Abstract
Background:
Co-morbid diabetes and eating disorders have a particularly high mortality, significant in numbers and highly dangerous in terms of impact on health and wellbeing. However, not much is known about the level of awareness, knowledge and confidence amongst healthcare professionals regarding co-morbid Type 1 Diabetes Mellitus (T1DM) and eating disorders.
Aim:
To understand the level of knowledge and confidence amongst healthcare professionals in Wales regarding co-morbid T1DM and eating disorder presentations, identification and treatment.
Results:
We conducted a survey of 102 Welsh clinicians in primary care, diabetes services and eating disorder services. 60.8% expressed low confidence in identification of co-morbid T1DM and eating disorders. Respondents reported fewer cases seen than would be expected. There was poor understanding of co-morbid T1DM and eating disorders: 44.6% identified weight loss as a main symptom, 78.4% used no screening instruments, and 80.3% consulted no relevant guidance. The respondents expressed an awareness of their lack of knowledge and the majority expressed willingness to accept training and education.
Conclusion:
We suggest that priority must be given to education and training of all healthcare professionals in primary care, diabetes services and mental health services who may see patients with co-morbid T1DM and eating disorders.
Keywords
Background
Since first being recognised in the 1980s (Colton et al., 2015), it has become increasingly clear that Type 1 Diabetes Mellitus (T1DM) with a co-morbid eating disorder can be a dangerous combination. They are difficult to treat effectively (Allan, 2015), and offer a high risk of complications (National Institute for Health and Care Excellence, 2017; Scheuing et al., 2014). These include retinopathy and nephropathy (Rydall et al., 1997), other medical sequalae of poor glycaemic control and an overall higher mortality rate (Goebel-Fabbri et al., 2008).
Patients with T1DM are unique in their ability to induce rapid weight loss through omitting their insulin injections as a purging weight loss behaviour (Goebel-Fabbri et al., 2008). This can give rise to specific complications associated with hyperglycaemia, such as diabetic ketoacidosis. Futhermore, literature suggests that the detailed focus on food necessitated by T1DM, such as carbohydrate counting and insulin dose adjustment (Diabetes UK Nutrition Working Group, 2018), can give these patients a higher level of concern regarding food, thus contributing to an elevated risk of development of an eating disorder (Toni et al., 2017).
These unique aspects of an eating disorder when it occurs in an individual with T1DM need to be considered during the process of identification and management of the patient’s eating disorder. Literature suggests that different parameters should be used to monitor eating disorders in people with T1DM. While a low body mass index (BMI) can often help clinicians identify an eating disorder in the normal population, diabetics omitting insulin may have deceptively normal BMI but high HbA1c levels as well as high ketones and blood glucose levels (Allan, 2017). This can make diagnosis of an eating disorder in an individual with T1DM particularly difficult for clinicians: research suggests that clinicians are generally not good at identifying eating disorders, let alone the complex presentations of eating disorders in patients with T1DM (Waller et al., 2014). Furthermore, there are a number of limitations with the current guidelines provided regarding recognition and management of eating disorders. Firstly, it has been suggested that only small numbers of clinicians actually access the published National Institute for Health and Care Excellence (NICE) guidelines which guide clinical practice in the United Kingdom regarding recognition and treatment of eating disorders, and of those, even fewer apply the recommendations of this guidance to their practice (Currin et al., 2007). However, there are also limitations from the guidelines themselves: as noted by the guideline committee behind the latest NICE guideline for eating disorders, there are currently no detailed guidelines to facilitate diagnosis and management for eating disorders in people with T1DM. Instead, current NICE eating disorder guidelines signpost the NICE diabetes guideline, which in turn signposts the NICE eating disorder guideline (National Institute for Health and Care Excellence, 2016a, 2016b, 2017). In other words, there is little guidance that can help clinicians with the challenges of identifying this specific patient group. At the same time, T1DM and eating disorders are being increasingly recognised as dangerous to health and worthy of particular care in treatment (All Party Parliamentary Group for Diabetes, 2018; Diabetes UK, 2017).
It has been shown that early diagnosis and treatment of eating disorders can significantly improve health, wellbeing and recovery (Arcelus et al., 2011; Schmidt et al., 2016). However, there has been little research into clinicians’ experiences, confidence and ability to identify and manage eating disorders in people with type 1 diabetes mellitus, whether working in diabetes services, eating disorder services or in primary care. This study aims to gain insight into clinicians’ experiences of identification and management of eating disorders in people with T1DM, surveying healthcare professionals across Wales who come into contact with this patient group. There are currently no clear pathways for the treatment of patients who have both T1DM and eating disorders, and it is possible that they may also be seen, identified and treated in a range of settings, particularly diabetes services, primary care and mental health services. Because of this lack of pathways and lack of knowledge of what awareness, experience or knowledge might be, we decided to do a general survey of all three groups. This was an unfunded medical student research project conducted by IS and supervised by JT, and it was intended to be a simple preliminary survey mapping the terrain of awareness and confidence, rather than an in-depth study.
