This workflow is for GPs and doctors caring for children who are preparing for the DCH theory paper and want to use BMJ OnExamination's personalisation without letting it quietly skip syllabus areas. It addresses the Foundation of Practice theory component only. The principal limitation to accept up front: any personalised feed optimises towards your recent activity, so a rising overall score can hide entire low-volume domains you have barely attempted. This plan sets floors so that cannot happen.
What BMJ OnExamination offers for DCH right now
At the last check on 20 July 2026, BMJ OnExamination runs a dedicated DCH revision product. The following figures are vendor-reported and should be confirmed on onexamination.com before you buy: 900+ questions matched to the DCH syllabus and authored by paediatricians; a mix of single best answer and extended matching formats; subscription plans running from roughly £34.99 for one month to £149.99 for twelve months, with intermediate durations priced in between; a free trial of ten questions per day; and a mobile app for offline practice. The platform advertises "smart" study features — customisable difficulty, mock tests curated by its editorial team, revision plans that target weak areas, and group leaderboards. Treat these as personalisation aids, not a proprietary adaptive engine; the vendor does not publish an algorithm you can audit, so this workflow does not assume one.
Crucially, this is a written-theory product. It reproduces the knowledge layer of the DCH; it does not reproduce the DCH Clinical examination.
Exam anchor: what the DCH actually is
The DCH (Diploma in Child Health, RCPCH) has two separate components that can be taken in any order. The theory exam is the RCPCH Foundation of Practice (FOP) paper: 100 single best answer questions in two hours, computer-based at a test centre or under online invigilation, one mark per question with no negative marking. This same FOP paper is shared with the MRCPCH membership route, so much FOP-labelled material is relevant. The second component is the DCH Clinical examination — a circuit of stations with examiners and role-players, run as a separate practical assessment for clinicians working with children. Confirm the current structure, sitting dates and syllabus on rcpch.ac.uk, and distinguish official RCPCH requirements from any third-party claim about "what comes up". BMJ OnExamination's DCH product maps to the FOP theory paper; the Clinical circuit needs separate, practical preparation this workflow does not cover.
Baseline week: measure before you personalise
Before you let any feature personalise your feed, take a blueprint-stratified baseline. In week one, complete a small unseen sample — around 60 to 80 questions — deliberately spread across every FOP domain, timed at roughly the exam's pace of a little over a minute per item. The point is not the score; it is a domain map showing where you actually stand. If you begin by letting a "weak areas" feature drive from day one, you optimise around whatever you happened to attempt first, and low-volume domains never surface. Record first-attempt accuracy per domain and keep it; this baseline is what you will measure improvement against, not your rising cumulative percentage.
First pass: set domain floors the feed cannot hide
Personalisation tends to over-serve domains you engage with and under-serve those you avoid. Counter it with explicit floors. Before starting the personalised feed, list every FOP domain and set a minimum number of attempted questions per domain — a floor — that you will hit regardless of what the algorithm suggests. Work the feed for learning, but each week check the floors: any domain below its floor gets a manually filtered block that week. This is the single most important habit in the workflow. A rising overall score with three untouched domains is not progress; it is a blind spot with good optics. Floors convert the platform's convenience into genuine coverage.
Error taxonomy: code every miss before you act
Do not treat all wrong answers the same. Tag each miss with one code:
- Knowledge gap — you did not know the fact or concept.
- Misread stem — you knew it but misread the question or missed a qualifier.
- Premature closure — you locked onto a diagnosis before reading the full vignette.
- Guideline error — your answer was outdated against current UK guidance.
- Calculation error — a dosing, growth or fluid calculation slip.
- Time-pressure error — you rushed under the clock.
The code dictates the fix. Knowledge gaps need a short source read then a transfer question; misreads and premature closure need a stem-reading discipline, not more content; guideline errors need a check against NICE, CKS or the relevant RCPCH guidance and the SmPC/eMC for medicines detail; calculation and time-pressure errors need drilling under the clock. Coding takes seconds and stops you re-reading full explanations you did not need.
Review interval: not everything deserves an immediate repeat
Decide what each miss earns. A knowledge gap earns a new transfer question — a different item testing the same principle — after a short source read, not an immediate repeat of the same stem you will now recognise. A misread earns a note and a fresh timed block. A recurring domain weakness earns spaced review at widening intervals. Reserve immediate repeats for almost nothing, because re-answering a question you just saw measures memory of that item, not mastery of the concept. Space the misses; test transfer.
Mixed-block switch: move off topic filters on evidence
Topic-filtered blocks are for building a weak domain; they also cue you to the topic, which the real paper never does. Switch from filtered practice to timed random blocks when three conditions hold: every domain is at or above its floor, your weakest domain's first-attempt accuracy has stopped falling, and you can articulate why each recent miss happened. From that point, increase the share of full-length, mixed, timed practice each week, so your final fortnight is dominated by unseen mixed blocks at exam pace. Do not wait for 100% completion to make this switch; coverage plus stability, not completion, is the trigger.
Exit criteria: what "ready" looks like
You are ready to stop acquiring and start consolidating when five signals line up: every FOP domain is at or above its floor; first-attempt accuracy on unseen mixed blocks is stable across at least three checkpoints; you finish a full paper inside the two-hour limit with time to review flagged items; your logged error types are shrinking, not just your percentage; and you have calibrated against official RCPCH sample material, not only third-party questions. Bank completion alone is not on that list, and neither is a single good mock.
