On a busy day shift in the emergency department, our seasoned triage nurse comes to me after I finish caring for a hallway patient, “Hey, can you come see this guy in the triage room? His vitals are fine…”. Seemingly unsure, she pauses, “I’m getting an interpreter, and I think he has belly pain, but something just doesn’t seem right.”

Coming into triage, I see a young man—Georgian-speaking—bracing himself with a hand against the wall and holding his lower abdomen. With what I can only describe as a “grimace” by a singular word, his face conveyed so much more. Without talking or touching the patient, the triage nurse picked up on a deep pain. She had subconsciously analyzed the orchestration of 43 facial muscles and the patient’s body language to create an impression, “This patient cannot wait hours in the waiting room to be seen.” Our nurse did not study Paul Ekman’s Facial Action Coding System for Action Units to code “fear” in the patient’s face.¹ Instead, she had a “gut feeling”—a feeling that appears quickly in consciousness, with unclear awareness of the underlying reasons, but is strong enough to act on.²

Coming by different names as “gut feeling,” “gestalt,” or “intuition,” medicine recognizes this skill in clinical decision-making and interpretation. However, it is often juxtaposed against objective evidence—like lab work and imaging—and deemed inferior. Understandably, it is hard to ask a consultant for admission/observation for a patient with unremarkable work-up and vitals just because “I have a hunch.” It’s inherently hard to define and difficult to explain to others. However, gestalt is something emergency physicians use every day and often in conjunction with our objective reasoning. Thus, gestalt should be recognized as a powerful skill to be honed, respected, and coupled often with our objective diagnostics to make timely decisions.

When is Gestalt Helpful?

Gestalt is useful in areas of time-sensitivity and uncertainty. Sound familiar? This is the essence of emergency medicine. Caring for critically ill patients with limited information requires snap assessments and judgements for timely resuscitation and efficient emergency department throughput.

In the age of big data, more information sounds like a boon. However, more data can be extra noise, which is both time-consuming and can be misleading. For example, experienced emergency physicians have great clinical gestalt and accuracy to predict sepsis in critically ill patients at just 15 minutes from patient arrival—more so than scoring tools like the qSOFA, MEWs, and even machine-learning trained artificial intelligence models.³ This clinical judgment is fast—prior to any lab work to help guide a clinician’s suspicions. In such cases, would you wait for a lactate, white blood cell count, bandemia, or other diagnostics to confirm a source of infection before starting antibiotics, fluid resuscitation, and/or pressors? In this study, clinical gestalt is not only fast, but accurate for the benefit of timely resuscitation and intervention.

Should I draw a venous or arterial blood gas before deciding on intubation? Should I wait for imaging results or labs for this trauma patient before deciding on starting blood or placing a chest tube? Does that normal troponin and ECG obviate the need for cardiology consultation for my patient with a concerning story for acute coronary syndrome? Procrastination to prevent risk—for fear of misdiagnosis or incorrect treatment—can contrarily increase risk for the patient. Furthermore, a contradictory data point can erroneously result in inaction (‘analysis paralysis’) or, worse, steer care in an incorrect direction.

Arguably, emergency medicine has the broadest range of patient presentations among all specialties coupled with a high degree of uncertainty. For example, syncope is a complaint with broad diagnostic uncertainty and up to 40 percent of the time there is no clear diagnosis for syncope.⁴ However, emergency physicians rely on gestalt to predict outcomes well. The externally validated Canadian Syncope Risk Score (CSRS) demonstrated no better prediction of 30-day adverse outcomes for “very low” and “low” risk patients than clinical judgment alone of emergency physicians.

In the diagnosis of pulmonary emboli, experienced emergency physicians’ gestalt performed similarly to prediction rules of PERC, Revised Geneva, and Wells.^5,6 Emergency physicians are not necessarily drawing troponins (only about 50 percent of patients in the syncope comparison study even had a troponin), calculating ECG axis, or asking every patient about hemoptysis as warranted by these decision instruments. Yet, they obtain similar results and arguably with less logistical and time hassle than these decision instruments warrant. Undifferentiated or vague complaints require gestalt to pursue a differential diagnosis and disposition.

Honing Gestalt

1. Recognize gestalt as a critical component to decisions

Gestalt is a skill that everyone uses, but seldom appreciates. Part of this problem stems from the misconception: Gestalt is antithetical to rationality. “Going by feeling” or intuition is often misconstrued as irrational or not dependable. However, gestalt is almost always at play in many decisions and works with reasoning to create rationality. Deliberate reasoning is crucial to make an educated decision, but there are unaccounted factors—our gestalt—that play into overall impression. Although we may reason that the elevated troponin is an NSTEMI, it is our gestalt that helps develop an overall impression that this is a Type I vs. Type II NSTEMI. We should respect this gut feeling as a contributory entity. Afterall, this gut feeling is built from years of perspectives, our unique experiences, and hidden cues.

We should respect our own and others’ gestalt assessments. This means deviating from protocols (when warranted) for sepsis because your initial impression is the patient would be harmed with that much fluid. This means listening to our teams’ words when they say, “I’m worried” or “something just doesn’t fit”. Respect this assessment as a valid concern on its own. This means withholding second-guessing our emergency medical services colleagues for their snap decisions in the field with little objective diagnostics.

