Uncertainty in medicine
BMJ 2017; 357 doi: https://doi.org/10.1136/bmj.j2180 (Published 11 May 2017) Cite this as: BMJ 2017;357:j2180All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
I totally agree that uncertainty is not being addressed properly in medicine. It is important however to distinguish between uncertainty associated with non-transparent or intuitive thought and uncertainty associated with transparent reasoning [1]. 'Non-transparent thinking' is what we do quickly and intuitively but which is susceptible to error from psychological factors. 'Transparent reasoning' is how we check our intuitive conclusions and explain how we reason with uncertainty to patients, colleagues and examiners.
Transparent reasoning with uncertainty makes use of an important process that involves setting up lists of possibilities (e.g. differential diagnoses) and then trying to show that all but one are improbable [1]. This depends on showing that findings occur commonly in a suspected diagnosis, but less commonly in any diagnosis that we are trying to show to be improbable [2]. The same approach can be used to analyze clinical trials in a more transparent way to arrive at treatment indication criteria and cut-off points that avoid over-treatment and under-treatment [2, 3]. Mental checklists can also be used to try to show that sources of error and bias are improbable when interpreting published findings, including clinical trials [4]. This allows us to assess whether clinically significant findings are likely to be replicated.
All this is different from applying Bayes rule to sensitivities and false positive rates or simply applying tests of statistical significance alone. I think that it is failure to progress from the latter that is preventing uncertainty from being addresses properly in medicine.
References
1. Llewelyn H, Ang HA, Lewis K, Al-Abdullah A. Oxford Handbook of Clinical Diagnosis 3rd edition. Oxford University Press, 2014, pp 1 – 23. http://oxfordmedicine.com/view/10.1093/med/9780199679867.001.0001/med-97...
2. Llewelyn H, Ang HA, Lewis K, Al-Abdullah A. Oxford Handbook of Clinical Diagnosis 3rd edition. Oxford University Press 2014, pp 615 – 642. http://oxfordmedicine.com/view/10.1093/med/9780199679867.001.0001/med-97...
3. Llewelyn H. Reasoning in medicine and science. Oxford University Press blogs, 2103. https://blog.oup.com/2013/09/medical-diagnosis-reasoning-probable-elimin...
4. Llewelyn H. Reducing over -diagnosis and over-treatment by improving their criteria and stratifying them. Preventing Overdiagnosis Conference, Oxford, 2014, poster 34. http://www.preventingoverdiagnosis.net/2014presentations/Board%2034_Huw%...
Competing interests: No competing interests
Re: Uncertainty in medicine
This interesting and relevant editorial reminded me of what Voltaire, the French philosopher, once said: 'Uncertainty is an uncomfortable position. But certainty is an absurd one.'
Competing interests: No competing interests