Decisions, Diagnoses & Uncertainty
Because medicine would be easy if every patient had read the textbook first — and every patient is a chapter you've never seen before.
Last updated: 16 April 2026 · Bradford VTS
📥 Downloads
Teaching resources, frameworks, and reference documents — everything you need to explore this topic deeply. Handouts, summaries, and teaching extras — ready when you are.
path: MAKING DECISIONS
- ambulance - how to use and categories of urgency.docx
- cognitive biases.ppt
- communicating a management plan.doc
- complexity (TEACHING RESOURCE).pdf
- complexity - how doctors think.pdf
- constructs and grids for the consultation.ppt
- deciding for the individual or the population - a story after candide.doc
- deciding what to do - RAPRIOP options.doc
- decision making and safety netting in acute presentations.docx
- diagnostic safety netting.docx
- experts - deciding to use one.ppt
- fast and slow thinking - system 1 and 2 thinking (TEACHING RESOURCE).ppt
- formulating a management plan.doc
- how doctors solve problems.doc
- how doctors think.pdf
- how we make decisions.doc
- illness vs disease.ppt
- microaggressions and therapeutic alliance - exploring our own biases.pdf
- patient management through RAPRIOP.doc
- pattern recognition in the consultation.doc
- pico - asking the right questions in ebm.ppt
- prioritisation - making decisions managing time covey matrix.ppt
- problem solving and achieving goals.doc
- probophilia - making decisions about quality or quantity.pdf
- recommending a strategy.ppt
- six category intervention analysis - facilitating interventions.docx
- six category intervention analysis.docx
- solving problems making decisions and managing crises.docx
- swot analysis form.doc
- swot on 2 sides of A4 plus the form.doc
- the diagnosis cycle and picot.pdf
- when listening is the therapy - the patients lament - hidden key to effective listening.pdf
- when listening is the therapy - the patients lament - turning moaning into therapy.pdf
- when no diagnostic label is applied.doc
- why patients go to doctors.doc
path: UNCERTAINTY
- 5 steps for managing uncertainty.docx
- coping with uncertainty questionnaire.pdf
- coping with uncertainty.pdf
- decision making and safety netting in acute presentations.docx
- diagnostic safety netting.docx
- handling uncertainty - 5 things all patients want to know.ppt
- medically unexplained symptoms - how to tell if organic or not - reducing uncertainty.ppt
- reducing uncertainty in medically unexplained symptoms.ppt
- uncertainty - separating zebras and horses exercise (TEACHING RESOURCE).pdf
- uncertainty - strategies (TEACHING RESOURCE).doc
- uncertainty by worcester vts.pdf
🌐 Web Resources
A hand-picked mix of official guidance and real-world GP training resources. Because sometimes the best pearls are not hiding in the official documents.
Quick Summary — If You Read Nothing Else
⚡ One-Minute Recall
- Good clinical decision-making = System 1 (fast/pattern-based) + System 2 (slow/analytical) — know which is running and whether it's appropriate
- Red patients (urgent) and green patients (reassuring) are relatively easy. It's the amber patients that demand your best thinking
- Your brain is wired for cognitive shortcuts (heuristics) — brilliant when right, disastrous when biased. Know your 10 key biases by name
- Uncertainty is normal in GP — it's not a sign of weakness; it's a sign of an honest doctor in a complex world
- RAPRIOP = 7 management options every GP should run through: Reassurance, Advice, Prescription, Referral, Investigation, Observation, Prevention
- Safety-netting is not optional — it's the mechanism by which you manage uncertainty safely and ethically
- The history is still your most powerful diagnostic tool — approximately 80% of diagnoses come from it alone
- Pattern recognition is fast and often right — but always ask: "What else could this be?" before closing the consultation
- Phronesis (medical wisdom) = doing the right thing, in the right way, at the right time — it comes with reflection and practice, not just knowledge
- The AKT tests real-world GP decision-making — think like a GP, not an exam machine
- In the SCA, showing your decision-making process out loud (explaining your reasoning to the patient) is what examiners reward
Why Decision-Making Matters in GP
Every consultation is a decision. Even doing nothing is a decision. Here's why getting this right matters more than any other clinical skill.
You see 30–40 patients a day. Every one requires a decision — sometimes several. You're doing this under time pressure, with incomplete information, while the previous consultation is still running in your head. Decision-making is not just a skill — it's survival.
Examiners are watching how you think, not just what you know. Every question you ask, every test you order, every management option you choose — all visible evidence of clinical decision-making. This is what separates a pass from a fail.
AKT questions are designed to mimic real GP decisions. The distractors are plausible exactly because they are decisions a less-experienced or biased clinician would make. Knowing how decisions go wrong = getting more questions right.
Cognitive error contributes to an estimated 75% of diagnostic mistakes. Poor decision-making isn't just an exam problem — it has consequences in real GP surgeries every day. Understanding this is the first step to changing it.
How Doctors Think
🧠 System 1 & System 2 Thinking
Nobel Prize-winning science. Genuinely useful in everyday GP. This is the most important framework for understanding how clinical decisions are made — and where they go wrong.
Your brain runs two decision-making systems simultaneously. System 1 is fast, automatic, and pattern-based. System 2 is slow, deliberate, and analytical. Most of the time they work together beautifully. But sometimes one gets in the way of the other — and that's where mistakes happen.
⚡ System 1 in Action
You see an unwell child with a non-blanching rash — you think meningitis before you've consciously processed the information. That's System 1. Fast, potentially life-saving. A shingles rash in a dermatome? Instant pattern. A crushing central chest pain in a 60-year-old man? System 1 fires MI before he's finished his sentence.
A patient calls asking for a home visit. Reason given: "a cough." Before you've said a word, your brain fires: cough = minor illness, not worth a home visit. That's System 1 — and it's already steering you.
Now watch what happens next. Because System 1 is in charge, you subconsciously start gathering data that confirms the impression you've already formed:
🔴 System 1 in charge — selective data gathering:
"Oh, so you've managed to eat okay today?" → "Yes…"
"And you're out of bed?" → "Well, yes…"
"Oh, that's good then." → ✗ Case closed. No home visit.
Now watch what happens when you slow down and use System 2 — asking open questions instead of leading ones:
✅ System 2 engaged — open questioning:
GP: "What's worried you today, Mrs X?"
Patient: "Well, I wouldn't normally call you about a cough. But I managed to get out of bed this morning — and then I collapsed. I'm up now, but I feel dreadful and I'm getting a bit confused. My partner never usually stays home when I'm ill. This time she hasn't left my side because she's worried too."
The same cough. A completely different picture. System 2 didn't just save the day — it potentially saved that patient's life.
🔬 System 2 in Action
A 67-year-old woman presents with chest pain. Parts of the history worry you (radiation to the left arm), but other parts seem reassuring (sharp character). You slow down, think systematically, weigh up the probabilities, and — because of her age and the overall picture — you refer. System 2 overrides your initial "probably musculoskeletal" hunch.
The best GPs don't always use System 1 or System 2 — they flex between the two. They know which system is running, and they can consciously switch when the situation demands it. This awareness is what separates a good GP from an unsafe one.
Have you ever met someone and immediately, instinctively decided you didn't like them — only to find, after actually spending time with them, that they're warm, thoughtful, and someone you genuinely like?
That first impression was System 1. The later reassessment was System 2.
It happens in clinical medicine too. System 1 forms its impression of a patient — their age, their manner, their presenting complaint — in seconds. If System 2 doesn't step in and ask the deeper questions, that first impression can become the whole story. Sometimes it's right. Sometimes it costs a diagnosis.
The goal is not to silence System 1. It's to know when to trust it — and when to ask it to wait while System 2 does its job properly.
System 1 gets better with every patient you see. Each time you encounter a presentation and reflect on it, you're building your pattern library. This is why experienced GPs can sometimes smell a diagnosis before they've asked a single question — their System 1 has been trained by thousands of consultations. Early in training, lean on System 2 more. With experience, trust System 1 more — but never switch off System 2 entirely.
Cognitive Biases — When Your Brain Betrays You
Cognitive biases are systematic errors in thinking that affect decisions and judgements. They are not a sign of stupidity — they are a feature of all human brains. The doctor who says "I'm not affected by bias" is, by definition, the most biased person in the room.
| Bias | What It Is | GP Example | How to Counter It |
|---|---|---|---|
| 🎯 Anchoring Bias | Locking onto initial information and not adjusting when new data arrives | Patient comes in labelled "anxiety" — you keep anchoring on that even when vitals suggest PE | "What if this first impression is wrong?" — explicitly ask it |
| 🔒 Premature Closure | Stopping the diagnostic search once a plausible diagnosis is found | Diagnosing UTI in an elderly woman with confusion — and missing the subdural haematoma | Complete the full history. Apply "one more question" rule before closing |
| 📺 Availability Bias | Judging likelihood based on what's easily recalled — often the last dramatic case you saw | After a meningitis case, you over-investigate every headache for two weeks | Base probabilities on epidemiology, not recent memory |
| ✅ Confirmation Bias | Seeking information that confirms your existing view; ignoring contradictory data | You think it's viral so you ask leading questions: "Has it been going on just a few days?" "No fever?" | Ask open questions. Actively seek disconfirming evidence |
| 🖼️ Framing Effect | Being unduly influenced by how information is presented to you | "This is the anxious patient in room 4" — before you've met them, your frame is set | Form your own clinical impression fresh; challenge inherited labels |
| 😐 Diagnosis Momentum | A diagnosis sticks as it gets handed from clinician to clinician without being re-evaluated | Patient referred as "fibromyalgia for 10 years" — but no GP has re-examined this in 3 years | Periodically re-examine established diagnoses, especially in undifferentiated presentations |
| 💡 Overconfidence Bias |
Believing your clinical judgement is more accurate than it actually is | "I've never been wrong with chest pain" — until you are | Maintain appropriate humility; use checklists and safety-nets |
| 🩺 Ascertainment Bias | Stereotyping patients based on demographics, labels, or previous encounters | Assuming weight-related symptoms in an obese patient are always obesity-related | Treat each presentation on its own clinical merits |
| 🔄 Commission Bias | Preferring action over inaction — "doing something feels safer" | Prescribing antibiotics for a clearly viral URTI because it "feels like doing something" | Watchful waiting is a legitimate clinical decision. Own it confidently |
| 🚫 Omission Bias | Preferring inaction to avoid causing harm, even when action is indicated | Not prescribing anticoagulation in AF because you're worried about falls | Quantify the risk of inaction versus action — both have consequences |
Anchoring + Premature Closure + Confirmation Bias — these three biases frequently occur together. You form a quick diagnosis (anchoring), stop gathering further information (premature closure), and then only ask questions that confirm it (confirmation bias). This is how missed diagnoses happen. This is how patients get hurt.
- Diagnostic timeout — pause deliberately and ask "What else could this be?"
- Speak to a colleague — sharing risk and getting a second set of eyes
- Use checklists — systematic tools interrupt intuitive shortcuts
- Reflective practice — review cases where your first instinct was wrong
- Mindful slowing down — especially at HALT: Hungry, Angry, Late, Tired
🔍 Pattern Recognition in GP
The closest thing to a clinical superpower — and the foundation of System 1 thinking.
What Is Pattern Recognition?
It's the ability to match a current patient's presentation to a mental template built from previous experience. Your brain says: "This pattern = this diagnosis." It bypasses conscious analytical thinking and arrives at an answer in milliseconds.
The history is still the most powerful diagnostic tool — approximately 80% of diagnoses come from it alone. Pattern recognition works on that history data.
