The transculturality of ‘gut feelings’. Results from a French Delphi consensus survey

Abstract

Background: General Practitioners (GPs) sometimes base their clinical decisions on ‘gut feelings.’ Research into the significance of this phenomenon with focus groups and a Delphi consensus procedure in the Netherlands provided a concept of ‘gut feelings:’ a sense of alarm, a sense of reassurance and several determinants. The transculturality of ‘gut feelings’ has been examined briefly until now as the issue is complex.

Objective: To determine whether a consensus on ‘gut feelings’ in general practice in France could be obtained. Using a similar Delphi consensus procedure and the same six initial statements as in the Netherlands, and compare the French results with the seven final Dutch consensual statements.

Method: Qualitative research, including a Delphi consensus procedure after a forward-backward translation (FBT) of the initial Dutch statements of ‘gut feelings.’ A heterogeneous sample of 34 French expert GPs participated. FBT of the final French statements was undertaken for a content comparison with the Dutch.

Results: After three Delphi rounds, French GPs reached agreement on nine statements. Many similarities have been found between the Dutch and the French defining statements, with reservations concerning the ‘sense of reassurance,’ which French GPs seemed to feel more cautious about.

Conclusion: ‘Gut feelings’ are a well-defined concept in France too. The Dutch and the French consensual statements seem very close. The transculturality of the concept is confirmed, which is a new indicator that ‘gut feelings’ are a self-contained concept.

Keywords::

• general practice
• medical decision
• qualitative research
• Delphi technique
• gut feelings

KEY MESSAGE:

• The French definition of ‘gut feelings’ is very close to that found in the Netherlands.

• The similarity between French and Dutch results confirms transcultural ‘gut feelings.’

• The concept is determined by a sense of alarm, a sense of reassurance and nine consensual statements to define them.

INTRODUCTION

General practitioners (GPs) are often faced with poorly defined syndromes and uncertain diagnoses and yet they must resolve these problems rapidly (1–3). Taking decisions in these situations of uncertainty is a characteristic part of medical practice (2,4). ‘Gut feelings’ can, therefore, play a determining role in the diagnostic process. Until recently, very poor information existed on the diagnostic and prognostic value of ‘gut feelings’ in general practice (5–8).

A qualitative study using four focus groups of 28 GPs in the Netherlands distinguished two types of ‘gut feelings:’ a sense of alarm and a sense of reassurance, along with six statements (9). As regards non-analytical diagnostic reasoning, it showed that ‘gut feelings’ could act as a compass in situations of uncertainty in everyday practice.

In the case of a sense of reassurance, the GP feels confident about the prognosis and the therapy even without a diagnosis. A sense of alarm causes the GP to feel that something is wrong even without objective arguments. He, therefore, wants to intervene to prevent serious health problems for the patient.

To validate the defining elements established by the focus groups and make the concept applicable in practice, a Delphi consensus procedure was carried out with 27 opinion leaders and experts in the Netherlands and Flanders (10). It showed, at the end of the consensus procedure, that a consensus existed in seven consensual defining statements on ‘gut feelings’ in general practice. A sense of alarm was defined as an uneasy feeling experienced by the GP, who was concerned about a possible adverse outcome for the patient, without having objective arguments to account for this: something did not fit in, ‘something is wrong here.’ This activated the GP's diagnostic reasoning as he felt the need for some kind of intervention to prevent serious health problems. A sense of reassurance was defined as a feeling of security experienced by the GP about the patient. Although he is not yet certain about the diagnosis, everything fits in. They concluded that the sense of alarm and the sense of reassurance were well-defined concepts.

Finally, nine semi-structured interviews with German GPs gave more detailed descriptions of the sense of alarm (11,12), which were added to the Dutch alarm statements. They described a feeling of discomfort in the presence of the patient even where there was no possible adverse outcome: ‘something does not fit in, here.’

The concept of ‘gut feelings’ with its two aspects of a sense of alarm and a sense of reassurance has been established in the Netherlands, Flanders and Germany. It seems to be transcultural, but those three languages all originate from the same proto-Germanic roots (13). If it were possible to extend it to another linguistic group such as the Roman languages, the transculturality of the concept could be revealed. In a preliminary study, the sense of alarm seemed to be transcultural (14). That study was done with a simplified questionnaire and had a low response rate. It was, therefore, necessary to confirm the results for the sense of alarm and establish them for the sense of reassurance.

