Acute Medicine SCE revision should sometimes begin with a presentation rather than a disease heading. A chapter title supplies the organising diagnosis. An undifferentiated case asks which information matters before that label is available.
The task is to build a concise problem representation, identify what cannot be ignored and choose information that could change the next decision. A longer differential is not automatically a more useful answer.
Alternate topic learning with presentation-led practice
The Federation's Acute Medicine SCE page provides the examination context. Its preparation material and samples help candidates inspect how clinical information is framed. This article uses the public preparation position reviewed for September 2026, not a claim about future assessment changes.
Topic-led work is appropriate when a mechanism or guideline is unfamiliar. Presentation-led work tests whether the learner recognises when that knowledge applies without the topic heading giving away the answer. Neither method needs to displace the other.
The approach below is a proposed study routine, not a validated predictor of an SCE result. Its value can be judged by whether your reasoning becomes more specific and easier to correct.
An original collapse case
A fictional adult is referred after an event described as a 'collapse'. The initial account does not establish whether consciousness was lost, what happened beforehand or how recovery occurred. The learning task is to identify what must be clarified before assigning a diagnostic label.
NICE CG109 on transient loss of consciousness provides an authoritative assessment framework. The important point for this exercise is that the referral word does not establish the mechanism. Relevant accounts of the event should not be silently replaced with assumptions.
Introduce a witness account that differs from the original summary. The learner now has to identify the discrepancy, attribute each account and explain its implications. Choosing whichever version supports the first hypothesis would conceal the problem rather than resolve it.
A later variation supplies an investigation finding. Ask what this changes and what it leaves unanswered. These original stages are not copied examination questions or a complete clinical pathway, and do not contain treatment doses or invented diagnostic thresholds.
Preserve uncertainty in the summary
Compare 'a patient with a faint' with 'an adult with an unexplained collapse, incomplete witness information and a recovery history that needs clarification'. The first is more compact, but may turn an unconfirmed interpretation into an established fact.
A useful problem representation includes the features that distinguish plausible explanations: the event, chronology, relevant context and unresolved information. It does not need to reproduce every sentence of the stem.
Revise the representation when information changes. An early summary is a working account, not a commitment to a diagnosis. If a later finding conflicts with it, record the conflict before trying to explain it away.
Choose information by its decision value
For every proposed question or investigation, state what a different answer would change. Where either result leaves your decision unchanged, reconsider whether it is the most useful next step in the particular educational task.
This is not a claim that real care always needs only one investigation. It distinguishes the priorities of a single-best-answer question from an exhaustive work-up. The options and wording constrain what the examination is asking you to choose.
In the fictional collapse case, clarifying whether the event was witnessed or resolving the recovery account may matter more than collecting an unrelated background detail. The precise choice still depends on the supplied facts. Do not promote one illustrative sequence into a universal protocol.
A worksheet for changing your mind
Record the current explanation, supporting evidence, evidence against it, missing information, next discriminating question and the finding that would make you revise the plan. Leave space for another explanation that remains plausible.
A useful entry could be: 'I accepted the referral label as a diagnosis; the missing event description could alter that interpretation.' Another might identify an investigation result that needs reconciling with the history rather than accepted in isolation.
The worksheet is an original learning aid, not an official marking scheme or an iatroX diagnostic feature. Use fictional or appropriately de-identified material. An empty field is preferable to an invented observation.
Before finishing, ask whether your proposed change of mind is genuine. 'I would reconsider if the diagnosis were wrong' supplies no discriminating evidence. Name the observation, history detail or properly interpreted result that would actually alter the decision.
Use the resources for different parts of the problem
StudyPRN's Acute Medicine resource and BMJ OnExamination's Acute Medicine SCE product advertise relevant preparation. Inspect the current package and samples for a September 2026 purchase rather than compare only question counts.
This guide is published by iatroX and includes its own learning tools. Its September 2026 specification describes acute-medicine questions, adaptive sequencing, spaced repetition and a Socratic Tutor that examines reasoning on an attempted question.
A useful Tutor conversation asks which assumption narrowed the differential prematurely or which new detail should change the summary. Producing an even longer list of possible conditions would not necessarily address the learner's mistake. Clinical recommendations introduced during the discussion should be checked against their sources.
Review the error, then choose the next session
Separate missing knowledge, premature closure, ignored contradictions and failure to answer the requested task. These are suggested learning categories, not formal diagnoses of performance.
For missing knowledge, return to teaching before another question block. For premature closure, practise with the diagnosis withheld. For ignored contradictions, require yourself to identify the finding that does not fit. For task errors, state the requested output before considering the options.
Keep commercial questions within their authorised platform and use your own notes to describe the uncertainty. A provider's complete paid explanation should not be copied into another tool without permission.
Keep StudyPRN or BMJ OnExamination when they provide useful practice and the real need is a better review method. Consider supplementary questions or tutoring when fresh applications or a focused challenge are missing. Real patient uncertainty belongs within appropriate clinical supervision, not an examination resource comparison.
Frequently asked questions
Should I stop revising by specialty topic?
No. Topic learning rebuilds knowledge, while presentation-led cases test whether you can recognise when to use it.
Is a longer differential always a better answer?
No. Prioritise explanations relevant to the supplied information and the decision being asked.
What should I record after an unfamiliar presentation defeats me?
Identify the assumption you made and the information that would discriminate between the leading explanations. That gives further study a specific purpose.