How GPs Use It
- Illness scripts — mental templates for how a disease typically presents
- Prototypical features — the "classic" picture that triggers recognition
- Contextual cues — age, sex, social history, time of year, past history
- Non-verbal signals — how a patient looks, moves, and speaks
- Gut feeling — the early warning signal of something amiss before you can articulate why
Pattern recognition can lead you to stop looking once you've found a plausible match — even when the full picture doesn't quite fit. This is called premature closure: accepting the first diagnosis that fits without asking "what else could this be?" Always complete the picture. A depression history isn't done after asking about mood and anhedonia. Ask about sleep, appetite, concentration, irritability, and — critically — suicidal ideation.
| Feature | Builds Pattern Recognition | Undermines Pattern Recognition |
|---|---|---|
| Experience | More patients = richer pattern library | Inexperience = fewer reference points |
| Reflection | Reflecting on cases deepens learning | Unreflective practice = same errors repeated |
| Feedback | Knowing outcome sharpens patterns | No feedback = patterns never corrected |
| Context | Full context makes patterns reliable | Incomplete history = pattern mismatch risk |
| Cognitive load | Low workload = clear thinking | Tired/stressed = pattern quality degrades |
📋 The Three Pillars of Good Clinical Decision-Making
There is no shortcut around a thorough clinical history. That said — a thorough history does not mean an exhaustive one. It means a complete enough history to make a safe clinical decision for this patient at this time.
Take depression as an example. Asking only about low mood and anhedonia is not enough. A complete depression history covers: mood, anhedonia (social and sexual), concentration, appetite, sleep (initial, interrupted, and early-morning waking), irritability, energy, motivation, and — without exception — suicidal ideation. Missing the last item is not a minor oversight; it's a serious patient safety issue.
Red flags exist because they change what you do next. Knowing your red flags by system is not optional learning. If you don't check for the serious things in a comprehensive and structured way, you have not completed a safe consultation — regardless of how good your rapport was or how elegant your explanation was.
- Every common presentation has a dangerous variant — know what it looks like
- Red flags must be explicitly elicited, not assumed absent
- Red flags must be explicitly documented — "checked and absent" is medico-legally important
- In the SCA, always mention key red flags — examiners do not give you credit for what they have to infer
You will not be asked to perform a "full general examination" in any GP assessment — it is both impractical and clinically inappropriate. Instead, the examination you choose must be the examination that your history suggests is needed. This is itself a test of decision-making.
- Choosing the right examination system demonstrates clinical reasoning
- Examining multiple systems indiscriminately suggests poor reasoning — it's visible to examiners
- Always signpost before examining: "I'd like to examine your chest now, because you mentioned breathlessness — is that okay?"
- A normal examination is a positive finding — document and use it
When you say "I'd like to examine X because of Y," you are demonstrating explicit clinical reasoning. This is exactly what SCA examiners want to see — connecting your history findings to your examination choice. Never just start examining. Say why.
Over-investigation is as much a clinical error as under-investigation. Ordering tests "to cover your back" is not good medicine — it reflects poor decision-making, wastes NHS resources, and can cause patient harm (false positives, incidental findings, unnecessary anxiety).
| Scenario | Appropriate? | Reasoning |
|---|---|---|
| TATT × 3 months — FBC, TFTs, LFTs, HbA1c, ESR, ferritin | ✅ Yes | Duration warrants investigation; multiple common causes need excluding |
| TATT × 1–2 weeks in previously well adult | ❌ Probably not yet | Too early; watchful waiting + safety-net is more appropriate |
| Chest pain in 30-year-old, typical musculoskeletal features, no risk factors | ⚠️ Targeted only | Clinical assessment sufficient if history and exam are reassuring |
| Any presentation where you're uncertain about diagnosis | ✅ Use it | Targeted investigations reduce uncertainty — this is legitimate use |
Every investigation has a financial cost, a time cost, and a patient-experience cost. The hypothetico-deductive model suggests investigations should confirm or refute hypotheses — not generate them. Ask: "What will I do differently based on the result?" If the answer is "nothing," don't order it.
When a doctor orders every test under the sun, it is — if we are being honest — a symptom of something. It signals a lack of confidence in your own clinical reasoning. If you are not sure what you are looking for, you cast a wide net and hope something comes back positive to tell you what to do next. That is the investigation driving the clinical thinking, rather than the clinical thinking driving the investigation. It should always be the other way around.
This is why developing the habit of focused, purposeful ordering matters so much — and the sooner the better. Every investigation you request should exist to answer a specific clinical question in your head. Before you click, ask yourself: "What exactly am I trying to find out, and how will the result change what I do?" If you cannot answer that clearly, stop and think before ordering.
Getting into this habit early does two things. It makes you a better, safer clinician. And it makes you look like a better, safer clinician — because examiners, supervisors, and colleagues can all tell the difference between a doctor who investigates with purpose and one who investigates out of anxiety.
Clinical Decision-Making Frameworks
🚦 The Red, Amber & Green Framework
An elegant mental model for clinical decision-making that works across virtually every presentation in primary care.
Red patients are relatively easy — you know what to do (admit). Green patients are also relatively easy — you know what to do (safety-net and reassure). It's the amber patients where the real decision-making skill lives. These are the ones that demand your best System 2 thinking, your best safety-netting, and your most careful follow-up planning.
The 7 Clinical Scenarios — And What To Do
| Scenario | Trajectory | Decision Complexity | Action |
|---|---|---|---|
| Green → stays Green | → 🟢🟢 | ⭐ Low | Reassure, safety-net, watchful waiting. E.g. viral URTI, mild back pain without alarm features |
| Green → turns Amber | → 🟢➡️🟡 | ⭐⭐⭐ Moderate-High | Start treatment, review shortly, safety-net for rapid deterioration. E.g. viral URTI now looking like early LRTI |
| Green → turns Red | → 🟢➡️🔴 | ⭐⭐ Moderate (urgent) | Admit. E.g. patient who presented with cough is now critically unwell on examination |
| Amber → stays Amber | → 🟡🟡 | ⭐⭐⭐⭐ Very High | Closely monitor, consider admission, further investigations, different treatment, second opinion. E.g. LRTI not settling after 3 courses of antibiotics → consider 2WW |
| Amber → turns Red | → 🟡➡️🔴 | ⭐⭐ Moderate (urgent) | Admit. E.g. previously borderline LRTI now looks dreadful despite antibiotics |
| Amber → turns Green | → 🟡➡️🟢 | ⭐⭐ Lower (but don't relax entirely) | Cautious reassurance + robust safety-netting. Colours can change again quickly — especially in children |
| Presents as Red | → 🔴 | ⭐⭐ Moderate (act fast) | Admit immediately. E.g. new-onset seizures, meningism, haemodynamic compromise → give benzylpenicillin if indicated, call 999 |
🧒 Paediatric Special: Amber in Children (Dr Edward Snelson)
Children deserve a special mention in the RAG framework — because they change colour faster than adults, and their amber presentations are uniquely challenging.
Consider a 2-year-old presenting with a cough, fever, and misery — temperature 39.5°C, HR 160, refusing to drink. But the parent reports the child looked much better 2 hours ago and was drinking a little. Unlike traffic lights, unwell children at this age swing from green to amber and back during uncomplicated infections. The normal physiological response to a viral illness can look alarming but usually resolves to reveal a reassuring baseline.
This age group is high-volume but actually lower risk for dangerous infections than very young infants — yet the challenge is identifying the small proportion with genuine serious illness within that amber presentation.
A 2-year-old presents with a cough and fever. The parent reports that a couple of hours ago the child was pale, lethargic, shivering, and had cold hands and feet with a hot core. Now they are walking, talking, and cheerfully interactive. The parent is visibly embarrassed — half-expecting you to tell them they overreacted.
This is one of the most common — and most important — paediatric scenarios in GP. Two things to recognise: the child is genuinely better right now, but what the parent described earlier was real and could happen again. Both matter.
What to do:
- Acknowledge how frightening the earlier episode was — don't minimise it or make the parent feel they were wrong to come
- Explain clearly why you are reassured now, and what makes this different from sepsis or meningitis
- Safety-net explicitly — the child can turn again at short notice
"What you described earlier sounds quite frightening, and I want you to know — you were right to come. At this age, children can respond very dramatically to infections: high temperatures, cold hands and feet, lethargy. That's their immune system working hard, and it can look alarming. What's really reassuring is how your child looks right now. The fact that they're walking, talking, and interacting tells me this isn't sepsis or meningitis — children with those conditions don't do what your child is doing now. I'm happy for you to continue managing at home with paracetamol and ibuprofen. But I want to be clear about when to come straight back..."
Follow with specific safety-netting: named symptoms, specific actions, clear timeframe.
Content contributed by Dr Edward Snelson, Paediatrician. See gppaedstips.blogspot.com
🔝 Clinical Decision-Making: Top Tips
- The decision to use watchful waiting vs. face-to-face review depends on: carer reliability, ability to monitor physiological improvement, and whether you need another clinician's assessment
- Many GPs in primary care do not need face-to-face re-assessment if the patient shows clear behavioural improvement — that itself demonstrates physiological change
- Good safety-netting empowers patients to be dynamic, continuous monitors of their own condition — it's a shared responsibility tool, not a defensive formality
- Amber patients take more work — but they are the core business of general practice. They are the group where your clinical reasoning, safety-netting, and follow-up planning matter most. Red and green patients largely manage themselves. Amber patients are where GP skill lives.
- 🚸 Infants and children change colour quickly — in both directions. A child who is fine one moment can deteriorate rapidly — and vice versa. Always safety-net for this explicitly. Write it down if the situation is complex. Make absolutely sure the carer knows what to look for and what to do.
- Availability bias is real: don't let "most kids with a cough get better" stop you assessing this child on their own merits today
- When genuinely uncertain: seek a senior opinion, document the discussion, share the risk safely — the Bolam principle protects reasonable peer-agreed decisions
- Trust your gut feeling — it is often an early warning signal from your System 1 that something is not right, even before you can articulate why. Investigate it with System 2
Decision-making: easy
Decision-making: relatively easy
Decision-making: the hardest
🗂️ RAPRIOP: Your 7 Management Options
One of the most useful frameworks in GP — and almost never taught explicitly. RAPRIOP gives you a mental checklist of every management option available in primary care.
RAPRIOP was developed by Brian McAvoy and described in Robin Fraser's seminal textbook Clinical Method: A General Practice Approach. It remains as relevant today as it was then — because the management options in GP haven't fundamentally changed.
Why RAPRIOP Matters
In the heat of a consultation, it's easy to default to "prescribe" or "refer" without considering all available options. RAPRIOP forces a structured pause: what is the best management option for this patient right now?
Notably, a prescription is only one of seven options — and often not the most appropriate one. Good GPs use the full range.
RAPRIOP in Practice
For a patient with low back pain (no alarm features, 2 weeks duration, mild):
- R — Reassure: not serious, will resolve
- A — Advice: stay active, avoid bed rest
- P — Prescribe: analgesia if needed (NICE ladder)
- R — Refer: not yet indicated
- I — Investigate: not indicated at this stage
- O — Observe: review in 4–6 weeks if not improving
- P — Prevention: posture, weight, exercise advice
AKT management questions are essentially asking: "Which RAPRIOP option is most appropriate here?" When you see a management question, mentally run through RAPRIOP. The correct answer is usually the most conservative appropriate option — not the most dramatic one. Prescribing when reassurance is sufficient, or referring when watchful waiting is correct, are common AKT distractors.
Managing Uncertainty & Medical Wisdom
🌫️ Understanding & Managing Uncertainty
Diagnostic uncertainty in general practice is not a sign of poor medicine. It is an inherent feature of primary care — where time is needed for a condition to declare itself, where multimorbidity blurs presentations, and where cultural and individual illness narratives shape what gets said and what doesn't. The skill is not to eliminate uncertainty, but to manage it safely and honestly.
🌫️ Why GP Is Different
In hospital medicine, patients often arrive pre-selected with a problem that is already partly characterised. In GP, you see undifferentiated illness at its earliest — before the diagnosis has declared itself. This is why uncertainty is part of the job description, not an exam failure.
