Theory and practice of making difficult clinical decisions

BMJ 2006; 333 doi: (Published 28 September 2006) Cite this as: BMJ 2006;333:694
  1. Ed Peile, head of institute1 (ed.peile{at}
  1. 1 Institute of Clinical Education, Warwick Medical School, University of Warwick, Coventry CV4 7AL

    The student patient described in this report excited widespread interest in the medical community.1 As well as heartfelt sympathy for the patient and her family over this crisis, respondents have expressed empathy for the clinical team faced with the dilemma of offering best care. Some interesting threads have emerged in the correspondence.2

    Doctors of many disciplines have modelled their thinking processes, and there are some wonderful examples of clinical reasoning. From the start, a majority of readers were highlighting the probability of adult onset Still's disease, but, as at the bedside, the twists and turns of clinical progression forced careful reconsideration. There was full endorsement of the decision to get a second opinion, and the value of multiple perspectives was mirrored in the wisdom of medical microbiologists, rheumatologists, and generalists in the web responses.2 The importance of optimal communication with the patient and her family amid the uncertainty was lost on none.

    How we make decisions

    What can we glean from the responses about how doctors approach clinical decision making in difficult cases? Many of the experts did not cite their sources, and indeed much of the knowledge that comes into play for them is stored in rapid access memory banks. For those who did cite sources, the computerised diagnostic aid was the most erroneous, coming up with a single definitive (wrong) diagnosis. We are not there yet in terms of artificial intelligence in diagnosis.

    Then there was the doctor who referred to, “My little Oxford Handbook of General Practice.” She was clearly on the right lines, showing the place of reliable pocket books in pointing us in an appropriate direction. A generalist commented that: “I have had to read more on Still's disease,” demonstrating the importance of the next stage in accessing “just in time” knowledge. Other respondents, particularly those of specialist backgrounds, went to relevant reviews and into the primary and secondary literature.

    Are we doing what experts on best evidence advise? It is now over 10 years since the Bandolier's 1996 new year resolution to “ensure that all decision makers can find the best available evidence on tests and treatments.”3 At about the same time the science of decision analysis was being promoted,4 and much has since been published about the application of best evidence to clinical decision making.5 The web discussion suggests that the processes used by the medical community for complex dilemmas are thoughtful and effective but probably not as systematic as they might be.


    • Competing interests None declared.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    View Abstract

    Log in

    Log in through your institution


    * For online subscription