Questions were designed to determine which consensus on ‘gut feelings’ in general practice could be obtained in France, using the Dutch results and the same consensus methodology, and compare the French results with the Dutch consensus.

METHODS

Participants

For the Delphi procedure sample, the participants were randomly selected from the list of Associated Teachers in General practice in France (obtained from the French National Association of Teachers in General Practice). That list contained 131 members. The Dutch procedure was performed with 27 participants in the final Delphi round. To get the same number at the end of the French consensus, one out of every four members was randomly selected. Only one respondent refused to participate because he disagreed with the concept of ‘gut feelings.’ The result was a heterogeneous sample of 34 French GPs associated with 25 universities.

Delphi procedure

The aim of a Delphi consensus procedure is to determine to what extent experts agree about a given issue (15–18). This anonymous process was organised via a series of statements sent by e-mail to all participants, inviting them to rate their agreement on a scale from 1 (total disagreement) to 9 (total agreement). Participants were asked to comment on their ratings, especially when rating 7 or lower. According to the chosen Delphi method, a statement was accepted when at least 70% of the participants agreed on a rating of 7 or higher (19). The research team (two French researchers) for further reformulation then used the comments and ratings of the statements, which had not been accepted. After each round, the ratings and comments were summarized and incorporated into a new version of the document. The participants then re-rated their agreement with each statement, with the possibility of changing their rating in the light of the group's response. The rounds were repeated until consensus was reached or appeared impossible.

Initial statements

To ensure the feasibility of this research, the starting point was the six initial statements from the first Dutch-Flemish work obtained through their first focus group research. These statements were translated into French using a careful performed forward-backward translation procedure. It was performed by a Dutch and English speaking French general practitioner involved in clinical research for the forward procedure. The Maastricht University's official French/Dutch translator assumed the backward procedure. The result of the backward translation was submitted for validation to the original Dutch-Flemish researchers who had created the definition contained in the six statements. Those statements are listed in Box 1.

Box 1. Definition statements on ‘gut feelings’ initially submitted to the French participants.

1. A GP's ‘sense of reassurance or alarm’ is mostly related to their degree of certainty about the prognosis of the complaints.

2. The ‘sense of reassurance or alarm’ has very little to do with formulating working hypotheses or establishing diagnoses.

3. A ‘sense of alarm’ implies that a GP is worried about a patient's health status, even though he or she has as yet no objective argument for this; it is a sense of ‘there's something wrong here.’

4. A ‘sense of alarm’ means that some form of intervention seems necessary to prevent imminent serious health problems.

5. A ‘sense of reassurance’ means that a GP feels secure about the prognosis, even though there are no objective arguments for this: everything fits in.

6. A ‘sense of reassurance’ implies that a GP feels secure about whether and what therapy needs to be started.

•Additional proposition (as in the Dutch study). Maybe you would like to formulate your own definition of gut feelings. Would you like to write it down below?

Comparison of French results with Dutch-Flemish results

To compare the results of both French and Dutch Delphi procedures, translation of the final French statements into English was necessary using a similar forward-backward translation (FBT) method. The original Dutch-Flemish researchers (two Flemish Dutch GPs, one with academic responsibilities and one PhD student) and the French researchers (two French GPs, one with academic responsibilities and one trainee) carried out a comparative content analysis of those final statements.

RESULTS

Participants

Thirty-four professors or associate professors in General Practice participated. They came from all over France and all regions were represented. Three dropped out in the first round (failed to respond), one in the second round (failed to respond) and one in the third round (failed to respond). The study was completed by 29 respondents.

Delphi rounds

Three rounds were needed to reach consensus. During the entire procedure, 10 statements were submitted to the participants: six original ones and four new ones that emerged from the comments (summaries of these comments are below). There was a high level of response (only five GPs out of 34 did not return their responses). These GPs were contacted by telephone to ascertain why they had dropped out. In each case, shortage of time was cited as the reason. A large number of comments were made (see Table 1).

