For a trainee in EM, it is useful to know about three types of cognitive practice that require caution.
While a knee jerk reaction may sometimes save time, a shotgun investigation may improve billing and a kitchen sink therapy may create the illusion of therapeutic rigor, arguably that’s all there is to it.
In reality, there is not much true value to any of these three missed approaches.
We will look at each one with a few examples and then briefly discuss below.
Knee Jerk
When I was rotating in the ED as an MS4, a visiting EM attending once told me that “adding a Type and Rh should become a knee jerk” for any patient with vaginal bleeding in early pregnancy. Whether or not taking the extra 30 seconds to scroll through the EMR for a previously documented Rh likely to be on file is a better strategy, this one is fairly simple.
Not all of our knee jerk reactions are equally simple or harmless.
I have seen adenosine being pushed before one could say “Mama” for anything from sinus tach to atrial flutter and A-fib with RVR: paramedics, physicians and even unsupervised nurses all being equally guilty. Why? Because a sustained heart rate above 180 is scary to some. And the reflex is to do something quickly because we don’t like to remain scared.
Nursing staff going straight for IV placement while forgetting not only the basic ABCs of resuscitation but even to disrobe the patient is another example. Starting any patient at 100% oxygen saturation who is short of breath on nasal cannula oxygen is yet another.
We like to do what we are trained to do well and/or what is easy. Our brains then compel us to prioritize doing it.
Once my ED team halted a verbal order for a whopping dose of colchicine blurted out to nursing by a careless consulting cardiology fellow – the patient had mentioned his ankle pain to the fellow in passing. The man was in acute renal failure and ended up with a septic ankle joint diagnosed later. Knee jerk is in part responsible for well-perpetuated ED mental formulas such as “gout = colchicine”, “fever = paracetamol”, “wheezing = albuterol” and “hypotension = 2 liter IV fluid bolus”.
The knee jerk is how we pick from our favorite antibiotics and how we generally prescribe, how we diagnose and order things on lobby and triage patients and how we even decide on CT scans and dispositions. Frequently, our hospitalist medicine colleagues will utilize the same reflex and unnecessarily or prematurely consult specialists.
On occasion, when the arrow released via a knee jerk reaction hits the bull’s eye, it feels and looks great. Knee jerk, unfortunately, is also how we assume, stereotype, over-simplify, ignore and ultimately miss.
Shotgun
This one does not have to be shot from the hip, though it certainly looks cooler that way. Often this is done thoughtfully, with a pseudo-scientific aroma to it.
I was on my MS3 internal medicine rotation when one day, the dreaded ED handed us an elderly female with a congratulatory thick paper chart, a bouquet of vague complaints and no clear diagnosis. When I asked my senior resident what we should do, the answer was a shoulder shrug and a confident “Lab ‘er up!”.
Shotgunning is not just about shooting out labs in the dark, however. It usually refers to a much wider “strategy” (actually, a lack thereof) of checking anyone for “anything” so as to not miss “something”.
Consider an ED evaluation of a headache involving some component of facial pain. Let’s order a migraine cocktail, CT and CTA of the head and neck, ESR to check for temporal arteritis; and when we find nothing, let’s do antibiotics in case of possible dental caries, otitis, mastoiditis or sinusitis. Sounds pretty thorough and terrific, doesn’t it? In fact, many patients would tend to think so. Clearly, after all that, we just could not miss something real badTM. We should remember that in EM you are worth every test that you order.
Hyperlaboratoremia and panscanosis are not the only clinical manifestations of the shotgun approach.
Though in all places, it is well-intended, there is a more buried shotgun in standardized chest pain workups, ED triage scales, pre-conceived clinical pathways and universal screenings than you may think.
Kitchen Sink
One might say that kitchen sink is the therapeutic twin of shotgun diagnostics, though one does not need to stem from the other.
