Exam fundamentals12 min read·25 July 2026

UKMLA Content Map Checklist: Build a Revision Tracker That Actually Works

The GMC publishes the content map free — don't pay for a repackaged copy. But a tick-box condition list is the wrong tracker now the list is explicitly non-exhaustive. A four-column tracker built on presentations, confidence and first-attempt accuracy, ready to paste into Sheets.

Revising this topic? See where you stand first.10 free AKT questions — no account

If you want a UKMLA content map checklist, start with the official source: the GMC publishes the MLA content map as a free PDF at gmc-uk.org. You don't need to buy anyone's version of it. What you do need is a way to turn it into something you can track — and, more importantly, a tracker built around the right unit.

Because here's the thing almost every "430 conditions checklist" gets wrong. The GMC now describes the condition list as indicative and non-exhaustive. It is explicitly not a closed syllabus, and a closely-related condition that isn't named on it is still fair game. A tracker whose only column is a tick box teaches you to chase completion of a list that, by the GMC's own framing, cannot be completed.

So this post gives you a tracker you can build in ten minutes, designed around coverage and confidence rather than ticks — plus a ready-to-paste spreadsheet you can save straight to Excel or Google Sheets.

Get the official map first — and the right version
Why a tick-box condition list fails
What to track instead
The tracker, ready to paste
How to score each row
What order to work through it in
Using the tracker in a weekly loop
Five ways trackers fail
FAQ

1. Get the official map first — and the right version

The map is free. The GMC publishes it as a downloadable PDF on their site under the Medical Licensing Assessment section. Don't pay for a repackaged version of a free document.

The version matters more than anything else this year. An updated content map applies to all MLA exams and assessments taken from September 2026 onwards. Papers sat up to and including August 2026 use the previous version. Since we're now past that first date for anyone booking ahead, most people reading this are revising for the updated map — check your own sitting date against it before you build anything.

Three things changed in the update, and each one changes how a tracker should work:

The mapping grid was removed. The previous map included a grid that effectively constrained which conditions could appear in which clinical contexts. It's gone. Any condition can now be tested in any setting — acute, primary care, follow-up. So tracking a condition as "done" because you know its textbook presentation in its usual setting is no longer enough.

The list became explicitly indicative and non-exhaustive. The named conditions are examples of the expected standard, not a boundary. This is the single most important sentence in the document for how you revise.

The scope expanded. The condition list grew substantially. We're deliberately not printing a headline number here, and you should be suspicious of posts that do — published figures vary between sources and between versions of the map, and the count is the least useful thing about it. Open the PDF that applies to your date and count it yourself if you care. What matters is the shape, not the total.

There's a peer-reviewed paper on exactly this, titled "The medical licensing assessment (MLA) content map: A list is not a syllabus, and a syllabus is not a curriculum." The title is the argument. The map tells you the territory you're expected to be safe across; it does not tell you what to learn, in what depth, or in what order. That's your job, and it's what the tracker is for.

For the full breakdown of the update see what changed in the September 2026 content map, and for how to read the map as a whole see the content map guide.

Establish your starting point before you build a tracker. Ten free AKT-style questions, no account — enough to tell you whether your self-assessment is calibrated before you start filling in a spreadsheet based on it.

2. Why a tick-box condition list fails

Three failure modes, all common, all expensive.

It measures exposure, not retention. You tick "pulmonary embolism — done" in week 2. By week 9 you've forgotten the Wells score thresholds. The tracker still says done. It is now actively lying to you, and you'll trust it because you made it.

It rewards breadth over discrimination. The exam does not ask "what is myasthenia gravis". It gives you a vignette where three answers look plausible and asks which is most likely, or what you'd do next. Knowing a condition exists and being able to separate it from its neighbours under time pressure are different skills, and only the second one scores. A tick box can't tell them apart.

It's the wrong unit of work. Patients don't arrive with a diagnosis. They arrive with chest pain, breathlessness, a rash, a fall. With the mapping grid gone, the link from presentation to the conditions behind it is the spine of the exam — so a tracker organised purely by condition is organised against the way you'll be tested.

The clinical presentations are the front doors. The conditions are the rooms behind them. Track the doors.

3. What to track instead

Four things per row, and you can fill them in honestly in about fifteen seconds each.

1. Coverage — have I actually studied this? Binary, and the least important column. It's the one everyone builds a whole tracker out of.

2. Confidence — could I discriminate this under time pressure? Rated 1–3. Not "have I heard of it" but "if this came up as a vignette with three plausible distractors and 70 seconds on the clock, would I get it right for the right reason?" This is the column that predicts your result.

