Comment Highlight: Medical Bricolage by Dr. Doug McGuff & Domain-Dependence of Checklists

Medical bricolage is how to care for the Patient of One.

It's a clinical craft.

It's taking that particular n=1 patient's specific symptoms and presentation at the time, embedded within the context of the care episode at hand (however that unfolds), and then gathering and cataloguing data points actively through careful examination, questioning, and observation in order to deduce how best to proceed iteratively (which always includes inaction as an option).

What does this craft look like in practice?

In response to my previous essay on craft, Dr. Doug McGuff, author of Body by Science, chimed in with an awesome real-world example of medical bricolage in practice:

This weekend I experienced an opportunity for Bricolage during the resuscitation of a septic 3 week old infant. This child presented in respiratory failure (later found to be due to pneumonia and sepsis) and suffered a respiratory arrest upon presentation. We moved the child to the major resuscitation room and performed bag-valve ventilation until it started breathing on its own. However, it quickly had another apneic spell that resulted in a near cardiac arrest (when infants stop breathing their hearts slow way down and will even arrest fairly quickly). I intubated the child (who arrested during the procedure) and then quickly ventilated him until the heart rate came up and the child's vital signs returned.

Typically, we will administer a paralytic so that we can completely take over the child's breathing in order to spare it of the work of breathing. However, as we were preparing to administer a paralytic, we began to notice an unusual pattern in this child. Whenever it was breathing under its own power, the vital signs were perfect. Whenever he would have an apneic spell, his oxygen saturation would plummet and he would arrest even though we were ventilating him optimally via the endotracheal tube. Everyone involved (me, nurses, respiratory therapists) noticed this pattern and all agreed to withhold paralytics. We then developed a perfect rhythm of matching our bag-valve ventilation to the child's spontaneous respirations. Within 5 minutes, we began to notice that the child was now matching HIS respirations to our bag-valve ventilation.

When I spoke to the pediatric intensivist, I explained our observations. I expected him to recommend the standard procedure of administering paralytics, but to his credit, he felt very strongly that we should continue in the same pattern that had been working for us. There was no good physiologic explanation for this, but unless we were augmenting the child's own breathing pattern, things would just crash.

It was a very stressful resuscitation, but the child did well. This was a great example of Bricolage and of how you (for unexplainable reasons) can kill someone with templated medical care.

Talk about a positive Black Swan in emergency medicine.

Dr. McGuff's exemplary experience of adaptive and creative deductivist decision-making (improvising, as Brett would say), as communicated in Dr. Sandeep Jauhar's excellent book review, "One Thing After Another" (thanks to Dave Lull), of Dr. Atul Gawande's latest masterpiece, The Checklist Manifesto, speaks to the domain-dependence of checklists/templates/standards/protocols/et al. --> They are sometimes applicable and effective support modalities in some known (concrete) engineering and repetitive (acute) intervention procedure domains, but they often prove ineffective and fail quickly in other more uncertain, novel domains (like in Dr. McGuff's case) in the following ways that the ever-insightful Dr. Jerome Groopman captures magnificently (hat tip to Dave Lull):

That may account for the repeated failures of expert panels to identify and validate "best practices"? In large part, the panels made a conceptual error. They did not distinguish between medical practices that can be standardized and not significantly altered by the condition of the individual patient, and those that must be adapted to a particular person. For instance, inserting an intravenous catheter into a blood vessel involves essentially the same set of procedures for everyone in order to assure that the catheter does not cause infection. Here is an example of how studies of comparative effectiveness can readily prove the value of an approach by which "one size fits all." Moreover, there is no violation of autonomy in adopting "aggressive" measures of this kind to assure patient safety.

But once we depart from such mechanical procedures and impose a single "best practice" on a complex malady, our treatment is too often inadequate. Ironically, the failure of experts to recognize when they overreach can be explained by insights from behavioral economics. I know, because I contributed to a misconceived "best practice."

In my personal reflections, I've contrasted Dr. Groopman, author of How Doctors Think, and Dr. Gawande on these clear differences for some time (non-surgeon vs. surgeon, unknown vs. known, population vs. individual, etc.), and I have seen Dr. Groopman acknowledge domain-dependence straightforwardly and insightfully, as above, but have yet to see Dr. Gawande do the same. From my view, Dr. Groopman spends more time--which seems wise--considering the messy/uncertain cognitive (heuristics & biases) deduction domains, before we decide to 'operate', as Dr. McGuff did perceptively when he responded appropriately to the feedback he kept receiving from that child. Once the decision has been made to 'operate', a concrete task/objective is at hand--a mediocristan procedure like landing a plane--that is more amenable to 'checklist manifesto' infusion. Deciding whether or not to put a plane in the air in the first place in the face of bad (as well as nonlinear & chaotic, extremistan) weather is an entirely different ball game.

Dr. Doug McGuff landed a plane in a hurricane.

That's craft in practice at its finest.

You can't template that.

Undoubtedly, checklists/templates can shape/reinforce habits based on known past failures to avoid false steps in certain concrete/repetitive procedural situations, but they cannot tame complexity, uncertainty, and provide prescriptions/proscriptions (especially in longitudinal, less controlled settings) for the care of individual patients in their specific cases. Respect for the limits of knowledge (epistemology) must rule the day then. We must always, as Nassim Taleb suggests, factor for the unseen.

Somewhere, there's a transition phase/step between these two fundamentally different domains--a transition that sage physician bricoleurs like McGuff, Jauhar, and Groopman understand intimately: it's their craft that they speak about, after all, born out of experience and extensive reflection and introspection. That's being human in the practice of medicine; it's caring for Patient of One human beings by being a humble human being in the first place.

Like Seth Roberts' Theory of Human Evolution reiterates (hat tip to Dave Lull and Professor Aaron Blaisdell), we can make our livings in diverse ways if we identify the roles that crafts play in different domains. When it comes to healthcare domains, clinical crafts provide the foundational backbone of medicine; a backbone that is best equipped to bend but not break when bricolage carries the day.

That's, at least, what I say.

m=1/n=1, as always.

To good health,

Brent