How do junior doctors working in a multi-cultural context make sense of somatisation ? 1

Background: Physicians’ explanatory models of commonly encountered clinical problems reflect not only formal medical training, but also personal and professional experience. In situations where formal training is absent, the clinical context of informal learning is likely to have an important influence on junior doctors’ evolving knowledge and practice. Objectives: The purpose of this study was to explore junior doctors’ explanatory models of somatisation in the absence of formal training, and in particular the influence of working with a large proportion of immigrants (asylum seekers and refugees) on their understanding and practice. Methods: Semi-structured interviews were conducted with all 14 junior doctors working at the Geneva University Hospitals general medicine outpatient clinic. Interviews explored junior doctors’ definitions, understanding and management of somatisation. Interviews were tape-recorded, transcribed verbatim and analyzed for key themes. Results: In the absence of formal training, somatisation evoked considerable uncertainty and frustration in terms of diagnosis and management. Junior doctors’ understanding of somatisation was heavily influenced by their contact with immigrant patients. They primarily attributed somatisation to psychosocial stresses related to migration and tended to label certain socio-cultural groups as more prone to somatisation than others. Conclusion: It is important to recognize that contextual factors have an important influence on physicians’ evolving explanatory models for commonly encountered problems, especially in the absence of formal training. Results from this study suggest that formal training in management of somatisation should be offered in order to broaden junior doctors’ understanding of somatisation and avoid cultural stereotyping.

Background: Physicians' explanatory models of commonly encountered clinical problems reflect not only formal medical training, but also personal and professional experience.In situations where formal training is absent, the clinical context of informal learning is likely to have an important influence on junior doctors' evolving knowledge and practice.
Objectives: The purpose of this study was to explore junior doctors' explanatory models of somatisation in the absence of formal training, and in particular the influence of working with a large proportion of immigrants (asylum seekers and refugees) on their understanding and practice.
Methods: Semi-structured interviews were conducted with all 14 junior doctors working at the Geneva University Hospitals general medicine outpatient clinic.Interviews explored junior doctors' definitions, understanding and management of somatisation.Interviews were tape-recorded, transcribed verbatim and analyzed for key themes.
Results: In the absence of formal training, somatisation evoked considerable uncertainty and frustration in terms of diagnosis and management.Junior doctors' understanding of somatisation was heavily influenced by their contact with immigrant patients.They primarily attributed somatisation to psychosocial stresses related to migration and tended to label certain socio-cultural groups as more prone to somatisation than others.
Conclusion: It is important to recognize that contextual factors have an important influence on physicians' evolving explanatory models for commonly encountered problems, especially in the absence of formal training.Results from this study suggest that formal training in management of somatisation should be offered in order to broaden junior doctors' understanding of somatisation and avoid cultural stereotyping.

Key words: somatisation; explanatory models; junior doctors
Somatisation has been described as "a tendency to experience and communicate somatic distress and symptoms unaccounted for by pathological findings, to attribute them to physical illness, and to seek medical help for them" [1].Although patients often present in primary care with medically unexplained symptoms [2,3], somatisation is given little attention in medical school and postgraduate training [4,5].As a result, physicians often feel unsure how to respond in such situations [6,7].
Several studies have shown that physicians develop their own working models for the problems they encounter regularly [8][9][10].These "explanatory models" are influenced not only by their formal medical training but also by factors such as personal experience and characteristics, the type of professional literature they read, the physician's clinical population, and the economic and political organization of the health care system in which the physician works [11,12].In the absence of formal training for a specific medical problem, such factors may take on a particularly important role in determining physicians' explanatory models.
The purpose of this study was to explore the explanatory models for somatisation of junior doctors working in a hospital outpatient general medicine clinic in Geneva, Switzerland.These doctors received no prior formal training in somatisation and most of them had only very limited contact with somatising patients before working at our clinic.For most junior doctors, it also was their first encounter with outpatient care and their clinical population was characterized by a large number of immigrants, asylum seekers and refugees.We were particularly interested in exploring how -in the absence of training -contact with socially and culturally diverse patients influenced their explanatory models of somatisation.