✅ The GP Uncertainty Compass
Ask yourself three questions before closing:
- Have I excluded the serious diagnoses for this presentation?
- Have I met this patient's needs today (including their emotional needs)?
- Is there a reliable safety net in place if things change?
All three = proceed with appropriate confidence. Any not met = address before closing.
Five Steps for Managing Uncertainty
Research shows patients with uncertain diagnoses broadly want to know five things: What might this be? What won't it be? What will you do about it? What should I watch for? When will I know more? Answer all five — even briefly — and uncertainty becomes manageable for everyone.
In GP, the passage of time is a legitimate and powerful diagnostic tool. A presentation that is unclear today may become much clearer in 48–72 hours. Watchful waiting with structured safety-netting is often the right answer — not a cop-out. Use time deliberately: "Let's see how this develops over the next few days and review then."
MUS represents the ultimate test of your ability to manage uncertainty. There is no clear diagnosis. The patient is suffering. Tests keep coming back normal. The temptation is to keep investigating to "find something" — but this often makes things worse, reinforcing illness behaviour and patient anxiety. The skill is to: acknowledge the reality of the patient's experience, explain the current negative findings honestly, develop a positive management plan, and avoid the trap of endless investigation as a substitute for a therapeutic relationship.
⚖️ Phronesis — Medical Wisdom
The intellectual virtue at the heart of wise doctoring. You can know the guidelines perfectly and still make the wrong decision. Phronesis is what bridges that gap — and it can be learned.
What Is Phronesis?
Phronesis (φρόνησις) is Aristotle's concept of practical wisdom — the ability to discern the right course of action for this patient, in this context, at this moment. It is not simply knowing what the guidelines say. It is knowing what to do when guidelines, values, patient preferences, and real-life complexity collide.
Aristotle described five intellectual virtues. Two have become thoroughly embedded in medical education. The other three are barely mentioned.
🔬 Dr Sabena Jameel — Researching Wisdom in GPs
Dr Sabena Jameel (GP, Associate Clinical Professor, University of Birmingham) spent seven years researching a deceptively simple question: what does wisdom actually look like in a working GP? Her 2021 PhD — Enacted Phronesis in General Practitioners — is the most comprehensive empirical study of practical wisdom in family medicine practitioners in the UK.
Starting by surveying 211 GP trainers across the West Midlands using the Ardelt 3-Dimensional Wisdom Scale, she identified the top 10% as potential wisdom exemplars and conducted 18 in-depth biographical narrative interviews. From that rigorous analysis, she derived 34 constituent features of enacted phronesis — and developed a striking new theory to describe it: The Fish School Theory of Practical Wisdom.
- 211 GPs surveyed across West Midlands
- 16 wisdom exemplars interviewed
- 34 constituent features of phronesis identified
- Wisdom correlates with job satisfaction and wellbeing
- Not all older doctors are wiser — age alone is not enough
- Some of the wisest were trainees who had overcome adversity
🐟 The Fish School Theory of Practical Wisdom
Jameel's most striking insight is that phronesis is not a single quality — it is a constellation of many qualities that move together. Like a school of fish, no single quality is dominant. They swim as a coordinated whole. Remove one and the formation changes; add one and the whole group responds. Wisdom is fluid, dynamic, and relational — not a fixed checklist.
Her research identified 34 constituent features, grouped into five key areas that together constitute enacted wisdom in a GP:
Resilience, compassion, integrity, emotional intelligence, curiosity, and a strong moral orientation towards good — the character virtues that underpin everything else.
Constructive reflection (growth-oriented, not ruminative), tolerance of uncertainty, willingness to be wrong, deliberate self-questioning, and pattern-recognition built over time.
Acute self-awareness. Knowing your own biases, triggers, and blind spots. Understanding how your background and experiences shape your clinical judgement — and compensating accordingly.
Deep listening. Giving patients the space and time to be heard. Building trust. Recognising that wisdom is not a solo pursuit — it is relational, developed in and through connection with others.
Life experience — including adversity. Personal illness, caring for relatives, systemic challenges. Jameel found that many wise exemplars had navigated significant personal hardships and turned them into insight. Wisdom grows from experience, but only when that experience is processed reflectively.
🩺 What Wise Doctors Actually Do — Themes from the Research
One of Jameel's clearest findings: wise GPs have a distinctively high tolerance for uncertainty. They do not panic when the picture is unclear. They gather more information, safety-net clearly, and use time deliberately — without either over-investigating or dismissing. The ability to sit with not knowing is a mark of wisdom, not weakness.
Wise doctors are driven by a clear sense of purpose: the desire for the patient to flourish. Jameel describes this as a "purposeful journey" — justice and compassion are highly prized. When clinical decisions are complex, it is this moral orientation that navigates the way. They ask not just "What does the guideline say?" but "What is right for this person?"
Wise doctors reflect — but not in a self-critical, ruminative way. Research by Meeks & Jeste (2009) and Jameel both emphasise growth-oriented reflection: learning from experience, understanding what happened and why, and using that understanding to improve. The reflection that leads to wisdom is purposeful, forward-looking, and non-self-punishing.
Jameel uses the image of a graphic equaliser to describe how wise doctors work. Every consultation involves adjusting multiple sliders — between doctor-centred and patient-centred, between firm and flexible, between clinical caution and human connection. Wisdom lies in knowing exactly where to set each slider for this patient, at this time. Not fixed. Responsive.
Research by Paton & Kotzee (2021) demonstrates that doctors develop phronesis primarily through narrative — listening to and telling stories about difficult cases. These stories become a shared bank of experience. Hearing how a colleague navigated an ethical dilemma is as formative as experiencing it yourself. This is why good tutorials and case discussions are not just revision: they are wisdom-building.
Jameel tested Aristotle's hypothesis directly: that developing phronesis leads to eudaimonia (flourishing). The correlation between high wisdom scores and a sense of wellbeing, satisfaction, and contentment in work was clear and striking. Developing practical wisdom is not just better for your patients. It appears to protect against the burnout and demoralisation that plague contemporary general practice.
🧪 The Psychology of Wisdom — Three Dimensions
Jameel used the Ardelt 3-Dimensional Wisdom Scale to identify exemplars. Monika Ardelt (2003) — drawing on psychology research — proposed that wisdom has three mutually reinforcing dimensions. Together these approximate closely to phronesis in clinical practice:
The ability to understand life and see situations clearly — including uncomfortable truths about oneself. Avoiding self-deception. Seeing beyond surface appearances. Understanding that most clinical situations are more complex than they first appear.
In practice: "I need to be honest with myself about why I made that decision — was it really clinical, or was I influenced by the patient's manner?"
Examining situations from multiple perspectives. Not taking your initial interpretation as the only one. Considering how the patient sees things, how another clinician might see it, and what you might be missing. The capacity for genuine epistemic humility.
In practice: "I felt certain of my diagnosis — but let me think about what else could explain this picture, and whose perspective I haven't fully considered."
Genuine empathy and care for patients as people. Not performed kindness, but actual interest in the patient's life and wellbeing. Jameel found that wise doctors experience genuine joy in helping people — it is not merely a professional duty.
In practice: noticing a patient's emotional state even when it is not what they are presenting with — and responding to the person, not just the problem.
⚖️ Aristotle's Golden Mean — The Virtue Equalisers
Birmingham University researchers (Conroy et al., 2021) built on Jameel's work by identifying 15 virtues relevant to wise medical decision-making. Crucially, they mapped each virtue as a midpoint between two extremes — echoing Aristotle's concept of the Golden Mean. Wisdom lies in finding that midpoint, not in sitting at either extreme.
| Too Little | ⚖️ Wise Midpoint | Too Much |
|---|---|---|
| Rigid, rule-bound | Principled but flexible | Inconsistent, rule-bending |
| Doctor decides alone | Negotiate with patients | Patient always decides, even unsafely |
| Cold, clinical, detached | Emotionally intelligent | Over-involved, boundary-less |
| Accepts things uncritically | Constructively reflective | Ruminating, self-flagellating |
| Reckless, overconfident | Appropriately courageous | Paralysed by caution |
| Dismisses patient concern | Takes it seriously without catastrophising | Over-medicalises everything |
| Never asks for help | Seeks advice at the right moment | Over-dependent on others' opinions |
The task of the wise GP is not to pick a fixed point on any of these spectrums — but to know where to sit on each one, for this patient, in this consultation, today.
Jameel's literature review identified clear barriers to phronesis development — and many of them are built into medical training:
- Assessment and competency-driven learning — a relentless focus on ticking boxes crowds out reflection on meaning and values
- The hidden curriculum — the informal attitudes, shortcuts, and dismissals modelled by senior clinicians that quietly teach the opposite of phronesis
- Anti-mentors — negative role models can both actively inhibit phronesis and, paradoxically, inspire it ("I will never be like that")
- System pressures and burnout — practical wisdom cannot flourish in a constraining, exhausting system. Phronesis needs space to breathe
- Over-reliance on guidelines — using guidelines as a substitute for judgement rather than a resource for it
🌱 How to Develop Phronesis — Evidence-Based Approaches
The good news is that Aristotle was right: phronesis can be taught and learned. Here is what the evidence suggests actually works:
Listening to senior colleagues' difficult case stories builds a bank of vicarious wisdom. Don't just ask "what did you do?" — ask "what was going through your mind? What made it hard? What would you do differently?" These are phronesis conversations.
Not just "what happened" but "what does this reveal about me?" Growth-oriented, not punishing. Schön's reflection-in-action (during the consultation) and reflection-on-action (after it) are both important — and both trainable habits.
Deliberately sit with ambiguous cases rather than immediately resolving them. Notice what it feels like. Use the diagnostic timeout. Safety-net explicitly. Learn that uncertainty does not require panic — it requires management.
Watch the clinicians you admire not just for what they do, but how they think. Ask them to describe their reasoning out loud. The biographic narrative approach used in Jameel's research works in tutorials too — the stories of wise doctors teach wisdom.
Your FourteenFish portfolio is not just a tick-box exercise. When you reflect on a difficult ethical case, a consultation that felt wrong, or a moment when you had to choose between what the guideline said and what felt right — that is phronesis training in writing.
Jameel's finding that wisdom correlates with wellbeing cuts both ways. Exhausted, demoralised doctors struggle to access their phronesis. Protecting your own flourishing is not selfish — it is a precondition for wise clinical care.
- The frail 92-year-old with AF — guidelines say anticoagulate, but she falls frequently and lives alone. What does wisdom say?
- The patient who needs hospital admission but has nobody to care for their children tonight
- The 45-year-old who wants to reduce antidepressants despite a history of severe relapse — because they feel fine
- The young man misusing controlled drugs — you feel compassion, but you cannot simply give what they want
- Choosing whether to follow the guideline or honour the patient's informed refusal
- Knowing when not to refer — when watchful waiting is the wiser choice than action
Jameel describes the consultation as containing a Hermeneutic Window — a space where doctor and patient make meaning together of what is happening. This is not the presenting complaint. It is not the examination findings. It is the human encounter itself.
Conventional medical education barely acknowledges the Hermeneutic Window exists. Phronesis is the virtue that equips you to function within it — to understand not just the disease but the illness, not just the patient's words but their meaning, not just what they say but what they need to hear.
You are already developing phronesis — whether you know it or not. Every difficult consultation, every case that kept you awake, every moment of uncertainty that you navigated rather than just survived: these are the raw material of practical wisdom.
The difference is whether you process those experiences reflectively or just let them pass. A wise clinician is not one who never feels uncertain. It is one who has learned to use uncertainty, to question their own assumptions, to listen to their gut feeling as data, and to hold the patient's humanity alongside the clinical complexity — all at the same time.
You cannot shortcut phronesis. But you can accelerate it — through deliberate reflection, through story-listening, through seeking out wise mentors, and through choosing to treat your hardest consultations as the best lessons you will ever have.
