The MRCGP AKT Q-Bank Content-Gap Checklist: What to Verify Before You Stop Doing New Questions

This is for MRCGP Applied Knowledge Test candidates deciding whether more questions will move their score. The honest answer is that a bank percentage cannot tell you, and the AKT has a well-known trap: roughly a fifth of the paper is not clinical medicine at all, and that fifth is where prepared candidates quietly lose the exam. You have covered the AKT when you can evidence six things — clinical breadth, the two non-clinical strands, the right cognitive level, current UK general-practice answers, GP-context data interpretation, and stable unseen performance. This checklist makes each verifiable rather than a feeling.

The six-part coverage test

Here is the whole checklist in one view. A question bank has done its job for the MRCGP AKT only when you can tick all six rows honestly. The rest of the article is how to evidence each one.

#	What to verify	You can tick it when…
1	Clinical breadth (~80%)	You have sampled every clinical topic area in the RCGP AKT content, in UK general-practice framing — not just the systems you enjoy.
2	Evidence-based practice (~10%)	You can handle statistics and critical appraisal, not just clinical facts.
3	Organisational (~10%)	You know UK GP administration and regulation — certification, driving, reporting, safeguarding processes.
4	Cognitive level	Your practice tests management decisions and next-best-step reasoning, not only recall.
5	Jurisdiction and currency	Your answers reflect current UK guidance, with each guidance-sensitive item dated and sourced.
6	Unseen measurement	Your readiness is proven on fresh, timed, mixed blocks — not on a re-tested bank percentage.

Rows 2 and 3 are the whole game. Almost everyone revises the clinical 80%; the candidates who pass comfortably are the ones who did not treat the statistics and organisational strands as optional.

What the MRCGP AKT actually tests

The Applied Knowledge Test is the machine-marked, single-best-answer component of the MRCGP, sat at Pearson VUE test centres. From October 2025 the paper is 160 questions in 2 hours 40 minutes, with four sittings a year — a change from the previous 200 questions in 3 hours 10 minutes. That works out at roughly one minute per item, and it tests knowledge relevant to independent UK general practice. The AKT is one of three MRCGP components alongside the Simulated Consultation Assessment and Workplace-Based Assessment; this checklist concerns the AKT knowledge paper only, and the Simulated Consultation Assessment is a separate skills exam a written bank cannot cover.

The AKT samples across three strands. About 80% is clinical medicine, spread across the breadth of the RCGP curriculum topic areas. About 10% is evidence-based practice — statistics, critical appraisal and research interpretation. About 10% is organisational — the administration and regulation of UK general practice, from certification and driving rules to reporting duties and practice systems. That 20% non-clinical share is small enough to neglect and large enough to fail on, which is exactly why it belongs on a coverage checklist. There is also a strategic argument for treating it as high-yield rather than as a chore. Unlike the clinical strand, where the breadth is enormous and each extra hour returns less, the evidence-based-practice and organisational items draw on a comparatively finite and stable body of material — a defined set of statistical concepts and a defined set of UK processes. A focused pass through them therefore tends to convert into marks more reliably than yet another sweep of clinical questions, which is the opposite of how most candidates actually allocate their final weeks.

Build a blueprint coverage matrix, not a percentage

The tool that makes this concrete is the blueprint coverage matrix, set out in the completion-is-not-coverage pillar. List each strand and clinical topic area as a row, then record questions attempted, first-attempt accuracy, the date last reviewed and a red/amber/green confidence flag. Keep the two non-clinical strands as their own explicit rows so they cannot hide inside a clinical average. Confirm the exact topic list against the current RCGP AKT content guide.

Strand / topic area	Approx. share	Questions attempted	First-attempt accuracy	Last reviewed	Confidence
Cardiovascular	Clinical	—	—	—	R / A / G
Respiratory	Clinical	—	—	—	R / A / G
Gastrointestinal	Clinical	—	—	—	R / A / G
Metabolic & endocrine	Clinical	—	—	—	R / A / G
Musculoskeletal	Clinical	—	—	—	R / A / G
Neurology	Clinical	—	—	—	R / A / G
Mental health	Clinical	—	—	—	R / A / G
Women's & reproductive health	Clinical	—	—	—	R / A / G
Children & young people	Clinical	—	—	—	R / A / G
Older adults	Clinical	—	—	—	R / A / G
Skin	Clinical	—	—	—	R / A / G
ENT, eyes & oral	Clinical	—	—	—	R / A / G
Kidney & urology	Clinical	—	—	—	R / A / G
Sexual health	Clinical	—	—	—	R / A / G
Infectious disease & travel	Clinical	—	—	—	R / A / G
Cancer & palliative care	Clinical	—	—	—	R / A / G
Evidence-based practice & statistics	~10%	—	—	—	R / A / G
Organisational & regulatory	~10%	—	—	—	R / A / G

A matrix like this stops a strong clinical average from masking two blank non-clinical rows. Any topic that is under-sampled and amber or red is a genuine gap; the evidence-based-practice and organisational rows are the two most candidates leave until it is too late.

