General practitioners' experiences of patients with eating disorders

Mia Holtet Aalmen, Jan H. Rosenvinge, Andreas Holund, Gunn Pettersen About the authors

Eating disorders include the conditions of anorexia, bulimia and overeating. In general practice, their prevalence varies from less than one per cent (anorexia) to 16 per cent (overeating and sub-clinical conditions) (1). The likelihood of a favourable long-term prognosis (2, 3), thereby preventing high mortality and reducing the potential for severe somatic injury (46), increases if the condition is detected, diagnosed and treated at an early stage. Here, the GP plays a key role, but there are at least four challenges associated with this.

The first challenge is communicative ‘noise’ in the doctor-patient relationship. Many patients describe their eating disorder only indirectly by referring to diffuse psychological, gastrointestinal or gynaecological afflictions (7, 8). Their number of consultations tends to be above average, but fear of stigmatisation and of not being understood by the doctor, as well as limited insight into their own need for treatment (9, 10) help explain why it may take up to four or five years before the patient brings up the topic of eating disorders in the consultation (11). The second challenge consists in how psychiatric comorbidity (12) can ‘mask’ the eating disorder and delay diagnosis and treatment. This applies to anxiety-related, depressive and personality disorders in particular (13, 14). The third challenge is the gradual transition between sub-clinical eating disorders and the statistically speaking more frequent variations in people’s relationship to food, their body and weight (2, 3). For example, dieting will not be an unambiguous sign of an eating disorder. All these three types of challenges can explain why no more than approximately every other patient with an eating disorder is identified by their GP (15). The GPs’ patient volume in this disease group can therefore be fairly small. There is no unambiguous association between volume and quality of clinical work (16), but national surveys (17, 18) indicate that GPs perceive it as challenging that they see so few patients with eating disorders.

A fourth challenge is associated with the organisation of treatment. Despite the recommendations in national guidelines (19), national surveys (17, 18) and international studies (20) show that collaboration with the specialist health service can be perceived as problematic. One consequence can be that the GP might need to handle all the degrees of eating disorders. In any case, the GP will be the patient’s fixed point of reference throughout the illness trajectory (15, 21, 22). Patients with the most common diagnosis (bulimia) seek out their GP for somatic ailments (23), and within the time constraints of individual consultations it can be challenging for the doctor-patient relationship to address the patient’s perceived needs while also focusing on reducing symptoms such as overeating and induced vomiting. A Norwegian mapping study indicates high patient satisfaction with the GPs’ follow-up practices (22). On the other hand, we know less about the GPs’ experiences of patients with eating disorders. Such knowledge is important in light of the key role that GPs play in treatment (23), but also in light of their total workload, partly as a result of the transfer of responsibilities from other agencies (24, 25).

Material and method

Participation in the study was conditional on at least three years’ experience in general practice, and on having encountered patients with anorexia or bulimia during this period. Attempts at recruiting participants were made by the first author, who informed 50 GPs in 12 medical centres in Nordland county about the study, orally as well as in writing. Initially none of them agreed to participate, but after some reminders five doctors participated, of whom one was a woman and four were men with 4, 16, 24, 30 and 40 years of experience as GPs respectively. The participants signed a declaration of consent to use of the data material both in a master’s degree thesis in medicine and in a later research article. A semi-structured interview guide was prepared by the group of authors. It was also discussed in a focus-group meeting at the University of Tromsø – The Arctic University of Norway in order to obtain feedback from others who were engaged in the same research area. The interview guide consisted of questions related to three main topics: 1) experience with patients with eating disorders; 2) how to identify eating disorders; and 3) the GPs’ responsibility for identifying eating disorders. Follow-up questions were asked as needed. The project was approved by the Norwegian Centre for Research Data (NSD), and the approval was conditional on the requirement that the interviews should not put the doctors at risk of breaching their duty of confidentiality.

The interviews were conducted in the period from August 2017 to January 2018, each of which lasting 40–60 minutes. The interviews were held at the doctors’ workplaces and were recorded. The material was transcribed shortly after the completion of each interview.

The transcribed interviews were condensed and summarised with the aid of systematic text condensation (26) to identify general impressions and meaningful text units, as well as illustrative quotes.


The textual analysis resulted in three topics; these are presented below.

Topic 1: A patient group which is hard to identify

One important reason why the doctors felt that discovering eating disorders was difficult, was that they had met relatively few patients who had this problem. One of the doctors noted that ‘unfortunately, it’s only a tiny part of our daily work’. Another estimated that in his 24 years as a GP he had seen perhaps eight or ten seriously ill patients with eating disorders, while a third noted that he saw perhaps one or two cases each year. Some explained this by the high average age of the patients on their lists. Despite the higher prevalence of bulimia than anorexia in the population (5), they had seen only patients with anorexia.

The difficulty in identifying the patients in question could be due to the fact that the doctors undertook little exploratory work and mostly addressed the issues that the patients presented with, or because they felt it as too sensitive to hint at an underlying eating disorder. As one of the doctors said:

‘I know there are more than those we see, since we see so few. No patient comes in to see a GP and says “good morning, I have an eating disorder”. You need to build trust. There must be some kind of motivation, some reason for the patient to take this further. After all, many of them are specialists in covering this up.’

Another put it like this: ‘It often happens that others express concern that something is going on here.’ As indicated by the quote, reports of concern could come from friends or family members, and the doctors found this challenging to handle. In such cases, it was difficult to contact a patient with a problem that had been reported by others, and in order to get through to the patient they needed to be open about why the patient had been called in to an appointment.