Worked example: a seven-day plan
This is one week for a GP preparing for the DCH theory paper, using BMJ OnExamination for one defined job — personalised weak-area theory drilling — and iatroX for unseen transfer measurement. No proprietary-algorithm claims are made about either tool.
- Day 1: 60-question blueprint-stratified baseline on BMJ OnExamination, timed; record accuracy per domain; set floors.
- Day 2: Personalised weak-area block (about 30 questions); code every miss; do a short source read for knowledge gaps only.
- Day 3: iatroX unseen block on the two weakest domains — a transfer test, not a repeat — to check whether yesterday's learning moved.
- Day 4: Manual floor top-up: filtered blocks in any domain still below its floor.
- Day 5: 40-question timed mixed block on BMJ OnExamination; log error codes; note pacing.
- Day 6: Spaced review of the week's misses plus a fresh iatroX unseen mixed block to measure first-attempt accuracy independently.
- Day 7: Rest or a single full timed paper if a sitting is near; update the domain map.
The division of labour matters: one product drives learning and personalisation, the other supplies genuinely unseen items so your measurement is not contaminated by questions you have already seen explained.
Decision checklist: continue, supplement, switch or stop
- Continue with BMJ OnExamination if your floors are filling, unseen accuracy is trending up and pacing is improving.
- Supplement with an unseen-measurement layer if your BMJ percentage is high but you cannot tell whether learning transfers — a classic sign you are measuring familiarity.
- Switch emphasis to timed mixed mocks once every domain is at floor and accuracy is stable, even with questions unattempted.
- Stop buying more resources when your gaps are known and consolidation, not acquisition, is what remains. Base each decision on a measurable gap, never on novelty or on how much you have already spent.
FAQ
Is BMJ OnExamination enough for DCH on its own? For the theory paper, a well-matched bank like BMJ OnExamination can carry most of your knowledge preparation, especially given its vendor-reported 900+ DCH-mapped questions and paediatrician authorship. But "enough on its own" is the wrong frame for two reasons: it does not prepare you for the DCH Clinical examination at all, and, used alone, its rising score reflects growing familiarity with its own items rather than proven transfer. Pair it with official RCPCH sample material for calibration and a separate unseen pool for measurement, and it is a strong core; treat it as your whole plan and you risk a blind spot.
Which DCH component does BMJ OnExamination not reproduce well? The DCH Clinical examination. BMJ OnExamination is a written-theory bank built for the Foundation of Practice paper, so it drills single best answer and extended matching knowledge effectively but does nothing for the clinical circuit of stations, examiner interaction, communication with children and families, or examination technique. Those require face-to-face practice, courses and supervised clinical exposure. Use the bank for the theory paper and prepare the Clinical component through the appropriate practical routes; do not let a high theory percentage lull you about a component the bank never tests.
How many BMJ OnExamination questions should I complete per day for DCH? There is no universal number, but a sustainable range for most working clinicians is 30 to 50 questions on a study day, always reviewed properly rather than rushed for volume. What matters more than the count is that the questions are spread to keep every domain at its floor and increasingly done as timed mixed blocks as the exam nears. Fifty questions skimmed without coding your errors is worth less than 25 reviewed with a fix identified for each miss. Prioritise review quality and domain coverage over a daily target.
When should I stop using BMJ OnExamination and move to mixed mocks? Move the emphasis to timed mixed mocks when every FOP domain is at or above its floor, your weakest domain's first-attempt accuracy has stopped falling, and you can explain why recent misses happened. You do not need to have completed the bank; completion is not the trigger, coverage plus stable unseen accuracy is. Keep the bank available for targeted top-ups on any domain that slips, but let full-length, mixed, timed practice dominate your final fortnight so you rehearse the real paper's pace and unpredictability.
How should I combine BMJ OnExamination with iatroX without duplicating practice? Give each tool a distinct job. Use BMJ OnExamination for learning and personalised weak-area drilling, and use a fresh, timed iatroX DCH-level block purely as an unseen check on whether that learning transferred — never re-answer a BMJ item you have already seen explained, because that measures recall of the item, not mastery. In practice, learn on one platform and measure on the other, following the logic of the two-Q-bank rule so the second bank adds unseen measurement instead of duplicating questions. That keeps your first-attempt accuracy an honest signal.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 20 July 2026. Vendor figures (question counts, prices, access periods and "smart" features) are vendor-reported by BMJ OnExamination and were correct at the last check; verify them on onexamination.com before purchase, as plans and counts change. Disclosure: iatroX operates a competing UK question bank, so its role here is confined to unseen transfer measurement and the underlying-knowledge layer — jobs BMJ OnExamination's product does not claim — and note that no dedicated iatroX DCH landing page exists, so the CTA points to the iatroX UK Q-bank. Corrections are welcome via the feedback route on iatrox.com. References: RCPCH, About the Diploma in Child Health and theory-exam structure (rcpch.ac.uk); BMJ OnExamination DCH product page (onexamination.com/exams/child-health/dch); the iatroX comparison hub and the note on why your Q-bank percentage is not your exam score.