This workflow is for a candidate who has bought — or is part-way through — a DGM revision course and wants to stop watching and start retaining. The method converts passive course consumption into active recall and timed question practice, aimed at the Part 1 KBA. The principal limitation to name first: a course is a knowledge-delivery tool, so however well you convert it, you still need unseen question volume to measure readiness, and none of this prepares you for the Part 2 OSCE.
What "DGM Revision Courses" are, in practice
There is no single DGM revision course to standardise against; the term covers assorted study days, webinars and small-group tutorials from NHS departments and private providers, none endorsed by the RCP or BGS. So treat any course figure — hours, lecture count, bundled questions — as vendor-reported and verify it. What every course in the category shares is the same weakness this workflow targets: they deliver knowledge efficiently but rarely make you retrieve it, and retrieval is what the KBA rewards.
| Attribute | Finding (last checked 21 July 2026) |
|---|---|
| Product type | A category: study days, webinars, small-group courses; not one product |
| Endorsement | None — BGS and RCP do not endorse any particular course |
| Bundled questions | Varies; often small or absent — verify per provider |
| AI / adaptive features | Rare; verify |
| DGM components covered | Mostly KBA content, sometimes OSCE technique — confirm per provider |
The exam your converted notes must serve
The KBA is one online paper of 100 best-of-five questions in three hours, once a year, no negative marking, Hofstee-standard-set (RCP, last checked 21 July 2026). The syllabus (February 2025) is syndrome-led — around 40% common geriatric syndromes, then other illnesses (~15%), pharmacology (~15%), rehabilitation and CGA (~8%), administrative/ethical/legal (~10%), demographics (~5%) and the biology of ageing (~2%). The OSCE is four live stations, roughly 1.5 hours, at Liverpool, after the KBA. iatroX supplies the KBA knowledge and unseen-MCQ layer only; it does not simulate the OSCE.
Before each module, answer a short diagnostic set
Do not open a module cold. First answer five to ten questions on its topic — from a bank, or from your own recall — and get some wrong on purpose. This does two things: it exposes what you already know, so you do not waste the session on the familiar, and it creates a reason to watch, because you are now looking for specific answers rather than passively receiving content. The failed pre-test primes the very memories the lecture then consolidates; it is one of the cheapest gains in the whole workflow.
Watch in bounded segments, then recall before checking
Break each lecture into segments of no more than 20 to 30 minutes, then close the resource and write a concise recall from memory — the key discriminations, the management steps, the caveats — before you look at your notes or the slides. The act of retrieving, and noticing what you could not retrieve, is what moves material into durable memory; re-watching feels productive but mostly builds familiarity, which the exam does not reward. Check your notes only after the recall, to correct and fill, not to copy.
Convert every objective into three prompts
For each learning objective, write three questions you can test yourself on later: one discrimination question ("what distinguishes delirium from dementia here?"), one management rule ("what is the first step in this faller's assessment?") and one "why not the alternative?" prompt ("why is this drug wrong for this older patient?"). This mirrors how the KBA actually asks — best-of-five items live or die on discriminating options and on knowing why the plausible wrong answer is wrong — and it forces the course content into the exam's own shape rather than leaving it as inert prose.
Test with fresh questions within 24–48 hours, and again after a gap
Do not let the lecture be its own revision. Within a day or two of a module, test the topic with fresh questions you have not seen — ideally in a timed, mixed set so the topic is not signposted — and then again after a longer interval of a week or two. Spacing the retrieval is what makes it stick; replaying the lecture as "revision" is the trap the whole workflow exists to break. If your course bundles questions, verify they are high-fidelity and current first; where they are thin or absent, draw the fresh volume from a dedicated bank such as iatroX.
Build a weekly mixed block so course order is not a cue
Courses run in a tidy order, and that order quietly becomes a crutch: you get the falls question right because you just did the falls module. Break the cue with a weekly mixed block that pulls questions from every topic covered so far, in random order, under the clock. If your accuracy holds when the topic is unsignposted, the knowledge is genuinely yours; if it collapses, you had been relying on context, and the mixed block has just saved you from discovering that in the exam.
Exit the course when exam-format performance improves — not at 100%
Completion percentage is a vanity metric. The signal to move on from a module, or from the course, is that your performance on fresh, timed, exam-format questions in that domain has improved and held across a spaced interval. A candidate who has "finished" every lecture but whose mixed-block accuracy is static has not converted anything; a candidate who has watched half the course but whose weak domains are now solid on unseen questions is ready to reallocate time. Track the outcome, not the progress bar.