🐟 The Fish School — Visualised
Jameel's 34 constituent features of practical wisdom don't form a list — they form a constellation. Here is how they cluster into five mutually reinforcing domains:
🌱 How Phronesis Develops — The Pathway
Aristotle was right: phronesis is not inherited — it is cultivated. But experience alone is not enough. The key is reflective processing of experience. Here is the evidence-based pathway:
🎛️ The Graphic Equaliser — Finding Your Clinical Sweet Spot
Jameel and colleagues describe wise consultations like a graphic equaliser: every clinical encounter requires setting multiple sliders between two extremes. Wisdom is not about picking a fixed position — it is about reading the situation and adjusting in real time.
📐 The Five Intellectual Virtues — Why Phronesis Is the Missing One
💬 What Trainees and Junior Doctors Say About Wisdom
These insights come from documented trainee voices in UK GP training research and VTS teaching sessions — what doctors at the start of their careers say about wisdom, once it has been explained to them.
One of the biggest misconceptions trainees carry is that wisdom belongs to consultants after 30 years of practice. Jameel's research showed otherwise — some of the wisest GPs in her study were trainees who had navigated significant personal adversity and processed it reflectively. Age correlates loosely with wisdom, but reflection correlates more strongly. You can begin accumulating wisdom from your first consultation.
Doctors consistently identify the difficult cases — not the straightforward ones — as the formative ones. The consultation that went awkwardly, the diagnosis that was missed, the patient who left unsatisfied. These are not failures to be buried. They are the raw material of phronesis, but only if you go back and examine them. A case that keeps you awake at night is a case waiting to make you wiser.
Kaldjian and colleagues note clearly: experience alone is not sufficient for wisdom. Without reflection, experience merely justifies repeating the same patterns with more confidence — including the wrong ones. The trainee who has seen 2,000 patients but never examined what they learned is not necessarily wiser than the trainee who has seen 500 but reflected deeply on each one.
Research by Paton & Kotzee (2021) on UK junior doctors found that hearing colleagues tell stories about ethically difficult situations was the primary way doctors developed practical wisdom. The stories became a shared bank of experience that could be drawn on later. This is why a good tutorial is not just revision — it is wisdom transmission. Ask your trainer for their hardest cases. The stories teach what guidelines never can.
A question raised spontaneously by trainees at one of Jameel's teaching sessions — and a fascinating one. The research does not confirm or deny this. What it does show is that GPs who score highly on wisdom indicators feel significantly more satisfied and fulfilled in their work. Whether GP selects for wisdom, or wisdom grows through general practice, or both — the correlation between wisdom and wellbeing is consistent and striking.
Many doctors, when first introduced to phronesis, respond with recognition rather than novelty. They already knew what it felt like to sit with a patient and sense that the guideline wasn't quite right for this person, at this moment. Phronesis gives language to something they were already doing — imperfectly, intuitively, without a framework. Naming it is the first step to cultivating it deliberately.
Insights consistently highlighted in UK GP training contexts — things that actually work in practice:
In tutorials, ask your trainer: "Tell me about the consultation that changed how you practise." These stories are worth more than ten case presentations. Narrative is the primary vehicle for transmitting wisdom.
Not just "what happened" but "what was I feeling? What made me uncomfortable? What was I avoiding?" The discomfort is data. Processing it with a trainer or in your portfolio is where wisdom grows.
Don't just document what you did — document what you were unsure about, what you wished you had done differently, and what you noticed about yourself. Those entries demonstrate phronesis development far better than polished summaries.
Saying to your trainer: "I'm not sure what to do here, and here's why it feels hard" is a wiser statement than presenting false confidence. Articulating your uncertainty clearly is itself a mark of developing phronesis.
GP training schemes consistently recommend reading medical humanities, patient narratives, and ethics alongside clinical texts. GP wisdom lives at the intersection of the scientific and the human. Read on both sides of that divide.
HDR (Half Day Release), tutorial groups, and peer discussion sessions are not just revision — they are communities of practice where tacit wisdom is transmitted. Engage actively, share cases, ask questions, be honest about what you don't know.
📊 What Contributes Most to Developing Phronesis?
Based on themes identified across Jameel (2022), Conroy et al. (2021), Kaldjian et al. (2023), and Paton & Kotzee (2021) — the relative contribution of different inputs to practical wisdom development in GPs:
✅ Phronesis Self-Audit — Where Are You on the Journey?
Not a formal tool — just an honest set of questions to prompt reflection. Read each one and ask yourself: is this something I do? Could I do it better?
- Do I notice when my emotions are affecting my clinical judgement?
- Can I hold a patient's difficulty with genuine compassion — not just professional courtesy?
- Do I take responsibility when I get something wrong — without self-punishment?
- When I leave a hard consultation, do I go back to it — or just try to forget it?
- Can I sit with clinical uncertainty without it becoming anxiety?
- Do I question whether a guideline is right for this particular patient?
- Do I ask senior colleagues to tell me their hardest cases — not just the correct answers?
- Do I genuinely listen in consultations, or am I already formulating my answer?
- Could my patients tell that I was interested in them as a person?
- Do my portfolio reflections show genuine uncertainty — or just competent performance?
- Have I written about a case where I felt the guideline was wrong for this patient?
- Have I discussed a consultation that kept me awake with my trainer?
If most of these feel unfamiliar — that is fine. That is exactly why this section exists. The audit is the beginning, not the verdict.
🎓 The WiseGP Programme — Practical Wisdom, Nationally Recognised
In 2021, Professor Joanne Reeve — GP and Professor of Primary Care at Hull York Medical School — launched the WiseGP Programme, supported by Health Education England Yorkshire and the School for Primary Care Research. It is one of the most significant national initiatives directly focused on developing practical wisdom in UK general practice. And it is free to access.
💡 The Core Idea: You Are a Knowledge Worker
Reeve makes a crucial distinction. GPs are not information workers — collecting and processing data against externally set targets. They are knowledge workers: using clinical knowledge, life experience, and wisdom to create tailored, contextual understanding of each patient's situation. Phronesis IS the knowledge work of general practice.
"In general practice there is a relentless trend to replace knowledge workers with so-called information workers." — Prof Joanne Reeve, BJGP Life
🔎 Mindlines: Wisdom in Your Head
Gabbay & Le May (2004) studied how GPs actually use knowledge in practice. Their finding: clinicians rarely consult guidelines directly in consultations. Instead they develop mindlines — internal, personalised, continuously updated guides built from evidence, experience, peer discussion, and patient feedback. Mindlines are practical wisdom in action. The GP who has excellent mindlines is the one you want treating your family.
Gabbay & Le May: Evidence based guidelines or collectively constructed "mindlines"? BMJ, 2004.
🧵 How Mindlines Work — From Evidence to Action
Gabbay & Le May (2004, BMJ) observed GPs in practice. Contrary to the EBM ideal of "consult the guideline," GPs actually navigate practice through constantly updated internal frameworks. This is not laziness — it is phronesis operating in real time:
🫂 Balint Groups — The Original Phronesis Workshop
Decades before phronesis entered medical education discourse, Michael Balint (psychoanalyst, Tavistock Clinic, 1950s) was already doing what Jameel would later research empirically. He created a space for GPs to reflect on the doctor-patient relationship — not to solve clinical problems, but to understand them as human ones. This is phronesis development in a group, live.
🩺 What Happens in a Balint Group?
- A trainee presents a patient from memory — no notes, just the felt experience of the consultation
- The group asks: "How did this patient make us feel?" — not "What should we have done?"
- The focus is the doctor-patient relationship — the emotional texture of the encounter
- No clinical solutions sought. Instead: insight, perspective, recognition
- "Difficult" or "puzzling" patients are often the most instructive
- All discussion is confidential — a safe space for honesty
💜 What Trainees Get From Balint Groups
- Catharsis — the relief of naming what you're carrying
- Deep reflection — structured space to understand what really happened
- Perspective — colleagues help you see what you couldn't see alone
- Reduced isolation — "It's good to get other people's opinions on how they would deal with it"
- Resilience against burnout — GPs who attend Balint groups report significantly better wellbeing
- Phronesis — developing the capacity to read the human dimension of every consultation
Balint described the doctor themselves as a powerful therapeutic agent — separate from any prescription or test. The doctor's presence, listening, and humanity is itself treatment. This is phronesis expressed in a relationship. The question is not just "what drug to prescribe?" but "how should I be, as a person, in this consultation?"
Balint noted that doctors hold beliefs about how patients should behave, and subtly push patients towards those expectations. Wise doctors recognise and actively counteract their own apostolic function — they meet the patient where they are, not where the doctor wishes they were. This requires self-awareness: a core constituent of phronesis.
🔭 The Future Doctor — Wisdom Is Now Official Policy
In July 2020, Health Education England published their Future Doctor Report, specifying what tomorrow's doctors need. The language is striking: GPs must have "generalist skills and the wisdom to make complex judgements, adapt, and be a catalyst for change." Phronesis — practical wisdom — is now an explicit expectation of the future GP, endorsed at national level.
Explicitly requires capability in tolerating uncertainty, making contextual clinical judgements, and acting with ethical wisdom. These are phronesis by another name.
Requires doctors to exercise professional judgement, balance competing values, and act in patients' best interests. This is the language of phronesis in regulatory form.
Reflective log entries, CbD discussions, and educational supervisor reviews all offer direct opportunities to document phronesis development — if you know what you're looking for and how to write about it.
🗣️ Voices From UK GP Educators and Trainers
Jameel and Turner — who have together presented on phronesis to GP audiences nationally — describe how studying phronesis changed their own practice: they now prescribe less. The mechanism is simple: when you listen more deeply to what the patient actually needs, you realise that the prescription is often not the most useful thing you can give. The consultation itself can be the treatment.
Dr Chris Turner, whose team devised the 15-virtue framework at Birmingham, notes that a core skill of the phronimos is narrative articulacy: being able to explain, to yourself and to others, the reasoning behind a wise decision. This is exactly what good portfolio reflections, tutorial discussions, and educational supervisor meetings ask of you. Phronesis training and GP training are aligned.
Turner describes the loss of informal professional spaces — senior staffrooms, communal lunch tables — where juniors used to absorb the wisdom of experienced colleagues without formal teaching. In those spaces, clinicians traded experience, processed difficult cases, and transmitted tacit knowledge. HDR, tutorial groups, and Balint sessions are now the structured replacements for what once happened naturally. Use them as if they are irreplaceable — because they are.
Prof Joanne Reeve describes her own practice: she estimates that only about half the patients she sees have an illness that can be fully addressed by a disease protocol. For the other half — the medically unexplained symptoms, the existential distress, the patients holding multiple long-term conditions alongside complex lives — she is working with pure phronesis: knowledge work without a map. This is everyday general practice. Being prepared for it is the purpose of wisdom training.
📅 Your Wisdom Timeline — Where You Are and Where You're Going
Phronesis is not a destination. It is a direction of travel. Here is how wisdom typically develops across a GP training career — and what characterises each stage:
📚 Go Further — Resources for Developing Your Phronesis
Prof Joanne Reeve's free national programme on knowledge work skills for GPs. Includes the WISDOM course (FutureLearn), GEM library, and newsletters.
Resources on Balint groups including finding a group near you, podcasts, and guidance for GP trainers wanting to run groups within their VTS scheme.
Her 2022 PhD (Enacted Phronesis in GPs), her 2025 JAMP paper on 12 themes in phronesis literature, and her chapter "The Doctor with the Golden Heart."
An accessible, readable interview with Jameel and Turner about their work. Excellent for sharing with trainees and trainers who want a non-academic introduction.
⚗️ From Algorithms to Artistry
One estimate in the clinical ethics literature suggests that doctors now face more than 7,000 deontological guidelines — with more added every year. No single clinician can hold all of them. No guideline can anticipate every patient. And yet, every consultation requires a decision.
