Issue 43 / 11 November 2013

IN the US, it is often claimed that medical practice is hamstrung by the threat of litigation. In Australia, many of us feel the weight of regulation.

Neither the law nor the rules should stop us from exercising clinical judgement for the benefit of patients. But is it possible to negotiate the multiple — and often conflicting — influences on our behaviour while maintaining integrity, safety and sanity?

We are in an era of clinical governance practice that emphasises standardisation. The number of guidelines, policies, directives and performance indicators is mounting, along with an increasing tick-box mentality that assesses compliance.

Adherence to the guideline seems to have become the outcome — rather than better health for patients.

In this setting, what is the role of basic clinical assessment and decision making based on what we used to call clinical “nous”?

Well, reassuringly, it’s still there. It seems that the many rules that have been created to help standardise a clinical assessment and risk estimation, if applied wisely, can be almost as good as the advice of a sound clinician. If applied inappropriately however, decision rules and pathways can lead to inappropriate care.

For example, research in Annals of Emergency Medicine compared the use of a clinical scoring system with clinical judgement in the diagnosis of appendicitis. It found that unstructured clinical judgement performed better.

Another study in the same journal compared unstructured “clinician gestalt” with two validated scores for assessing the pretest probability for suspected pulmonary embolism. That research also found that clinician judgement performed better than decision rules for both low and high probability patients.

Clinician gestalt is not magic. Like other types of so-called intuition, clinical judgement is a subconscious appreciation of learned patterns or rules, based on both knowledge and experience.

What clinicians do better than any alternative is decide which patients should enter a particular diagnostic or therapeutic pathway. When decision rules lead to the inappropriate application of these pathways to a low-prevalence population of patients it can produce too many false-positive results. This causes harm.

The authors of a paper titled “Medical decisionmaking: let’s not forget the physician” hit the nail on the head. Just as new drugs need to be tested against existing therapy, clinical decision rules need to be tested against physician judgement without the tool.

As the authors say: “Risk-stratification tools do not eliminate subjectivity; they merely shift it from one domain (eg, does this patient need this test?) to another (eg, is this patient eligible for this decision rule?)”.

Clinical governance systems that pay lip-service to clinical judgement, but then punish individuals for not following specific steps, are sending mixed messages. We are fooling ourselves if we think that a decision rule can convert a novice into an expert.

The place for experience and judgement is at the beginning of any pathway.

Every guideline should start with: “Is this guideline appropriate for this patient? If unsure, consult a good clinician”.

 

Dr Sue Ieraci is a specialist emergency physician with 30 years’ experience in the public hospital system. Her particular interests include policy development and health system design, and she has held roles in medical regulation and management.

15 thoughts on “Sue Ieraci: Clinical nous

  1. Matthias Maiwald says:

    In response to Bill. I must admit I was exaggerating the situation — deliberately. I am well aware of the formal requirements of a systematic review, and the differences between a systematic review, a narrative review and an opinion piece. For clarification, I was not implying in my comment that a systematic review should be written like an opinion piece. Nevertheless, apart from good clinical and scientific reasoning independently being important tools and important skills to learn, and apart from knowledge of physiology and pathophysiology being essential for good medicine, it is also clear that both clinical and scientific reasoning are necessary for evidence assessments, including systematic reviews. Simple example is the formulation of appropriate selection criteria that make clinical sense. And so on. Evidence assessments are not the “reasoning-free zones” that they are often purported to be, and the often-implied notion that one can just follow the prescribed steps (cookbook style) and then “automatically” arrive at the correct evidence is not tenable. We have recently uncovered and then published about a massive, large-scale medical literature error that arose precisely because trial reports and systematic reviews — in droves — violated basic rules of scientific reasoning, despite most of these having adhered to the formal requirements of evidence assessments.

