Her intuition was her favorite superpower.
Experienced surgeons have always believed that they are good judges of surgical risk. Some assert that their “eyeball” test upon first meeting a patient is as good—or better—than objective measures that require physiologic testing. The more judicious among us may at least wait until the end of our history and physical examination rather than the beginning to declare our judgement.
How good is surgical intuition, however? Do we have objective evidence to support or refute the value of a respected surgeon's sense about a patient?
THE EYEBALL TEST: INTUITION IN GENERAL
It is reasonable to believe that some individuals are born with greater powers of intuition than others. Iconic military leaders such as Napoleon were said to have coup d’oeil, literally “stroke of the eye,” able to glance at a prospective battlefield and know immediately where to optimally deploy resources. Malcolm Gladwell, in his best-selling book Blink: The Power of Thinking Without Thinking, asserts that first impressions can and should be educated and controlled.1 That is, we need to recognize and control the subtle influences and biases that can undermine our best instincts. If we can do this, such predictions can be quite accurate.
Shtulman makes a similar point in his book, Scienceblind: Why Our Intuitive Theories About the World Are So Often Wrong,2 when he points out that professional scientists reason more accurately than do nonscientists not because they have relinquished the nonscientists’ misconceptions but because they have learned to inhibit those misconceptions. A number of authors describe a distinction between intuition and deliberative thought,3 or unconscious thought and conscious thought,4 or clinical method and actuarial method.5 Brunelli, in assessing thoracic surgeon decision making, concluded that simple choices produce better results after conscious thought but that complex matters require unconscious thought.
Making a similar point in metaphor, the educational theorist Donald Schön wrote that straightforward problems of a technical nature—solvable through the application of research-based theory—were the high, hard ground overlooking a swamp. “In the swampy lowland, messy, confusing problems defy technical solution.”6 The clinician may use evidence-based, technical knowledge but not be able to put his or her rationale into words. Schön terms this “knowing-in-action” and it is best developed over time through feedback, or “reflection.”7,8
Dawes et al,5 however, in their review of clinical versus actuarial judgment, assert that guideline or rule-based decisions are consistently better than clinical ones. They point to studies of the Goldberg Rule, an algorithm based on the Minnesota Multiphasic Personality Inventory, showing its superiority in distinguishing neurosis from psychosis compared with that of 29 judges, some PhD psychologists with extensive experience. Even with 4000 practice judgements and later with knowing the outcomes of the Minnesota Multiphasic Personality Inventory the judges were inferior to the Rule. The authors point out that the actuarial method equaled or surpassed the clinical in nearly 100 social science studies as well. They also conclude that clinical experience may be better for rare events but in general guidelines or rules eclipse instinct.
Common sense, and all of the authors above with the exception of Dawes et al, would agree that experience helps one develop better intuition. Crebbin proposes that the expert moves effortlessly along the continuum between subconscious intuition and conscious analysis, increasingly employing pattern recognition as experience increases9 (Fig. 1). Ferguson, using clinical vignettes and videos rather than actual patients, found that surgeons and trainees differed in frailty assessment and risk estimation, with the eyeball factor helping most when vignettes and videos were discordant.10
Is there evidence that supports the power of surgical intuition in general as well as the role of experience in it?