Table 1. Synthesis of the three Delphi rounds.

	Round1		Round2		Round3
Statement	n = 31	CP (%)	n = 30	CP (%)	n = 29	CP (%)	Result	Round
1	25	26.7	NAs	NAs	NAs	NAs	Rejected	1
1a	NA	NA	13	86.2	A	A	Accepted	2
1b	NA	NA	9	93.1	A	A	Accepted	2
2	21	35.5	13	75.9	A	A	Accepted	2
3	17	90.3	A	A	A	A	Accepted	1
4	22	41.9	11	86.7	A	A	Accepted	2
5	22	48.4	13	63.3	6	79.3	Accepted	3
6	1	77.4	A	A	A	A	Accepted	1
7	NA	NA	10	80	A	A	Accepted	2
8	NA	NA	11	82.1	A	A	Accepted	2
9	NA	NA	16	66.7	M5	M5	Rejected	2

n, number of participant's comments for accepted or rejected statement; CP, consensus percentage (number of votes equal or above 7) for accepted statement should be above 70 % to be accepted; NA, not applicable: statement was not formulated at that time; NAs, not applicable: statement split in 1a and 1b; A: accepted (no further round); M5, merged with statement 5.

First round. In the first round, statements 3 and 6 were accepted. Statement 1 had to be split up into statements 1a and 1b to be more specific. Indeed, participants stressed emphatically the cohesion between the patient's complaints and the GP's observations.

Statement 4. Participants noted that a sense of alarm was not always linked to serious health problems that might occur in the patient: the GP just did not understand the whole situation at that point. In regard to statement 5, the GPs were very cautious concerning the definition of the sense of reassurance and felt ‘that it was dangerous for a GP not to go further in his diagnostic reasoning when lacking objective arguments.’ Participants were also invited to add their own statements about ‘gut feeling:’ three new statements emerged that were submitted to the participants in the next round (statements 7, 8 and 9). Statement 7 again emphasized pattern recognition, but added the element of ‘deep conviction,’ and the element of informed consent. Statement 8 was a concise formulation about the position of ‘gut feelings’ in GPs’ diagnostic reasoning process. Statement 9 emphasized that the sense of reassurance needed for the GPs to ‘feel confident about the lack of serious outcome for the patient.’

Second round. Statements 1a, 1b, 2, 4, 7 and 8 were accepted. In regard to statement 5, participants made two major comments: the word ‘diagnosis’ seemed too conclusive: they preferred ‘management plan.’ They also stressed emphatically that, if the GP felt secure about his patient, he had to continue to be very cautious and be prepared to review his position if in any doubt. They stressed the point that this ‘secure’ feeling was only temporary. Comments made on statement 9 showed a great similarity to statement 5 and that is why the two statements were merged into one only new statement 5.

Third round. Statement 5 was accepted. After three rounds, a consensus was reached on nine statements (see Table 2) compared with the Dutch consensus on seven statements (see Table 2).

Comparison of French results with Dutch-Flemish results (Table 2)

The conclusion of the comparative content analysis of the final statements between Dutch-Flemish and French researchers proved very favourable. As shown in Table 2 some small differences could be found, but in general, there was considerable similarity.

Table 2. French results and comparison with Dutch-Flemish results.

