When the bake fails

One of my favourite kitchen appliances is the small weighing scale I use for baking. It gives a precise measure of how many grams of each ingredient is required to bake a delicious treat. Prior to even making it onto the scales, I’ve considered the constituents of each ingredient in the recipe – the percentage of milk solids in butter, the temperature of the eggs and the several different types of sugar. The correct measure of ingredients, the appropriate ratios to achieve the desired consistencies and a tried and true method will lead to the outcome of a skilled bake. This is all very precise. There is a structure to it. But this is only the how and the what of the bake. The why is very different. We bake sweet treats to celebrate, to mark endings and to share with friends. We also risk the development of diabetes but at least the soul is happy. The point is that we don’t just bake to simply sustain but for more soul fulfilling reasons. One of the reasons why I like baking so much is there is a certain degree of predictability in outcomes, given a set of variables. (Remind me to tell a story at some point that involves my love of cooking – which dangles dangerously on the other end of this wild spectrum and follows no recipes but sprinkles spices with much more reckless abandon). But back to predictability. This predictability in baking affords pattern recognition when certain ratios of well-chosen ingredients are combined to produce a desired outcome. Pattern recognition is at the core of medical training. We learn how to develop these almost knee-jerk reactions quite early on. It is a design of the training that is meant to help us safely and consistently recognise dangerous presentations. It is meant to be the built-in failsafe for when we’re unbearably tried and short on time. But how do we become more skilled and less reactive when it matters the most?

It was a Sunday evening in the Emergency Department and as always in these environments, time either inches along or speeds away from you like that train you were running for but missed in the morning commute. It was one of those shifts where it was the latter. 10pm approached and little 5-year-old Brooklyn* was brought in by her mother. A quick glance at the computer revealed key words typical of many a paediatric presentation: fever, general lethargy, rash. I pulled back the all too familiar blue curtains that we use to somehow convince ourselves give our patients a shroud of privacy. Surely no-one else will hear the questions about their bowel movements outside of the sacred pulled across blue curtains. Surely they’re sound proof. As I did, I saw a little girl lying on the bed. She looked unwell. Actually, she looked absolutely terrible.

She looked unwell.

There is much to be said about a gut instinct or skilled intuition. Experts are said to use cues in the environment and pattern recognition derived from prior experience in order to exact their skilled intuition that looks very much like ‘gut instinct.’ The intuitive judgement of some professionals in medicine is remarkable. Clinical decision-making on any given day can range from the mildly effortful to the high-stakes decisions that wrestle with life’s fragility. There are many models of decision-making that have been studied throughout the years. One of the models pertinent to medical decision making, is the recognition-primed decision-making model where the judgement involves both an automatic process primed through prior experience and the execution of the decision which is mentally simulated. As I am absolutely no expert by any stretch of the imagination, I find it interesting how these skilled intuitions are developed and honed. It is said that it requires both high validity environments (that is, a place where repeated presentations of the same are likely to occur) and the opportunity to learn and receive feedback. No wonder medicine feels like a marathon that you need to sprint.

I went on to examine Brooklyn. As I knelt down to gently hold her hand and explain what I was doing she whispered as though it would be our little secret, although I strongly suspected it was because she was fatiguing. She whispered to me that her heart hurt. So like any paediatrically-inclined person worth their salt, I pulled out the power of the sticker. I offered her a sparkly unicorn sticker to pop over her heart and it make it hurt less. Unfortunately, as much as I’d like to believe that stickers and magic can cure the world of all its ills, sometimes you just need to bring out the vasopressors and inotropes and antibiotics.

Brooklyn continued to be worked up and treated long after I had left for the evening. Days later she would go on to deteriorate and require much more extensive support than could be offered on the ward. I’ve thought a lot about little Brooklyn since that evening. I have not only a visual memory of what she looked like lying on the bed but also the feeling of the ‘gut instinct’ I had. Now, our memories are notoriously fallible and we remember peaks and troughs of events and feelings much better than the everyday consistencies. But this was one trough that I feel as though is very important for me to learn from. Sometimes the formula for the bake fails. Sometimes a singular ingredient is missed or the incorrect oven temperature goes unrecognised until it is past the point of correction. Like in baking, I’d like to believe that skills in recognising the deteriorating patient can be learned and practiced. So I’ll continue to learn. I’ll also continue to bake sweets. Both are hope-filled endeavours.

*All patient names have been changed and details omitted/changed for the sake of confidentiality

Proper readings instead of the drivel you just read:

Kahneman, D. and Klein, G., 2009. Conditions for intuitive expertise: a failure to disagree. American psychologist, 64(6), p.515.