Deliver a Difficult Medical Diagnosis: Interviews Analysis to Design a Particular Training Support

Copyright: © 2016 Saint-Dizier de Almeida VA. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Introduction
This paper deals with patient-physician consultations in contexts of serious medical diagnoses. It presents a research supported by the French "Ligue Contre le Cancer". The goal was to design a Web product which is particularly addressed to future physicians. The product design process comes from ergonomic psychology. Based on the presentation of the design process, we show how ergonomic psychology, with its posture, allowed designing a particular product -while classical training focuses on physicians' practices and uses an injunctive format, we employ a descriptive format and presents the collective activity in consultation from different points of view. In order to design the content of web product, we mainly used findings arising from discourse analyses. To collect this corpus (patient-physician consultation of serious medical diagnosis in simulated situations), we used role-playing technique. Based on the theoretical-methodological framework used to study activity accomplishment in consultation from various angles, product provides a particular content in accordance to medical recommendations (behavioral prescriptions).

Methods and Materials
Defining the scope The web-based package was supplied by the French Ligue Contre le Cancer through the Centre Alexis Vautrin (Cancer Hospital, Nancy, France) and the multimedia service of the University of Nancy (Vidéoscope). The overall project sought to develop a website available to all, but targeting students in medicine and early career physicians in particular. The site was designed to address the doctorpatient relationship. The site was already under construction when we joined the project. There were four sections. The first dealt with laws and regulations, the second to defense mechanisms that could emerge among patients and physicians, the third to media, patients' associations and health networks, and the fourth gave general information on the pathologies, and on patients' needs and expectations. The site was driven by resources extracted from legislation, written material and interviews with physicians and one psycho-oncologist; it was based on professional experience and/or the literature.
Our task as researchers specialized in the psychology of communication was to design a fifth section on interpersonal communication. We set out to design a product that (i) addressed students in medicine or early career physicians, (ii) dealt with the delivery of a difficult diagnosis, (iii) was easy to use, (iv) was in multimedia format, (v) was useful, interesting and appealing for its beneficiaries.
To this end, the content of diagnosis delivery training programs was analyzed. By examining programs summarized in educational resources and training materials we identified what content not to replicate. Studies in linguistics and the psychology of linguistics on doctor-patient consultations [1][2][3] and our knowledge of ergonomics revealed dimensions of diagnosis delivery training that were absent from the literature, or had received scant attention, such as the structural dimension of the activity (operational structure of delivery), the relational dimension (the roles and status relationships in place) and the linguistic means by which recommendations for appropriate behaviors are accomplished.
This study is original as it lends diagnosis delivery new dimensions hitherto unexamined in the literature.