French consensus results	Dutch–Flemish consensus results	Discussion
Statement 1a: The ‘sense of alarm’ is particularly related to the non-concordance between the patient's complaints and the signs (semiological or clinical) found by the GP.	Statement 1a: A ‘sense of alarm’ means that a GP perceives an uneasy feeling, as he/she is concerned about a possible adverse outcome.	French GPs explain how the sense of alarm arises, while the Dutch GPs describe the sense of alarm. The latter one can be found in the French statement 3. Meaning is almost the same.
Statement 1b: The ‘sense of reassurance’ is linked to the similarity between the patient's symptoms and the clinical picture found by the doctor.	Not found	This is the exact opposite of the French statement 1a. The French GPs seem more cautious about the sense of reassurance and want to describe it in a different way from statement 5. In the Netherlands, the Departments of General Practice are parts of universities and did a great deal of research over a 40-year period. This process is just starting in France. Perhaps the French GP has lower self-confidence.
Statement 2: The ‘sense of alarm’ triggers the general practitioner's diagnostic process, pushing him to formulate and test hypotheses linked to serious illnesses that could have serious consequences for the patient.	Statement 2: A ‘sense of alarm’ activates the diagnostic process by stimulating a GP to formulate and weigh up working hypotheses that might involve a serious outcome.	Similar
Statement 3: ‘The sense of alarm’ implies that the GP mistrusts the patient's health status even though he does not have any objective arguments—something is wrong here.	Statement 3: A ‘sense of alarm’ implies that a GP worries about a patient's health status, even though he/she has found no specific indications yet; it is a sense of ‘there's something wrong here.’	Similar
Statement 4: The ‘sense of alarm’ means that the GP will go further in his diagnostic or communicational process until he understands the clinical picture and/or the patient's fears.	Statement 4: A ‘sense of alarm’ means that, if possible, the GP needs to initiate specific management to prevent serious health problems	French GPs seem more focused on comprehension and diagnosis while Dutch GPs express the need for action in initiating specific management. In the end, French comprehension and understanding seem close to Dutch specific management.
Statement 5: The ‘sense of reassurance’ means that the GP, even though he lacks objective data, feels sufficiently sure of himself to be able to propose an adapted management plan. This certitude is, however, temporary, as he needs to be cautious and prepared to review his position if in any doubt.	Statement 5: A ‘sense of reassurance’ means that a GP feels secure about the further management and course of a patient's problem, even though he/she may not be certain about the diagnosis: everything fits in.	Again, the French GPs emphasize the ‘principle of cautiousness.’ See comments on statement 4.
Statement 6: The ‘sense of reassurance’ means that the GP is sufficiently sure of himself to start the patient's treatment.	Statement 5: A ‘sense of reassurance’ means that a GP feels secure about the further management and course of a patient's problem, even though he/she may not be certain about the diagnosis: everything fits in.	French GPs emphasize the treatment aspects but apart from that this statement is close to the Dutch–Flemish statement 5.
Statement 7: The ‘sense of alarm’ and the ‘sense of reassurance’ indicate the GP's conviction, both concerning the coherence of the clinical picture and the quality of the doctor patient relationship, within the clinical picture.	Statement 8: The ‘sense of reassurance’ and the ‘sense of alarm’ constitute a dynamic element in a GP's diagnostic process.	French GPs added the term conviction in an intimate way and the need to share the decision with the patient. That difference could arise from a constraint of the health's system as the French system has a ‘gate keeping’ system maintained by GPs for only the last five years.
Statement 8: The ‘sense of alarm’ and the ‘sense of reassurance’ help the GP in his decision making in uncertain clinical situations.	Statement 8: The ‘sense of reassurance’ and the ‘sense of alarm’ constitute a dynamic element in a GP's diagnostic process.	The meaning is the same. French GPs stayed within a concise formulation, whereas Dutch GPs were more informative about the decision-making process.
Statement 4: The ‘sense of alarm’ means that the GP will go further in his diagnostic or communicational process until he understands the clinical picture and/or the patient's fears.	Statement 9: A ‘sense of alarm’ will decrease as the diagnosis and the right management become clearer.	Was only developed by Dutch GPs to describe the whole process. It is quite similar to French statement 4, which is merely expressed the other way round.

DISCUSSION

Main findings

The study reached an overall consensus on a precise description of two types of ‘gut feelings,’ a sense of alarm and a sense of reassurance, in the diagnostic reasoning of French and Dutch-Flemish GPs. The transculturality of the ‘gut feelings’ concept between proto-Germanic and Roman languages has now been revealed.

A ‘sense of alarm’ means that the GP is worried about his patient while lacking objective arguments: ‘something is wrong here.’ The non-concordance between the patient's complaints and the GP's observations makes the GP go further in his diagnostic reasoning until he has understood the situation. The ‘sense of alarm’ makes the GP use more hypothetico-deductive reasoning.

A ‘sense of reassurance’ means that the GP feels sure about the management of the patient even while lacking objective arguments. He is not worried: everything fits in. This secure feeling is still temporary and needs to be reviewed if in any doubt. This was particularly important to the French GPs who emphasized the need to remain alert even when they felt a sense of reassurance. It is considered that uncertainty avoidance may be stronger in Latin culture, which could explain this response (20).