A seven-day worked example
Give the course one job — deliver a domain — and iatroX the measurement job, in a spaced loop. No proprietary-algorithm claims are involved.
- Day 1: Pre-test your weakest domain (five to ten questions), then watch only that module in bounded segments with recall after each.
- Day 2: Convert the module's objectives into discrimination, management and "why not?" prompts.
- Day 3: Sit a fresh, timed 30-item mixed block in iatroX; do not re-watch first.
- Day 4: Code errors by domain and cause; repair knowledge gaps from the module or BGS resources.
- Day 5: Space-test the same domain with new questions; compare with day 3.
- Day 6: Weekly mixed block across every domain covered so far, unsignposted and timed.
- Day 7: Review the error log; choose next week's single module by weakest accuracy.
Decision checklist: continue, supplement, switch or stop
- Continue the course if converting a module reliably lifts your unseen, timed accuracy in that domain.
- Supplement with a question bank wherever the course's bundled questions are thin or dated.
- Switch your time from watching to testing once a domain's mixed-block accuracy has stabilised.
- Stop re-watching lectures as revision; if the progress bar is full but accuracy is flat, the conversion has failed and questions must take over.
The bottom line
A DGM revision course only pays off if you convert it, and conversion is a specific discipline: pre-test before you watch, recall before you check, reshape every objective into exam-style prompts, and test on fresh, spaced, mixed questions rather than replaying the lecture. Do that and the course becomes a genuine knowledge engine feeding an active-recall system; skip it and you have paid for a very expensive way to feel busy. Either way, measure readiness on unseen questions, and remember the OSCE is a separate task the course cannot complete for you.
Frequently asked questions
Is DGM Revision Courses enough for DGM on its own? No, because a course delivers knowledge but rarely makes you retrieve it under exam conditions, and the KBA rewards retrieval and pace, not viewing hours. Even a well-converted course must be paired with fresh, timed, mixed question practice to measure whether the knowledge holds, and no course prepares you for the four live OSCE stations. Treat the course as the input and an active-recall-plus-questions system as the thing that turns it into marks.
Which DGM component does DGM Revision Courses not reproduce well? The OSCE above all — recorded or live lectures cannot rehearse examining a frail patient, communicating a capacity decision or performing a comprehensive geriatric assessment against the clock in front of examiners. Many courses also under-reproduce the KBA's timed, unsignposted, mixed-domain conditions, which is why the weekly mixed block in this workflow matters: it recreates the pressure the course itself does not.
How many DGM Revision Courses questions should I complete per day for DGM? If your course bundles a verified, current question set, 20 to 40 per day in timed, mixed sets is a reasonable target, rising toward the exam. But course-bundled banks are often small or absent, so in practice most of your daily quota should come from a dedicated bank such as iatroX, with the course supplying teaching rather than volume. Quality and freshness matter more than the raw daily count.
When should I stop using DGM Revision Courses and move to mixed mocks? Begin the shift once converting a module reliably improves your unseen accuracy, and complete it in the final few weeks when timed mixed mocks should dominate your schedule. The concrete trigger is behavioural: if you catch yourself re-watching lectures instead of sitting fresh blocks, you have overstayed in the course. From there, use it only as a reference to repair specific gaps your mocks expose.
How should I combine DGM Revision Courses with iatroX without duplicating practice? Assign one job each and keep the flow one-directional. The course teaches a domain; iatroX measures whether that domain survives fresh, timed conditions. Watch and convert a module, test it in iatroX within a day or two and again after a gap, and let your error log — not the course running order — pick what you study next. You avoid duplication by never re-watching a lecture you have already tested well and never re-answering an item you can already recall.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 21 July 2026. DGM revision courses are a fragmented, unendorsed category, so every course figure here is vendor-reported and should be verified with the provider; bundled-question counts and currency in particular vary widely. Disclosure: iatroX operates a UK question bank that competes with course-bundled banks; this article confines iatroX to the unseen-measurement job a lecture course does not claim to do, and it is not a substitute for the OSCE. Corrections are welcome via the feedback route on iatrox.com.
References: RCP Diploma in Geriatric Medicine exam page and syllabus/blueprint (February 2025), rcp.ac.uk; British Geriatrics Society statement on course endorsement and resources, bgs.org.uk; iatroX DGM revision guide, iatrox.com/blog/dgm-revision-2026-question-banks-diploma-geriatric-medicine-study-resources; "Your Q-Bank Percentage Is Not Your Exam Score," iatrox.com/blog/qbank-percentage-not-your-exam-score.