The honest answer to "what are the best RACP Adult Medicine resources by candidate type" is that no single product wins for everyone. The right resource depends on how many weeks you have, what you can spend, and where your errors actually cluster. This is a decision tree for the RACP Divisional Written Examination in Adult Internal Medicine. It sorts the main question banks, revision courses and official materials by the one job each does well, then uses an unseen baseline to point you at the correct branch — not at whichever bank markets hardest.
The RACP Adult Medicine written exam: what you are preparing for
The RACP Divisional Written Examination (DWE) for Adult Internal Medicine is delivered as two papers of predominantly single-best-answer multiple-choice questions, roughly 170 questions in total. The larger paper, Clinical Applications, carries about 100 questions; the shorter paper, Medical Sciences, carries about 70. Current public descriptions note that a small minority of items are extended-matching in style, and each paper runs a little over its nominal window — so confirm the current question counts, item types and durations on racp.edu.au before you plan around them. Delivery is computer-based through Pearson VUE test centres; verify the current mode for your diet.
Two points matter for resource selection. First, the written papers reward broad, blueprint-wide internal-medicine knowledge and disciplined single-best-answer discrimination under time pressure — not encyclopaedic depth in one favourite subspecialty. Second, the DWE is only the written stage: a separate Divisional Clinical Examination, built from a long case and short cases, follows it. No question bank, course or knowledge platform — iatroX included — reproduces that clinical examination. When this article talks about resources, it means resources for the written papers, and it says so plainly wherever the clinical stage is the real question.
Start here: six candidate profiles
Before you buy anything, place yourself in one of six profiles. Your branch of the tree follows from the profile, not from a product review.
- First attempt, adequate runway. You need broad coverage and calibration, not novelty. One good bank plus official material will carry most of the work.
- Retake after a near miss. You need a diagnosis of why you missed — coverage gap, pacing, or a specific weak domain — before you spend again on content.
- Busy trainee revising around clinical work. Your binding constraint is time, so every activity has to earn its slot; you cannot afford duplicated effort.
- Weak foundations. You need teaching input — a course or a core textbook — before high-volume question practice becomes efficient.
- Strong knowledge, poor pacing. You need timed mixed blocks and mock conditions, not more content to read.
- Strong on MCQs, weak on the practical stage. You need to protect written momentum while you begin clinical-case work elsewhere; the written banks will not prepare you for the long case.
The minimum viable stack
Most candidates over-buy. A defensible RACP written stack has only four slots, and two of them are optional:
- One primary question bank for volume, breadth and mixed practice.
- Official calibration material — the RACP sample papers and practice questions — used to check that your bank's difficulty and phrasing track the real examination.
- One teaching or reference source, added only where your profile shows genuine foundation gaps (a revision course, or a core text such as Harrison's or Davidson's).
- One measurement or transfer layer — a pool of unseen questions — used to test whether learning has actually stuck, kept separate from your primary bank so it stays genuinely unseen.
More tools do not equal more coverage. Two overlapping banks bought at the same time usually duplicate a large share of their content and destroy your ability to tell learning from memory. If you are tempted to add a second bank, read the two-Q-bank rule first and add it deliberately, as a measurement layer, rather than as more of the same.
Budget bands (verify all prices on publication day)
Prices move, and several of these products bill in different currencies, so treat every figure below as something to confirm on the day you buy.
- Free or low-cost. The RACP sample papers and practice questions are free, and they are the single most important calibration asset you own. FRACP Vault advertises 50 free sample questions and Ace the Exam offers a trial, both vendor-reported. Hospital and training-network revision sessions are often free to enrolled trainees, and a borrowed core textbook costs nothing.
- One premium resource. A single subscription bank is the highest-yield paid purchase for most candidates. As a labelled example, Ace the Exam lists tiered access at NZD $89 (one month), $139 (three months) and $189 (six months) — vendor-reported on acetheexam.co.nz as at 20 July 2026; verify the current pricing and currency yourself. FRACP Vault does not publish a price on its homepage, so check its pricing page directly.