  2. Bill Slater says:

    Mathias, although the term “Systematic Review” is not (to my knowledge) trademarked, I think that whoever writes and publishes an article using that term should be adhering to the “rules” of the genre. If you don’t like the rules then by all means call your article something else – “opinion piece” is quite popular.

  3. Sue Ieraci says:

    Thanks for all the interesting comments. I should clarify that I don’t believe medicine should be ALL ”gestault” – many things that we intuitively think should be a certain way can turn out to be incorrect when tested in a controlled method. What I do think is that evidence should inform practice wherever possible or relevant – but the evidence should be applied with intelligent judement. Standardisation has its benefits – everyone should, for example, use recommended antibiotics for specific clinical situations, treat acute asthma in a standardised way. The real dilemma is where there are multiple conflicting ”pathways”, based on individual risks or symptoms, that all could apply to a single patient. The recent discussion about statins is a case in point – there are multiple risk factors for coronary disease, most modifiable to some extent, but not all patients are ”modifiable”. Kees is right – emergency medicine (medicine in general) has moved along from the great cowperson achievements of the twentieth century. What we are doing now is fiddling around the edges, in a risk averse and demanding culture where what we have is never enough, and new therapies have diminishing returns for cost. In the background, we need to remember that vaccines, antibiotics, steroids, IV fluids, anticoagulants and reassurance are still there in the background – even if the community takes them for granted.

  4. CKN Queensland Health says:

    The early years of any pioneering endeavour requires strong individuals with passion, gestalt and courage along with a certain distain for regulation and the status quo. Think aviation and the early air-cowboys. Inevitably a crunch time will arrive, perhaps in the form of the 1977 Tenerife airport disaster. Then regulation and attendant tick-boxes will drive the cowboys out and invite practitioners with different traits.

    Emergency Medicine was once a decidedly pioneering endeavour. Now mainstreamed, let’s hope it never completely looses contact with its roots. Maybe, those with still a bit of cowboy or cowgirl in them can pioneer ways of getting patients safely off pathways.

  5. Dr John B. Myers says:

    Rules and regulations that are arbitrary are self-serving in the Medical Board’s and AHPRA’s interest, and bureaucratic interests, only. We, competent and caring professionals, need to promulgate the message and obtain collegiate support for common sense and ability as professionals to treat our sick patients according to need, not according to dogma. The Tribunal’s decision in support of the Medical Board’s dogma does not serve as a guide to doctors and does not in any way protect the public. Promoting arbitrary dogma is self-serving. It is misguided and generates fear. It is duplicitous. It stifles creativity, downplays rights and spurns criticism and objective analysis. It denies natural justice.  Oral submissions to the Legislative Assembly Inquiry into AHPRA will be heard at 2000h, at Parliament House, Melbourne, Wednesday 27 November 2013. To stand up to do what is both good and right, roll up on submission night and have your say.  

  6. Matthias Maiwald says:

    In response to Sue: Again I completely agree with you. Yet, what you and I are proposing is so totally politically incorrect that it couldn’t be any more politically incorrect. Just picture the following scenario. Someone writes a systematic review (yes, a systematic review, the core tool of EBM) and admits — yes, openly admits — having used common sense, proper clinical and scientific reasoning and the knowledge of the underlying physiology and pathophysiology while preparing the review. I just imagine the outcry from some people when seeing this admission. It would go something like: “What??? What have you used for your systematic review? Common sense? Scientific reasoning? Clinical reasoning? Knowledge of physiology? You cannot possibly do that! That is not evidence-based!!!”. (You can see I am imagining a Monty-Pythonesque sketch here). To be fair, I have seen articles from some of the really top EBM guys emphasising that clinical and scientific reasoning is indeed necessary while doing EBM, but that is not recognised or heeded by a large number of EBM followers. P.S. Have a look at what we write about biological plausibility on this healthcare blog: http://haicontroversies.blogspot.sg/2013/04/the-environment-and-hai-where-does_25.html

    I was tasked with preparing a systematic review for the 2010 NHMRC National Infection Control Guidelines, and I was gobsmacked (yes, I liked the word “gobsmacked” from the Johnny English movie) by how little emphasis was placed on actual subject knowledge and how important the formal, ritualistic process of evidence assessment in a “cookbook” style was.