Developing the training support
To design the training support, we mainly used findings from interviews analyses. For ethical reasons, and to get the best from our interviews, we used a role-playing technique. The physicians involved in this role-play were at the end of their career; they had an extensive experience in the medical diagnosis delivery; they said that they used an empathetic approach and managed emotions. They knew that interviews will be analyzed and later used to design on-line training support.
The role of the male patients is supported by a student in the fifth year of a psychology curriculum; the role of the female patient was played by a psychologist practicing in an Oncology service.
The delivery interviews (patient-physician consultations of serious medical diagnosis in simulated situations) were transcribed and validated [4]. circumstances in the 'here and now' takes precedence over scenario monitoring [5].
• As opposed to the studies carried out by Health addressing ordinary doctor-patient interviews, our results show that the high status of dominance associated with physicians and the correspondingly low status associated with patients were significantly less pronounced or even almost reversed [3]. Physicians are at the service of the patients, and listen to them. The patients are active. They show they are listening, show their understanding or their difficulty in understanding, and they share their knowledge and express their need for information, their malaise, their worries and fears, and the reasons behind their worry. Some of these behaviors may initiate phases and command the structure of discourse. Respecting the recommendations for behaviors to exhibit and to avoid in the literature (cf. 2.1.) helped to implement an empathic approach. A physician shows empathy by having a comprehensive, nondirective attitude which results in a modification of the roles and status relationships at the relational level [6].
In this type of interview, the physician put a lot of effort into the modal register reflecting his/her concern with regard to the patient's emotional state.
• An original item not defined in the medical literature emerged from the analysis: in potentially stressful situations, before informing the patient, the physician inquired about his/her mental and emotional state in order to plan appropriate discourse.
• In designing training to impart communication skills that can help to improve patients' experiences during a diagnosis delivery, the following elements are considered fundamental: • The structure of the diagnosis delivery does not reflect the strict implementation of a scenario. The physician must be ready to backtrack. If the patient interrupts the physician to ask a question about a point brought up at the beginning of the interview ("Ah yes, excuse me, I wanted to ask you... just now when you said... "), the physician agrees to suspend the intervention. Once the response has been made, the physician comes back to the point in hand.
• The patient is active in the process. The hierarchical asymmetry (dominant/dominated) is mitigated [3]. The fact that the physician displays attentiveness causes the patient to be more forthcoming and express fears or lack of understanding.
• Being listened to requires identifying what the patient will understand; this requires inquiring about the patient's level of knowledge; to comfort and support the patient, and dedramatize the situation, it is necessary to accurately identify the object of his/her fears.
• Identifying the intentions of others in communication cannot be taken for granted. For example, during the first diagnosis delivery interview, just because a patient asks: "How long have I got?" does not mean that a straight answer to the question is what is actually wanted. It may be that the patient is seeking reassurance. These different aspects of content appear in the training.

Results
The internet product.
The content of the different sub-categories: The analyzed interviews were gathered as part of patient information collection. The interview analyses addressed three different levels of activity: (i) the structural operating level, (ii) the operating/behavioral level and (iii) the relational level.
The study of the structural operating level identified the successive phases of each interview (e.g. presenting the results of a medical examination, ensuring compliance with a diet, allaying a patient's fears, agreeing on subsequent therapy, setting the next appointment, etc.). The study of this level aimed to show that the difficult diagnosis delivery interviews did not reflect the strict application of the physician-patient consultation scenario as classically taught in medicine to manage ordinary interviews [2]. The scenario for ordinary consultations takes the form of a sequence of phases conducted by the physician: • Phase 1: opening, • Phase 2: identifying the problem (questioning), • Phase 3: examination, • Phase 4: diagnosis, • Phase 5: discussing the diagnosis, additional examinations, • Phase 6: recommendation, • Phase 7: closing [2].
We postulated that a strict application would be incompatible with medical recommendations for difficult diagnosis deliveries. Instead, we expected these phases to be repeated and some to be initiated by patients. For example, if a patient communicates a problem of comprehension on a point brought up in an earlier phase ("Oh, but that's not what I understood just now"), the physician will not hesitate to go back and clarify the information given earlier. In this example, it is the patient (and not the physician) who introduces a new phase for the purpose of clearing up a point presented in a preceding phase.
The study of the operating/behavioral level identified the recommendations that the physician makes through his or her behaviors, e.g. when the physician declares, "Don't worry, we've got it in time", we identify the accomplishment of the recommendation, i.e., attenuate an emotional shock. This study was essential to delimit the sequences that illustrate how communication skills can be materialized in discourse.
The study of the relational level identified the positions held in the course of the activity. Through this analysis, the aim was to show that the recommendations such as listening, being attentive to the patient, helping the patient to overcome fear, etc. will cause the physician to show the patient more care and consideration. These behaviors should thus induce a less asymmetrical social relation than that seen in classical interviews. We would expect the physicians, through their behaviors (they do not interrupt, they listen to, inform and reassure the patient), to hold a position in which they make no claim to power status.
The results of the interview analysis show that: • The interviews conducted were unlike ordinary interviews reflecting the consultation scenario taught to students in medicine. There was a recurrence in the different phases of the interview. Following instructions when conducting delivery (listening to the patient, ensuring that he/she understands, paying attention to his/ her psychological state, keeping emotions under control, etc.) led to interviews with structures that revealed scenario flexibility. Thus in Suchman's words, in this type of consultation, the processing of social This sub-category focuses on the multifunctional dimension of communication. Communication is not only about sharing information, it also makes it possible to complete a task jointly, share mental states, build relationships, and express and manage emotions. It focuses on the fact that the doctor-patient interview is not a monologue occasionally ratified by the patient, but a genuine interaction where the physician and the patient are active and play complementary roles in a situation. This interview also suggests a communication contract where the classically high asymmetrical social relationship is mitigated. It is notably through the designation of roles (identical formulations as highlighted below) that we sought to mitigate this asymmetry.
The patient's roles: Make enquiries (on the disease, checkups, treatment…), inform (talk about experiences, his/her family history…), express his/her fears, etc.