The kitchen sink is how you and I treat most non-threatening and hence not easily identifiable ED rashes. As one of my professors once said: the rule of dermatology is that “if it is dry, use a wetting agent, if it is wet, use a drying agent, plus steroids and antibiotics for everyone”.
At its core, any kitchen sink approach violates two key pillars of modern medicine – evidence-based practice and personalized therapy.
Another example is the kitchen sink phase of resuscitation in a soon to be aborted CPR effort. While in the beginning, we do tend to follow certain parameters and algorithms, towards the end and well into the “futile” stage of CPR remedies like calcium, magnesium, bicarbonate, second and third anti-arrhythmic and so on all inevitably flow one after another regardless of the suspected cause of cardiac arrest or objective facts known.
While benign rashes are benign, and futile CPR is futile, most of the kitchen sink does not involve such obvious extremes. In fact, some of it is perfectly legitimized and even justified – have you ever thought of what “broad-spectrum antibiotics” in sepsis really implies?
Reasons For Need To Know
Why is knowing about the knee jerk, the shotgun and the kitchen sink ahead of time important?
First, the cognitive action patterns described are unavoidable and inescapable. It is precise because we will not be able to fully stop using all three on occasion, that we should know about them ahead of time.
Second, there is something positive and well-thought-out corresponding to the other side of each of the three behaviors:
Fundamentally, knee-jerk reactions rest on pattern recognition as the predominant cognitive pathway at work – something that physicians start to rely upon more and more as they mature. While risking the error of premature diagnostic closure (among others), pattern recognition does save time and resources. This mode is why, as some studies suggest, senior-most providers may be more effective in triage.
On the opposing side of the shotgun coin are the well-accepted mantras of keeping one’s differentials broad and of thinking outside the box. Such forced mental efforts help avoid anchoring among other cognitive errors.
Last, kitchen sink elements may indeed be acceptable in salvage type of situations or in uncharted waters, given multiple paucities in our scientific evidence and in our full understanding of physiologic processes. In such select cases, we humbly admit our limits and hope that something unknown may work at the last minute, while there is no further harm that can be done.
It would be a mistake, however, to confuse each of the positives described with the three patterns we started with when taken in their pure form.
Third, the limitations and harms encountered by not keeping the three tendencies in check are real and immediate:
- Knee-jerk reactions do not yield beneficial results when the situation encountered is new and principally different from those experienced before, yet it has the external appearance of something familiar. Think of COVID.
- Shotgun-galore practices subject multiple patients to unnecessary tests and to potentially harmful procedures and interventions that inevitably follow, further inflating the costs of healthcare.
- Perpetuating myths and unmerited traditional practices, kitchen sink therapies also coach our patients into expecting both the unreasonable and the unnecessary for the next visit, undermining any accepted standard of care at its very core.
What Next?
A more in-depth discussion of all three phenomena presented would indeed be appropriate, including an investigation into any viable alternatives.
For now, I encourage all trainees to look further into the general and well-researched topic of cognitive errors in emergency medicine.
We should also all strive to practice based on best available evidence and not to be coerced into questionable behaviors by external pressures such as performance metrics that may lurk as false substitutes for quality.
References and Further Reading
- Frye KL, Adewale A, Martinez Martinez CJ, Mora Montero C. Cognitive Errors and Risks Associated with Provider Handoffs. Cureus. 2018;10(10):e3442. Published 2018 Oct 12. doi:10.7759/cureus.3442
- Oliver G, Oliver G, Body R. BET 2: Poor evidence on whether teaching cognitive debiasing, or cognitive forcing strategies, lead to a reduction in errors attributable to cognition in emergency medicine students or doctors. Emerg Med J. 2017;34(8):553-554. doi:10.1136/emermed-2017-206976.2
- Schnapp BH, Sun JE, Kim JL, Strayer RJ, Shah KH. Cognitive error in an academic emergency department. Diagnosis (Berl). 2018;5(3):135-142. doi:10.1515/dx-2018-0011