POWER OF THE GLANCE: SURGICAL INTUITION IS GOOD
For over a decade a group of New Zealand and British surgeons has studied the accuracy of surgeons’ clinical prediction of postoperative complications. Their surgeons assessed the risk of a major complication on a visual analog scale before and after surgery. In 2007 they reported that although a surgeon's prediction alone was not sensitive or specific enough to reliably describe surgical risk, his or her prediction did make an important contribution when incorporated into a multifactorial model.11 Their more recent study was more favorable to surgeons, finding their clinical assessment comparable to the Physiological and Operative Severity Score for the Enumeration of Mortality and Morbidity (POSSUM), with an area under the receiver operating characteristic curve of 0.822.12 There was, however, consistent over-prediction of complications in low-risk patients. Feedback did improve performance but level of surgeon experience did not, possibly because the predictions were made only for operations appropriate to the doctors’ level. Interestingly, Markus also compared surgeons’ gut feeling to the Portsmouth POSSUM (P-POSSUM), finding that the former was a good predictor of postoperative outcomes and that the P-POSSUM over-predicted; the surgeons were questioned after the operation, however.13
Glasgow et al14 studied experienced surgeon prediction of postoperative risk after consultation and before operation, comparing their predictions to those of a statistical model. Both experienced surgeons and model were good and were in good agreement. In a comparison of surgeon estimation versus statistical risk model for coronary bypass surgery, the surgeons routinely overestimated risk of 30 day mortality but their estimate more accurately reflected 180-day mortality since some patients died between 30 and 180 days.15 Another study of mortality risk after open heart surgery concluded that subjective assessment was accurate at the extremes of risk but was inaccurate for intermediate risk levels.16 Ferguson also reported that surgeons were better at assessing frailty at the extremes, but this study employed videos of patients and not actual outcomes.17 The selective power of instinct, the fact that subjective judgement is better in some areas compared with others, is further emphasized by Smith and McCahill18 who reported that Surgical Oncologists could accurately estimate their patients’ survival time after surgery but not their degree of symptom palliation.
One could argue that these studies do not describe a pure eyeball test, as the surgeon is armed with all of the preoperative data and examination findings, and in 1 study with the findings at operation.
SCHÖN'S SWAMP: SURGICAL INTUITION IS MEDIOCRE OR POOR
Farges et al19 studied 946 patients undergoing liver resection by 26 hepato-biliary surgeons in 9 teaching hospitals. They found that surgeons’ preoperative anticipation of patients’ postoperative outcomes had low sensitivity, specificity, and accuracy, did not differ by center, and were not improved by surgeon experience; their prediction after completion of the operation was equally poor. Objective statistical prediction models were better and were not improved by the addition of surgeon intuition.19 Jain also studied statistical models versus eyeball test in 5099 cardiac surgery patients. Both methods modestly overestimated mortality but the model did better.20
Neurosurgeons did no better in Sagberg's study of expected functional level following intracranial surgery, generally overestimating this level at 30 days after operation; consultants did no better than residents. All clinicians were poor at predicting survival.21 General surgeons lacked predictive accuracy for anastomotic leakage after completing an anastomosis22 and plastic surgeons, as individuals, were poor at estimating aesthetic outcome after breast reconstruction surgery; experienced surgeons performed no better.23 Trauma surgeons were poor at predicting the need for massive transfusion based on clinical gestalt.24
In 3 studies involving virtual patients, subjective judgement was mediocre or poor. Timmermans found that surgeons did not distinguish between high and low-risk patients sufficiently when presented with written vignettes.25 Ferguson reported that experienced surgeons were better than trainees in predicting complications after lung resection but the accuracy of all was only fair.26 Healy found that both Surgery and Internal Medicine residents similarly overestimated risks of postoperative morbidity and mortality and, unsurprisingly, the surgery residents were more confident in their predictions.27 The results of these studies do not support the confidence in one's judgement that even experienced practitioners manifest.28
Again, these studies cannot speak to the power of pure, isolated surgical intuition—the power of the glance—unaided by knowledge of physical examination findings, laboratory data, imaging, and even operative findings. No one researched the intuition of the surgeon upon first greeting the patient.
All surgeons would like to believe that their clinical intuition is excellent, and their patients would like to believe this even more, but the evidence does not support such a global conclusion. A few statements, however, are reasonable:
- Surgical intuition can be good but this is far from guaranteed, as the evidence is mixed.
- Intuition may play a greater role in complex or confusing situations as opposed to straightforward or purely technical ones.
- Feedback about a surgeon's predictions improves his or her future ability to predict.
- A great deal of evidence supports the power of statistical models; some have concluded that the actuarial method always equals or surpasses clinical intuition.
- Surgeons are more confident in their intuition than the evidence justifies.
Nothing can be concluded about the power of pure, raw surgical intuition—when the surgeon first meets the patient, unaided by data, examination findings, or operative findings—as this has not been studied. Conclusions about surgeons’ coup d’oeil must wait for another day.
