1. Introduction

---

‘Mental health literacy’ is defined as ‘knowledge and beliefs about mental disorders which aid their recognition, management or prevention’[1]. Health professionals are usually thought to have a higher level of mental health literacy than the general population, although there appears to be a continuum from lay beliefs to professional knowledge rather than a dichotomy[2]. It is well recognized that the general population has highly stigmatizing attitudes towards individuals with eating disorders, often blaming them for their problems[3,4]. Studies have shown that general psychiatrists, general practitioners and hospital physicians have poor knowledge of eating disorders whilst non-psychiatrists have overly pessimistic views about their outcomes[5-9]. Furthermore, hospital specialists, general practitioners and medical students consistently rank anorexia nervosa (AN) as an illness with low prestige when compared with other diagnoses[10]. This is a major concern given that eating disorders have the highest mortality rate of any psychiatric condition with a standardised mortality ratio of 6[11]. Little is known, however,about the knowledge and attitudes of gastroenterologists towards individuals with eating disorders. This is a concern given that gastroenterologists often play a crucial role in the management of these patients when they require urgent inpatient medical treatment. Furthermore, a gastroenterologist is identified as a key member of a MARSIPAN (Management of Really Sick Patients with Anorexia Nervosa) expert working group which outlines standards of treatment for MARSIPAN patients[12].

With this in mind the aim of this study was to first, assess gastroenterologists’ knowledge and attitudes towards individuals with eating disorders and second, assess gastroenterologists’ experiences of interface working with mental health services and the use of MARSIPAN guidelines[12].

2. Methods

---

2.1. Design And Participants
The study was approved by the Leeds and York Partnership NHS Foundation Trust R&D Department. Gastroenterologists were contacted on three separate occasions via the British Society of Gastroenterology online newsletter which included a web-link to an online questionnaire.

2.2. Measures And Procedures
The online questionnaire (see appendix 1) was designed by the authors and was adapted from previous measures in this field. The questionnaire was piloted by two consultant gastroenterologists and two consultant eating disorder psychiatrists.

2.2.1. Knowledge Questions
These included four extended matching items and three true/false items. Questions were based on the academic literature[13,14] and DSM-IV diagnostic criteria[15] and addressed both AN and bulimia nervosa (BN). In keeping with previous work in this field a negative marking scheme was utilized to allow for questions having multiple correct answers, where scores were reduced by random answers or guesswork.

2.2.2. Attitude Items
There were 17 attitude items and gastroenterologists were asked to indicate their agreement by using a five-point Likert scale. Many of the attitude items were derived from previous work on stigma and eating disorder mental health literacy. Additional attitude items explored experiences of mental health support for individuals with eating disorders whilst receiving treatment in inpatient medical settings as well as clinicians’ awareness and experience of MARSIPAN guidelines[12].

2.2.3. Data Analysis
Data were anonymised and collected using the web-based survey tool SurveyMonkey (http://www.surveymonkey.com) and analysed using SPSS V20.0 (Chicago, Illinois, USA). Sample characteristics were compared using Fisher’s exact tests for categorical variables and independent sample t-tests for continuous variables as these were all approximately normally distributed. All significance tests were performed using a two-tailed five percent significance level.

3. Results

---

3.1. Respondents
In total the questionnaire was sent to 3118 gastroenterologists and 77 responded giving a response rate of 2.5%. Of the respondents, 57.1% were men and 42.9% were women. Ethnic background was predominantly ‘White British’ (74.0%) with 9.1% of respondents describing themselves as ‘Indian’. 46.8% had worked in gastroenterology for more than 10 years.

3.2. Levels Of Knowledge Among Gastroenterologists
The maximum score possible for knowledge questions was 21 and individual scores were normally distributed ranging from 3 to 19 (mean=11.8, SD=3.1). There were no statistically significant differences in knowledge scores in relation to age, gender, ethnicity or clinician seniority.

3.2.1. Diagnostic Criteria
‘Fear of fatness’ was the most commonly identified diagnostic criterion for AN (82.8%). 43.1% identified the correct diagnostic body mass index (BMI) threshold for AN as being below 17.5 kg/m2 but 34.5% believed the BMI threshold to be even lower (16.0 kg/m2). Finally, 3 months of amenorrhoea was recognized as a diagnostic criterion for AN by only 29.3% of gastroenterologists.

‘Extreme weight-control behaviours’ and ‘recurrent episodes of binge eating at least twice a week for 3 months’ were the most commonly identified diagnostic criteria for BN being identified by 71.2% and 61.0% of gastroenterologists respectively. ‘Over-evaluation of shape and weight’ was recognized by only 40.7% of gastroenterologists and only 10.2% identified the correct diagnostic BMI threshold for BN as being 17.5 kg/m2 or above.