The ten blind spots self-selected practice hides

When candidates pick their own questions, the same weak points recur. Treat these ten as high-suspicion gaps deserving deliberate attention, ideally checked against the RCGP AKT content guide and official sample material rather than trusted on your own read.

Statistics and critical appraisal, the evidence-based-practice strand — likelihood ratios, number needed to treat, study design and how to read a paper.
Certification and fitness for work, including the Statement of Fitness for Work ("fit note").
DVLA fitness-to-drive rules, a perennial source of dependable, learnable marks.
Death certification and referral to the coroner, and the surrounding processes.
Notifiable diseases and public-health reporting duties.
Immunisation and travel-health schedules, checked against current UK sources.
Safeguarding processes, the referral pathways and thresholds as much as the clinical cues.
Prescribing safety and monitoring, verified against the SmPC/eMC and NICE/CKS rather than memory.
Regulation and practice systems, including the roles of the GMC and CQC and significant-event processes.
End-of-life prescribing in the community, anticipatory medicines and symptom control.

Format checklist: coverage, statistics, administration

Doing clinical questions is not the same as covering the AKT. Verify the following.

You have sampled every clinical topic area, in UK general-practice framing.
You have deliberately practised the evidence-based-practice strand — statistics and critical appraisal — not just clinical items.
You have deliberately practised the organisational strand — certification, driving, reporting, regulation and practice administration.
Your NICE, CKS and SmPC/eMC knowledge is current, not carried over from an earlier edition.
You can hold roughly one minute per item across 160 questions without a back-half collapse.

Interpretation checklist: can you read the data?

The AKT rewards interpretation in a primary-care context. Confirm you can do each from the stem alone.

ECGs and spirometry at the level a GP is expected to read.
Blood results and trends, including which values prompt action in the community.
Growth and development charts in child health.
Prescribing calculations and monitoring, including in older and paediatric patients.
Statistics and appraisal data, interpreting a study's numbers and its relevance to practice.
Administrative scenarios, reading who is responsible and what the correct process is.

Recency checklist: is your knowledge in date?

The AKT is strongly guidance-sensitive, and the exam format itself changed in October 2025.

You have flagged your guidance-sensitive topics — for example hypertension, diabetes, asthma, anticoagulation and contraception.
For each, you can name the source and date, from the UK stack: NICE, CKS, SIGN, the SmPC/eMC for medicines, the immunisation guidance and NHS content.
You have confirmed your bank reflects the 160-question, 2-hour-40-minute format and current guidance, not the older paper.
You have noted the jurisdiction of each source, so nothing outside UK general practice is imported into the exam.

Performance checklist: prove it on unseen questions

A percentage on a bank you have partly seen is inflated by memory of the items — the mechanism is explained in why your Q-bank percentage is not your exam score. Readiness needs a clean, unseen signal.

You have sat at least one fresh, unseen mixed block under timed conditions.
Your unseen score is stable across two or three blocks, not one good run.
Your pace-adjusted accuracy holds across all 160 items.
You have reviewed your high-confidence errors, the ones you were sure of and got wrong.
You have checked retention, re-testing topics from earlier revision.
You have calibrated against official RCGP AKT sample material, including its statistics and organisational examples.

Revising on one bank and measuring on a separate, unseen bank is the two-Q-bank rule, and it keeps recall from inflating your readiness estimate. Because the AKT is entirely single-best-answer, a written bank — iatroX included — can genuinely cover and measure all three strands, provided you insist it samples the non-clinical ones; the same is not true of the Simulated Consultation Assessment, which needs supervised practice.

Worked example: Tom, four weeks out

Tom has done 85% of a large AKT bank at 78% and thinks he is nearly ready. His coverage matrix tells a sharper story. The clinical rows are mostly green, with strong cardiovascular and respiratory numbers. But the evidence-based-practice row shows 12 questions at 50%, and the organisational row is almost blank — he has done a handful of certification and driving questions and nothing on death certification, notifiable diseases or regulation. On the 20% he neglected, he is close to guessing.

He sits a fresh, unseen, timed block and scores 70%, weighed down almost entirely by the non-clinical items, which he has never systematically practised. His clinical strength did not rescue him, because the marks he was dropping were elsewhere.