Setting
The study was conducted at the Geneva University Hospital general medicine outpatient clinic, known locally as the "Policlinique".The Policlinique provides care to patients from diverse social and cultural backgrounds, and a review of medical records conducted by the first author showed that between October 2001 and March 2002 the Policlinique saw patients representing 127 different nationalities.Only 30% of patients had Swiss nationality.
The Policlinique offers a one-year residency training program in general medicine for junior doctors planning to work as general practitioners.Sixteen junior doctors are employed yearly, all at the end of their residency training, which consists of at least two to three years on internal medicine hospital wards.For most junior doctors, the Policlinique is their first experience with general medicine outpatient care.

Data collection and analysis
The study consisted of semi-structured interviews with 14 junior doctors who had been working for at least 9 to 12 months during the academic year October 2001-September 2002 at the Policlinique.Two junior doctors who had started in April 2002 were excluded, since it was felt that their experience with patients was too limited.These 14 junior doctors included 11 women and 3 men, ranging in age from 28-45 years.Twelve were Swiss nationals who had studied medicine in Switzerland and trained in Swiss hospitals for at least 2 to 4 years.Two were over 40 years of age and had already worked as general practitioners in France and Algeria for several years before joining the Policlinique.All agreed to take part in the interviews.
Interviews were conducted using a short interview guide consisting of open-ended questions followed by prompts (table 1).Interviews explored junior doctors' understanding of and responses to somatisation, and junior doctors were encouraged to narrate their personal experiences and talk about issues that were most relevant for them.All interviews were conducted in French by the first author who worked part-time as a clinical supervisor during the study period.Interviews occurred between June and September 2002 and lasted from 45 to 60 minutes.Interviews were tape-recorded, transcribed and analyzed using Winmax ® software for qualitative data analysis [13].All translations of junior doctor' utterances are the authors'.
Analysis of transcripts was conducted by both authors, and followed the "editing style" approach described by Miller and Crabtree [14].Each interview was read first and then each author independently identified passages that reflected the doctors' explanatory models of somatisation.Particular focus was placed on identifying the doctors' ideas concerning the definition, causes, symptoms, risk factors and treatment of somatisation.The authors discussed each interview, identifying and resolving any discrepancies.NJP coded all interviews, and then PH reviewed coded segments.Disagreements were discussed and texts were recoded where necessary.Once all texts were coded, data display tables were created to facilitate the examination of similarities and differences across respondents and to identify cross-cutting themes.
The study was approved by the Geneva University Hospital Ethics Committee.

What is somatisation?
While all of the doctors said they frequently encountered somatisation in their work at the Policlinique, they were first hesitant and sometimes embarrassed to provide a definition of somatisation.
"What is somatisation?(Silence) Well, in fact it's a question that I've often asked myself.Well, I think I understand a little, I don't have a very clear definition but ..

." (R12)
Nonetheless, they all described somatisation as the presence of vague, repetitive symptoms in the absence of evident objective physiological or biological malfunction that might explain the patient's suffering.It was considered a "diagnosis by elimination".All of the physicians defined somatisation as the physical expression of psychological or social distress, or as a sign of underlying psychiatric morbidity.Somatising patients were often described as being unable to verbally express their emotions or make the link between psycho-social events and physical symptoms.
"So, for me it's physical pain that translates another type of pain, bereavement or a state of depression, or any other pain that can't be expressed through the psychic route, and is expressed through the body."(R7) However, most of them (10/14) were afraid of "missing something" and unsure how to proceed with patients for whom no physiological or biological problem could be found.
" the end you do the exams anyway, and you go around and around."(R3).

Why do patients somatise?
Most of the doctors (11/14) spontaneously referred to migrant patients (generally refugees, asylum seekers seasonal workers) when asked about the causes of somatisation.
"They're foreign patients.That is, they're from Africa or Kosovo or they're traumatised by war." (R5) "It's migrant patients who seem to me to present the most somatisation."(R4) They tended to classify somatising patients according to their symptoms' patterns, origin, sex or socio-cultural background.Some groups of patients were more prone to develop somatisation than others.
"I don't have that impression with Africans, for example.They come, they have a health problem, it's clear.OK, they have their problems, they have their PTSD, but it's rare that they have pains.For example, with people from the Balkans, it's often ... they have chronic pains ... without any reason, I mean" ..

