Behaviour, Cognition, Investigation and Learning

Earlier I wrote a short piece in relation to investigations, motivations and behaviours that lead to error.  Pending the future PSII syllabus I have since adopted a model that enabled me to obtain a wider understanding of cognition and how the journey of decision making is affected by different levels of bias within that journey. The effect of behaviours and system safety will be a key driver, often bias is referenced just as an end point, but like most decision making journeys there is more often than not a beginning, a middle and an end, the failure point (the root cause).

Using this model I have been able to identify that within 100% of diagnostic errors over a period of time these cognitive behaviours are present on "each and every occasion", within that there is also a high incidence of communication failure, which you can also associate in the main to those cognitive behaviours.  That poses the question in terms of patient safety how do you improve cognitive behaviour?  Changing policy, pathway, procedure will not remove cognitive error alone?

Diagnosis is made by some interacting combinations of intuitive automatic processing [Kahneman system 1 thinking] and deliberate rational consideration [system 2 thinking].  The majority of clinical diagnosis are derived intuitively by acknowledging that most conditions are common and easily identified.  These judgments cannot be taught they emerge subconsciously but intuition can be strengthened and improved.

Bias affects decision making and the model adopted is now being built into the investigative doctrine to enable the identification of bias within patient safety incidents but importantly the journey of decision making will be able to be identified as part of the wider "system" to improve patient safety and being undertaken by:

Shared learning template to communicate the behaviour influences, a review of the patient communication approaches, developing a post incident review into "reflection" to mirror the identification of the cognitive errors within the incident, short term post incident e-learning and longer term educational approaches to cognition and metacognition. 

This approach is to promote warnings about possible bias, show how bias distorts good decision making and at what point, when an individual has made a bias related judgement error providing effective feedback, repeating extended coaching.  

This is an important area for consideration and happy to receive information on any other approaches. 

Keith

 

    

A fascinating post, thanks @Keith Bates I’d love to find out more. Maybe we could do an interview with you and share on the hub?

Would also be interesting to connect with AI people, how these issues are considered in systematising of decision making. @Richard Jones @Clive Flashman Might you be interested too? 

I've seen a lot recently on work being done around rarer NCDs, the 'long-tail' of diagnostic go'to's.

I know that a company call Volv based in Switzerland has been trying to create an AI system smart enough to bring these into the normal diagnostic process, not sure how successful they have been yet.