1. Gladwell M. Blink: The Power of Thinking without Thinking. New York: Little, Brown and Company; 2005.
2. Shtulman A. Scienceblind: Why Our Intuitive Theories About the World Are So Often Wrong. New York: Basic Books; 2017.
3. Miller C. Medicine is not science: guessing the future, predicting the past. J Eval Clin Pract
4. Brunelli A, Pompili C, Salati M. Patient selection for operation: the complex balance between information and intuition
. J Thorac Dis
5. Dawes RM, Faust D, Meehl PE. Clinical versus actuarial judgment
6. Schon D. Educating the Reflective Practitioner. San Francisco: Josey-Bass; 1987.
7. Schon D. The Reflective Practitioner: How Professionals Think in Action. New York: Basic Books; 1983.
8. Mickleborough T. Intuition
in medical practice: a reflection on Donald Schon's reflective practitioner. Med Teach
9. Crebbin W, Beasley SW, Watters DA. Clinical decision making: how surgeons do it. ANZ J Surg
10. Ferguson MK, Farnan J, Hemmerich JA, et al. The impact of perceived frailty on surgeons’ estimates of surgical risk. Ann Thorac Surg
11. Woodfield JC, Pettigrew RA, Plank LD, et al. Accuracy of the surgeons’ clinical prediction of perioperative complications using a visual analog scale. World J Surg
12. Woodfield JC, Sagar PM, Thekkinkattil DK, et al. Accuracy of the Surgeons’ clinical prediction of postoperative major complications using a visual analog scale. Med Decis Making
13. Markus PM, Martell J, Leister I, et al. Predicting postoperative morbidity by clinical assessment
. Br J Surg
14. Glasgow RE, Hawn MT, Hosokawa PW, et al. Comparison of prospective risk estimates for postoperative complications: human vs computer model. J Am Coll Surg
15. Cornwell LD, Chu D, Misselbeck T, et al. Predicting mortality in high-risk coronary artery bypass: surgeon versus risk model
. J Surg Res
16. Pons JM, Borras JM, Espinas JA, et al. Subjective versus statistical model assessment
of mortality risk in open heart surgical procedures. Ann Thorac Surg
17. Ferguson MK, Thompson K, Huisingh-Scheetz M, et al. Thoracic surgeons’ perception
of frail behavior in videos of standardized patients. PLoS One
18. Smith DD, McCahill LE. Predicting life expectancy and symptom relief following surgery for advanced malignancy. Ann Surg Oncol
19. Farges O, Vibert E, Cosse C, et al. “Surgeons’ intuition
” versus “prognostic models”: predicting the risk of liver resections. Ann Surg
20. Jain R, Duval S, Adabag S. How accurate is the eyeball test
?: a comparison of physician's subjective assessment
versus statistical methods in estimating mortality risk after cardiac surgery. Circ Cardiovasc Qual Outcomes
21. Sagberg LM, Drewes C, Jakola AS, et al. Accuracy of operating neurosurgeons’ prediction of functional levels after intracranial tumor surgery. J Neurosurg
22. Karliczek A, Harlaar NJ, Zeebregts CJ, et al. Surgeons lack predictive accuracy for anastomotic leakage in gastrointestinal surgery. Int J Colorectal Dis
23. Sun CS, Reece GP, Crosby MA, et al. Plastic surgeon expertise in predicting breast reconstruction outcomes for patient decision analysis. Plast Reconstr Surg Glob Open
24. Pommerening MJ, Goodman MD, Holcomb JB, et al. Clinical gestalt
and the prediction of massive transfusion after trauma. Injury
25. Timmermans D, Kievit J, van Bockel H. How do surgeons’ probability estimates of operative mortality compare with a decision analytic model? Acta Psychol (Amst)
26. Ferguson MK, Stromberg JD, Celauro AD. Estimating lung resection risk: a pilot study of trainee and practicing surgeons. Ann Thorac Surg
27. Healy JM, Davis KA, Pei KY. Comparison of internal medicine and general surgery residents’ assessments of risk of postsurgical complications in surgically complex patients. JAMA Surg
28. Croskerry P, Norman G. Overconfidence in clinical decision making. Am J Med
2008; 121 (5 suppl):S24–S29.