What I learnt this week: Confirmation bias – the cousin of over-confidence #WILTW

This is the 56th #WILTW

During my medical training I sat in many outpatient clinics. The vast majority I don’t remember specific details about but there is one that has always stood out for me. I was in a clinic with a Professor of Paediatric Cardiology. The Professor was nearing retirement, in fact, this may have been his last ever clinic. I has been asked to listen to a child’s heart and describe what I found. I remember not really being able to either describe the heart murmur or diagnose the problem. The Professor kindly explained to the family what the sound he was hearing was and that he would send the child for a scan to confirm his diagnosis. As the parent left the room he confidently explained the key features of this heart condition and what the plan would be to treat the patient.

You see, Damian, it is not always possible to hear the fixed splitting characteristic of an Atrial Septal Defect but this child has all the other obvious presenting features

The patient returned from the echo-cardiography suite with scan findings completely different heart problem than that suspected by the Professor. He was surprised and a little perplexed. You could see a range of emotions briefly come over his face before he professionally explained to the family the next steps.

After the clinic the Professor was clearly still mulling over the patient. What was troubling him seemed to be the unexpected nature of this error. As a clinician on the edge of retirement I got the sense that his confidence had been rocked. I think the reason I remember this case so clearly amongst the hundreds of other clinic cases I have seen was his disbelief and palpable disappointment.


16 years later I was asked to review a child with breathing difficulty. His parents were concerned about him and the junior wasn’t sure what was causing the problem. The child looked moderately unwell but on listening to his chest and reviewing the history it seemed they needed treatment with inhalers.

You see, John, it’s not always possible to hear wheeze, but if you notice he is having difficuly breathing out and the history of persistent cough despite regular antibiotics is consistent with viral wheeze

I was subsequently called into the resuscitation room but on returning found that the child had had an X-ray taken. While I had been away the child had detriorated and had started grunting. The X-ray is below:


This took me somewhat by suprise as it isn’t the typical X-ray of a child with viral wheeze and is consistent with a severe chest infection.

Every clinical shift brings a new opportunity to learn and develop skills in medicine. Sometimes you need a reminder of biases inherent in decision making. I had fallen victim to confirmation bias 

Confirmation bias: the tendency to look for confirming evidence to support a diagnosis rather than look for disconfirming evidence to refute it, despite the latter often being more persuasive and definitive.

Constantly re-evaluating decision making is not something that should dissipate with experience. I hope I am open to the possibility of making mistakes right up to retirement.

What have you learnt this week? #WILTW


For medical readers the Echo-cardiogram revealed a VSD

Many thanks to the family of the child in question who were happy for me to discuss this story but I have also made some subtle alterations to the clinical events. 

Leave a Reply