across four papers of 50, one hour each. The clinical-area weighting (June 2025 blueprint) is led by Cardiovascular 8%, Acute and emergency care 7% and Gastrointestinal 7%, down to 2% for areas such as ophthalmology, palliative care and research/statistics; the cognitive split is Clinical management 40%, the two diagnosis tasks 20% each and Biomedical knowledge 20%. The OSCE is a separate 16-station exam (14 scored, two rest) at The Spine in Liverpool. Verify the current structure on rcp.ac.uk. Every branch below serves this blueprint; what changes is which tools you buy to cover it.
Step 1: Segment yourself
Pick the profile that fits — most people are a blend, so choose the dominant one.
- First attempt, adequate time. You need broad coverage, calibration and unseen measurement, in that order.
- Retake. Your priority is diagnosing why you failed — a coverage hole, a pacing problem, or an error type — before buying anything new.
- Weak foundations. You need teaching or a reference source before high-volume question practice will stick.
- Strong knowledge, poor pacing. You need timed, mixed blocks and mock conditions, not more content.
- Strong MCQ, weak practical performance. Your written prep may be fine; your investment belongs in OSCE practice.
- Short of time. You need ruthless prioritisation toward the high-weighted domains and away from anything optional.
Step 2: Define the minimum stack
The minimum viable PARA stack is smaller than most candidates think:
- One primary question bank for volume and review.
- The official RCP material — blueprints and released sample questions — for calibration, spent unseen and once.
- One teaching or reference source, only if your foundations are weak.
- One modality tool — OSCE course or data-interpretation practice — only where a measured gap justifies it.
Adding a second bank is worthwhile mainly for fresh, unseen items once your first bank becomes familiar; do it deliberately, following the two-Q-bank rule, not by collecting subscriptions.
Step 3: Budget bands
All prices are vendor-reported and move; verify each on the day you buy. As of 20 July 2026 several PA vendors did not publish transparent pricing on their homepages, so treat the bands below as structure, not quotes.
| Band | What it buys | Best for |
|---|---|---|
| Free / low-cost | Free trials (PLABable for PAs), free/low-cost UK question access (iatroX lists PARA on /quiz-landing), the official RCP blueprints and sample questions | Baseline, calibration, tight budgets |
| One premium resource | A single paid PA bank — RCP Revise (RCP-authored), Matrix Education, PassMAP or PLABable for PAs | Most first-attempt candidates |
| Comprehensive stack | A primary bank + a second unseen bank + an OSCE course/textbook (e.g. Matrix Education) | Retakes, weak foundations, or candidates wanting OSCE coverage |
Step 4: Time bands
State what to omit, rather than cramming everything into every plan.
| Weeks to exam | Do | Omit |
|---|---|---|
| Under 4 | High-weighted domains only (cardiovascular, acute care, GI, respiratory, the diagnosis/management tasks); daily timed blocks; one calibration sitting | New teaching courses; exhaustive low-yield domains |
| 4–12 | Full blueprint pass on one primary bank; weekly mixed timed blocks; mid-point calibration | A second bank until the first is familiar |
| More than 12 | Foundations/teaching if needed, then broad question practice, then a second unseen bank and OSCE work | Nothing structural — but avoid drilling the same items to recognition |
The platform-to-job decision matrix
Match each platform to the job it does best. This summarises the choices; for the detailed evidence, follow the child audits linked below rather than re-reading long descriptions here.
| Platform | Best job | KBA | OSCE | Notes (vendor-reported, 20 Jul 2026; verify) |
|---|---|---|---|---|
| RCP Revise | Blueprint-faithful practice + calibration | Yes | No | RCP-authored; count/price not confirmable at check |
| Matrix Education | Teaching + OSCE + reference ecosystem | Yes | Yes | App, textbook, OSCE courses; large vendor-reported item/flashcard counts |
| PLABable for PAs | Straightforward KBA volume | Yes | No | Free trial; count/price not published at check |
| PassMAP | KBA practice with spaced repetition + analytics | Yes | No | "Adaptive" difficulty engine not confirmed — verify |
| iatroX | Unseen written measurement, free UK-core | Yes | No | Lists PARA on /quiz-landing; measurement/second-bank role |
For the detailed platform evidence, see the child articles on RCP Revise calibration, Matrix Education, PLABable for PAs and PassMAP.
Cannibalisation guardrail
This hub deliberately keeps platform descriptions short. If you want the first-pass plan for PLABable, the calibration protocol for RCP Revise, or the low-volume-domain method for PassMAP, use the child articles above. Duplicating their detail here would only blur which page answers which question.
The one-line decision table
If you read nothing else, read this. Each row maps a measurable condition to a branch — never novelty, never sunk cost.
| If this is measurably true of you | Then take this branch |
|---|---|
| Failed a previous diet with blueprint holes | Audit coverage first; add a second unseen bank before buying teaching |
| Failed because you ran out of time | Timed mixed blocks only — add no new content |
| Failed on one error type (e.g. management) | Targeted transfer drilling in that cognitive task |
| Weak foundations and more than 12 weeks | Teaching/reference first (e.g. Matrix textbook and app), then questions |
| Strong knowledge but poor pacing | Mock conditions and timed blocks; skip buying more content |
| Strong written but weak practical | An OSCE course (e.g. Matrix Education); maintain a KBA baseline only |
| First attempt with under 4 weeks left | High-weight domains + daily timed blocks + one calibration; omit low-yield |
| Baseline shows a 2–3% domain at zero attempts | Manual top-up in that domain now, regardless of overall score |
| Your bank's items feel familiar | Add a fresh unseen bank (e.g. iatroX) for measurement |
Three worked candidate profiles
Profile A — first attempt, ten weeks, moderate budget. Start with a blueprint-stratified baseline on unseen questions. Buy one primary bank (RCP Revise or PLABable for PAs) and use the free/low-cost tier of a second for measurement. Reserve the RCP official sample and sit it once, unseen and timed, at around week six as your calibration point; add iatroX for fresh unseen items in the final month. Weekly allocation: four question-and-review sessions of 40–50 items, two timed mixed blocks, and one calibration or mock. Exit criterion: every blueprint domain practised to floor, pacing at roughly 60 seconds per item, and a recent unseen block holding up under timing.
