What about GPs’ gut feelings when faced with dyspnea and thoracic pain?

Friday, August 12th, 2016

Invited comment:

In a letter to the editor of the Annals of Family Medicine, Marie Barais, Erik Stolper and Paul Van Royen give their comments on an article comparing the results of the Wells rule with the Gestalt. Their main criticism is that what the authors called Gestalt did not differ from what they measured in item 2 of the Wells rule. The first step in the diagnostic reasoning process of GPs, when faced with a patient with dyspnoe and thoracic pain, may be a gut feeling that there is something wrong. The Gestalt is the next step: recognizing a possible pattern of pulmonary embolism and estimating its likelihood. The third step is applying the Wells rule.

See the Letter to editor

Comment Ruling Out Pulmonary Embolism in Primary Care: Comparison of the Diagnostic Performance of “Gestalt” and the Wells Rule

In their article “Ruling Out Pulmonary Embolism in Primary Care: Comparison of the Diagnostic Performance of “Gestalt” and the Wells Rule”, Janneke M.T. Hendriksen et al. compared the accuracy of “Gestalt” and the Wells rule for ruling out Pulmonary embolism (PE) in suspected cases in primary care (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4868561/). Combined with d-dimer testing, both Gestalt, which uses a cut-off of less than 20%, and the Wells rule, which uses a score of 4 or lower, are safe for ruling out PE in primary care. The number of patients who need to be referred for further testing was substantially lower when using the Wells rule instead of gestalt probability (efficiency = 45% vs 25%), as well as when using the stepped approach, that is combining gestalt, the Wells rule and D-dimer testing (efficiency = 47%).  This conclusion is certainly helpful for daily practice in primary care.

However, what do the authors understand by gestalt since a clear definition is missing in the article? They described gestalt as an “implicit physician’s estimate” and asked the GP participants to rate an estimated probability of PE being present using a visual analogue scale ranging from 0% to 100% for consecutive adult patients seeking care with symptoms raising suspicion of PE. What they called “gestalt measurement” was in fact the assessment with a Visual Analogic Scale of the probability of one pre-defined pathology. In our view, the authors measured the same diagnostic process with the gestalt scale and item 2 of the Wells score, i.e. that there is “no alternative diagnosis” than PE with a high impact on the final score. Moreover, for the theorists who developed the concept, gestalt is a holistic top-down approach with pattern recognition as opposed to the atomistic approach where each element is individualized [1, 2].

In fact before using any prediction rule oriented toward a particular diagnosis, the GP should have some suspicion of PE and it is precisely this initial stage which is unclear.

The major point is the perspective of the PE diagnosis directly put to the GPs. What is crucial in daily practice? Before using any prediction rule oriented towards this particular diagnosis, the GP should think about PE or have some suspicion of PE when faced with the patient’s symptoms and it is precisely this initial stage which is unclear. Which elements drive the practitioner to suspect PE in the first place? In our qualitative study, the main determinants of PE suspicion were the absence of indicative clinical signs for diagnoses other than PE, a sudden change in the condition of the patient, the GP’s experience of previously failing to diagnose PE, and a sense of alarm preventing diagnostic error in missing PE [2]. We think that gut feelings, and the sense of alarm, in particular, do intervene at an early stage in diagnostic reasoning. This occurs earlier than Gestalt as it was described in this study. The sense of alarm is an uneasy feeling about a patient’s health status, even though he/she has found no specific indications yet [3]. So, according to the definition of Gestalt theory, the sense of alarm is a holistic approach to the case, leading to a prime hypothesis. Gestalt as an estimation of the probability of a PE diagnosis is used at the second stage: the GP is already considering PE – he did recognize the pattern of PE after asking questions to the patient- and focused on this one diagnosis, as a reflective process following a possible sense of alarm. The Wells rule intervenes as a third step in diagnostic reasoning: when considering this hypothesis, which individual elements consolidate or contradict the PE diagnosis?

We consider the study to be important in diagnostic reasoning within the process of PE diagnosis. However it did not address the crucial question of the suspicion of PE, at an early stage, in daily primary care practice. Gut feelings act as a compass steering the practitioner through the diagnostic process. It may prevent the GP from too early excluding an important working hypothesis. A holistic approach to diagnosis seems to be more sensitive than specific but, at this stage of the diagnostic reasoning process, it is important not to lose sight of hidden diagnoses too early. Gut feelings and then Gestalt could prevent diagnostic errors at an early stage in the diagnostic process. We want to know the contribution of the gut feelings in that process. So we designed a study aiming to calculate the diagnostic test accuracy of the sense of alarm when applied to dyspnoea and chest pain, using the gut feelings questionnaire [3, 4, 5]. The findings of this study complete the description of the sense of alarm by contributing an essential quantitative component.

1. Cook C. Is Clinical Gestalt Good Enough? J Man Manip Ther. 2009;17: 6–7.

2. Cervcellin G, Borghi L, Lippi G. Do clinicians decide relying primarly on Bayesian principles or on Gestalt perception?

3. Barais M, Barraine P, Scouarnec F et al. The accuracy of the general practitioner’s sense of alarm when confronted with dyspnoea and/or thoracic pain: protocol for a prospective observational study. BMJ Open. 2015;5:e006810. doi:10.1136/bmjopen-2014-006810

4. Stolper E, Van Royen P, Van de Wiel M, et al. Consensus on gut feelings in general practice. BMC Fam Pract.2009;17;10:66. doi: 10.1186/1471-2296-10-66.

5. Stolper CF, Van de Wiel MW, De Vet HC, Rutten AL, Van Royen P, Van Bokhoven MA, Van der Weijden T, Dinant GJ. Family physicians’ diagnostic gut feelings are measurable: construct validation of a questionnaire. BMC Fam Pract. 2013 Jan 2;14:1. doi: 10.1186/1471-2296-14-1. PMID: 23281961; PMCID: PMC3565882.