The statements showed different aspects of ‘gut feelings:’

• The significance of a ‘sense of alarm’ and of a ‘sense of reassurance’ and how they occur (statements 1a, 1b, 3, 5 and 7). The notions of pattern recognition and congruity between the complaints and the medical observations play a key role.

• The vague and uneasy feeling concerning the prognosis while the GP is lacking objective arguments (statement 3).

• The consequences of the sense of alarm (statements 2, 4 and 6) emphasize the key role of ‘gut feelings’ in decision-making.

• Statement 5 introduces the notion of cautiousness concerning the sense of reassurance, as it is temporary. The participants have largely emphasized these aspects of cautiousness and permanent questioning.

• Finally, statement 8 is a concise formulation of the place of ‘gut feelings’ in the GPs’ diagnostic reasoning: it acts like a dynamic compass in situations of uncertainty.

Role of expertise

French GPs who participated in the Delphi procedure were all associated with a university. They were all experienced GPs and even considered being experts. A question arises: is there a link between ‘gut feelings’ and experience? At this stage, the assumption is the more extensive the GP's experience is, the more reliable his ‘gut feelings’ should be. They provide a useful compass, as coping with uncertainty is part of expertise.

The Dutch team pointed out in their first research that ‘a sense of alarm’ was a feeling that something was wrong, that something needed to be done. This sensation was also frequently experienced by young GPs (3). The German team noticed that the ‘sense of alarm’ arose when the GP felt that there was something unusual about the patient. For them, experience formed a background and ‘gut feelings’ caused the GP to reflect on his procedures (11). Those results again highlight the question of the role of experience in ‘gut feelings.’

Strengths and limitations

The Delphi consensus technique has been used widely in health care research, but its validity and reliability have been the topic of many debates. However, guidelines for the use of this consensus technique were followed (17,18,21). The main advantage of the Delphi technique is that consensus is reached without any leader effect. It also provides additional data, which emerges from the expert's knowledge. The main disadvantage is that it is only a consensus, in that it qualifies but does not quantify a concept.

Although this Delphi consensus procedure only included 34 participants, they were all well-known experts from all over France with wide experience, both as GPs and researchers or medical educators. The extent to which they represent general practice and their ability to implement the findings could be influential in deciding how far we can generalize from the results and again pinpoint the role of experience in ‘gut feelings.’

Implications

The results of the Dutch consensus procedure (9,10) were confirmed. ‘Gut feelings’ should be considered a useful transcultural concept for general practice that works as an alerting or a reassuring factor. It causes the GP to pause and reflect on his practice (5,8,22,23). It could be linked to a theoretical framework for decision making and judgement in situations of uncertainty as described by Kahneman. This could lead to an increased awareness of the qualitative conditions required if intuitive expertise is to be useful to the patient (24).

The precise description of ‘gut feelings’ undertaken in the Netherlands, Flanders, Germany and France may allow further research on this decision-making compass for the GP (23).

To evaluate the role of expertise, two studies using the same Delphi consensus procedure are currently being undertaken with GP trainees and GPs, who are not involved in university programmes. If these two populations tend towards a similar consensus, this would indicate that the concept is self-contained, independent of the individual GP's level of experience and that the concept of ‘gut feelings’ could be extended to an average population of GPs.

It is now time to validate the concept's usefulness for patients in a more quantitative way (does it help or not the patient's health outcomes?). Further research is needed for validation data (incidence, prevalence, specificity, sensitivity, reproducibility).

Conclusion

‘Gut feelings’ are recognized by expert GPs (whether they are native speakers of a Germanic or a Roman language) as playing a substantial role in their decision-making process. More collaborative European research is needed to continue with the validation process.

ACKNOWLEDGEMENTS

The authors thank Alain Mercier (MD), Laurence Coblentz-Baumann (MD), Dominique Gras (MD), Isabelle Aubin-Auger (MD) for their help in collating the participants’ comments.

Declaration of interest: The authors declare no conflict of interest, and the study had no other source of funding than the department of general practice of Brest University (France).