2nd category: Dialogues
The selection of a process: This draws on studies that argue that it is not possible to provide ready-made formulas with regard to diagnosis deliveries owing to the multiple determinants specific to each consultation; this is to avoid suggesting that formulas in this field could be provided.
A progressive procedure: This involves recalling a recommendation with regard to the diagnosis delivery process: this must be progressive so as not to shock the patient, but also because an appropriate behavior requires the physician to inquire about the patients' knowledge, his/her desire for information, his/her fears and what they are based on.
One of the sequences used makes it possible to show how the physician will bring the patient, using different questions, to communicate what he/she understands by cancer, thereby enabling the physician to adjust his/her discourse.

The structure of the dialogue:
The third sub-category shows that the different phases of discourse do not correspond to the strict application of a scenario. The scenario as taught at university is only a guide for conducting interviews. The physician thus plans his/ her interview minimally, and the structure of the interview largely depends on the interventions, and notably the emotional reactions of the patient.

3rd category: Behaviors
Target: This entails showing that the patient's words are rich and can inform different registers and guide further action.

Specific features:
The action taken will depend on the illness (curable or incurable, known or uncertain prognosis, etc.) and the degree of progress in the diagnosis delivery process (whether the checkups have already been carried out, whether the diagnosis is already established, whether the treatment has already been initiated, etc.)

Invariants:
The physician must bear in mind that irrespective of the interview and the advancement of the process, whatever is thought to be accomplished at a given time may later be unaccomplished. From one interview to another, the patient may have obtained additional information with regard to his/her pathology, treatment, and so on; from one occasion to the next, his/her desire for information can have changed.

4th category: Skills
The selected sequences show how recommendations that are understandable, reassuring, de-dramatizing, etc. can be accomplished discursively.
Be understood: This sub-category informs on how to be accessible and clear. In this category, the sequence where the physician uses the medical term "PSA" is used, suggesting that disseminating information must not lead the physician to renounce the use of technical terms.
Be heard: This suggests that for the physician's words to be heard, the patient must be willing to hear them. It is the physician's role to inquire about what the patient hears in the "here and now".
Manage emotions: This category shows, using interview excerpts, how emotions can be expressed (through verbal, paraverbal and nonverbal forms) and how these can be managed discursively. The sequences illustrate the accomplishment of operational actions (such as reassuring or de-dramatizing).
The content of each category is often illustrated -when relevantby one or more sequences extracted from the interviews. The sequences that are essential in our scheme can be read or listened to. They are described and commented on (Table 1).

D63b
"What do you mean, cancer" P64 "Er, as you've just said, it's…cells that…keep on growing relative to, well…er…everything…to me."

