I thought, “This is heavy-handed on the epinephrine.” Intramuscular epinephrine, racemic epinephrine, and now a dirty epinephrine drip* were running wide open through the IV. However, the patient who came from the contrast echo bubble study to our resuscitation bay was still stridulous, diaphoretic, and vomiting. We had this irrational shred of hope linked to the blood pressure cuff. Like its climb, brief oscillation around 90 mmHg, and subsequent deflation, our team felt our hopes mirrored.

BP 78/34. 

At the time, I was a fresh attending working with our chief resident. When our chief resident placed the laryngoscope, the patient quickly became hypoxic, first to 90 percent and then 85 percent. We bagged the patient up and tried again. It led to the same result, but this time, when we tried bagging, resistance was met; he remained hypoxic. The chief resident asserted, while lifting on the laryngoscope blade and cradling the occiput, “I can get it.”

I trusted him and our process, so we persisted. However, I heard the downtrending tone of the pulse oximeter. The heart rate started to soften — we had to cric. I cut into the neck with the scalpel while the pulse oximeter read 62 percent.

Should we have moved to a different modality earlier—supraglottic, delayed-sequence intubation? Should we have aborted the second intubation attempt? Should we have moved to cricothyrotomy earlier?

When is the best time to quit a clinical course of action? And why is it so hard to quit?

Richard Thaler, in 1980, described the “sunk cost fallacy.”¹ In perfectly rational decision-making, we should only consider future costs and benefits and we should only persist if there is a positive expected outcome. However, this is not how people function in reality. Inextricably, we take into account the past investment — what we’ve already spent on an endeavor. This sunk cost can make us persist when there is no possibility for a positive outcome. The time and effort invested in waiting for the delayed CT abdomen/pelvis scan may make one feel a need to continue with the imaging study even when the patient passes a large bowel movement and no longer feels symptomatic. That loss of time, effort, and identity is strong — oftentimes referred to as loss aversion — and is felt twice as intensely as an equivalent gain when scientists have been able to quantify it.²

When we quit, we are not just quitting an action or course but also quitting the associated investment or sunk cost. It is difficult to decouple quitting a course of action from quitting something more — the trust of our patients, our perceived competence in our procedural skills, or our self-image of a resolute physician.

Why Can Quitting Be Beneficial?

Imagine if all decisions you made were final. This would be detrimental in a specialty where constant change is the norm. New information from laboratory tests changes our diagnostic considerations. An incoming critical patient changes our workflow and prioritization.

Quitting, amid these circumstances, creates capacity and opportunity. Although you might have initially performed a therapeutic paracentesis for the stable patient with ascites, deciding to ask interventional radiology to perform the procedure instead can create capacity to prepare for the incoming patient in cardiac arrest.

How to Quit Better

Quitting means not knowing an outcome at times. Although difficult to accept, this is a fact of practicing emergency medicine: We must accept ambiguity. We accept ambiguity to avoid harm, inefficiency, and overreach. For example, a patient presents with chest pain. You have a low pretest probability of pulmonary embolism and acute coronary syndrome based on your initial interview and workup. The opportunity cost of pursuing a definitive workup for a pulmonary embolism (PE) and/or acute coronary syndrome can bring potential harm to that patient and diverts resources and time that could be used on other patients in the department or the hospital. Quitting allows us to avoid the potential harms from unwarranted anticoagulation in a patient with a <2 percent probability of PE and found to have a subsegmental PE or a <2 percent risk of complication for coronary angiography.^3,4

Reframing to focus on the positive gains, rather than sunk costs, is important. Although you may not have achieved the goal of return of spontaneous circulation or pacemaker placement, this shouldn’t overshadow the gains made. The embittered resuscitation you run that you decided to terminate was not for nothing. Perhaps you gained information for the family to give them a potential reason for the cardiac arrest or gave them the opportunity to say goodbye. Aborting a right internal jugular and left subclavian pacing catheter you cannot advance due to calcifications provides crucial information for the electrophysiologist to plan for a femoral access temporary pacer.

Establish Criteria to Quit

Create deliberate criteria to quit, and share this with stakeholders. I establish “kill” criteria — which probably should be renamed in our line of work — of when to quit. During intubation prebriefs with the team, I ask to be alerted when the pulse oximeter drops to 90 percent or notified when one minute has elapsed before we abort the intubation attempt for bag valve mask.

Admittedly, there is limited evidence of when to stop a procedure or terminate a resuscitation. These decisions depend on many factors such as communication with family, workflow, and individual patient circumstances. However, having predefined criteria can prevent blind overcommitment. Set the criteria conservatively and adjust based on your growing experience.⁵

Quit Coach

We are more likely to make rational decisions on behalf of others than for ourselves.⁶ Thus, have a trusted colleague be a “quit” coach, someone who is kind and honest. It could be as gentle a reminder as, “The waiting room is getting busy,” or directly saying, “What is the end point?” when you have been aggressively trying to raise the core temperature of a patient found hypothermic in the water.

In this article, I purposely circumvent the contrasting idea of grit. I believe that as emergency physicians, we become “gritty” individuals during the hidden curriculum found in our daily practice environment and training. Americans laud “grit” as a gold standard, and in a society that appreciates the benefits of persistence, we must ask ourselves, What do we sacrifice when we endure? What harm are we causing?

We can be more tactful in the way that we approach a plan of action through understanding the hubris of persistence, an openness to reevaluating a situation, and employing some strategies of quitting. In a specialty defined by uncertainty, knowing when to quit may be one of the most honest skills we have.

Dr. Koo is a faculty member and an emergency physician at MedStar Washington Hospital Center in Washington, D.C., and St. Mary’s Hospital in Leonardtown, Md.

References

1. Ronayne D, Sgroi D, Tuckwell A. Evaluating the Sunk Cost Effect. JEcon Behav Organ. 2021;186:318-327. Available at: https://doi.org/10.1016/j.jebo.2021.03.029.
2. Tversky A, Kahneman D. Loss Aversion in Riskless Choice: A Reference-Dependent Model. Q.J.Econ. 1991;106(4):1039-1061. Available at: https://doi.org/10.2307/2937956.
3. Freund Y, Cohen-Aubart F, Bloom B. Acute Pulmonary Embolism: A Review. JAMA. 2022;328(13):1336-1345. Available at: doi:10.1001/jama.2022.16815.
4. Sandau KE, Funk M, Auerbach A, et al. Update to Practice Standards for Electrocardiographic Monitoring in Hospital Settings: A Scientific Statement From the American Heart Association. Circulation. 2017;136(19):e273-e344. doi:10.1161/CIR.0000000000000527.
5. Levitt SD. Heads or Tails: The Impact of a Coin Toss on Major Life Decisions and Subsequent Happiness. Rev Econ Stud. 2021;88(1):378-405. Available at: https://doi.org/10.1093/restud/rdaa016.
6. Lin H, Zheng H, Wang F. Do bystanders always see more than the players? Exploring Solomon’s paradox through meta-analysis. Front Psychol. 2023;14:1181187. Available at: doi:10.3389/fpsyg.2023.1181187.
7. Duke, A. Quit: The Power of Knowing When to Walk Away. New York: Portfolio, 2022.

*Editor’s note: A “dirty epinephrine drip” as described by the author is a code dose epinephrine (1 mg 1:10,000) into a 1 L bag of saline and run wide open without a pump. It provides 1 mcg/mL. Through an 18 Ga. IV, this drip will probably run over approximately 30 minutes. When the author doesn’t have enough time for pharmacy to mix a bag, this is how he creates his own epinephrine drip.