This is the central tension of modern medicine. The infrastructure of healthcare has become algorithmic — guidelines, protocols, pathways, audit criteria, QOF targets. But the human encounter at the heart of it remains irreducibly complex. You cannot code for a patient who is frightened, a family in conflict, a diagnosis that fits no algorithm, or a clinical decision that must balance six competing goods.
Jameel and colleagues have framed this challenge precisely: medicine must move from algorithms to artistry. The algorithm is necessary. The artistry — phronesis — is what makes it wise. One without the other produces either chaos or harm.
Ubuntu — "I Am Because We Are"
Jameel draws on Ubuntu — the Southern African philosophy of shared humanity — as a lens for medical wisdom. In medicine, Ubuntu means recognising that healing is not a solo act. The doctor brings knowledge; the patient brings their life. Neither can navigate the consultation wisely without the other.
Ubuntu also underpins the communal dimension of wisdom development: we grow wiser through connection — with patients, with colleagues, with our community of practice. The isolated clinician does not develop phronesis as richly as the one embedded in relationship.
From Jameel's appearance on Flourishing Leaders Podcast, Episode 34: "Human-Centered Practice: Kindness, Burnout, and the Moral Core of Medicine"
Burnout, Moral Injury & Wisdom
Jameel is explicit: burnout in medicine is not simply about workload. It is often moral injury — the distress of being unable to do what you know is right, of being constrained from acting wisely by a system that does not value wisdom.
Her research suggests that developing phronesis protects against moral injury. Wise doctors are more resilient — not because they care less, but because they understand better what they are doing and why. Gratitude, boundaries, and a clear sense of purpose are not soft add-ons to clinical practice: they are constitutive of wisdom itself.
"We can get so tied up in micromanagement, bureaucracy, and audits that we don't have time to hold a patient's hand." — Jameel, BMA Doctor Magazine
🔄 How Wisdom Protects You — The Virtuous Cycle
🎙️ Where to Hear More — Podcasts, Talks & Video Resources
Dr Sabena Jameel: Human-Centered Practice: Kindness, Burnout, and the Moral Core of Medicine
Jameel speaks directly about Ubuntu, enacted wisdom, moral injury, gratitude, boundaries, and whether AI can ever be truly wise. Available on Spotify, Apple Podcasts, YouTube.
Flourishing Leaders Podcast (Dr Scott Parsons, host) — Episode 34
"The Malady of Medicine: Practical Wisdom as the Antidote"
Jameel's keynote address at the Association for the Study of Medical Education Annual Scientific Conference, Aberdeen, July 2022. Summary published in The Clinical Teacher, October 2022.
The Clinical Teacher, Vol 19, October 2022
"Can You Learn to Be Wise?"
An accessible, engaging interview with Dr Jameel and Dr Chris Turner. Describes how phronesis can be understood, taught, and applied in UK general practice. Excellent starting point for trainees new to the concept.
BMA Doctor Magazine (The Doctor), January 2022 — free to access at thedoctor.bma.org.uk
Sabena Jameel — Ethics and the Good Doctor: Character in the Professional Domain
Jameel's extended contribution to the literature on practical wisdom, character and medical professionalism. Directly relevant for GP trainees wishing to explore phronesis beyond this page.
Available from major booksellers
Jameel's podcast discussion raises a question that feels increasingly urgent: in an era of AI-assisted diagnosis, decision-support tools, and algorithmic medicine — can artificial intelligence ever be phronimos?
The tentative answer from the phronesis literature is: probably not. Here is why. Phronesis requires embodied experience — the kind of knowing that comes from being present in the room with a frightened patient, from the weight of moral responsibility, from having made mistakes and processed them. AI can recognise patterns. It cannot sit with uncertainty, carry moral responsibility, or make meaning. It can inform phronesis — but it cannot replace it.
This has a direct implication for GP trainees: as AI becomes more embedded in primary care, the specifically human qualities of wisdom — empathy, moral reasoning, contextual judgement, the ability to hold a patient's hand — become more, not less, important. Phronesis is the part of doctoring that AI cannot automate.
From the RCGP GP Registrar blog, trainees at all levels reflect on what wisdom actually looks and feels like in training:
"Always stay hungry for knowledge. Ask questions and make the most of your time in your learning environment — every encounter will teach you something new. Most experienced GPs are still learning new things even after more than 25 years in the profession."
— RCGP GP Registrar Co-Chair (ST4 academic trainee, Exeter)
"If you're struggling with anything, speak up. The more you open up to your supervisors and colleagues, the more comfortable you will feel with your interactions — and the more genuine you will come across."
— RCGP GP Registrar Co-Chair (INT GP trainee, Devon)
"Despite challenge, the exemplars [wise GPs] remain positive. Many could relate to the stories of the exemplars and felt inspired. The case studies showed the common themes in their lives that had led to wisdom. It is interesting to note that wisdom seems to be correlated with job satisfaction — is there causality?"
— GP trainee feedback after Dr Sabena Jameel's teaching session at a VTS event
The Practical Wisdom Network
Co-founded by Prof. Sabena Jameel and colleagues, the Practical Wisdom Network was formally launched at St Catherine's College, Oxford in April 2025 under the auspices of the Collaborating Centre for Values-Based Practice. It is an interdisciplinary community of practitioners, educators and researchers united by a commitment to developing phronesis in health and social care professions.
For UK GP trainees, this represents a growing community of practice specifically focused on the dimension of wisdom that medical training consistently underserves. Following the network's work is a way of staying connected to the evolving literature on practical wisdom in medicine.
valuesbasedpractice.org/network/practical-wisdom-network/
Safety-Netting — The Decision-Making Safety Net
🔴 Jess's Rule — "Three Strikes and We Rethink"
A major UK patient safety initiative introduced by NHS England and the RCGP in September 2025. If you trained before this was implemented — or haven't heard of it yet — this is essential reading.
Jessica Brady died at the age of 27 in December 2020. In the five months before her death, she attended her GP practice more than 20 times with symptoms including weight loss, night sweats, fatigue, a persistent cough, and swollen lymph nodes. Her symptoms were attributed to long COVID; she was reassured she was "too young for cancer." Her family arranged a private referral. She was found to have stage IV adenocarcinoma and died three weeks after diagnosis. She was 27. Jess's Rule is her legacy.
Developed in partnership with the Jessica Brady CEDAR Trust, NHS England, and the RCGP. Implemented nationally in GP practices in England from September 2025.
Jess's case is a textbook example of multiple biases operating simultaneously: ascertainment bias ("too young for cancer"), confirmation bias (seeking data confirming "long COVID"), anchoring bias (sticking to the first working diagnosis), and diagnosis momentum (each new clinician seeing the same label and perpetuating it). The rule exists specifically because these biases are common — not rare. They will happen to you. Jess's Rule is the structural safeguard.
- Continuity of care matters — a patient seeing multiple different GPs is high risk for missed diagnoses
- Remote consultations cannot replace a face-to-face assessment when a patient is not improving
- Young age does not protect against serious illness — do not let this be your anchoring cue
- Non-specific symptoms persisting without explanation deserve escalation, not reassurance
- Collaborative decision-making (including the patient's own advocacy) is built into the framework
Jess's Rule is now implemented guidance from NHS England and the RCGP. While it does not have statutory force, departing from it without clear documented reasoning is likely to support a clinical negligence claim if a patient has a poor outcome. Document your clinical reasoning explicitly when a patient presents repeatedly without a clear diagnosis.
Jess's Rule embeds exactly the kind of decision-making examiners look for: systematic re-evaluation, appropriate escalation, honest uncertainty management, and collaborative patient care. A candidate who recognises the pattern ("this patient has been seen three times with similar symptoms and no clear diagnosis...") and verbally acknowledges it, then applies structured thinking, is demonstrating exactly the GP reasoning the SCA rewards.
🛡️ Safety-Netting — The Complete Framework
Safety-netting is not just a phrase at the end of a consultation. It is a clinical decision-making tool, a patient empowerment mechanism, and a medico-legal safeguard — all at once.
Roger Neighbour, the GP educator whose "Inner Consultation" framework underpins UK GP training, originally defined safety-netting as answering three questions at the end of every consultation:
- "If I'm right, what do I expect to happen?" — defining the expected course of the illness you've diagnosed
- "How will I know if I am wrong?" — identifying the warning signals that would tell you your diagnosis is incorrect
- "What would I do then?" — the explicit plan for what happens if those signals appear
Neighbour R (2004) The Inner Consultation: How to Develop an Effective and Intuitive Consulting Style. Radcliffe Publishing, Oxford.
📊 What the Research Shows
A large-scale UK study (BJGP, 2025) of out-of-hours primary care records found that safety-netting advice was documented in approximately 78% of consultations. Reassuringly, this has been rising. However, there is one persistent gap that stands out:
Despite the overall improvement, only about 1 in 5 consultations recorded a specific timeframe for when the patient should return or seek further help. Telling a patient "come back if you're not better" without saying "in 5–7 days, specifically" is incomplete safety-netting. This is one of the most common and easily correctable gaps in GP practice.
✅ The Gold Standard Safety-Net Script
Based on the research and training guidance, a high-quality safety-net statement covers five elements in one clear statement:
"This is likely viral and should start improving in 5–7 days. If it doesn't improve by then — or if you develop a high fever above 39°C, difficulty breathing, or you feel significantly worse at any point — please contact us that day or call 111. Don't wait for your review appointment if those things happen."
✅ When to return (5–7 days) ✅ Specific symptoms (fever, breathlessness) ✅ What to do (call same day/111) ✅ Explicit permission to return sooner ✅ Distinguishes routine review from urgent escalation
"Come back if things don't improve" — three GPs used these words with Jess. None of them described what worsening would look like. None gave a timeframe. None connected the dots between visits. This is the real-world consequence of vague safety-netting. It is not a theoretical risk. In the SCA, it fails marks. In real practice, it fails patients.
Voices From the Trenches
🏥 The Single Most Important Mindset Shift in GP Training
If you've come from hospital medicine — and almost all GP trainees have — this is the insight that changes everything. It's the first thing experienced GP educators want you to understand, and the last thing most trainees fully grasp.
The RCGP curriculum describes this distinction precisely: "When solving problems, GPs have to tolerate uncertainty, explore probability and marginalise danger, whereas hospital specialists have to reduce uncertainty, explore possibility and marginalise error." This is not just exam theory — it describes a fundamentally different clinical job. If you're still approaching GP patients with a hospital mindset, this is the root cause of most of the mistakes trainees make in consultations and assessments.
UK GP trainers consistently report the same pattern: trainees from hospital backgrounds over-gather data, spend too long on history-taking, and then rush through the management plan — or skip it entirely. Research confirms the single most common cause of SCA failure is an inability to manage conditions according to current guidelines. Not poor communication. Not inadequate empathy. Insufficient clinical management knowledge — applied under time pressure.
A widely shared practical tip from trainees who've sat the SCA: aim for 6 minutes of data gathering, 6 minutes of management. The clinical management domain carries extra weighting in the mark scheme. Many trainees spend 9–10 minutes gathering data and have almost nothing left for management — this is a direct route to a borderline or failing score.
💬 What Trainees Wish They Had Known Earlier
These insights come from the collective experience of UK GP trainees preparing for — and sitting — the SCA and navigating real GP consultations. They are consistent, recurring themes — not individual opinions.
"I spent ages doing a beautiful history. I explored ICE brilliantly. Then I had two minutes left for management. I got the consultation badly wrong — not because I didn't know what to do, but because I ran out of time. The examiner never got to see my clinical management. Time management is as important as clinical knowledge."
"I kept hedging. 'We could do X, or maybe Y, or it depends...' Examiners don't want a wishy-washy candidate — they want to see that you can make a decision, own it, and explain it clearly. You're allowed to say 'I think X is the most appropriate approach here, because...' That confidence, backed by reasoning, is what scores marks."
