How Junior Doctors Think: A Guide for Reflective Practice


In the UK, junior doctors will rotate through emergency medicine in their second year post-graduation (Foundation Year 2). They’re granted autonomy to make independent decisions and ‘own’ patients for the first time.

Elsewhere in the hospital, a junior’s role is largely secretarial, and generally within the confines of ‘normal working hours’. In the ED, the hours are brutal, the pace is relentless, and the sudden spike in responsibility ED is daunting. The learning curve is steep, but rewarding.

‘My first shift in the ED was the first time I felt like a real doctor.’

Perhaps the most unique element is the density of decision-making. Each shift serves up a broad menu of undifferentiated patients ‘fresh’ from the community, often requiring multiple investigations/interventions. When you consider the wider landscape of an ageing population, over-burdened health service, and the much maligned 4-hour target, it’s an undeniably tough job for rookies; a pressure-cooker workplace that’s fertile ground for misdiagnosis and clinical error.

Therefore, in the interests of patient safety (and junior doctors overall wellbeing) it’s important to understand the mechanisms by which junior doctors collect/process information, and make decisions. Otherwise known as ‘clinical reasoning’.

Published in the EMJ in June 2016, Adams et al have qualitatively investigated clinical reasoning in junior doctors rotating through the ED for the first time. Thirty-seven doctors were interviewed and took part in focus groups. Questions and discussion were based around retrospective recall of two cases (one straightforward and one difficult case).

The authors used ‘dual cognition theory’ (DCT) as a framework. DCT describes two distinctive cognitive approaches to decision-making: ‘Type 1’ thinking (T1) is automatic and intuitive; ‘Type 2’ (T2) is deliberate and analytical.

This system was, of course, pioneered by Daniel Kahnemen in his book ‘Thinking Fast and Slow’ (essential reading). Below is a video which provides a useful visual explanation.

Key findings from the paper:

  1. There are 3 phases of clinical reasoning in junior doctors – each is briefly explained below, but for more complete explanations please read the original paper.
  2. During all 3 phases, both thinking pathways (T1 and T2) were working in parallel.

*Please note: indented sentences in italics are my own interpretation of the research, not actual quotes from the data.

Phase 1: Case Framing

Initially, a decision needs to made whether to go into clerking mode (i.e. systematic enquiry) or ‘resuscitation’ mode (call for help, ABC approach etc).

The information processed to make this decision are clues from the patient demographics, triage note and ‘end-of-the-bed-o-gram’.

Phase 2: Evolving Reasoning

The next phase involves establishing a diagnosis. This happened in two ways:

1) Diagnosis instantly recognised (T1 predominant)

Usually from a single cue in the assessment.

‘This patient has right iliac fossa pain. It’s probably acute appendicitis.’

These reflex diagnoses are then interrogated for error via:

  • Screening for ‘red flag’ features
  • Diagnostic timeout’ to organise thoughts
    • Reflection whilst writing patient notes
    • Informal discussion/presentation to a colleague
  • Begin another task, allow opportunity for spontaneous thought (passive diagnostic timeout)

2) Diagnosis not recognised, further analysis commences (T2 predominant)

With no immediate diagnosis reached, ‘hypothetico-deductive reasoning’ is employed. This is where multiple possible hypotheses are generated, and then the history, physical examination, and investigations are used to test these hypotheses, with a view to eliminating them one-by-one.

The ‘SOCRATES’ mnemonic is a useful tool for challenging hypotheses in chest pain patients.

‘Site? Onset? Character? Radiation? Associated symptoms? Timing? Exacerbating/relieving factors? Severity?’

Ongoing observation is also employed to test diagnostic hypotheses, and to screen for an evolving clinical scenario.

‘The inflammatory markers are higher than I thought they’d be. This patient is likely to be septic. Let’s start IV antibiotics and refer to the medics.’