3. Evidence — what's my actual accuracy? Your first-attempt percentage on fresh questions in this area. The crucial word is first-attempt: accuracy on questions you've seen before measures memory of a question bank, not clinical reasoning. This column is the antidote to the confidence column, which is self-reported and therefore optimistic.

4. Last touched — when did I last see this? A date. Forgetting is a function of time, and a tracker without dates can't tell you what's decaying. Anything untouched for three weeks needs re-testing regardless of how green it looked when you wrote it.

The interaction between columns 2 and 3 is where the value is. High confidence with low accuracy is the most dangerous cell in the whole tracker — it means you believe something wrong and feel certain about it. Those are the marks you'll lose without ever knowing why. Low confidence with high accuracy is the opposite and much less urgent: you know it, you just don't trust yourself yet.

4. The tracker, ready to paste

Copy the block below, save it as ukmla-tracker.csv, and open it in Excel, Numbers or Google Sheets. Then add a row per presentation from the map PDF that applies to your sitting date — we've seeded it with a few of the highest-frequency front doors so the format is obvious.

Area of clinical practice,Presentation,Key conditions behind it,Covered,Confidence (1-3),First-attempt accuracy %,Last touched,Next action
Cardiovascular,Chest pain,"ACS, PE, dissection, pericarditis, GORD, MSK",N,1,,,Read + 20 Qs
Respiratory,Breathlessness,"Asthma, COPD, PE, heart failure, pneumonia, anaemia",N,1,,,Read + 20 Qs
Neurological,Headache,"SAH, meningitis, GCA, migraine, raised ICP, tension",N,1,,,Read + 20 Qs
Gastrointestinal,Abdominal pain,"Appendicitis, obstruction, AAA, pancreatitis, ectopic",N,1,,,Read + 20 Qs
Mental health,Low mood,"Depression, bipolar, hypothyroid, grief, risk assessment",N,1,,,Read + 20 Qs
Renal/urology,Acute kidney injury,"Pre-renal, ATN, obstruction, drugs, sepsis",N,1,,,Read + 20 Qs
Endocrine,Hyperglycaemia,"DKA, HHS, new T1DM, steroid-induced",N,1,,,Read + 20 Qs
Paediatrics,Fever in a child,"Sepsis, UTI, meningitis, Kawasaki, viral",N,1,,,Read + 20 Qs
Obs/gynae,Bleeding in early pregnancy,"Ectopic, miscarriage, molar",N,1,,,Read + 20 Qs
Emergency,The acutely unwell adult,"Sepsis, shock, anaphylaxis, arrest rhythms",N,1,,,Read + 20 Qs

Two deliberate design choices. The "key conditions behind it" column is what keeps this presentation-led rather than condition-led — you're rehearsing the differential, which is the actual exam skill. And "next action" is a single concrete verb, not a topic name, so opening the tracker tells you what to do rather than what to feel guilty about.

Build it once, properly. Sit down with the PDF for one evening and enter a row per presentation. It's dull and it's the last time you'll have to think about structure. Don't try to enter every condition — they live in the third column as differentials, which is where they're useful.

If you'd rather have the marking done for you, our free 50-question diagnostic returns a per-specialty breakdown you can transcribe straight into the accuracy column — it's the fastest way to populate it with real numbers instead of guesses.

5. How to score each row

Confidence, 1–3:

1 — Would guess. Can't reliably separate this from its neighbours.
2 — Would probably get it, might be slow. Knows the core, shaky on edges, atypicals or first-line UK management.
3 — Would get it fast and could explain why every distractor is wrong. That last clause is the test. If you can't say why the wrong answers are wrong, you're a 2.

Then colour by combining confidence and accuracy:

Red — confidence 1, or first-attempt accuracy under 60%. Active work needed.
Amber — confidence 2, or accuracy 60–75%, or anything untouched for 3+ weeks.
Green — confidence 3 and accuracy above 75% and touched within 3 weeks. All three, or it isn't green.

Green decays. A row that's been green and untouched for a month is amber whether you like it or not. Build that in with a date formula rather than trusting yourself to remember, and re-test rather than re-read — testing yourself is what makes it stick.

The dangerous cell, again: confidence 3, accuracy below 60%. Mark those in a different colour entirely and do them first. Every one is a mark you're currently confident about losing.

6. What order to work through it in

Not alphabetically, and not by whichever specialty you enjoy.