3.2.2. Physical Complications And General Knowledge
The majority of respondents correctly identified electrolyte abnormalities (91.5%), bradycardia (89.8%), hypoglycaemia (81.4%), osteoporosis (93.2%) and anaemia (79.7%) as common physical complications of AN. However, fewer knew about the possibility of proximal myopathy (64.4%) and oedema (57.6%). Likewise, the majority identified electrolyte abnormalities (88.1%), dental erosion (96.6%) and parotid swelling (61.0%) as common physical complications of BN, however, only 49.2% knew about the possibility of oligomenorrhoea.

Most gastroenterologists (78.0%) correctly stated that the MHA can be used to enforce nasogastric feeding in individuals with AN whilst roughly half (54.2%) correctly stated that BN has a higher prevalence rate than AN.

3.3. Attitudes Among Gastroenterologists
3.3.1. Confidence Levels And Training
Of the respondents, less than half (43.6%) were confident in their ability to diagnose eating disorders whilst only 36.4% were confident in their ability to manage these conditions in their current practice. Moreover 61.8% of respondents were unsatisfied with the level of eating disorder training they had received during their medical training.

3.3.2. Attitudes Towards Eating Disorders
32.7% saw AN as being ‘culturally determined by woman’s role in society’, whereas 56.4% saw AN as ‘representing a form of neurotic mental disorder’. 18.2% of respondents viewed AN as representing ‘abnormal behaviour in the context of a weak, manipulative or inadequate personality’ whilst 10.9% saw AN as being ‘essentially untreatable’. 41.8% believed that AN was ‘a neurophysiological disorder of unknown origin’.

34.5% saw BN as being ‘culturally determined by woman’s role in society’, whereas 50.9% saw BN as ‘representing a form of neurotic mental disorder’. 25.5% of respondents viewed BN as representing ‘abnormal behaviour in the context of a weak, manipulative or inadequate personality’ whilst 5.5% saw BN as being ‘essentially untreatable’. 41.8% believed that BN was ‘a neurophysiological disorder of unknown origin’.

There were no statistically significant differences in attitude towards individuals with AN or BN in relation to age, gender, ethnicity or seniority.

3.3.3. Use Of The Mental Health Act In Anorexia Nervosa
80.0% felt it ‘appropriate that the MHA enables compulsory re-feeding of patients with AN’ whereas 83.6% believed that ‘the MHA should be used to enforce admission to hospital for patients with AN’. Only 9.1% supported the statement that ‘the MHA should not be used when patients clearly believe that the advantages of AN outweigh the disadvantages’ and 54.5% believed that ‘the MHA should be used more frequently’.

There were no statistically significant differences in attitude towards the use of the Mental Health Act in induvial with AN in relation to age, ethnicity, gender or clinician seniority.

3.3.4. Interface Working And MARSIPAN
45.5% of gastroenterologists surveyed stated that ‘access to liaison psychiatry services and support was readily available’ whilst 41.8% felt that ‘access to specialist eating disorders services and support was readily available’. 78.2% felt that ‘there was often a delay in transferring patients with eating disorders from a medical unit to a specialist eating disorder unit’ whilst 45.5% felt that ‘the needs of patients with eating disorders were not being met in their clinical practice’. 29.1% felt that ‘patients with severe AN should not be treated on a medical unit’.

38.9% were aware of the use of a formal clinical guideline for the management of patients with eating disorders in their hospital whilst 24.7% stated that their team met regularly with a MARSIPAN group. Of those that met with a MARSIPAN group, 76.9% felt that the group had ‘a positive impact on patient care’, 76.9% said that it had ‘improved communication between medical and eating disorder services’ and 53.8% felt that the group had ‘enhanced knowledge of eating disorders amongst medical staff’.

4. Discussion

---

To our knowledge this is the first study to examine the knowledge and attitudes of gastroenterologists towards individuals with eating disorders. Individuals with eating disorders, particularly those with AN, sometimes require admission to a gastroenterology unit during the course of their illness. These admissions are often prolonged and can involve lengthy periods of nasogastric feeding often under duress and under the legal framework of the MHA. Likewise gastrointestinal symptoms are common in eating disorders and patients often present for treatment of chronic symptoms[16-18] which can often prove challenging in an outpatient setting and such patients often carry profound psychosocial morbidity in excess of those presenting to eating disorder services[19]. Hence it is crucial that gastroenterologists should receive adequate training in both the diagnosis and management of these conditions and that information on eating disorders should be made readily available when patients with eating disorders present in medical settings.