Phase 3: Ongoing Uncertainty

This was predictably common in an inexperienced cohort of doctors, and dealt with via:

  • Delaying discharge and continuing to observe
  • Simplifying the overall decision: ‘is this patient too unwell to go home?’
  • Sharing responsibility
    • With seniors
    • With peers via informal discussion
    • With patients via safety netting

Three points I’ve taken from the paper…

1) Inexperience can lead to ‘misframing’

T1 judgement was the dominant thinking pathway during ‘case framing’, particularly when it came to first impressions – a ‘gut-feeling’ assessment of acuity.

A lack of experience might prevent juniors from picking up on subtleties (e.g. sweating, mild agitation), or get falsely reassured by certain details (e.g. normal vital signs) that a senior doctor would not, and ‘misframe’ the patient’s level of acuity. Clearly, this has the potential to be detrimental to patient outcomes.

For this reason, focused reflection on this crucial phase of the assessment via case-based discussion with a senior colleague is critical. It will encourage juniors to gain some insight into how their own intuitive thoughts play a prominent role, and perhaps encourage them to interrogate those thoughts for biases.

‘Did you make any assumptions about this patient before seeing them?’

‘Were there any clues or triggers when you first saw them that changed how you felt about the case?’

‘Can you think of any other clues that might subtly point to the patient being more unwell than the triage note suggests?’

2) Diagnostic time-outs should be encouraged

It was established that junior doctors would utilise diagnostic time-outs whilst writing notes or via informally presenting to peers. This is an important part of the clinical reasoning process, particularly in complex patients. It can safeguard against premature ‘closure’ of a case (i.e. jumping to conclusions, and then sticking with them incorrectly) .

Again, this part of the process needs to be reflected on, so that it’s utility is appreciated.

‘It was when I was writing the notes about the 68 year-old male that I thought had renal colic, that the possibility of ruptured AAA came into my head. It’s a useful moment to think about the case.’

Diagnostic timeouts should be encouraged by higher ups, despite being potentially time-consuming, and particularly if the shop floor is manned heavily by junior doctors (a not-uncommon scenario). Despite the overburdened and target-driven climate of UK  emergency medicine, departments must avoid falling into the trap of pushing their staff to work faster. Patients are safer when junior doctors are given the chance to slow down and think.

3) Juniors should be given protected time for case follow-up

The transient nature of our patient encounters in the ED can lead to an ‘out of site, out of mind’ culture, where we fail to follow-up uncertain or particularly interesting cases. Again, this is re-enforced by the pressure to work quickly. This represents a glaring missed opportunity for learning, and the lack of diagnostic feedback potentially leads to the propagation of flawed clinical reasoning, particularly in inexperienced doctors.

Perhaps juniors should be given protected time to follow-up on cases they’ve seen (read discharge summaries/visit ward/call patient at home if discharged). They could then log this process, and formally reflect on notable cases with a supervisor.

This paper has highlighted that junior doctors have a tendency to make judgements on single cues, as opposed to pattern recognition, and can draw premature conclusions from insufficient clinical information. There is no doubt that regular feedback on real cases will serve as a powerful tool to improve clinical reasoning. It will gradually nudge them towards the realm of expertise.

Much like the encouragement of diagnostic timeouts, the key is likely to be departmental culture change. This will require brave consultants and senior nurses.

Final Thoughts

This paper should serve as a guide for junior doctors (and their supervisors) for more focused, effective reflective practice.

It’s not just about reflecting on the pathology encountered and decisions made, it’s also about the clinical reasoning process that led to those decisions. The journey is just as important as the destination. Junior doctors should be reflecting on how they think.

Additionally, departments should strive to create a healthy environment for regular reflective practice, and not to prioritise targets over the development of junior doctors clinical reasoning skills.

The Paper

Clinical reasoning of junior doctors in emergency medicine: a grounded theory study
E Adams, C GoyderC HeneghanL BrandR Ajjawi
Emerg Med J emermed-2015-205650
Published Online First: 23 June 2016 doi:10.1136/emermed-2015-205650

Many thanks to Dr. Emily Adams, the primary author, for her assistance in the creation of this blog post.

Robert Lloyd