Topic 2: Working with eating disorders in general practice

Even in a busy clinical work situation, time was set aside for longer consultations:

‘I set aside extra time first, then I make a new appointment. Patients with eating disorders require a lot of time, far more than the time available. I therefore book a double appointment, preferably at the end of the day. Then I avoid the pressure from the waiting room. I need to see how serious it is and form an opinion of what I can handle myself and what I need to refer to others.’ In less serious cases, the doctors could provide close follow-up themselves rather than referring, but even then they wanted to continue the patient contact to stay informed about the treatment process, also because they were usually made responsible for the somatic follow-up:

‘I believe that eating disorders quickly become associated with a situation where one is charged with following up certain things while the patient is undergoing psychiatric treatment. This is perhaps the most difficult situation, when someone has been hospitalised or is in outpatient treatment, and then the GP has had his task defined accordingly. We have to check specific things and then report them to the therapists. I believe that this is the relationship most of them would like to have.’

Some of the doctors had positive experience of having been called in to collaborative meetings with the specialist health service and highlighted their usefulness, because GPs are often charged with the responsibility for further follow-up. Many of the participants related actively to this responsibility:

‘Treatment in general practice can be extremely beneficial for patients with eating disorders, because they need long-term follow-up, and this follow-up is often largely about repeatedly talking to people, taking some tests, examining. We in general practice are very good at having long-term relationships. One becomes almost like a life companion.’

The doctors also made frequent references to safe settings and confidentiality that make for openness:

‘It’s much safer to go and see your GP. Going to sit in the GP’s waiting room is totally innocuous, because people with all sorts of ailments sit there. If you sit in the waiting room of a psychologist or psychiatrist, then it’s quite obvious why you’re there.’

Topic 3: Challenges and measures

Time constraints were mentioned as a major challenge to the establishment of long-term, secure doctor-patient relationships. For example, one of the doctors had 20 consultations per day on average, with a time limit of 10–20 minutes with each patient per consultation, paperwork included.

‘Again I believe that for a person to communicate such a sensitive matter as this, it requires a totally different setting from what we have on a daily basis, in well-defined consultations in general practice. I believe that we cannot avoid seeing that sometimes, we actually need to attend to this particular patient for a little longer.’

One of the doctors specified that time constraints could be solved by having more permanently salaried GPs with greater capacity and fewer patients on their lists. Others referred to the need for better basic competence in eating disorders.

‘I studied medicine in the 70s, and I cannot recall hearing anything about eating disorders. When you haven’t learned anything about it in medical school, nor come across it on the foundation course or later – it’s not a matter that’s especially close to your heart. Perhaps you feel that this is not your special talent and you therefore have little to contribute, then it’s OK that someone else takes care of it. In general practice, you need to know something about most things. You cannot know everything about everything, so some things must be excluded.’

It was noted that a lack of basic competence could be compensated for by further education and training, but that eating disorders were perhaps not something to be prioritised in a cost-benefit perspective. On the other hand, many of the doctors called for a simple screening tool: ‘Perhaps some tools that don’t necessarily have to be applied in a fully standardised way, but at least could help raise awareness of this patient group among us doctors.’

The doctors also mentioned the limited opportunity to refer patients to the specialist health service because of the long waiting times and a strict selection of only the very sickest patients. One doctor wanted the specialist health service to exercise its duty to provide guidance more actively:

‘I would like the specialist health service to live up to its duty to provide guidance to a greater extent. If they could teach us more about this kind of group that they are dealing with a lot, and we perhaps little, it would be a great help for us.’

Some called for better municipal services, where the doctor could collaborate with a nutritionist and a psychologist as well as with group-based programmes arranged by the Healthy Living Centres.


The GPs’ experiences with patients with eating disorders showed that the time that this patient group needs to establish a good relationship fits poorly into the busy work schedule of the GPs. The summarised experiences also point to challenges in developing competence gained through volume of patients. Some studies may indicate that health personnel give low priority to this patient group and have a negative attitude towards it (2729), but the consequences for clinical practice are debatable (30). With the reservation that doctors with positive attitudes to this patient group may have self-selected for the study, the results rather give an impression of a desire to provide better proactive and preventive services, more collaboration and consultations with the specialist health service, and the ability to set aside more time for each consultation and provide long-term care for the patient. The latter has been shown to be crucial for patients in this group to perceive the GP as a support (22, 31, 32). Other qualitative studies have also shown that in spite of the challenges, GPs have been and continue to be perceived by their patients as having a focus on ‘the good doctor’ as the ideal for a holistic patient perspective (22, 33).

The vision behind the contract GP scheme is that the patients should be followed up by one and the same doctor, and that a doctor who knows his or her patients well can do a better job. The time constraints that are described in the results can be seen in the context of the debate over the contract GP scheme, in which many have argued that the administrative burden is now so great that it undermines the vision and purpose of the scheme (2325).

The results highlight the need for a clearer distribution of responsibilities and tasks between the service levels. This distribution is described in national guidelines (19), but is poorly reflected in how GPs and other health personnel perceive the collaboration with the specialist health service (17, 18).

The results can also lend support to the argument that medical students and specialty registrars need more competence in clinical assessment and treatment of eating disorders, beyond sporadic information on somatic aspects of these disorders. The study must be interpreted in light of its low number of participants, whereby self-selection may impact on the results. We also need to add the proviso that the experiences of GPs in Northern Norway may differ from those of GPs in other professional, administrative and geographical contexts.

The article has been peer reviewed.

Anbefalte artikler