The RCGP tells you exactly where candidates struggle in the AKT, because it publishes examiner feedback after every sitting, and the striking thing is how consistent the weak areas are from one cohort to the next. Neurology has been flagged in four consecutive sittings. Data interpretation, prescribing in older adults, and ophthalmology recur repeatedly. If the same domains catch cohort after cohort, they will catch you too unless you target them deliberately. Here is what the reports actually show and how to attack each area.
Key takeaways
- The RCGP publishes an AKT feedback report after every sitting, in general terms to protect the questions.
- Neurology was flagged in four consecutive AKTs, making it the most persistent weak area.
- Prescribing in older adults, data interpretation, and ophthalmology recur across recent sittings.
- These domains persist because of low clinical exposure, low base rates, and admin-heavy content.
- Untargeted question volume does not fix a personal weak area; targeted, diagnosed practice does.
What the feedback reports are, and where to find them
After each AKT sitting, the AKT core group publishes a feedback report on the RCGP website, giving candidate performance under curriculum headings. The feedback is deliberately general, to preserve the security of the questions, so it names domains rather than specifics. Since AKT 41, the RCGP has also produced a visual summary of areas needing improvement over the previous five years, which makes the recurring patterns easy to see. These reports are the single most authoritative guide to where candidates lose marks, and they are free to read.
The recurring domains across recent sittings
Reading the recent reports together, several domains come up again and again:
- Neurology. This is the most persistent, flagged in four consecutive sittings, with candidates in AKT 57 (October 2025) struggling to identify significant neurological conditions and those in AKT 58 (January 2026) struggling with common gait disturbances.
- Prescribing in older adults and drug monitoring. Flagged after three of the last four sittings, covering drug side effects, prescribing in older adults, and monitoring, with AKT 57 noting difficulty with side effects of medications for long-term conditions.
- Ophthalmology. AKT 54 (January 2025) and AKT 58 flagged difficulty recognising common eye presentations, with the RCGP noting that this remains core knowledge even though eye care is often delivered by others.
- Data interpretation and statistics. Recurring across sittings, including study design terminology and, in AKT 58, charts of practice-level data.
- Neurodevelopmental and genetic conditions. AKT 54 flagged difficulty recognising genetic conditions associated with intellectual disability, a rare but important area.
- Paediatrics. AKT 57 flagged childhood cancer presentations and acute illness in children.
Across the wider run of reports from AKT 47 onward, ECG interpretation and asthma management have also featured. The pattern is consistent enough to plan around.
Why the same domains recur cohort after cohort
These are not random. They cluster around three structural reasons. Low clinical exposure explains several: neurology, ophthalmology, and childhood cancer are often seen elsewhere in the system, so a GP trainee may simply encounter them less. Low base rates explain others: rare genetic conditions and childhood cancers are important but uncommon, so pattern recognition is hard to build from practice alone. And admin-heavy content explains the rest: data interpretation and practice-level statistics are unlike day-to-day clinical work, so candidates get little incidental practice. The domains persist because the exam tests important things that ordinary training does not reliably expose you to.
The individual version of the same problem
There is a personal parallel to the cohort pattern. Just as the same domains catch cohort after cohort, you have your own recurring weak areas that catch you attempt after attempt, and the reason they persist is the same: you are not exposed to them enough, and undirected revision quietly skips over them. Doing more questions in general does not fix this, because volume tends to reinforce what you already know while leaving your genuine gaps untouched. The fix is to identify your specific recurring domains and target them, exactly as the RCGP advises candidates to target the cohort-level ones.
How to attack each domain
Match your revision to the flagged areas:
- Neurology. Focus on combinations of symptoms and signs typical of specific conditions, and on common gait disturbances, as the RCGP repeatedly advises, and ask a trainer to help you find clinical exposure.
- Prescribing in older adults. Work through the STOPP and START criteria, polypharmacy, and the monitoring requirements and common side effects of drugs used for long-term conditions.
- Ophthalmology. Drill recognition of both normal and abnormal eye signs and common presentations, since this is core knowledge regardless of who usually manages eyes.
- Data interpretation. Practise the chart types the exam uses, and, as the RCGP itself suggests, attend a practice cluster meeting or sit with your practice manager to see how practice-level data is presented, and use the RCGP's own examiner videos on data interpretation.
- Neurodevelopmental and genetic conditions. Learn to recognise the signs and symptoms of the range of genetic conditions that can first present in general practice.
For all of these, verify management against UK guidelines such as NICE and CKS, since the exam is set in the UK NHS context.
Where targeted practice fits
The single most efficient way to act on all this is to let your practice find your recurring weak domains and drill them, rather than revising everything evenly. An adaptive engine does exactly that, surfacing the topics you personally keep missing and scheduling them for repeated review, which is the individual version of attacking the cohort-level weak areas. iatroX is built to do that, with free sample questions to try at iatroX. For context on how the AKT is scored and how it changed, see the AKT's move to IRT and how pass marks are set.
Frequently asked questions
What topics do candidates struggle with most in the AKT? Neurology most persistently, flagged in four consecutive sittings, followed by prescribing in older adults, data interpretation and statistics, ophthalmology, neurodevelopmental and genetic conditions, and paediatrics, according to RCGP feedback reports.
Where can I find the AKT feedback reports? On the RCGP website, published after each sitting under curriculum headings, plus a visual five-year summary produced since AKT 41. They are free and are the most authoritative guide to weak areas.
Why do the same AKT topics keep being flagged? Because of low clinical exposure to areas like neurology and ophthalmology, low base rates for rare conditions, and the admin-heavy nature of data interpretation, so ordinary training does not reliably build these.
How should I revise the hardest AKT topics? Target them specifically rather than revising evenly: symptom-sign combinations in neurology, STOPP and START in older-adult prescribing, eye-sign recognition, and the exam's chart types, verifying management against NICE and CKS.
Does doing more questions fix my weak areas? Not on its own. Undirected volume reinforces what you already know and skips your genuine gaps. Identifying and targeting your specific recurring weak domains is what closes them.