Aims
The aim of the survey was to understand the level of knowledge and confidence amongst healthcare professionals likely to see either eating disorder or T1DM regarding co-morbid T1DM and eating disorder presentations, identification and treatment.
Methods
Setting and design
A short, written questionnaire was developed by the authors, and distributed amongst healthcare professionals across Wales from three disciplines who come into contact with patients with T1DM and/or eating disorders: primary care, specialist diabetes services (both paediatric and adult) and eating disorder mental health services (both child and adolescent and adult). The survey was intended to be quick and simple to use and was disseminated using SurveyMonkey (Copyright 1999-2019 SurveyMonkey, San Mateo, CA, USA), and featured eleven questions, including open and closed questions, as well as allowing free text comments (see Supplemental File 1 for the questionnaire). Questions asked included number of cases seen, self-reported confidence in identification, symptoms and signs expected, use of screening tools and guidelines, the importance of measuring physical health parameters, as well as questions regarding the most suitable clinician to refer patients to, to monitor physical health and make management decisions. Respondents were also asked to comment on what might help them feel more confident identifying and managing eating disorders in people with T1DM.
Participants were recruited during the study period (January to September 2018) by email. In this email, they received information regarding the study and gave written consent to participate, as well as receiving a link to the online questionnaire.
Analysis
Survey data were collated in and analysed using Microsoft Excel. Where appropriate, statistical tests were carried out.
Participants
Inclusion criteria included healthcare professionals of the following categories: primary care practitioners (general practitioners (GP)/family physicians, GP practice nurses and allied health professionals), paediatric and adult specialist diabetes teams (endocrinologists, paediatricians, diabetes specialist nurses, diabetes dietitians and allied health professionals such as diabetes dietitians and psychologists) and general and specialist mental health teams (adult general psychiatrists, specialist adult eating disorder psychiatrists, child and adolescent psychiatrists, psychiatric nurses and allied health professionals such as occupational therapists, dietitians and clinical psychologists), working in an NHS setting across Wales.
Results
One hundred and two clinicians completed the survey. Considering the breakdown of respondents by specialty, 43.1% (n = 44) were from primary care, 31.3% (n = 32) were from diabetes services and 25.4% (n = 26) were from mental health services. Clinicians were categorised by their profession: 26.7% (n = 27) were allied health professionals, 49.5% (n = 50) were doctors and 23.8% (n = 24) were nurses. Table 1 shows the distribution of types of professional amongst the services for all respondents.
Distribution of types of professionals amongst services for respondents.
Experience with cases of co-morbid T1DM and eating disorders
Clinicians were asked to estimate the number of cases of eating disorders in people with T1DM that they had come across in the previous year. Responses ranged from zero to ten cases, with 43.1% (n = 44) of clinicians having seen no cases. The median was 1 case with a mean number of cases seen of 1.38, with a 95% confidence interval of 0.98–1.79 (see Table 2).
Number of cases of eating disorders with type 1 diabetes mellitus seen.
Confidence making diagnosis
Considering clinicians’ self-reported levels of confidence in their ability to identify eating disorders in people with T1DM, 61.8% (n = 63) of clinicians said they were not confident (either ‘not at all confident’ or ‘not very confident’) (see Table 3). Unsurprisingly, those working in mental health were more confident in identifying an eating disorder – even so, 34.3%(n = 11) of respondents working in mental health lacked confidence in diagnosing an eating disorder. There was no correlation between an increasing number of cases experienced and an increased level of confidence in identification (Correlation Coefficient .11, p > .05).
Confidence in identification of eating disorders in people with T1DM by specialty.
Symptoms/signs
Analysis of free text comments regarding symptoms and signs that would make a clinician suspicious of an eating disorder in a person with T1DM revealed a large range of responses (see Figure 1). In terms of the three most common responses, the most common was weight loss, cited by 44.6% (n = 45) of respondents, followed by body image and self esteem concerns, cited by 22.8% (n = 23) of respondents, in turn followed by insulin omission, cited by 19.8% (n = 20) of respondents.

Signs/symptoms of an eating disorder in people with T1DM as understood by clinicians.