First — anything that kills. The acute, time-critical presentations: chest pain, breathlessness, the acutely unwell adult, sepsis, the unwell child, reduced consciousness. These carry disproportionate exam weight because they carry disproportionate clinical weight, and the exam is built around what a doctor completing F1 must be safe at. Start here even if you feel you know them.

Second — your reds by volume. Rank red rows by how commonly the presentation appears, not how much they scare you. A red on chest pain matters more than a red on a rare endocrine presentation.

Third — the systematically neglected. Nearly everyone under-revises the same areas: psychiatry, ethics and law, safeguarding, health promotion, evidence and statistics, prescribing. They are all on the map and they all appear. If your tracker has no rows for them, your tracker is incomplete — see ethics and law and prescribing.

Fourth — ambers, for maintenance. Short, frequent re-tests rather than long re-reads.

Never — greens, unless they've decayed. Re-reading things you know is the most comfortable and least productive revision there is.

For which conditions repay attention first, see the highest-yield UKMLA conditions.

7. Using the tracker in a weekly loop

Fifteen minutes a week, same time each week.

Update "last touched" dates and let anything older than three weeks fall back to amber.
Enter this week's first-attempt accuracy per area from your question bank's analytics.
Compare confidence against accuracy and flag every mismatch. These are your priorities regardless of what you'd planned.
Count your reds. One number, tracked weekly. It should fall. If it isn't falling, your method is wrong, not your effort.
Set next week's three actions — three, not ten.

Then leave it alone. A tracker you edit daily is a procrastination surface with a spreadsheet interface.

A fortnightly mock keeps the accuracy column honest, because self-assessment drifts optimistic without external measurement. How to sit one properly.

8. Five ways trackers fail

Building it instead of revising. Colour-coding is enormously satisfying and teaches you nothing. One evening to build, fifteen minutes a week to maintain. Anything more is avoidance.

Self-assessment with no external check. Confidence columns drift optimistic — reliably, in everyone. Without an accuracy column fed by real first-attempt data, a tracker just records your mood.

Treating it as a completion race. The map is non-exhaustive; there is no finish line. The goal is no reds and no decay, not every box ticked.

Tracking conditions instead of presentations. Organises your revision against the way the exam is written, especially now the mapping grid is gone.

Abandoning it at week six. Almost always because it was too detailed to maintain. If yours has thirty columns, it will be dead by mid-October. Four columns survive.

9. FAQ

Where do I download the official UKMLA content map? Free from the GMC at gmc-uk.org, in the Medical Licensing Assessment section. Take the version that applies to your sitting date — updated map from September 2026 onwards, previous version up to and including August 2026.

How many conditions are on the content map? Published figures vary between sources and between versions, which is why we don't headline one. More usefully: the GMC describes the list as indicative and non-exhaustive, so the number is a measure of scope rather than a syllabus to complete. Open the PDF for your date if you want the current figure.

Should I make a checklist of every condition? No — make one per clinical presentation, with the conditions as the differential behind each. That matches how you're tested, and it's more maintainable.

Is there a ready-made tracker I can download? The CSV in section 4 is ready to paste and takes a minute to set up. Third-party trackers exist, including paid ones — but they're built on a free GMC document, and building your own is genuinely part of the learning.

How is this different from a revision timetable? A timetable allocates time; a tracker measures state. You need both, and the tracker should drive the timetable rather than the other way round. The 12-week study plan is the timetable half.

What if my question bank doesn't give first-attempt accuracy? Then it's hiding the number that matters, because overall accuracy inflates as you re-attempt seen questions. Find a per-specialty first-attempt breakdown somewhere — our free 50-question diagnostic produces one, and you can transcribe it.

Does the September 2026 map mean my old notes are useless? No. Medicine didn't change. What changed is scope, framing, and the removal of the constraint on which conditions appear in which settings. Keep your notes; widen your coverage and stop revising conditions only in their usual context.

How many rows should the tracker have? One per presentation on the map that applies to you. That's a manageable evening's work, and far fewer rows than a condition-level list — which is rather the point.

Populate the accuracy column with real data, not guesses. Sit the free 50-question diagnostic — map-aligned, every answer referenced to NICE or the BNF, and a per-specialty breakdown you can copy straight into your tracker. No card. Or try ten questions with no account.

The summary: get the free official map for your sitting date, then build a four-column tracker around presentations, confidence and first-attempt accuracy rather than a tick box per condition. Score it honestly, work reds first, let greens decay on a timer, and check the whole thing against a mock every fortnight.

The map tells you the territory. It doesn't tell you where you are on it — that's what the tracker is for, and it only works if you're honest in the confidence column.