"I used to panic when I didn't know what was going on. Then my trainer told me: just say it. 'I'm not completely certain yet — here's what I'm thinking, and here's what I'd like to do next.' That honesty, plus a plan, is exactly what an examiner wants to see. The worst thing you can do is pretend to be certain when you're not — and get it wrong."
"I wasted months trying to have 'perfect' consultations in practice. My trainer eventually said: the exam doesn't want perfect — it wants real. Just be yourself on a good day. That shift in mindset — from performing to being — changed everything. I passed on my next attempt."
"The three minutes before each SCA case are gold. Most trainees skim the information and walk in unprepared. Train yourself to use those minutes properly: note the patient's age and gender, identify the likely presentation category, think about one or two red flags you'll need to check, and consider which ICE elements are most likely to be relevant. Walking in with a plan — even a rough one — is transformative."
"I used to tell patients what I was going to do. Then I learned: involve them. 'We've got a couple of options — let me talk you through them and see what feels right for you.' That single change improved my 'Relating to Others' score dramatically. The examiner is watching whether the plan is patient-centred or doctor-imposed."
🎓 GP Educator Insights — From Those Who've Trained Thousands
These themes come from experienced GP trainers, TPDs, deanery educators, and published GP education research. They represent the distilled wisdom of a profession that has been teaching clinical decision-making for decades.
A qualitative study of UK GPs (BJGP, Goyder et al.) found that GPs typically "think in patterns rather than in diagnoses" — which works brilliantly for common presentations, but breaks down in two important ways:
When symptoms don't fit a known pattern, errors occur. One GP in the study described missing an abdominal aortic aneurysm because the patient had atypical leg pain that kept coming back: "I've seen an awful lot of sciatica... but this one had something extra that I just didn't pick up on."
The lesson: When a presentation keeps coming back, or feels slightly "off" even if you can't say why — that's your early warning System 1 signal. Act on it.
Multiple GPs reported that patients or carers expressing concern was an important signal — and that overriding it based on pattern recognition alone was a recurring source of near-misses. One GP changed their approach after a parent pushed back, applied analytical reasoning, and avoided a diagnostic error.
The lesson: When the patient or carer says "but something's not right" — slow down. That's data.
Experienced GPs in the study described a strategy of systematically considering all potentially serious conditions for any given presentation — not to investigate all of them, but to consciously rule out the ones that matter. This goes beyond relying solely on pattern recognition. The question is not "does this fit my known pattern?" but "have I consciously considered the serious diagnoses and actively excluded them?"
Dr Avril Danczak is a GP with a special interest in managing uncertainty in medicine, and co-author of an RCGP-published book on the topic. Her framework — discussed extensively in UK GP training circles — classifies uncertainty into distinct types and offers structured strategies for each.
Uncertainty is not one thing. Identifying which type of uncertainty you're dealing with points you to the right strategy. The GP who says "I'm not sure" without diagnosing why they're not sure is stuck. The GP who says "I'm not sure because the evidence is genuinely equivocal here" — or "I'm not sure because I need more history" — or "I'm not sure because this feels beyond my current confidence level" — has a path forward.
Resource: Mapping Uncertainty in Medicine: What To Do When You Don't Know What To Do — Danczak, Lea and Murphy (RCGP, 2016)
An important BJGP article (2015) drew attention to a bias in clinical settings that receives far less attention than anchoring or availability bias — groupthink: the tendency to conform to majority clinical opinion rather than applying independent analytical reasoning.
- Copying a hospital discharge's medication list onto the repeat prescription without independent review
- Trusting a TTO without checking it against the patient's existing medications
- Prescribing based on a nurse's request without performing your own clinical assessment
- Accepting a diagnosis that has been handed down through multiple clinicians without ever re-examining the evidence
- Doing a focused examination guided entirely by information from a previous clinician, rather than your own history
GPs often receive patients "pre-labelled" by previous clinicians. Groupthink is the mechanism by which bad diagnoses persist — and good ones get missed. The antidote is simple but requires deliberate effort: always form your own clinical impression first, before reading the referral letter, the previous notes, or the nurse's summary. Then use those resources to add information, not to replace your thinking.
One of the most effective self-directed learning strategies for GP trainees — endorsed by GP trainers and deaneries across the UK — is the PUNS/DENS approach:
Every time you see a patient where you weren't sure how to manage the problem — write it down. Straight after the consultation, read about the management. Write down what you learned. Discuss it with your trainer. This converts every moment of uncertainty from a source of anxiety into a targeted learning opportunity.
PUNS/DENS also generates the most authentic, relevant material for CbD discussions and FourteenFish portfolio entries — because it's rooted in your actual practice, not a theoretical curriculum. The RCGP SCA toolkit explicitly endorses this approach for filling management knowledge gaps.
🏛️ Deanery & VTS Collective Wisdom
These tips are drawn from preparation guidance shared by UK deaneries, VTS schemes, and the North West England SCA consultation toolkit — endorsed by the RCGP and used by trainees across the country.
⏱️ The Clock Is Your Friend (Not Your Enemy)
Trainees who fail to manage time in the SCA typically make the same error: they're so focused on gathering data that they lose track of where they are in the consultation. Bristol VTS trainees recommend actively monitoring the 6-minute mark — if you're still in pure history mode, it's time to begin transitioning to explanation and management.
In the 3 minutes before the case, glance at the BNF summary for the likely presentation (e.g. treatment options, drug interactions in pregnancy). Walking in with management options already in mind means you can spend more time listening and less time thinking under pressure.
🎭 "Be Awkward" in Practice
A recurring tip from Bristol VTS trainees and others: when rehearsing SCA cases with colleagues, deliberately play difficult patients — unreasonable, upset, avoidant, or asking for something inappropriate. The SCA specifically includes cases requiring negotiation, persuasion, and compromise. If you've only ever practised with cooperative simulated patients, a challenging actor in the exam will catch you completely off guard.
How you handle the moment when a patient says something unexpected, asks for something you can't provide, or becomes upset — that's when your decision-making and communication under pressure become visible. Train for these moments specifically.
🗺️ The North West SCA RAG Tool — A Practical Self-Assessment Framework
Developed by RCGP examiners in the North West (Dr Anne Hawkridge and Dr David Molyneux) and endorsed by the RCGP, the SCA consultation RAG tool allows trainees to self-assess consultations across all 29 competencies in the three marking domains. It's particularly useful for identifying consistently "red" areas that need focused attention before the exam.
From Dr Hawkridge (SCA examiner since 2007): "Clinical management domain marks receive extra weighting — this is where trainees can gain or lose the most marks. Know your management cold."
📖 Study Group Structure That Works
Across multiple VTS schemes, trainees consistently report that study groups of 3–5 people work best. The key is having an active observer — not just a passive timekeeper. The observer should watch for specific decision-making moments: when does the candidate commit to a diagnosis? When do they offer options? When do they safety-net? That level of observation develops clinical reasoning awareness faster than just practising.
- One person = doctor
- One person = simulated patient (and deliberately be demanding)
- One or more = observers, each watching a specific competency domain
🗂️ The Decision-Making Hierarchy in the SCA Consultation
The North West consultation toolkit describes a clear hierarchy for how good clinical management decisions are made in the SCA — adapted here as a visual framework.
Adapted from the North West England SCA Consultation Toolkit (Hawkridge & Molyneux, RCGP-endorsed). Each level is a decision — skip any one of them and the consultation, or the marks, suffer.
🩺 Experience-Based Clinical Decision Tips
These insights are grounded in UK primary care research and the collective experience of GP educators — translated into the language of everyday clinical practice.
Experienced GPs name one situation where their comfort becomes dangerous: when they've seen a patient several times for the same problem and it "still hasn't changed." Familiarity breeds comfort — and comfort breeds framing effect bias. The patient who has been seen six times for "anxiety" may be having their seventh consultation for the same reason because the diagnosis is wrong. Re-examine established presentations with fresh eyes at every significant review.
UK GPs in qualitative research consistently describe one of the most important decision-making habits: if something feels wrong, investigate that feeling — don't dismiss it. Several diagnostic errors in the literature involved GPs who had a nagging sense that something was off, but overrode it with logic: "the tests are normal, the examination is fine, it must be fine." The gut feeling — your System 1 raising an alert — deserves a System 2 response. Ask: "What is making me uneasy, and what would I need to do to exclude it?"
Research with UK GPs in high-risk primary care settings identified one of the strongest protective behaviours: when uncertain and it matters, always take advice from a colleague. "I usually discuss it with a sensible colleague and just say 'look, neither of us think this should be acted upon'" — this shared risk is not weakness; it is sound medico-legal and clinical practice. Document the conversation.
Clinical decision-making research consistently shows that cognitive bias is amplified when you are Hungry, Angry, Late, or Tired. Your System 1 becomes dominant, System 2 slows down, and errors become more likely. At the end of a long surgery, with the last patient who has a complex presentation — that's the highest-risk moment of your day. Slow down deliberately at HALT moments.
Borrowed from decision science, the pre-mortem is a simple debiasing tool. Before closing a consultation with a diagnosis you're reasonably confident in, ask: "Imagine I'm wrong about this. What would be the most plausible alternative? What would I need to see to make that alternative more likely?" This interrupts confirmation bias and premature closure without requiring a complete re-assessment. It costs 20 seconds and has caught multiple near-misses in documented GP practice.
Research on clinical expertise makes a sobering point: experience without reflection does not produce expertise — it produces entrenched habits, including bad ones. A GP who has seen 20,000 patients without ever reflecting on their decisions may have 20,000 opportunities to reinforce their biases. The GP who reflects on 2,000 patients builds genuine expertise. Reflection — on cases, errors, near-misses, and unexpected outcomes — is how expertise is constructed.
💎 The Hidden Gem in the RCGP Curriculum
This is one of the most important things the RCGP says about GP decision-making — and one of the least-read pieces of curriculum guidance. It reframes everything.
The RCGP curriculum explicitly states that GPs should "use an understanding of probability, based on the prevalence, incidence, natural history and time course of illness, to aid decision-making." This means every clinical decision in GP should be anchored in epidemiology — not just pattern recognition. How common is this diagnosis in a patient like this? What's the base rate? What does the natural history tell me about how urgently I need to act?
This is the probabilistic thinking framework. It's what separates a good GP from a diagnostically anxious one — and it's what AKT questions test when they show you a presentation and ask you to choose the most likely diagnosis from a list of five plausible options.
🧮 A Worked Example of Probabilistic Thinking
Scenario: A 45-year-old woman presents with a 6-week history of fatigue, mild low mood, and weight gain of 3kg. No alarm features. Not pregnant.
System 1 response: "Sounds like depression — screen and start antidepressant discussion."
Probabilistic System 2 override: In this demographic, what are the base rates? Hypothyroidism is common in women of this age, often presents exactly this way. TFTs are simple, cheap, and will substantially change management if positive. Fatigue + low mood + weight gain in a middle-aged woman = hypothyroidism until excluded. The pre-test probability is high enough to investigate before labelling.
Decision: Request TFTs, screen for depression (PHQ-9), review when results available. This is probabilistic thinking applied correctly.
📊 Probability in AKT Questions
AKT "most likely diagnosis" questions are almost always answered correctly by asking: "What is the most probable diagnosis in a patient like this, based on demographics, presentation pattern, and primary care epidemiology?"
Common errors:
- Choosing a diagnosis because it's the most serious (availability + overconfidence bias)
- Choosing a diagnosis because you just revised it (recency availability bias)
- Choosing based on one prominent symptom without considering the full picture (anchoring)
Correct strategy: What is statistically most likely in this patient, in a GP setting, given everything described?