D65
"That grow on their own, you mean." P66 "Right, and..." D67 "...and that start growing, well, a bit independently." P68 "Er... so it's true that often when they talk about cancer they talk about cells that have a problem of degen..., what is it, degeneration."  Commentary as it appears in the section: In the first part of the interview, the physician does not use the word "cancer" but talks about cells that grow abnormally. But in P62b, the patient deduces from the physician's discourse that it is cancer: "its cancer?" The term that has been studiously avoided has been pronounced. In response, the physician is not going to meet the patient's request, i.e., say "Yes it's cancer", or "It might be". No. The physician will start working on the mental picture the patient has of the word "cancer" in order to identify the ideas the term carries. This joint work will enable the physician to find out at what level to intervene and begin a process of de-dramatization that will be efficient, because it will be based on the patient's thought world. Looked at more closely, the physician's first attempt (D63b) is not satisfactory because the patient remains in the medical register (P64). The second attempt is efficacious, in particular because the physician invites le patient to shift from a medical register to an emotional one via D69. This invitation lets the patient formulate fears and the object of those fears. In this case, what the patient fears is not the relatively common association of cancer with death, but one of the treatments against cancer, namely chemotherapy and its consequences (hair loss). This work will then enable the physician to find an appropriate discourse reviewing the different treatments currently available and stating that chemotherapy is not inevitable. We must remember that this de-dramatization phase is essential. It should help the patient to be forward-looking and commit to the therapeutic process more calmly. Without this initiative, the physician would doubtlessly have emphasized that "cancer" does not always mean "death", which would have led the patient to take a morbid view of the situation against an inner thought world that did not elicit such a view. And the patient's actual fears would have persisted and not been allayed by the physician.
The formalization of the content is in line with ergonomic recommendations for electronic documents. This section is accessible at: http://www.infos-patients.net/ (section: "role-playing"). For additional information on the process underlying the development of this section [4].
The multimedia model of the section was presented at a meeting with Professors of medicine at the Centre Alexis Vautrin and representatives from the French Ligue Contre le Cancer. The presentation was praised by the Ligue and the physicians validated its content. The sponsors therefore decided to incorporate it into the site as a fifth section termed "role-play" in order to avoid misleading users about the nature of the sequences used to illustrate the section's content.
It is also noteworthy that the global site is currently supported by the National Conference of the Deans of the Faculty of Medicine, the National Order of Physicians, the French Ligue Contre le Cancer, the UMVF (the Francophone Virtual Medical University), the Centre Alexis Vautrin, and the Faculty of Medicine of the University of Nancy. It was also certified on 5 January 2010 by the Health on the Net foundation in collaboration with the French National Authority for Health in accordance with law No. 2004-810 of 13 August 2004.

Conclusion
Having to deliver a difficult medical diagnosis to a patient is a situation that every physician will have to face sooner or later. Preparing and training to manage this type of interview is thus essential for both patients and physicians. For patients, it has been shown that a delivery correctly made has a positive effect on their commitment to the therapeutic process and on their propensity to adopt an active forward-looking attitude [7][8][9][10]. For physicians, being trained in managing the delivery interview helps them approach these grueling and dispiriting situations with greater equanimity [11]. We can distinguish two forms of difficult diagnosis delivery training: classical and interactive. The first form imparts information on legislation, defense mechanisms (Ruszniewski, what should be done during the interview (managing emotion, answering questions, etc.) or should not be done (interrupting or making a judgment on the patient, etc.) [12][13][14][15]. The second form focuses on acquiring the communication skills needed to smoothly conduct a delivery interview [16,17]. The first form has a low implementation cost (one-way communication), unlike the second form, which generally involves courses lasting several days, and requires trainers versed in role-play management and collective discussion. Accordingly, with the support of the French Ligue Contre le Cancer, we developed a training website devoted to delivering a difficult diagnosis. It has the following features: (i) it is freely accessible, being web-based, (ii) it is a self-training package and so it needs no trainers, (iii) an online training session takes at most half an hour, and so it is not costly or time consuming for the trainees, and (iv) its content is based essentially on the results of analyzing physician-patient interviews obtained by role-play.