Symptoms and signs reported by clinicians were further categorised (see Figure 2), and suggested that diabetes specific symptoms (such as increased HbA1c, poor glycaemic control, complications of diabetes) and weight changes (including both weight gain and weight loss) were the most commonly cited, each respectively comprising 76.5% (n = 78) of responses.

Categories of signs/symptoms of an eating disorder in people with T1DM, reported by clinicians.
Screening tools
78.4% (n = 80) of respondents reported using no screening tool to aid identification of eating disorders in people with T1DM. Of those that did use screening tools, SCOFF (Morgan et al., 1999) was used the most, cited by 13.7% (14) of respondents. Other screening tools mentioned included four mentions of the Eating Disorder Examination Questionnaire (EDE-Q) (Fairburn & Beglin, 1994; Luce & Crowther, 1999; Mond et al., 2004), and single mentions of the Brief Screening Tool for Disordered Eating in Diabetes (DEPS-R)(Markowitz et al., 2010), the Clinical Impairment Assessment Questionnaire (CIA) (Bohn & Fairburn, 2008), the Development and Well-Being Assessment (DAWBA) (Goodman et al., 2000) and the Eating Attitudes Test (EAT 26) (Garner, 2004). Two respondents (both working in eating disorder services) stated that they use a full eating disorder assessment instead of a screening tool (See Table 4).
Use of screening tools to help identify an eating disorder in people with T1DM.
Guidelines
When questioned on whether they had ever consulted published guidance to assist in identification and management of eating disorders in people with T1DM, 80.3% (n = 82) had used none (see Figure 3). Of the 19 respondents who had consulted publications, guidelines from the National Institute for Health and Care Excellence (NICE) were reported as used the most (n = 8), followed by local guidance (n = 5). Other guidance consulted included information disseminated from teaching sessions, MARSIPAN guidelines for the recognition and management of very medically unwell patients with anorexia nervosa (The Royal College of Psychiatrists, 2012, 2014) and organisations such as Diabetes UK and Diabetics with Eating Disorders (DWED).

Use of published guidance regarding eating disorders in people with T1DM.
What would help?
Clinicians provided free text answers to the question, ‘What do you think could help you feel more confident in identifying and managing eating disorders in people with T1DM?’ The most common theme was training and education, cited by 51.9% (n = 53) of respondents (see Figure 4). Other common themes included closer MDT working (18.6%, n = 19), the publication of guidelines (14.7%, n = 15) and better access to specialists (13.7%, n = 14). Responses also highlighted that the clinicians surveyed do not see many cases of eating disorders, with 10.7% (n = 11) of respondents suggesting that more experience of these cases could help them feel more confident.

Clinicians’ suggestions of what would improve confidence in identification and management of eating disorders in people with T1DM.
Discussion
There appear to be a multiplicity of guidance both from eating disorder and diabetes treatment perspectives which highlight the importance of understanding physical co-morbidities in patients seen for eating disorders in mental healthcare settings. The importance of understanding mental health issues and co-morbidities in patients seen for Type 1 diabetes mellitus cannot be understated. At the same time, our survey, while not representative, suggests that the level of knowledge and confidence that professionals within primary care, diabetes services or mental health services would know how to identify and eating disorder in a patient with T1DM is generally not high.
This lack of confidence and knowledge about co-morbid T1DM and eating disorders is demonstrated by a large majority of healthcare professionals identifying weight changes as a marker they would look for as indicative of a possible eating disorder in someone with T1DM, when in fact this is a poor marker for eating disorders in general and particularly poor for people with T1DM and eating disorders. Furthermore, the fact that over 80% of respondents did not use any guidelines in treating these patients, when the majority are unfamiliar and unconfident regarding identification and management, is further cause for concern. Sixty-one percent of respondents were ‘not at all confident’ or ‘not very confident’ in identifying eating disorders in T1DM; 44% had never seen a single case and a further 38% had seen only one to three cases before. There was low awareness of the MARSIPAN guidance which is well established and intended to assist any healthcare professional regardless of specialty to recognise and manage medical risk amongst patients with severe eating disorders, which would be particularly important for patients with T1DM.
To our knowledge, our survey is the first study of its type exploring clinicians’ experiences of eating disorders in people with T1DM, considering their confidence in making these diagnoses and gaining information about what could improve practice.