🕳️ Common Pitfalls & Trainee Traps
- Running on System 1 at the end of a long surgery when the last patient needs System 2
- Not completing a comprehensive history because the diagnosis seemed obvious from the door
- Accepting an inherited diagnosis without re-examining the evidence yourself
- Safety-netting vaguely ("come back if you're not better") instead of specifically ("if X or Y happens, do Z")
- Ordering all-possible investigations instead of targeted ones based on hypothesis
- Prescribing to avoid conflict, not because it's the right decision
- Having a beautiful consultation but never explicitly naming the diagnosis or reasoning
- Not asking about red flags — and hoping the examiner assumes you considered them
- Exploring ICE but then ignoring it when making the management plan
- Giving the patient a plan rather than agreeing one with them
- Rushing past uncertainty by offering a confident-sounding diagnosis you're not sure about
- Closing without a safety net — the consultation isn't finished until this is done
- Choosing the most recently revised drug (availability bias)
- Choosing the most impressive-sounding investigation (complexity bias)
- Forgetting that GP management is more conservative than hospital management
- Ignoring the contextual clues in the stem (age, sex, comorbidity, specific finding)
- Treating extended matching vignettes as if they are independent SBA questions
- Not reading the unit labels in "enter your answer" questions
- Confusing pattern familiarity for pattern correctness — the patient who looks like your last diagnosis might not be
- Mistaking patient assertiveness for clinical evidence ("but they're sure it's X")
- Closing the diagnostic space too early, especially in undifferentiated presentations
- Not pausing to consider — "what am I missing here?"
💎 Insider Pearls — What Trainees Wish They'd Known Earlier
The trainee who says "I just knew" after a correct diagnosis has a well-trained System 1. The trainee who says "I wasn't sure but I worked through it" has a well-used System 2. The consultant who says "I knew, but I always check" is using both. That's the goal.
You can't stop cognitive biases from occurring — they are hardwired. What you can do is recognise when conditions favour bias (you're tired, it's your 30th patient, you've just seen three similar presentations) and slow down deliberately at those moments.
"I don't know" said confidently, honestly, and followed by "here's what I'm going to do to find out" — is one of the most trusted things a GP can say to a patient. Patients don't expect certainty. They expect honesty and a plan.
The amber patient is your most important patient. Red and green patients mostly manage themselves. The amber patient requires your best judgment, your best safety-netting, and your most careful follow-up. This is where GP training actually lives.
The history is still king. When trainees miss diagnoses, it is almost always because they stopped asking questions too early — not because they lacked knowledge. The knowledge is usually there. The questions weren't.
Examiners want to pass you. They are watching for evidence that you think like a GP. Show your reasoning, name your uncertainties, involve the patient, safety-net specifically. These are signals that say "I will be a safe GP." That's all examiners want to hear.
🧩 Memory Aids & Cheat Sheets
🧠 The BIASED Mnemonic — Cognitive Biases to Watch For
- B — Blinding anchoring (fixating on first impression)
- I — Information availability (recent cases colour your thinking)
- A — Ascertainment (stereotyping by demographics)
- S — Search cutoff premature (closing before complete)
- E — Estimation overconfidence (you're not as certain as you feel)
- D — Diagnosis momentum (inherited label not re-examined)
🗂️ RAPRIOP — The Seven Options
- R — Reassurance and Explanation
- A — Advice and Counselling
- P — Prescribing
- R — Referral
- I — Investigations
- O — Observation and Follow-up
- P — Prevention and Health Promotion
Run through all 7 before defaulting to prescription or referral.
🚦 The RAG Decision Rule
- 🔴 Red — Urgent action needed. Decision = admit/emergency refer
- 🟡 Amber — Treat + watch + robust safety-net. Decision = complex
- 🟢 Green — Reassure + safety-net. Decision = watchful waiting
- 🔑 Key rule: Always ask "Has the colour changed?" at review
- ⚠️ Children can switch colours rapidly — safety-net explicitly
🌫️ Uncertainty Management: 5 Steps
- 1. Name it — acknowledge uncertainty to yourself and the patient
- 2. Explore it — gather more information (history/exam/tests)
- 3. Share it — communicate honestly with the patient
- 4. Contain it — create a robust safety net
- 5. Tolerate it — use time as a diagnostic tool when appropriate
Before closing every consultation with an uncertain diagnosis, take 10 seconds and ask yourself three questions: "What am I assuming here? What's the worst this could be? What would make me wrong?" This one habit, applied consistently, could prevent the most common diagnostic errors in GP. It costs 10 seconds. It could save a patient's life.
💡 Bradford VTS Curriculum Insight — Uncertainty Is a Portfolio Strength
One of the most underappreciated insights in GP training: the consultations where you were uncertain are not weaknesses to hide — they are some of the most valuable learning entries in your FourteenFish portfolio.
Good GP practice routinely involves diagnostic uncertainty, risk calibration, and honest conversations with patients about what is and isn't known. Showing that you can hold uncertainty safely — acknowledging it, communicating it clearly to the patient, managing it with appropriate safety-netting and follow-up — is a direct demonstration of the capabilities the curriculum is testing.
Include uncertainty in your portfolio logs. Say it out loud in your consultations. It is not a weakness to be hidden.
📋 What to Write in Your CbD Log
From the RCGP guidance on CbD selection: "Cases where there was an element of uncertainty or where a conflict in decision-making are particularly good ones to choose."
An entry like this scores well in multiple capability areas:
"I was uncertain about X, because of [specific clinical features]. I considered the differentials of A, B, and C. I decided to [action] because [reasoning]. I communicated my uncertainty to the patient by explaining [what I said]. I safety-netted by [specific instructions]. I arranged follow-up for [timeframe and specific reason]."
This covers: data gathering, clinical management, communication — and documents Jess's Rule thinking implicitly.
🔄 The Mindset Shift That Changes Everything
The RCGP curriculum's core message on this topic:
"The single most important mindset shift in GP training is moving from being diagnosis-focused (the hospital model) to being risk-focused, continuity-aware, and context-sensitive (the GP model)."
A GP who always sounds completely certain is a GP who isn't being honest. Acknowledging uncertainty — and then describing how you managed it safely — is a mark of professional maturity. ARCP panels are looking for this.
👨🏫 Trainer & Educator Pearls
How to Teach Decision-Making
Decision-making cannot be taught by lecture alone — it must be practised, reflected on, and refined through supervised clinical experience. The trainer's role is to create the conditions for this to happen.
- Bias case-study game — present a case and ask the trainee to name which bias could derail it at each stage
- RAG rating — take 5 real recent cases and have the trainee classify them as red/amber/green and justify their reasoning
- RAPRIOP walk-through — take any case and work through all 7 management options explicitly, even if only to rule them out
- The "what else could it be?" challenge — after a diagnosis is made, generate 3 alternative diagnoses and say what would need to be true for each to be correct
- Uncertainty mapping — ask: "In this case, what don't you know? What does that uncertainty mean for the patient? How are you managing it?"
- System 1 vs 2 debrief — after a difficult case, ask: "Which system were you using? Was it the right one? What made you switch?"
- "Walk me through how you arrived at that diagnosis — what were you thinking at each stage?"
- "Was there a moment when you could have gone in a different direction? What made you choose this one?"
- "What would you have done if your initial impression had been wrong?"
- "How certain were you? What would have made you more certain?"
- "What did you observe about your own thinking in that consultation?"
- "If this patient came back worse tomorrow, would you be surprised? Why / why not?"
Common Learner Blind Spots on This Topic
- Trainees often know about System 1 and 2 intellectually but don't apply the framework to their own clinical decisions
- Many trainees avoid acknowledging uncertainty with patients, fearing it will undermine confidence — help them reframe uncertainty as honesty and professionalism
- RAPRIOP is underused — most trainees default to prescribe or refer. Explore why, and practise working through all 7 options
- The concept of phronesis is often new to trainees — but they are already demonstrating it in many consultations; naming it helps them recognise and develop it
- Cognitive biases feel abstract until they see a case where one clearly contributed to a mistake — use real (anonymised) examples wherever possible
🎙️ UK GP Training Podcasts & Video Resources — Decision-Making & Uncertainty
These are verified UK GP training resources — produced by UK GP educators, examiners, and primary care academics. They are not brand or commercial channels.
🎙️ Primary Care Knowledge Boost (PCKB) — Episode: Dealing with Uncertainty in General Practice
With Dr Avril Danczak — GP, primary care educator, co-author of the RCGP-published Mapping Uncertainty in Medicine (2016).
- Uncertainty in GP is not failure — it's the natural condition of first-contact, undifferentiated medicine
- The key skill is not eliminating uncertainty but navigating it in partnership with the patient
- Classifying what type of uncertainty you're dealing with (scientific / practical / personal / ethical) points you to the right strategy
- "What do I do when I don't know what to do?" is one of the most important questions a GP trainee can sit with and learn to answer
- The language you use with a patient about uncertainty matters enormously — it can either increase their anxiety or help them feel held and safe
PCKB Podcast, 2020 (replay 2022). Available at pckb.org. Endorsed by GP Excellence Greater Manchester.
🎙️ PCKB — Episode: Tips to Pass the SCA for GP Trainees
With Dr Anne Hawkridge FRCGP — MRCGP examiner since 2007, GP trainer Bolton, co-author of the RCGP-endorsed North West Consultation Toolkit.
- The SCA tests GP consulting skills — which means it is testing decision-making in real time
- The most common reason candidates underperform is leaving insufficient time for clinical management
- Good exam technique isn't about memorising scripts — it's about internalising good GP consulting habits so they're automatic
- The three-part consultation template: data gathering → clinical management → relating to others; all three run simultaneously but clinical management has extra weighting
- Observe your trainer: watch how they handle uncertainty, how they make decisions, how they explain their reasoning to patients. That's your SCA preparation happening in real time
PCKB Podcast, 2024. Available at pckb.org. Also featured at gmpcb.org.uk.
📺 BJGP Life Podcast — Episode 196: Safety-Netting in General Practice
Based on landmark BJGP research (2025) — a large-scale longitudinal analysis of safety-netting documentation in UK OOH primary care.
- Safety-netting is improving in UK primary care — but timeframe specificity remains the persistent gap
- Face-to-face consultations produce better safety-netting documentation than telephone encounters
- GPs need to move from verbal to written safety-netting for complex or high-risk patients — verbal-only has significant continuity limitations
- Safety-netting is part of the decision-making process, not an add-on — it is how you manage the uncertainty you're leaving the patient with
BJGP Life Podcast, Episode 196, February 2025. Available at bjgplife.com.
📺 RCGP YouTube & Dr Matthew Smith (YouTube) — SCA Consultation Skills
The RCGP has an official YouTube channel (@RCGPVideos) with curriculum-related content including video examples for SCA preparation. Dr Matthew Smith has a free YouTube series on SCA consultation skills specifically recommended by Bristol VTS.
- Making your clinical reasoning visible and audible — narrating your thinking as you work through a case
- Transitioning from data gathering to management with a clear pivot phrase: "Based on what you've told me, here's what I'm thinking..."
- The difference between a "doctor-led" consultation and a "patient-centred" one — and how to tell which you're doing in real time
- How a good consultation looks from the outside — watching yourself on video is one of the most powerful learning experiences in GP training
Search YouTube: @RCGPVideos (RCGP official) and "Dr Matthew Smith SCA GP" for the trainee-oriented series. Recommended by Bristol VTS, 2024.
📚 InnovAiT — RCGP Journal for GP Trainees
InnovAiT is the RCGP's journal written specifically for GP trainees and early-career GPs. The March 2026 issue featured a landmark piece revisiting Jess's Rule as a decision-making and safety-netting framework, alongside articles on diagnostic uncertainty and ethical complexity in everyday GP practice. The "Crammer's Corner" section provides AKT practice questions monthly.
❓ Frequently Asked Questions
Decision-Making in the Exams
🔥 AKT: Decision-Making in the Applied Knowledge Test
The AKT is not a memory test. It is a decision-making test disguised as a memory test. Every question asks: "What would a competent GP do here?"