Profile B — retake, six weeks, higher budget. Do not buy first — diagnose. Take a blueprint-stratified baseline and identify whether you failed on coverage, pace or a specific error type, because each points to a different branch. If coverage: add a second unseen bank and top up the empty domains. If pace: timed blocks only, with no new content bought. If a single error type: targeted transfer drilling in that cognitive task. If the practical exam was the problem: invest in a Matrix Education OSCE course and only maintain your KBA. Weekly allocation follows the diagnosis rather than a fixed template — the point of a retake is to fix the identified fault, not to repeat the whole cycle. Exit criterion: the specific failure mode is measurably corrected on fresh material, not merely revisited.
Profile C — strong MCQ, weak practical, twelve weeks. Confirm your written strength with an unseen, timed KBA block before you assume it; a good historical percentage on a familiar bank is not proof. If it holds, redirect the bulk of your spend and time to OSCE preparation — courses, simulated stations, and communication and procedural practice — because no written bank, iatroX included, trains the 16-station OSCE. Weekly allocation: one KBA maintenance block of 40 items to hold your knowledge, with the remaining sessions on OSCE stations and feedback. Exit criterion: consistent station performance under timing across the mandatory specialties, plus a maintained KBA baseline that has not slipped while you focused on the practical.
Evidence hierarchy
When sources disagree, rank them: official RCP material first for format and blueprint; primary UK guidance (NICE, CKS, SIGN, the SmPC/eMC for medicines) for clinical content; vendor pages for product facts, clearly labelled as vendor-reported; and independent user experience last, for usability only. Never let a vendor's claim about coverage override the official blueprint, and never let your Q-bank percentage stand in for a readiness judgement — read why the percentage is not your score.
Reading your iatroX baseline to pick a branch
A baseline is only useful if it is blueprint-stratified and unseen. Sit a mixed, timed block, then read the domain profile it returns: which clinical areas fell below floor, whether your errors cluster in management or interpretation, and whether pace or knowledge is the limiting factor. That profile chooses your branch. A cardiovascular and acute-care gap sends you to targeted coverage; a pacing gap sends you to timed blocks; a practical gap sends you to OSCE work. The tool informs the decision; you make it.
Bottom line
Choose PARA resources by profile, not by ranking. Take one primary bank, calibrate against the official RCP material, add a second unseen bank only when the first goes stale, and invest in OSCE work separately because no written product covers it. Let a stratified baseline — not marketing and not anxiety — decide which branch you follow.
Frequently asked questions
How do I know whether I have covered the full PARA blueprint? Map your practice against the RCP KBA blueprint domain by domain and check that every area, including the 2–3% ones such as ophthalmology, palliative care and research/statistics, has appeared in your recent unseen questions. A finished bank is not the same as full coverage; build a simple coverage matrix, as set out in completion is not coverage, and treat any empty cell as an instruction rather than a rounding error.
Can one question bank be enough for PARA? For many first-attempt candidates one strong, blueprint-aligned bank plus the official RCP sample material is enough for the written KBA, provided it gives sufficient unseen volume and you review every miss. It is rarely enough on its own once the items become familiar — at that point a second bank supplies the fresh questions you need — and it is never enough for the OSCE, which requires separate practical preparation. Decide by measured gap, not by collecting subscriptions.
What should I measure instead of my overall Q-bank percentage for PARA? Measure blueprint coverage (have all domains appeared, including low-weighted ones), first-attempt performance on unseen items, pacing at roughly one minute per question, retention across spaced intervals, and how you score on the official sample sat once under timing. An overall percentage on a familiar bank mostly measures memory of that bank; the five signals above measure readiness. See why your percentage is not your exam score.
When should I stop doing new PARA questions? Stop adding new material when your blueprint floors are met, your first-attempt unseen performance is stable, your pace holds, and a recent official calibration sitting supports your readiness. Past that point, more new questions add fatigue rather than information; switch to consolidating your error log, light spaced review and rest. Doing questions to feel busy is not the same as doing them to close a measured gap.
Which PARA resource should I use for my weakest component? Match the tool to the component: for a weak written knowledge area, a blueprint-aligned bank (RCP Revise, PLABable for PAs, PassMAP or iatroX for unseen measurement); for weak foundations, a teaching or reference source such as Matrix Education's textbook and app; for a weak OSCE, an OSCE course such as Matrix Education's, because no written bank prepares the stations. Identify the weakest component from a stratified baseline first, then buy only for that gap.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 20 July 2026. Prices, question counts and access periods are vendor-reported, were not transparently published by several PA vendors at check, and change frequently; verify on each product page before you buy. Disclosure: iatroX operates a competing UK question bank that covers the PARA written/SBA layer only, not the OSCE; this hub keeps its role to unseen measurement and second-bank use, and links out to independent options positively where warranted. Corrections via the feedback route on iatrox.com.
References: RCP — PARA information for candidates, KBA and OSCE blueprints (June 2025), rcp.ac.uk; GMC — PA registration assessment content map, gmc-uk.org; vendor homepages for RCP Revise, Matrix Education, PLABable for PAs and PassMAP. Internal: completion is not coverage; Your Q-Bank Percentage Is Not Your Exam Score; the iatroX comparison hub.