Blog

For some time now I’ve been interested in how we make clinical decisions. There has been plenty of discussion about evidence based medicine over recent decades but little discourse about metacognition.

What is it that drives us to adopt or refute practices? It is often not based on scientific rigour and certainly not always a best evidence model.  I was entertained and intrigued by this article in the Christmas edition of the 1999 BMJ covering some of the practices employed by clinicians in guiding decisions.  Although tongue in cheek, it certainly strikes a chord and does stimulate thought on the subject.

Over the last couple of years as an intensivist, having invested innumerable hours studying and formulating my views in order to make the daily decisions required, I have had many of those views challenged.  I have realised that often they are based on little more than a necessity to have a decision ready to go and a justification should that decision be questioned.  The evidence is simply not there to answer many questions and that which is there has a myriad of flaws in translation to the patient in front of us.

There is something truly beguiling and simultaneously frustrating about working in a scientific field such as critical care.  No matter how much we learn and think we know, there is always uncertainty.  Given that science is all about asking questions and seeking answers, it is amazing how little progress we have made towards definitive answers. 

So what is it that drives the decisions that we make? 

Firstly, the role of evidence based medicine should be considered. It is important to acknowledge the importance of incorporating the best evidence that we have into our decision making schema wherever possible.  However, flaws in the evidence and its application to our individual patients should be (and often aren’t) acknowledged.  

I think that although all of us have some grounding in statistics over the course of our training, this has been underemphasised as a keystone in our understanding of the scientific literature. As clinicians we discuss the literature often with black or white assessments of studies. It is human to seek the bottom line of key “good” studies to incorporate into our daily practice without remembering that the populations studied may differ from the patient in front of us or the conclusions may be weaker than remembered. 

There is no “one size fits all” solution for patient management and in the quest for an evidence base to our decisions, this can be easily forgotten. While pragmatic to adopt consistent practices for recurrent problems based on an interpretation of current evidence, it is important to acknowledge weaknesses in the quality and applicability of such evidence when applying it to the patient in front of you. No single study is a panacea.  We all want the “definitive” study to answer our questions but that is unfortunately not how science works.  

Common questions may have been approached by groups like the Cochrane Collaboration or Up-to-date. However, when it comes to more difficult questions involving complex patients or rare conditions (or both) and figuring out what to do within our clinical environment, we must figure out our own answers.

It is at those times that we need to appreciate a few things.  Our searches are time limited. We are steered towards certain answers depending on search engines and search terms. There is so much information out there that it can be extremely difficult to know what is out there that we have missed.  

We are also guided by our current world view. It is normal to seek out answers which confirm our preconceptions and to be more rigorous in questioning those which challenge those preconceptions. This is actually quite efficient in practice. It enables us to put our efforts into closely assessing new pieces of data rather than spending all our time going over familiar territory.  The risk, though, is that dogma are formed and once established, these are not rigorously questioned again. It is important to remain skeptical even of the foundations of our standard practice. When new information comes to light challenging our dogma, be prepared to revise the foundations which formed that decision alongside the new information. As scientists, we tend to downplay emotion as part of the basis for our decisions. The emotional investment we have in our formulated views can be hard to breech.

It is also easy to be caught up in the excitement of a new technological or treatment breakthrough. Innovation is, of course, an essential part of medical advancement. However, for every breakthrough there are a multitude of failures. 

Opinion leaders are another filter used to help with decisions. However, although they have often reached trusted positions through hard work, knowledge and extreme cognitive ability, one must still be aware that they will have their own biases. Often such leaders will be asked to give (or will put forward) opinions on matters in which they are not expert. There is a common adage in medicine that half of what we do now will be wrong in 10 years - we just don’t know which half.  This has even been assessed and shown in this fantastic article.  The medical history books are littered with fads which come and go from common practice which seem crazy in retrospect. 

Likewise, it is normal to incorporate logic and knowledge of basic physiology into decision making processes. However, there have been plenty of instances where sound rational deduction has been shown to be confounded by reality.

There is also, of course, the effect that our own personal experience contributes to our decisions. Although clinical experience is not to be eliminated, it also comes with a warning. We are designed to spot patterns and our anecdotes may hold far more weight than they should in a decision making algorithm.

Fortunately, a spectrum of personalities exist. To moderate the enthusiasm of innovators/early adopters there are usually late adopters/laggards filling the converse, and still important, role of questioning new advances/ideas.  This tension allows medical progress tempered with skepticism. It is important to identify where one sits on the spectrum (generally and for each individual discussion).  Shifting to the innovation/early adoption end of the curve must come with a warning that without appropriate knowledge and experience one risks far more failures than successes.  Some will argue that early adoption causes harm through experimentation. Others will argue that late adoption causes harm through missed opportunity to realise a medical advancement. I would contend that both are harmful, but unavoidable, and the important part is recognising that every decision we make has opportunity cost. It is important to frame the benefits and risks to your patient and outcomes important to them, make a decision, and move on. 

I think that the key to our job is to remember to put every decision in context. Use multiple sources and tools to assist in decision making and remember that any one study/investigation/monitor/case is only a single piece of the puzzle - you are unlikely to make out the whole picture with that single piece.

At the end of the day, despite all of the risks and flaws outlined above in our decision making approach, there is some good news. Regardless of the path taken, cumulatively we must be doing something right as outcomes continue to improve over time. 

Good work people - keep it up!