🎯 The Core AKT Insight
The AKT mirrors the decisions you make in real GP. That means all the decision-making skills you use in actual consultations apply here. The distractors are exactly the choices a biased or under-thinking GP would make. Know the biases. Beat the distractors.
🧠 Decision-Making Skills the AKT Tests (That Real GPs Use Daily)
- Pattern recognition — "does this fit a known clinical picture?"
- Probabilistic thinking — "what is the most likely diagnosis given the context?"
- First-line vs second-line decision-making — "what does NICE say first?"
- Threshold decisions — "when do I investigate, treat, refer, admit?"
- Prioritisation — "of these options, which is most important right now?"
- Safety — "which choice avoids harm to the patient?"
- Context sensitivity — "how does age / sex / comorbidity change the answer?"
- Eliminating cognitive bias — "am I choosing this because it's right, or because it's familiar?"
📋 Single Best Answer (SBA) Questions — Decision-Making Framework
SBA questions test your ability to make a single, optimal decision from a list of plausible options. Every option is designed to look at least partially reasonable — because in real GP, most decisions are genuinely between plausible alternatives.
- Read the question stem carefully — identify the single most important piece of clinical information (often hiding in plain sight: age, sex, duration, a specific symptom pattern)
- Generate your own answer first — before reading the options. What would you actually do in real GP? Then look for your answer in the options
- Apply the NICE hierarchy — the correct answer is almost always the most evidence-based, guideline-aligned, first-line option. Flashy or complex options are usually wrong
- Eliminate obvious distractors — remove anything that is clearly too aggressive, too passive, or not primary care-appropriate
- Choose the most conservative appropriate option — AKT tests that you don't over-investigate or over-treat. When in doubt, the "watchful waiting + safety-net" answer is often correct
- Check for cognitive biases — "Am I choosing this because I just revised it? Because it's familiar? Because it sounds impressive?" That is availability or overconfidence bias
- Choosing hospital management in a GP scenario (wrong context)
- Choosing investigation before completing the clinical assessment
- Choosing second-line treatment before first-line has been tried
- Choosing the most recent drug you revised (availability bias)
- Choosing admission when watchful waiting is appropriate
- Ignoring a critical contextual clue (the patient is pregnant / elderly / has a specific allergy)
- Choosing based on hospital-medicine logic rather than GP decision-making
📋 Extended Matching Questions (EMQ) — Decision-Making Framework
EMQs present a list of options and multiple clinical vignettes, each requiring you to match the best option. The challenge is that the same option may be correct for multiple vignettes — or never correct at all. This tests your ability to discriminate between similar clinical scenarios.
- Read ALL the options first — get an overview of the decision space before reading individual scenarios
- For each scenario, identify the key discriminating feature — the clinical detail that makes this scenario different from similar ones (duration, severity, specific finding, contraindication)
- Think about what makes each option unique — when would option A be correct vs option B? Build a decision tree in your head
- Avoid anchoring on the first scenario answer — each scenario is independent; don't let your previous answer bias you
- If stuck between two options — ask "which is more GP-appropriate?" and "which is the more conservative, appropriate first step?"
- Check for deliberate distractors — options that are nearly identical to the correct answer except for one word (dosage, route, frequency, drug class)
- Age — same symptom in a 20-year-old vs 70-year-old may have entirely different correct answers
- Duration — "has lasted 2 weeks" vs "has lasted 6 months" radically changes management
- Severity — mild/moderate/severe grades often determine which treatment tier is appropriate
- Contraindications — often hidden in the stem to make Option A wrong and Option B right
- Special populations — pregnancy, breastfeeding, renal impairment, elderly — change the answer
✏️ Enter Your Answer / Free-Text Questions — Decision-Making Framework
These questions ask you to supply a specific value, threshold, name, or calculation. There is no option list to help you. They are the most demanding question type because there are no distractors — it's just you and the correct answer.
- Identify exactly what is being asked — a number? A drug name? A threshold? A score? Read twice before answering
- Think about the decision context — what clinical decision would a GP face here? The answer is the number or fact that informs that decision
- Recall the relevant guideline threshold — these questions love NICE thresholds: target HbA1c, BP targets, CHA₂DS₂-VASc scores, FRAX thresholds, CKD staging, weight thresholds for medication
- Don't second-guess a clear answer — if you know the number, write it. Overthinking "are they asking this?" wastes time
- If uncertain, reason from first principles — what would be physiologically or clinically logical?
- Check units if relevant — mmol/L vs mg/dL, mmHg vs kPa — missing the unit can mean a wrong answer
- Blood pressure targets: general (<140/90), diabetic (<130/80), over 80 years (<150/90)
- HbA1c: diagnosis ≥48 mmol/mol; type 2 DM management targets (48 or 53 per NICE)
- CKD staging: eGFR thresholds for G1–G5
- Anticoagulation in AF: CHA₂DS₂-VASc score ≥2 (men) or ≥3 (women) → offer anticoagulation
- FRAX: 10-year fracture probability thresholds for bisphosphonate prescribing
- Weight loss cut-offs for investigations (e.g. 5–10% unintentional loss in 3–6 months)
- Folate for pregnancy: 400 mcg daily (5 mg if high-risk)
- QTc thresholds for drug safety decisions
The single most important AKT insight: the exam is trying to predict how good a GP you'll be. Every question is asking "would this be a safe, effective, evidence-based GP decision?" Think less about "what's the right answer?" and more about "what would the patient need from me here, according to current UK guidance?" That framing wins marks.
🎯 SCA: Decision-Making in the Simulated Consultation Assessment
The SCA is 23 filmed consultations in which you are assessed on your clinical reasoning, communication, and decision-making. Examiners are not watching what you know — they're watching how you think.
🎯 The Core SCA Insight
The SCA mirrors every real consultation you've ever had. Every decision-making skill you use in actual GP applies here — and examiners can tell the difference between candidates who are performing a "consultation script" and candidates who are genuinely thinking and responding to what's happening in front of them. Be the latter.
🧠 Decision-Making Skills the SCA Tests (That Real GPs Use Daily)
- Hypothesis generation — forming and testing diagnoses during the consultation
- Prioritisation — identifying what matters most in this specific encounter
- Context sensitivity — adjusting your approach based on patient cues and background
- ICE integration — using the patient's ideas, concerns and expectations to inform your decisions
- Shared decision-making — involving the patient meaningfully in management choices
- Safety awareness — recognising red flags and acting on them explicitly
- Toleration of uncertainty — managing not-knowing honestly and safely
- Appropriate use of RAPRIOP — choosing the right management option for this patient
- Phronesis — knowing when the guideline doesn't quite fit this individual
🗺️ The SCA Consultation Navigation Framework
Use this as a mental GPS for every SCA consultation. You don't have to follow it rigidly — but knowing where you are at any moment means you can recover when things go off course.
As you gather history, you are simultaneously running a mental diagnostic process. Each answer from the patient either confirms or refutes your working hypotheses. Skilled candidates are visible to examiners in this phase — they ask purposeful questions, not a scripted list. They follow the patient's agenda, not their own checklist.
When you explain your thinking to the patient, you are also demonstrating it to the examiner. Saying "from what you've told me and what I've found, I think this is most likely..." shows clinical reasoning in action. This is one of the most scored moments in the SCA — don't rush through it.
🎯 SCA Decision-Making: Tips & Tricks
- Deciding on management before exploring ICE — and getting it wrong because you missed the real concern
- Arriving at a diagnosis without explaining your reasoning to the patient (invisible decision-making)
- Choosing management based on the textbook rather than this specific patient's circumstances
- Forgetting to explicitly ask about and document red flags
- Giving a plan without involving the patient in the decision
- Over-investigating in the consultation (anchoring to hospital medicine habits)
- Under-safety-netting — leaving the patient without clear guidance on when to return
- Always name what you're thinking: "I'm wondering whether this could be..." — shows active diagnostic reasoning
- Explicitly check red flags by name — don't assume the examiner knows you thought about them
- When uncertain, say so: "I'd like to find out more before we decide..." — this scores marks for honest uncertainty management
- Offer a choice of options, not a single plan: "We have a couple of approaches here..." — this is shared decision-making
- Summarise your reasoning at the end: "So my thinking is... and the plan we've agreed is..." — consolidates the decision for examiner and patient
- Use RAPRIOP mentally — have you considered all 7 options before defaulting to prescribing?
- "I want to be honest with you — I'm not completely certain yet, and here's what I'd like to do to find out..."
- "There are a few possibilities here. Let me talk you through what I'm thinking..."
- "I'd like to get a test back before we commit to a plan — is that okay with you?"
- "Sometimes in medicine we need a little more time for the picture to become clearer — here's what to watch for..."
Honest uncertainty management scores marks. Pretending to be more certain than you are — and getting it wrong — does not.
📋 SCA Decision-Making: What Actually Happens in the Room
Compiled from trainee accounts of sitting the SCA, and from SCA examiners describing what they observe. These are the moments that determine the outcome.
| Moment in Consultation | What Examiners See in Failing Candidates | What Examiners See in Passing Candidates |
|---|---|---|
| Making a diagnosis | Never names a diagnosis — keeps gathering data indefinitely, or gives a vague "it could be several things" | Names a working diagnosis explicitly: "From what you've described, I think this is most likely X" |
| Facing uncertainty | Pretends to be certain (and often wrong), or becomes paralysed and doesn't make a plan | Acknowledges uncertainty honestly and offers a structured plan for resolving it |
| ICE in management | Explores ICE in the history, then ignores it completely when making the management plan | Explicitly incorporates the patient's concern into the plan: "I know you were worried about X — I want to address that directly..." |
| Referral decisions | Either over-refers (reflects anxiety, not judgment) or under-refers (to avoid conflict) | Explains the referral decision with clinical reasoning: "The reason I'd like to refer you is because..." |
| Safety-netting | Vague: "Come back if things don't improve." Never defines what to watch for, or how urgently. | Specific: "If you notice X, Y, or Z — especially Z — please contact us that day, don't wait for the review" |
| Handling a difficult moment | Loses control of the consultation, becomes defensive, or rushes to resolve the discomfort | Acknowledges the difficulty, pauses, responds with empathy — then guides the consultation forward |
| Shared decision-making | Tells the patient the plan; may ask "Is that OK?" as a formality | Genuinely presents options, invites preference, adjusts the plan based on patient response |
Across deaneries, VTS schemes, and SCA preparation courses, one piece of examiner feedback stands out as the most repeated: "The candidate didn't take ownership of the management plan." Dithering, endless presenting of options without committing, and leaving the decision entirely to the patient are all forms of not owning the decision. Patients need your clinical judgment — not just a menu. Offer options, involve them, but commit to a recommendation based on the clinical picture.
🗣️ Consultation Phrases for Decision-Making
Ready-to-use phrases for the moments in consultations where decision-making becomes visible. Read these once; use them tomorrow.
🏁 Final Take-Home Points
- Every consultation is a decision. Understand the two systems you use to make those decisions — and flex between them.
- Amber patients are where the real work happens. Red and green manage themselves; amber is where GP skill lives.
- Cognitive biases are not a character flaw — they're human brain hardware. Name them. Counter them. Build habits that compensate for them.
- A thorough history is still the most powerful diagnostic tool in your kit. Don't shortcut it.
- Uncertainty is not failure — it's honesty in the face of complexity. Manage it actively and communicate it clearly.
- RAPRIOP: before you prescribe or refer, run through all seven options. Most of the time there's a better one.
- Phronesis — practical wisdom — is what separates a competent doctor from an excellent one. It comes through reflection, not revision.
- Safety-netting is never optional. It is how you manage the patient you can't be certain about, and it is how you protect them when you're wrong.
- In the AKT: think like a GP, not an exam machine. The correct answer is always the safest, most guideline-aligned, most GP-appropriate one.
- In the SCA: show your reasoning. Say it out loud. Examiners can't mark what they can't see — make your thinking visible.