SCA Marking Criteria Explained: What Examiners Are Looking For in Every Case
A Deep Dive Into the Three SCA Domains — What They Test, How to Pass Each One, and What Causes Candidates to Fai
How the SCA Is Marked
Every one of the 12 cases in the SCA is marked across the same three domains by a trained RCGP examiner. The marking is not a checklist — examiners make a holistic judgement about your performance within each domain across the full 12-minute consultation.
| Domain | Abbreviation | When It Applies |
|---|---|---|
| Data Gathering and Diagnosis | DG&D | Primarily the first part of the consultation |
| Clinical Management and Medical Complexity | CM&MC | Primarily the second part of the consultation |
| Relating to Others | RTO | Throughout the entire consultation |
Domain weighting: The RCGP confirms that CM&MC carries more weight than the other two domains, to reflect the breadth of capabilities it covers. However, weighting is determined separately for each diet — you cannot predict or game it. Demonstrate all three domains in every case.
The Four Grades
Each domain in each case is awarded one of four grades:
| Grade | Meaning |
|---|---|
| CP — Clear Pass | This domain is demonstrated above the standard of a newly qualified, independent GP |
| P — Pass | This domain is sufficiently demonstrated at the standard of a newly qualified, independent GP |
| F — Fail | This domain is insufficiently demonstrated at the standard of a newly qualified, independent GP |
| CF — Clear Fail | This domain is clearly demonstrated below the standard of a newly qualified, independent GP |
The standard you are being assessed against is the level of a newly qualified, independent GP — not a consultant, not a specialist registrar. Examiners are not expecting perfection. They are looking for safe, competent, patient-centred consulting at the right level for someone completing GP training.
After the Domains: Global Judgement
After marking the three domains, the examiner records an overall global judgement for each case. This judgement is carried forward into the Borderline Regression Method — the process used to set the pass mark for each sitting. There is no fixed pass mark; it is calculated fresh for every diet based on examiner judgements across all candidates.
Domain 1: Data Gathering and Diagnosis (DG&D)
What It Tests
This domain assesses how effectively you gather the information needed to safely assess the patient's problem — and how you use that information to reason towards a diagnosis.
According to the RCGP, the passing standard for DG&D requires you to:
- Systematically gather and organise relevant and targeted information to address the needs of the patient and their problem
- Adopt a structured and informed approach to problem-solving, generating an appropriate differential diagnosis — or relying on first principles where the presentation is undifferentiated, uncertain, or complex
Within this domain, examiners assess two core capabilities:
| Capability | What It Means in Practice |
|---|---|
| Data gathering | Using targeted questions to gather relevant information, making effective use of existing data, establishing the presence or absence of red flags, and eliciting relevant psychological and social context |
| Making a diagnosis | Using a structured, evidence-based approach to diagnostic reasoning — including generating a differential and managing uncertainty appropriately |
How to Pass DG&D
- Ask targeted, purposeful questions — not an exhaustive systems review
- Actively establish the presence or absence of relevant red flags for every presentation
- Use the information already in the candidate brief — the notes, previous history, medications
- Elicit the patient's psychosocial context — how symptoms are affecting their life, work, and relationships
- Reason aloud or through your actions towards a differential diagnosis, not just a single answer
- Ask enough to distinguish serious illness from something self-limiting — and be able to demonstrate that you have done so
The RCGP's own guidance puts it clearly: with each consultation, ask yourself — what else could this be, and what do I need to exclude? For example, if you are managing mechanical back pain, have you excluded cauda equina syndrome? If a patient is bleeding vaginally, have you considered pregnancy?
What Makes Candidates Fail DG&D
The most common DG&D feedback statement from the RCGP is: "Data gathering was insufficient to enable safe assessment of the condition or situation." This means the history was not broad or deep enough to safely assess the condition and its severity.
The failure mode in DG&D is almost never asking too many questions — it is asking too few of the right ones. Candidates who fail this domain tend to either jump to management without adequate history, or ask generic questions that do not target the specific clinical presentation.
Example
A patient presents with three weeks of lower back pain. A candidate who moves quickly to management after establishing the location and duration of pain — without asking about bladder or bowel changes, bilateral leg weakness, saddle anaesthesia, or the patient's occupation and functional impact — has not gathered sufficient data to safely assess this presentation. The history is too shallow, and the differential has not been adequately explored.
Being curious and interested in the severity of the patient's presentation — and how their symptoms are affecting them — is itself a mark of a safe clinician.
Domain 2: Clinical Management and Medical Complexity (CM&MC)
What It Tests
This is the most heavily weighted domain. It assesses the safety and appropriateness of your management — including how you handle complexity, uncertainty, and the individual patient in front of you.
The passing standard for CM&MC requires you to:
- Formulate a safe and appropriate management plan that is evidence-based, guideline-aligned, and tailored to the individual patient
- Recognise and appropriately manage multi-morbidity — adjusting care as necessary across multiple problems
- Demonstrate effective shared decision-making — explaining options, risks, and benefits clearly
- Safety-net appropriately — explaining to the patient who to contact, when, and why
- Handle clinical uncertainty and complexity in a way that keeps the patient safe
How to Pass CM&MC
- Always offer a management plan — even in complex or uncertain cases, a safe plan with appropriate safety-netting is better than no plan
- Tailor your plan to the patient in front of you, not a textbook version of their diagnosis
- Check for comorbidities and medications that affect your management — do not treat in isolation
- Explain your reasoning to the patient in accessible language — shared decision-making is part of this domain
- Safety-net every case: who to contact, what to look out for, and when to seek help urgently
- Acknowledge uncertainty honestly when it exists — this is a sign of clinical maturity, not weakness
The RCGP confirms that CM&MC is weighted more than DG&D and RTO to reflect the breadth of capabilities it covers. This does not mean you should deprioritise the other two — but it does mean that a weak management plan will cost you more marks than a weak opening question.