. (R3) "With the Hispanics, it's almost always bone and joint or abdominal pains. And then, with regards to Kosovars, it's almost always headache or epigastric pain ..." (R7)
Individuals with low levels of education were thought to be more prone to somatisation because they lacked the ability to understand and verbally express their emotions.
"Yea, often it's people who don't have a degree, a sufficient level of education or culture that would allow them to express their suffering differently, or, to help themselves in a different way."(R2) Junior doctors believed migrant patients developed somatisation more often than others because of their increased psycho-social vulnerability.Commonly mentioned predisposing factors included war, rape, exile or grief.Notions of loss and trauma due to migration were very present.

"Well, I'd say that I'm thinking mainly about all those patients from the Balkans who lived through the war, who have sometimes lived through rape, who have lived through the assassination of a loved-one -whether or not they witnessed it -who have had to flee and sometimes have been exiled, who have had to leave everything behind, abandon everything, who often don't know what has happened to their family." (R1)
"So, is it for them a way of existing, because there's nothing else but that.They've lost everything, they've lost their house, they've lost their previous life, they '

ve lost family members, they've lost, I'd say, respect for their bodies in the case of women who have been raped, and well, they've got nothing else to live for, to express, because there's nothing else anymore." (R10)
Only a few of them (5/14) acknowledged that their context of work may influence their perception of somatising patients and even fewer (2/14) considered somatisation to be a universal expression of distress.
"Well, we're biased because we have a lot of asylum seekers, but it's true that it's often asylum seekers [that somatise]."(R8) "Well, it's all mixed -the very young, middle aged, different categories too, diverse economic classes ..

. but it's true that among people coming from the Balkans, there are a lot [who somatise]." (R14)
What should be done?
In terms of management, most junior doctors (10/14) reported no prior specific training in how to care for somatising patients.They felt lost and helpless and did not like the "on-the-job" nature of their training.
"No, I feel that we're really ill-prepared to deal with these patients because first, in my case, no one ever told me about [these sorts of problems] ..

. well, I knew they existed but we don't have any book, or, I don't know, a course to know how to care for these patients. I have the impression that I just go on my feelings, and I often make mistakes ..." (R1)
However, all of the doctors stressed the importance of helping patients to "make the link" between their psychosocial problems and their physical symptoms, and creating an environment in which patients could express their emotions and worries.

"I try to put things in context, to explore thingsas well as I can, because often it's hard to get started, to see how [the pain] handicaps them in their day-to-day life, to learn how it started and whether it's linked to ... I have a lot of people who have had some really awful experiences ... you manage to retrace things a bit, and even if they don't make the link, at least you talk." (R9)
They (8/14) observed that their patients' health problems improved when they obtained a job or a permit, learned French or when their family situation improved.

Wishes for future training
When asked about ideas to improve training in somatisation, half of the junior doctors wished more information about geopolitical, cultural and legal context of migrants and asylum seekers, as if the first step in providing appropriate treatment to somatising patients was to know more about their patients' cultural and legal background.They focused on difficulties linked to cross-cultural communication issues, ignorance of patients' background and asylum administrative procedures. "

