As a senior GP registrar about to sit exams I felt that I was brimming with knowledge. So why then was I still second guessing myself about certain patients at the end of the day? I spoke with my learned supervisor about this – he’s been in the game a long time and has a much more complex gang of patients than I.
“Practice medicine in a way such that you can lie in bed each night and not worry, ethically or medically”.
To put it simply, there are three issues that need consideration regarding risk:
1. Not recognising it in the first place – that’s the whole unconscious incompetence argument.
2. Recognising risk but allowing biases to distort it.
3. Recognising the risk, defining it but not mitigating it.
Risk tolerance is an important concept related to cognitive bias in clinical practice.Reflective practice is the key to managing risk. If we are able to identify the critical steps in our decision-making, we can regularly calibrate our knowledge and justify it at each of the steps (Balla et al, 2009). So therefore, pattern-recognition, or System 1 thinking (see Removing your clinical blinkers) is not great for acknowledging risk.
We should also remember that risk tolerance is also influenced by our skills and experience, context, resource availability, and patient factors. We are much more likely to give the patient the benefit of the doubt if they have good health literacy, easy access to healthcare, and if our own skills and ability are reliable (i.e. we are not overconfident). In general practice, the chance of serious conditions presenting is relatively low, and so risk tolerance is higher. In an emergency department, the chance that a patient presenting with chest pain will have an outcome with high-morbidity is much higher, making risk tolerance lower. Availability bias (see When you can’t see the forest for the trees…) also influences our risk tolerance.
So what tactics do we use to mitigate against risk intolerance? Sadly it is investigations. We all know that the history will give us the diagnosis in 90% of cases, and the examination should confirm our findings (or help to exclude other differentials). Investigations are really there to guide management. But how many times have you found yourself (especially when time-pressured) taking a quick history, doing a cursory examination, and ordering the spectrum of bloods ‘just to make sure’? This is not great practice – for the public purse, the patient, or for clinical decision-making. Investigations shouldn’t be the ‘catch-all’ sieve for our risk management. That’s where guidelines are great – evidence-based guidelines are just that! What should we do when, for what diagnosis, and why. Yes, we use them for management but how often do we use them for diagnosis!
I challenge you this week to think about how you identify and mitigate against risk, and what determines how well you sleep at night. Investigations should not be the “colander for your clinical decision-making”. How many investigations do you order unnecessarily? Look back at your last five pathology request – how many investigations weren’t really defining your differentials or adding to your management?
Don’t forget about the clinical challenge for the week. Feel free to engage and comment on the case on my Facebook page.
Balla, J., Heneghan, B., Glasziou, P., Thompson, M., & Balla, M. (2009). A model for reflection for good clinical practice. Journal of Evaluation in Clinical Practice, 15, 964-969.
Brown, T. B., Cofield Ss Fau – Iyer, A., Iyer A Fau – Lai, R., Lai R Fau – Milteer, H., Milteer H Fau – Queen, B., Queen B Fau – Schwab, M. H., . . . Schriger, D. L. Assessment of risk tolerance for adverse events in emergency department chest pain patients: a pilot study. (0736-4679 (Print)).
Dushenski, D., Hicks, C., & Himmel, W. (2015). EM Cases Summary – Episode 62 – Diagnostic Decision Making Part 2: EMB, Risk Tolerance, Over-testing & Shared Decision Making. Retrieved from https://emergencymedicinecases.com/wp-content/uploads/filebase/pdf/Episode%20062%20Diagnostic%20Decision%20Making%20Part%201.pdf