- Comprehensive stack. A premium bank plus an intensive revision course plus a core textbook plus an unseen measurement layer. This is justified for weak-foundation or retake profiles that still have time, and wasteful for a well-prepared first-timer sitting in four weeks.
Time bands: what to omit, not what to add
The commonest planning error is trying to do everything regardless of runway. The discipline is to state what you will omit.
- Under four weeks. Run mixed, timed blocks from your existing bank, review errors hard, and calibrate against the official sample papers. Omit starting a new course and omit opening a second bank — both cost setup time you do not have.
- Four to twelve weeks. Do a blueprint-stratified first pass through one bank, weakest domains first, then convert to mixed timed blocks for the final third. Add a course only if your profile shows real foundation gaps; otherwise omit it and put the hours into questions.
- More than twelve weeks. Front-load teaching input while time is cheap, then move to bank volume, then to mixed blocks and unseen measurement in the closing weeks. Protect at least one full mock and a pool of unseen questions for the final fortnight; do not burn them early for reassurance.
The decision matrix: which resource for which job
Map each resource to the one job it does best, and stop asking it to do the others.
| Resource | Best single job | Notes (verify current specifics) |
|---|---|---|
| RACP sample papers and practice questions | Official calibration of format and difficulty | Free; the gold standard for phrasing and standard |
| Intensive revision course (Physed Written Exam Course, Dunedin, DeltaMed, RPA BPT, Otago) | Teaching weak foundations and structure | Lecture-based; counts, dates and fees vary — verify on the provider page |
| FRACP Vault | Timed mock realism plus breadth | 2,200+ questions and three full-length timed mocks, vendor-reported |
| Ace the Exam | Volume plus per-category analytics | Close to 4,000 MCQs and category tracking, vendor-reported |
| Other banks (PassFRACP, FRACPractice, LearnPhysician, Top Physician) | Additional volume or a second opinion | Verify coverage and counts before buying a second bank |
| Core textbooks (Harrison's, Davidson's) | Reference depth for foundation gaps | Slow to read cover-to-cover; use as targeted lookup |
| iatroX (Adult Medicine hub) | Unseen measurement and transfer practice | Blueprint-mapped unseen items kept separate from your primary bank |
No single row is "the best resource" — each is the best answer to a different question. The iatroX comparison hub lays the same logic out side by side.
Where the detailed evidence lives
To keep this hub readable, it summarises rather than re-audits each platform. For the detailed, platform-specific workflows, follow the child articles: the AceTheExam first-pass, review and exit plan; the FRACP Vault topic-block, mixed-block and error-review workflow; and the guide to converting a written-exam revision course into active recall. Each of those holds the granular current-state detail and the seven-day loops; this page holds the decision logic that tells you which of them you actually need. If you find yourself reading three platform pages in full, you have already broken the minimum-stack rule.
Three worked candidate profiles
Profile A — first attempt, ten weeks, moderate budget. Stack: one premium bank, the official sample papers, and iatroX for unseen measurement. Weeks 1–6: a blueprint-stratified first pass, weakest domains first, roughly 40–50 questions per study day with same-day error review. Weeks 7–9: mixed timed blocks that ignore topic labels. Week 10: official sample papers under exam conditions plus one full mock. Exit criterion: first-attempt accuracy on unseen mixed blocks stable across two sittings, and pacing inside the real per-question budget.
Profile B — retake, six weeks, strong MCQ but a near miss. Do not re-buy content blindly; diagnose first. Week 1: an unseen baseline mapped to the blueprint to locate the two or three domains that actually failed. Weeks 2–4: targeted repair of those domains only, plus mixed timed blocks to protect breadth elsewhere. Weeks 5–6: full mocks and official calibration. Exit criterion: the previously failing domains now sit at or above your personal floor on unseen items, with no new domain slipping in exchange.