What Makes Candidates Fail CM&MC
The most dangerous failures in CM&MC are patient safety failures — offering a management plan that could cause harm. Examples include continuing an NSAID in a patient with known chronic kidney disease, prescribing without considering drug interactions, or failing to escalate a genuinely urgent presentation. These can result in a Clear Fail for the domain regardless of performance elsewhere in the case.
Beyond safety failures, common CM&MC pitfalls include:
- Offering a generic plan without checking the patient's understanding or agreement
- Over-investigating or under-investigating — not proportionate to the clinical scenario
- Failing to safety-net at all, or safety-netting in vague, unhelpful terms
- Not adjusting the management plan for comorbidities identified during the consultation
- Running out of time before reaching management — leaving the case incomplete
Example
A patient with type 2 diabetes and CKD presents with worsening knee pain. A candidate who prescribes an NSAID without checking renal function or considering the patient's existing medications has failed to manage the medical complexity of the case safely. Equally, a candidate who identifies the issue correctly but then offers only a vague plan — "we'll arrange some bloods and go from there" — without explaining what, when, and what to do if things worsen, has not safety-netted adequately.
Domain 3: Relating to Others (RTO)
What It Tests
RTO assesses how you communicate and relate to the patient throughout the entire consultation — not just in isolated moments. It is not a separate phase of the consultation. It runs from the moment you open to the moment you close.
The passing standard for RTO requires you to demonstrate:
- Effective and appropriate verbal and non-verbal communication throughout the consultation
- A genuine patient-centred approach — exploring the patient's ideas, concerns, and expectations, and actively involving them in decision-making
- Empathy and sensitivity — particularly in cases involving difficult news, emotional distress, or sensitive topics
- Professionalism and appropriate boundaries throughout the consultation
- Adaptability — adjusting your communication style to the individual patient and their needs
How to Pass RTO
- Explore Ideas, Concerns, and Expectations (ICE) — naturally, not mechanically
- Respond to emotional cues when the patient gives them — acknowledge feelings before moving forward
- Use open questions early, and closed questions to clarify — not the other way around
- Involve the patient in decision-making genuinely, not as a formality at the end
- Check understanding at the end — summarise, invite questions, confirm the patient is happy with the plan
- Adapt your language — avoid jargon, adjust to the patient's level of health literacy
ICE should be explored conversationally and early — not saved as a separate box to tick towards the end of the consultation. A natural way in: "What were you hoping we might be able to do today?" or "Is there anything in particular you were worried this might be?"
What Makes Candidates Fail RTO
RTO is the domain where clinical knowledge cannot save you. A candidate who knows the right diagnosis and management plan but delivers it in a rigid, mechanical, or dismissive way will still fail this domain. Examiners are practising GPs — they recognise what a good consultation feels like, and they notice when it does not.
Common RTO pitfalls include:
- Consulting like you are running through a checklist — ticking ICE off quickly without genuinely engaging with the patient's response
- Missing or ignoring emotional cues — the patient hints at distress and the candidate moves straight on to clinical questions
- Explaining a management plan without checking the patient understands or agrees
- Using medical jargon without explanation
- Appearing rushed, disengaged, or not making appropriate eye contact with the camera
- A good opening and closing, but poor rapport throughout the middle of the consultation
Example
A patient presents with new-onset palpitations. Midway through the history, they mention they have been under a lot of stress at work since their father died. A candidate who acknowledges this briefly — "I'm sorry to hear that" — and moves on has responded adequately. A candidate who ignores it entirely and continues with the cardiovascular history has missed an emotional cue that the examiner will have noted. A candidate who pauses, explores how the patient is coping, and then connects the stress to the clinical picture will likely score well in RTO even if the clinical management is not perfect.
RTO is not about being warm for the sake of it. It is about ensuring the patient is genuinely heard, understood, and involved — throughout the whole consultation, not just at the beginning and end.
How the Three Domains Work Together in a Consultation
The RCGP's consultation toolkit makes the structure explicit. A 12-minute consultation is not divided equally between three domains — the domains have a natural flow:
- DG&D is primarily active in the first half — history taking, exploring the problem, generating a differential
- CM&MC is primarily active in the second half — management planning, shared decision-making, safety-netting
- RTO runs throughout the entire 12 minutes — from your opening to your closing
The RCGP advises candidates not to try to predict which domain matters most in any individual case. The domain weighting varies between cases and cannot be gamed. The correct approach is to demonstrate all three domains consistently in every case.
Understanding Your Feedback
After every SCA sitting, every candidate receives domain-level feedback. Any domain that is failed receives a feedback statement from the examiner. These statements are taken from a standardised list published by the RCGP and are linked to the official capability framework used in WPBA.
Feedback statements are not a justification of the mark — they are a guide for your future preparation. They may also be given to candidates who passed a domain, to highlight areas for further development. If the same statement appears across multiple cases, that is a strong signal about a consistent weakness in your consulting.
The RCGP recommends sharing your feedback with your GP Trainer or Educational Supervisor and using it alongside your WPBA capability framework to target your development before any resit.
The SCA is marked across three domains: Data Gathering and Diagnosis (DG&D), Clinical Management and Medical Complexity (CM&MC), and Relating to Others (RTO). Each is graded Clear Pass, Pass, Fail, or Clear Fail. CM&MC carries the most weight. DG&D failures typically involve insufficient or untargeted history taking. CM&MC failures often involve unsafe or incomplete management plans. RTO failures stem from mechanical, dismissive, or poorly patient-centred consulting. All three domains must be demonstrated consistently in every case — the pass mark is set fresh for each diet using the Borderline Regression Method.