Tom's next moves are specific: dedicated statistics and critical-appraisal practice; a deliberate sweep of the organisational strand — certification, DVLA, death certification, reporting duties, regulation; and timed full-length practice with all three strands mixed. Doing more cardiology questions would have changed nothing about the marks he was losing.

The stop-or-continue decision tree

Choose the next activity from the measured gap, not the calendar.

Keep doing new questions where a clinical topic or a non-clinical strand shows low attempts — you have not sampled it enough to trust your score.
Consolidate where attempts are healthy but accuracy is amber; focused review beats more new items on a half-known topic.
Prioritise the non-clinical strands whenever the evidence-based-practice or organisational rows are blank or red, regardless of your clinical score.
Simulate full timed papers when coverage is broadly green but your unseen or end-of-paper performance sags.
Seek teaching or a written resource for any area that stays red despite repeated attempts — statistics especially rewards a clear explanation over repetition.
Rest and protect retention once all three strands are green and your unseen scores are stable.

Your one-page MRCGP AKT content-gap checklist

Copy this and keep it beside your revision. You are ready to ease off new questions only when every box is ticked.

Frequently asked questions

How do I know whether I have covered the full MRCGP AKT blueprint? You know it from a coverage matrix with three explicit sections — clinical, evidence-based practice and organisational — not from a percentage. Sample every clinical topic area in UK general-practice framing, and give the statistics and organisational strands their own rows so they cannot hide inside a clinical average. You have covered the AKT when all three sections are genuinely green, including the roughly 20% that is not clinical medicine, rather than when your clinical number looks reassuring.

Can one question bank be enough for MRCGP AKT? Because the AKT is entirely single-best-answer, one strong, current bank can in principle cover all three strands — but only if it samples the evidence-based-practice and organisational content properly, which many banks under-serve. For measuring readiness, one bank is still a weak instrument once you have seen its items, so revise on one bank and keep a second, unseen bank purely to measure yourself. And remember the AKT is only one MRCGP component; the Simulated Consultation Assessment needs separate, supervised preparation.

What should I measure instead of my overall Q-bank percentage for MRCGP AKT? Measure coverage across all three strands, calibration on your high-confidence errors, pace-adjusted accuracy across a full 160-item block, and unseen performance on questions you have never met. Pay particular attention to your accuracy on the non-clinical items in isolation, because a strong clinical average routinely hides a weak statistics and organisational score that the headline percentage never reveals.

When should I stop doing new MRCGP AKT questions? Stop expanding when your matrix is green across clinical topics and both non-clinical strands, your unseen timed scores are stable, and your remaining errors are careless rather than conceptual. Before that point, keep doing new questions but aim them at confirmed gaps — usually the statistics and organisational strands — rather than your strongest clinical systems. After it, more volume mostly adds fatigue and your time is better spent on retention and rest.

Which MRCGP AKT resource should I use for my weakest component? Match the resource to the weakness. For clinical gaps, use a broad, current UK GP bank and target the topics your matrix flags. For statistics and critical appraisal, a focused evidence-based-practice resource with worked examples usually beats more mixed questions. For the organisational strand, use material built around UK GP administration and regulation — certification, driving, reporting, regulation — rather than hoping clinical practice covers it. Confirm any resource is current and UK-referenced, and compare options honestly on the iatroX comparison hub.

Editorial notes and references

Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026. The 160-question, 2-hour-40-minute format reflects the arrangements introduced in October 2025; the strand proportions are approximate and the content areas are periodically revised, so verify the current structure and content guide on the RCGP site before relying on specifics. Any third-party question counts or features referred to here are vendor-reported and should be confirmed on the relevant product page.

Disclosure: iatroX operates its own question bank and clinical-knowledge tools and therefore competes with the resources discussed here. It is positioned in this article only for the job it suits — building and measuring knowledge across all three AKT strands and providing fresh, unseen questions for baseline and readiness measurement — and it does not replace supervised preparation for the Simulated Consultation Assessment. Corrections are welcome through the feedback route on iatrox.com. This is the exam-level hub for the MRCGP AKT content-gap intent; companion checklists exist for MRCP Part 1 and MSRA.

References: Royal College of General Practitioners — AKT content guide and sample questions (rcgp.org.uk); NICE, CKS, SIGN, SmPC/eMC and UK immunisation guidance for currency; DVLA fitness-to-drive guidance; iatroX, why your Q-bank percentage is not your exam score and the completion-is-not-coverage blueprint-matrix method.

Complete a fresh MRCGP AKT baseline in iatroX →