Profile C — busy trainee, fourteen weeks, revising around clinical work. Time is the binding constraint, so cap study at what a clinical week genuinely allows — say five sessions of 45–60 minutes. Weeks 1–8: one bank, weakest domains first, small daily blocks, ruthless error coding. Weeks 9–12: mixed timed blocks twice weekly. Weeks 13–14: mocks and sample papers. Exit criterion: every blueprint domain covered at least once and stable unseen performance — not 100% bank completion, which this profile will never reach and does not need.
The evidence hierarchy behind these recommendations
Weigh your sources in order. Official RACP material outranks everything for format, blueprint and standard. Primary clinical guidance and core internal-medicine references outrank vendor content for what is actually true in medicine. Vendor pages are the correct source for product facts — question counts, features, prices — but must be read as vendor claims, labelled and dated, not as independent findings. Independent testing and peer experience (colleagues who recently sat the diet) are the right source for user experience and difficulty realism. When these tiers conflict, trust the higher one.
Frequently asked questions
How do I know whether I have covered the full RACP Adult Medicine blueprint? Your bank completion percentage will not tell you, because completion is not coverage. Map every question you attempt to a RACP blueprint domain and track the proportion of domains you have hit at least a threshold number of times — not the raw number of questions done. A blueprint-coverage matrix, of the kind set out in the completion-is-not-coverage guide, makes the blank cells visible; those blank cells, not your overall percentage, are what should drive your next block.
Can one question bank be enough for RACP Adult Medicine? For the written papers, one well-built bank can carry the bulk of the work if it clears two tests: it covers every blueprint domain at adequate depth, and its difficulty and phrasing calibrate against the official sample papers. What one bank cannot honestly give you is a genuinely unseen measurement of whether learning has transferred, because once you have worked through it, nothing in it is unseen. That is the specific gap a small separate pool of unseen questions fills — and the only reason to add a second resource at all.
What should I measure instead of my overall Q-bank percentage for RACP Adult Medicine? Measure per-domain first-attempt accuracy on unseen items, pacing against the real per-question budget, and retention across a spacing interval. Your overall percentage is inflated by repeats and by easy domains, and it hides the two or three areas that will actually sink you; the reasoning is set out in "Your Q-Bank Percentage Is Not Your Exam Score". A domain that reads 55% on unseen first attempts is a louder signal than an 82% aggregate.
When should I stop doing new RACP Adult Medicine questions? Stop opening large volumes of new questions when fresh items stop revealing new gaps — when your error review keeps surfacing the same handful of principles rather than new blind spots. At that point, retention work and mixed timed mocks matter more than volume. In the final week, most candidates should not be meeting large numbers of brand-new items at all; they should be consolidating known weak spots and rehearsing pacing under exam conditions.
Which RACP Adult Medicine resource should I use for my weakest component? It depends on what "weakest" means. If it is a knowledge gap, a course or textbook to teach it, then a bank to drill it. If it is pacing, timed mixed blocks and full mocks — content will not fix a timing problem. If it is the practical stage, none of the written banks help; you need clinical-case preparation and examiner-led practice. If it is transfer — you know the fact but miss the applied question — an unseen bank used for measurement is the right tool.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 20 July 2026. Vendor-reported figures — question counts, features and prices for FRACP Vault, Ace the Exam and the revision courses named — are the vendors' own claims as at that date and can change without notice; verify each on the product page before relying on it, and confirm the current DWE format on racp.edu.au. Disclosure: iatroX operates a competing question bank, and its role in this article is confined to unseen measurement, transfer practice and spaced retrieval — jobs the audited products do not claim — and it does not reproduce the RACP Divisional Clinical Examination. Corrections are welcome via the feedback route on iatrox.com. References: RACP Divisional Written Examination (racp.edu.au); Pearson VUE RACP testing (pearsonvue.com); FRACP Vault (fracpvault.com.au); Ace the Exam (acetheexam.co.nz); FRACPractice course listing (fracpractice.co.nz); plus the iatroX Adult Medicine hub, the comparison hub, "Your Q-Bank Percentage Is Not Your Exam Score", the completion-is-not-coverage blueprint matrix and the two-Q-bank rule.