“But what’s wrong with him, doctor?”

The constant refrain from many a consulting room is not “How can you make her better?” but “Can you name the problem so I can own and understand it?”. When addressing this need we will each develop our  own approaches; some of us will explore differentials, others state the top of our list, others delve into the concerns sitting behind the inquiry. When addressing the content – what’s actually wrong – we’ll probably want to come up with a diagnosis.

Now to make a diagnosis, we might go for the slap-me-it’s Shufflebottom’s Syndrome – “because it is” – (or if there’s a medical student to explain it to we may go all scientific and pretend it’s not instinctive but “because of of feature A,B and C”) – and this comes from a think-1 cognitive frame. Or we might go for a differential probability and shifting one way or another in a deductive Holmsian approach of think-2 systems.

We can try and find evidence that will support us to do this diagnostic thinking. The evidence for think-1, leap-in, its-that-’cause-it-is diagnostics is helpful in demonstrating how often it’s correct – or incorrect – but doesn’t really help us develop that artistic flair. Where we can find (more) evidence to help us will be along a path of rational deduction – not that this will be better that think-1 – just that there will likely be more of it.

In this set of linked Archimedes blog posts, we’ll be looking at how to appraise the evidence to underline our diagnostic procedures. We’ll be moving from an understanding of diagnostic test accuracy, through Bayesian statistics, to seeing if implementating a diagnostic procedure makes a difference to patients…

But next time we’ll start with a differential diagnosis.

– Archi