2. Developing rules of thumb

Heuristics are “rules of thumb” or mental shortcuts to make fast decisions. They are derived from gestalt to help translate practice patterns into concrete methods. Examples would be as below:

• The “walking wounded” triage heuristic in mass casualty incidents—patient able to walk to you can be initially triaged as “green” or “minimal”.
• The “worst first” mental model—thinking of the most dangerous etiologies first.
• The “bundling” heuristic—if I send a troponin, then an ECG is needed.
• The “escalation” heuristic—when a patient is decompensating with vital signs or mental status, two intravenous catheters and monitoring is needed.

3. Learning from the gestalt of others

Watching and asking other experienced clinicians reveal nuanced insights and check your gestalt. Learning gestalt from others is both a passive and active process. Passively, I watch another clinician’s cursory examination of a critically ill patient to determine how they deem a patient “toxic-appearing” or their bedside neurological assessment to activate a code stroke. Actively, I ask them on their patients that we both had the opportunity to see: What abscesses would they prefer to pack? What do you think the prognosis is for that patient we achieved ROSC?

4. Reflective practice

Deliberate testing of your gestalt prior to confirmation or knowing an outcome can help hone your gestalt for successive cases. An example would be categorizing a patient’s likelihood of pulmonary embolism (PE) into percentages or likelihoods (i.e. a 25 percent chance of PE or low likelihood) prior to any testing, and then reflecting on your predictions after diagnosis. Patterns may emerge on your gestalt. While PERC rule does not differentiate between estrogen use vs. immobilization, reflective practice may reveal that a recent ORIF of a lower leg appears to be a more severe risk factor than oral contraceptive use for deep venous thrombosis and PE.

5. Seeking new perspective and knowledge

Seeing perspectives and experiences from within and outside medicine are instrumental to honing intuition. Speaking and learning from other disciplines may give insight into previously unknown perspectives or skills. Peritonsillar abscesses that I traditionally drained may be more appropriate for antibiotics and steroids alone given otolaryngology literature or perspectives from the outpatient experience of my ENT colleagues.⁷ Playing video games—with the ability to understand spatial relationships and toggle control—may be beneficial to understanding directionality during fiberoptic intubation.⁸ The range of perspectives, skillsets, and experiences we have create an environment for expanded gestalt thinking.

Gestalt is a valuable component of our thought process and sometimes, an impetus strong enough to lead us to action. The astute triage nurse—based on gestalt—moved to earlier physician evaluation. The young man ended up having a left testicular torsion for the last three hours. After some fentanyl, applying traction, and “opening the book,” he improved. A testicular ultrasound confirmed restored blood flow. Urology took him for orchiopexy.

Respecting gestalt is owning it as a skill that we can improve. However, respect for gestalt is also recognizing its limitations. Quick judgements are prone to inaccurate and harmful bias. Having a recent patient with an aortic dissection may make one pursue excessive “aortic dissection” workups with their subsequent chest pain patients (recency bias). One may overlook counter examples or conflicting evidence in search of confirming evidence for their initial impression (confirmation bias). One may inadequately assess and treat pain in a Black patient with a sickle cell crises (implicit bias).⁹ It’s imperative to create “cognitive pauses” to assess our clinical gestalt and ask ourselves: Is detrimental bias involved?

In addition, gestalt is only as strong as a foundation of clinical knowledge. If you haven’t learned it, you can’t diagnose it or treat it. It’s important to continually learn changing clinical foundations and review evidence-based clinical knowledge. The subconscious has been an important part of medical decision making for millennia and should continue to be integral to our clinical care.

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

References

1. Donato G, Bartlett MS, Hager JC, Ekman P, Sejnowski TJ. Classifying Facial Actions. IEEE Trans Pattern Anal Mach Intell. 1999;21(10):974.
2. Gigerenzer, G. (2007). Gut feelings: the intelligence of the unconscious. New York: Penguin Books.
3. Knack SKS, Scott N, Driver BE, Pet al. Early Physician Gestalt Versus Usual Screening Tools for the Prediction of Sepsis in Critically Ill Emergency Patients. Ann Emerg Med. 2024;84(3):246-258.
4. Thiruganasambandamoorthy V, Kwong K, Wells GA, et al. Development of the Canadian Syncope Risk Score to predict serious adverse events after emergency department assessment of syncope. CMAJ. 2016;188(12):E289-E298.
5. Freund Y, Cachanado M, Aubry A, et al. Effect of the Pulmonary Embolism Rule-Out Criteria on Subsequent Thromboembolic Events Among Low-Risk Emergency Department Patients: The PROPER Randomized Clinical Trial. JAMA. 2018;319(6):559-566.
6. Penaloza A, Verschuren F, Meyer G, et al. Comparison of the unstructured clinician gestalt, the wells score, and the revised Geneva score to estimate pretest probability for suspected pulmonary embolism. Ann Emerg Med. 2013;62(2):117-124.e2.
7. Battaglia A, Burchette R, Hussman J, Silver MA, Martin P, Bernstein P. Comparison of Medical Therapy Alone to Medical Therapy with Surgical Treatment of Peritonsillar Abscess. Otolaryngol Head Neck Surg. 2018;158(2):280-286.
8. Ankay Yilbas A, Canbay O, Akca B, et al. The effect of playing video games on fiberoptic intubation skills. Anaesth Crit Care Pain Med. 2019;38(4):341-345.
9. Anderson KO, Green CR, Payne R. Racial and ethnic disparities in pain: causes and consequences of unequal care. J Pain. 2009;10(12):1187-1204.