  7. craranjo@optusnet.com.au says:

    True indeed.The bad outcomes from using ‘best practice’ rules are usually hidden from view, this helps promote adherence to ‘best practice’ over and over again.An Anecdotal scenario:How many are aware that the use of ‘best practice’ of prescribing the long-term use of a statin drug in a patient with a history of a cardiac event, is exposing that patient to a potential  fatal cardiac arrthymia, as in ventricular fibrillation, rather than ‘preventing’ another heart attack?

  8. Dr John B. Myers says:

    Well said! We’re getting somewhere! Thank you. Your statement: Rules and regulations can never replace an experienced doctor, needs to be sanctified. What this statement means is that context is taken out of the equation, as is specificity, and as far as quality and safety is concerned so are Rights and responsibility. Medicine is both an art and a science. Rules and regulations make room for neither and remove the clinician from clinical practice. My definition of Clinical nous is the finesse that characterises the conscious art and science of the clinically experienced physician whose skill is based on objectively tested and reviewed evidence, basic and acquired knowledge, including cause and effect, probabilities and interactions or context, associations and outcomes and critical assessment and review in day to day clinical practice and the ability to thank and attribute all of this to G-d, with joy. Thank You.   

  9. Joe Kosterich says:

    Time to reclaim medicine from the faceless bureaucrats and academics who promote cookie cutter and tick box medicine based on so called evidence.

  10. Sue Ieraci says:

    Hi, Matthias – yes, I have thought a lot about the respective significance of the concepts ”evidence based”, ”science-based” and ”evidence-led”. In my view, the most important thing about a health care therapy is that it should be based on the clinical sciences – ie it is plausible, based on what we know from anatomy, physiology etc. There are many therapies like this – from diuretics to chest decompression for pneumothorax. That is also how we know that many ”CAM” therapies – from homeopathy to aspects of chiropractic to iridology – are not plausible. We may not know everything about every aspect of how the body works – but so many things have now been directly observed and measured – from the potassium concentration inside cells to the action potential duration of a cardiac cell. Next, one needs clinical judgment in order to apply evidence rationally. SO, for example, we use judgment to decide whether to test someone for acute coronary syndrome or PE, knowing that application of a test to a low-risk population leads to excessive false positives (Bayes theorem). What could be more rational than combining scientific plausibility with the intelligent application of decision-rules to the right clinical scenario, based on known risk factors and measured probabilities? 

  11. Matthias Maiwald says:

    I agree completely. However, I can picture many evidence-based medicine fundamentalists who would argue that this is so totally non-evidence-based that one cannot possibly do medicine that way. It would be interesting to hear from Sue who she is thinking of reconciling that dilemma, which, by the way, is popping up everywhere in medicine (you just have to look). Also, by the way, important work on Gestalt recognition has come from Nobel Prize winner Konrad Lorenz. It is sometimes interesting to look into these other areas of science, too.

  12. Bryan Walpole says:

    Well said Sue, now let’s do it for low risk chest pain, and put an end to the troponitis epidemic!

  13. Huw R Davies says:

    Well said Sue. Too often guidleines & protocols are treated as Holy Writ, never to be questioned or modified if there is a compelling reason. They don’t replace reasoned analysis by an experienced and competent clinician, lthough they can supplement that process.

  14. Department of Health Victoria Clinicians Health Channel says:

    Thankyou Sue – wise words indeed.

  15. University of New South Wales says:

    Dear Sue,

    Great article!  Yes, it seems oft forgotten that guidelines are meant to be a guide for the clinician, not an algorithmic protocol.

Leave a Reply

Your email address will not be published. Required fields are marked *