Responses to the survey highlight the low number of cases of eating disorders being recognised in people with T1DM. While there are few statistics regarding prevalence of eating disorders in people with T1DM, literature suggests that 60% of women with T1DM will develop a clinically diagnoseable eating disorder by the age of 25 (Colton et al., 2015). The problem is not unique to women: an Australian study found that 11% of male adolescents with T1DM omit insulin for weight control purposes (Araia et al., 2017). This developing literature could suggest that the prevalence of eating disorders in people with T1DM is much higher than the numbers of cases reported by respondents to this survey: this suggests that clinicians in primary care and diabetes services are failing to recognise eating disorders and missing diagnoses due to lack of understanding and awareness of the problem; and that these patients may therefore not be referred to eating disorder services. This survey suggests that clinicians lack confidence in identifying and managing eating disorders in people with T1DM, regardless of setting. Interestingly, even the clinicians seeing a relatively higher number of cases do not report higher levels of confidence. There is clearly room for more training in this.
It is reassuring that there was recognition of low skills and knowledge in the subject with over half of respondents stating that more training and education would be welcomed.
In the service structures of the United Kingdom’s National Health Service (NHS), there is traditionally little if any communication between diabetes services and mental health services. There are some clinical psychologists working in some relatively well resourced diabetes services within Wales, which is a need particularly acknowledged in children’s diabetes services but much less so in adult diabetes services. However, even when there are clinical psychologists within diabetes services, they may lack the training and skills to manage eating disorders, particularly as eating disorders in T1DM are highly complex and difficult to treat. Furthermore, diabetes service colleagues need to be able to recognise cases in the first place in order to refer patients who may have eating disorders to their team clinical psychologists, who in turn need knowledge of how to assess severity and links to refer onwards to specialist eating disorder colleagues who work in mental health rather than physical health divisions.
There appears to be low awareness and confidence in mental health services in managing the particular issues associated with eating disorders in T1DM. Again, this suggests a need for closer co-working between eating disorder and diabetes services in order to achieve the wholistic and safe management of a complex disorder.
This research suggests that there is a real need for more sharing of skills and co-working, as well as mutual training and education between diabetes and eating disorder services, for the benefit of patients with co-morbid T1DM and eating disorders, who are a subgroup who are particularly complex in terms of treatment as well as vulnerable to serious medical sequalae and high mortality. The recent Welsh Government Eating Disorder Service Review (Health Minister Vaughan Gething AM, 2019; Tan, 2019) has recommended a national dual-diagnosis specialist team to assist the treatment of these patients in Wales, and there are some services beginning to specialise in treatment of ‘diabulimia’, increasingly known as T1DE (Type 1 diabetes with eating disorders) in England (Diabetes UK, 2017; McKee, 2019).
Conclusion
Co-morbid Type 1 Diabetes Mellitus and eating disorders are both significant in numbers and highly dangerous in terms of impact on health and wellbeing. There is a growing recognition of the complexity and importance of expertise and knowledge of both diabetes and eating disorders in order to prevent, recognise and manage this co-morbid condition. We have done a survey of Welsh clinicians in primary care, diabetes services and mental health services which show a worryingly low identification of cases, poor understanding of the co-morbid condition and lack of consultation of the relevant guidance. On the positive side, the respondents expressed both an awareness of their lack of knowledge and the majority expressed willingness to accept training and education. We suggest that this survey demonstrates a need for closer co-working between diabetes and eating disorder specialists to address this need for training and priority given to education to all healthcare professionals who may see such patients.
Research Data
sj-pdf-1-ccp-10.1177_13591045211009591 – for Identifying and managing eating disorders in type 1 diabetes mellitus: An all-Wales study
sj-pdf-1-ccp-10.1177_13591045211009591 for Identifying and managing eating disorders in type 1 diabetes mellitus: An all-Wales study by Jacinta O A Tan and Imogen Spector-Hill in Clinical Child Psychology and Psychiatry
Supplemental Material
sj-pdf-2-ccp-10.1177_13591045211009591 – Supplemental material for Identifying and managing eating disorders in type 1 diabetes mellitus: An all-Wales study
Supplemental material, sj-pdf-2-ccp-10.1177_13591045211009591 for Identifying and managing eating disorders in type 1 diabetes mellitus: An all-Wales study by Jacinta O A Tan and Imogen Spector-Hill in Clinical Child Psychology and Psychiatry
Footnotes
Acknowledgements
The authors would like to thank all the participants who took the time and trouble to complete the survey.
Author contributions
IS conducted the research fieldwork under supervision from JT. Both IS and JT conducted the analysis. JT and IS wrote this article.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.
Ethical review
The study received ethics approval from Swansea University Medical School’s Research Ethics and Governance committee (2017-0047).
Data accessibility statement
The authors have made the survey and the grouped data available through Supplemental Files. Detailed anonymous free text responses and full anonymised data of responses the survey are available upon reasonable request.
Supplemental material
Supplemental material for this article is available online.
Author biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
