The difference between a strong reflection and a weak one is not length — it is depth. Panels want evidence that you learned something, not a narrative of what happened.
What Panels Look For
Evidence of genuine learning — not description of events. Appropriate capability linking — with descriptor-based justification. Identification of learning needs — what you plan to learn next as a result. Progression over time — early entries show awareness, later entries show independent competence.
The Gibbs Cycle Applied Practically
Description (brief — 3-4 sentences maximum). What happened? Who was involved? What was the clinical context?
Feelings (honest). What were you thinking during the encounter? Were you confident, uncertain, anxious?
Evaluation. What went well? What could have been better?
Analysis (bulk of the entry). Why did it go that way? What knowledge or skills were relevant? What would the guideline say?
Conclusion. What would you do differently next time?
Action plan. What specific learning will you undertake? (This becomes a PDP entry or DEN.)
Strong vs Weak Example
Weak: "I saw a patient with chest pain. I took a history and examined them. I referred them to A&E. I learned that chest pain can be serious." (Descriptive. No analysis. No specific learning.)
Strong: "A 55-year-old presented with exertional chest pain and breathlessness. I was uncertain whether to refer urgently or arrange outpatient investigation. I calculated a QRISK3 of 18% but was unsure about the threshold for urgent referral. After discussing with my supervisor, I learned that the NICE NG95 pathway recommends urgent referral for suspected ACS based on clinical presentation rather than risk score alone. I plan to complete the CKS chest pain module and do 20 iatroX cardiovascular questions this week to consolidate." (Reflective. Specific. Actionable.)
AI Scaffolding
Learner+: Structured prompts tailored to clinical encounters, FourteenFish integration. Purpose-built.
iatroX CPD module: Maps learning to professional domains, scaffolds reflection structure.
ChatGPT (with caveats): Can help improve clarity — but never enter patient information. Output tends to be generic. Use for structure, not content.
The Authenticity Line
AI should scaffold your thinking, not generate it. Panels can and do explore entries. Red flags: formulaic language, identical structures across all entries, generic learning points that could apply to any trainee.
Where iatroX Fits
iatroX's CPD module maps learning activities to professional domains — structuring the reflection process so you spend time thinking, not formatting.