4.1.1. Mental Health Literacy Of Gastroenterologists
Our results suggest that gastroenterologists’ knowledge of eating disorders is variable with specific gaps in knowledge relating to diagnosis and the use of the MHA and higher knowledge scores in relation to physical complications compared to other doctors5,6. Certain diagnostic criteria for both AN and BN were recognised by a minority of gastroenterologists and some physical complications were not recognised. This is a concern as recognition of clinical signs and symptoms is crucial in making an accurate diagnosis, assessing physical risk and determining the most appropriate treatment plan. These findings could be explained by the fact that the diagnosis and management of eating disorders is not included in the Royal College of Physicians training curriculum for gastroenterology although nutrition does have its’ own sub-speciality curriculum which covers this area.

Attitudes towards AN and BN were largely similar with most gastroenterologists believing that both conditions represent a form of ‘neurotic mental disorder’ and are ‘culturally determined by woman’s role in society’. Whilst most disagreed with statements relating to eating disorders being ‘essentially untreatable’ and ‘representing abnormal behaviour in the context of a weak, manipulative or inadequate personality’ our results mirrored previous findings in this field, i.e. non-psychiatrists and hospital doctors have more pejorative attitudes regarding the aetiology and treatability of eating disorders when compared to psychiatrists. Moreover, our finding that gastroenterologists generally support a role for compulsory measures under the provisions of the MHA in the treatment of patients with AN reflects previous work in this field[20].

4.1.2. Confidence LEVELS AND TRAINING
Confidence levels amongst gastroenterologists were also variable, most feeling more confident in diagnosing eating disorders than managing these conditions. Indeed only 36.4% were confident in their ability to manage eating disorders in their current practice and the majority were dissatisfied with the level of training in eating disorders they had received which could be explained again by gaps in gastroenterology training. Interestingly confidence levels of managing eating disorders were much higher than confidence levels of psychiatrists (14.9%)5 which could reflect psychiatrists lack of knowledge in relation to physical risk assessment. Our finding that gastroenterologists were generally dissatisfied with the level of training in eating disorders they had received during their medical training could also be a contributing factor and mirrors previous work in this field[21,22]. Moreover, the paucity and uneven distribution of dedicated eating disorder services in the UK[23,24] means that it is likely that the majority of patients with eating disorders are managed by health professionals who do not have special expertise in treating these conditions. This is reflected in the views of service users who refer to a lack of professionals experienced in eating disorders in the UK[25] and who consistently value receiving treatment from professionals who are specialists in this field[26-28].

4.1.3. Interface Working And MARSIPAN
More than half of the respondents felt that access to mental health services was not readily available with access to liaison psychiatry services (45.5%) and specialist eating disorders services (41.8%) being similarly poor. Our results also suggest that most gastroenterologists do not feel that gastroenterology services are best placed to meet the needs of individuals with eating disorders and that there are often significant delays in transferring patients to specialist services. Indeed the MARSIPAN report highlighted these issues as being major contributing factors to a number of avoidable hospital deaths which have been highly publicised in the general media over recent years[29,30].

Our results also suggest that most gastroenterologists are not aware of clinical guidelines for eating disorders which could potentially reflect a lack of local policy or a lack of knowledge in this area.

5. Limitations

---

Although this study is the first of its kind to examine knowledge and attitudes of gastroenterologists towards eating disorders and their experiences of MARSIPAN, our findings must be considered in the context of several limitations. First, our response rate of 2.5% is low but given that responses for email and web-based surveys often fail to match those of others survey methods[31] and the fact that we collected a relatively large number of responses compared with previous work in this field, it is possible that these findings are indeed representative of gastroenterologists as a whole. Nevertheless, as our sample was sourced via a web-based newsletter from only one specific gastroenterology interest group, there is a possibility of sampling bias and our results may not be generalizable to all gastroenterologists working within the UK. Second, the self-report data collection procedures used in this study may have limited the validity and reliability of our findings. Third, as there is no agreed gold standard of ‘adequacy of mental health literacy’ against which questionnaires such as the one used in the study can be validated, the validity and reliability of our findings may be limited. These limitations could be addressed in future studies.

6. Clinical Implications

---

This study clearly highlights the need to improve knowledge levels of gastroenterologists to ensure that their clinical actions are appropriately informed and carried out in line with national guidelines. Likewise it is important for gastroenterologists treating patients with eating disorders to receive adequate training in both the diagnosis and management of these conditions and information on eating disorders should be made readily available to gastroenterologists when patients with eating disorders present in medical settings. Rosenvinge et al[32] described an educational programme which raised health professionals’ competence in treating eating disorders and the implementation of such programmes could contribute to addressing some of these issues, improving treatment outcomes and increasing service users’ satisfaction with treatment. Action is required to improve access to Liaison Psychiatry and Eating Disorder service support in gastroenterology settings and further dissemination of the MARSIPAN12 report is required within gastroenterology and other medical settings so that these guidelines can be incorporated into local hospital protocol.