Discussion
The results suggested first that, in the absence of training, junior doctors experience much uncertainty in terms of definition, diagnosis and management of somatisation.Second, their explanatory models of somatisation are heavily influenced by the patient population in which they first encounter somatisation.Junior doctors attributed somatisation primarily to psychosocial stresses related to migration and tended to classify certain socio-cultural groups as more prone to somatisation than others.Similarly, when asked about wishes for training, they tended to talk interchangeably about strategies to manage somatisation and those to manage migration or asylum related problems.
According to experts, formal training about somatisation requires at least five components: embracing of a biopsychosocial model, ability to identify relevant psychosocial issues, mastery of basic psychiatric diagnostics, specific interviewing skills and understanding of counter transference [4].In Switzerland as in many other countries, junior doctors tend to train almost exclusively in hospital settings before practicing outpatient general medicine, and their training emphasizes the importance of evaluating and treating symptoms according to well-established strategies and remaining alert for unusual disease presentation [16].Such training is less appropriate for many kinds of complaints seen in primary care which tend to reflect the burden of illness rather than specific diagnostic entities [17,18].In many ways, somatisation exemplifies what junior doctors face when switching from hospital work to ambulatory care: a reorientation of medical practice -away from the classification and treatment of disease and towards the care of the whole patient.Without appropriate training, this sort of professional re-orientation can at times be experienced negatively [17].Not surprisingly, junior doctors reported uncertainty and frustration when dealing with somatising patients like general practitioners elsewhere [6,7,[19][20][21][22].
Several studies have demonstrated that primary care physicians who participate in cognitive-oriented training programmes focused on assessment, treatment and management of somatisation increased their comfort, understanding, and use of interviewing skills with somatising patients [23][24][25].Main elements of such training programmes (reattribution model) focus on making the patient feel understood, broadening the agenda and negotiating a new understanding of the symptoms which includes psychosocial aspects.
In the absence of training, the uncertainty experienced by junior doctors may have emphasized the way they relied on the context of work to construct their explanatory models of somatisation.Junior doctors' explanatory models focused on somatisation as a problem of migrants, especially of those coming from the Balkans or having experienced war-related trauma.Junior doctors tended to develop a rather narrow and oversimplified understanding of somatisation derived from the contact with their clinical population.It has been shown that lack of confidence and knowledge in understanding and managing patients' suffering can lead to the use of unpleasant labels or stereotypes [20,26].Social cognition theory claims that all people create and use categories in order to simplify and structure the amount of information they must process [27].Social categories are used to quickly classify individuals, often in an unconscious and involuntarily way and, in the absence of guidance, such a categorization process can lead to the formation of oversimplified social perceptions and judgments [28].While it is true that somatic symptoms are common among refugees [29,30], they are also the most common clinical expression of emotional distress worldwide [31,32].Although doctors' tendency to develop cultural stereotypes about somatisation may result from lack of formal training, such findings are of concern since we believe that the circumstances and context of junior doctors' initial contact with somatising patients is likely to shape their knowledge, attitudes and practices on a long term basis.
We feel that our findings suggest the need for earlier and better structured education about somatisation in medical school and post-graduate training in order to help doctors in training move from a simplistic and categorical view of knowledge and values to a more complex pluralistic perspective [33], increase their tolerance to uncertainty and avoid stereotyping.
Our study has several weaknesses.First, the small size of our sample may not have allowed us to identify all potential sources of variation in junior doctors' perceptions and experiences.Second, our study is limited to a single outpatient clinic and our findings may not be transferable to other contexts.However, in Switzerland as in many other countries, university outpatient clinics tend to provide care to vulnerable populations from diverse social and cultural backgrounds and are important training centers for junior doctors planning to become general practitioners.Therefore, we think that our findings may be relevant for other similar settings.However, it would be important to further explore the influence of clinical context and patient charac-teristics on junior doctors' evolving explanatory models in other contexts.Third, the fact that interviews were conducted by a clinical supervisor working at the outpatient clinic may have influenced doctors' narratives.It is likely that doctors would have felt constrained to provide standard definitions of somatisation and to downplay their uncertainty.This was not the case, however, and we had the impression that the doctors were eager to discuss their thoughts and experiences with the interviewer.
Nonetheless, we feel that this exploratory study points to the important influence of clinical context and patient population characteristics on physicians' developing clinical models for somatisation, especially in the absence of formal training.Given the importance of somatisation for primary care medicine, formal training in somatisation should be provided early in physicians' training in order to broaden junior doctors' understanding of somatisation and avoid patient stereotyping.

Swiss Medical Weekly
The many reasons why you should choose SMW to publish your research Med Wochenschr (1871-2000) Swiss Med Wkly (continues Schweiz Med Wochenschr from 2001) Editores Medicorum Helveticorum

Table 1
Interview guide.
We're a bit lost really.It's true that you're always afraid that's it's a real ... a real pain.And so you lose a lot of time, you're hesitant, you try to focus in but then in What I think could help here is ... is to explain to us the context in the Balkans, the war, in fact the whole context of the war, how it came about, what they see, those people."(R4) "Well, there are things, for example, at the cultural level.Know people's cultures better.Have information on, for example, what ... what myths they have, what their ideas are with regards to certain types of pain in their country ... yes, that would help to better understand them."(R7) "If we knew at the beginning who to contact, the telephone numbers, how to reach the social worker, where they get the N permit ... When they (the asylum seekers) arrive, they go here, they go there ...And then you could look up a particular key word, or a major pathology.I don't know, maybe a few words about Srebrenica for example, which is really an important date ..." (R9)