U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Mental Health (UK). Eating Disorders: Core Interventions in the Treatment and Management of Anorexia Nervosa, Bulimia Nervosa and Related Eating Disorders. Leicester (UK): British Psychological Society (UK); 2004. (NICE Clinical Guidelines, No. 9.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Eating Disorders

Eating Disorders: Core Interventions in the Treatment and Management of Anorexia Nervosa, Bulimia Nervosa and Related Eating Disorders.

Show details

5Identification of eating disorders in primary care

5.1. Introduction

Although new cases of clinical anorexia nervosa are not a common occurrence in primary care (an average GP list of 1900 will have only one or two sufferers), eating disorders including bulimia nervosa and EDNOS reach a prevalence of five per cent in young women. Community studies show that less than half of clinical cases of eating disorders are identified in primary care. Despite this, patients with eating disorders consult more frequently prior to diagnosis with a variety of symptoms, psychological, gynaecological and gastroenterological (Ogg et al., 1997). The difficulties facing primary care clinicians in diagnosis derive in part from illness-related factors, such as ambivalence, denial, secrecy and shame, which make it difficult for sufferers to be open with their doctors. GPs may have little experience with eating disorders, and feel anxious about their management or unsympathetic towards an illness that may appear in part self-inflicted. Patients may report that the problem is not always taken seriously enough if presented at an early stage. Practitioners risk failing to identify eating disorders if they do not consider the fact that these illnesses also occur in groups not traditionally considered to be at risk, such as children, men and those from ethnic minority groups, and lower social classes.

The effective management of anorexia nervosa depends on a full assessment of physical status, psychological features, risk and capacity to consent to treatment. The details of appropriate physical assessment are detailed in Section 5.2.4, but included in the assessment of high physical risk will be an assessment of degree of emaciation (BMI/BMI centile), the presence of purging and any fluid restriction. The rate of weight loss is also an important indicator. In children and younger adolescents, reference to BMI norms is necessary, but it is also of note that children’s relatively small fat stores render them at risk with relatively less weight loss. Psychosocial assessment should take account of motivation and social support.

5.1.1. Current practice

Little is known about how eating disorders are actually assessed or managed in primary care. Referrals to secondary care services (including specialist eating disorder centres) vary widely as does the information contained in them. Current practice is not underpinned by an evidence base derived from primary care settings and guidance has largely been extrapolated from secondary or tertiary care settings with experience with more severely ill clinical populations.

5.2. Screening

Early detection and treatment of eating disorders may improve outcomes in eating disorders. General practitioners and other members of the primary care team are in a good position to identify patients with eating problems early. Screening tools may facilitate this process. The most effective screening device probably remains the general practitioner thinking about the possibility of an eating disorder.

It would be impractical for general practitioners to try and screen all their patients for eating disorders, as the prevalence of eating disorders in the general population is low. It might be possible to screen new patients when they register. One or two screening questions could be used to raise the index of suspicion, either verbally during the registration health check or in writing as part of the registration questionnaire.

High-risk groups within the general practice population could be targeted opportunistically. Such groups include young women, patients with a low or high BMI, adolescents consulting with weight concerns, menstrual disturbances or amenorrhoea, gastrointestinal disorders and psychological problems. A brief screening questionnaire could be used for such high-risk groups.

5.2.1. Current practice

Eating disorders may be difficult to detect in primary care settings. Patients may be slow to self-present and many remain undetected by general practitioners (King, 1989; Whitehouse et al., 1992). Adults with eating disorders appear to consult their general practitioner more frequently than controls, presenting particularly with psychological, gastrointestinal and gynaecological problems (Ogg et al., 1997). Consultations of this nature present an opportunity to screen for eating disorders. At present, no formal screening tool for eating disorders is widely used in primary care.

5.2.2. Screening for eating disorders

The aim of screening is to facilitate detection so that treatment can be offered early in the course of the eating disorder.

A systematic review of the literature did not identify a significant body of work in this area, nor were any high quality existing systematic reviews identified. The absence of high quality evidence in this area inevitably limits the conclusions that can be drawn from the review. The relevant studies that were identified are described below.

Existing screening instruments

A range of questionnaires exists of which the Eating Attitudes Test, EAT (Garner & Garfinkel, 1979) is probably the most widely used as a screening tool in epidemiological studies. In addition there are a number of other pencil and paper measures to assess eating disorder psychopathology (e.g. the Eating Disorder Inventory, EDI – Garner et al., 1983). However, these take a long time to administer and may need to be interpreted by specialists. Such instruments may be well suited for evaluating treatment progress in patients with eating disorders, but may not perform well in screening for eating disorders in community samples due to symptom denial and low prevalence (Williams et al., 1982; Carter & Moss, 1984).

Questionnaires of this type may have a role for screening in very high-risk groups in special settings, e.g. in ballet schools, fitness and sports facilities.

They may have occasional application in general practice, when a patient with a probable eating disorder has already been identified.

Several brief screening questionnaires, more suitable for use in community samples, have been developed and evaluated. These include the SCOFF (Morgan et al., 1999), Anstine and Grinenko (2000), the BITE and the BES (Ricca et al., 2000), the EDS-5 (Rosenvinge, 2001), Freund et al. (1993), the ESP (Cotton et al., 2003), Ri-BED-8 (Waaddegaard et al., 1999), the EDDS (Stice et al., 2000), the EAT-12 and the EDE-S/Q (Beglin & Fairburn, 1992, 1994). The most promising to date is the SCOFF.

The SCOFF questionnaire (Morgan et al., 1999; Luck et al., 2002; Perry et al., 2002) was developed and validated in the UK. It consists of five questions designed to clarify suspicion that an eating disorder might exist rather than to make a diagnosis. The questions can be delivered either verbally or in written form and there is one study validating the use of the SCOFF in adult women in a general practice population (Luck, 2002). Further research is needed to evaluate the SCOFF questions in general practice populations before they can be recommended for use in primary care.

5.2.3. Clinical summary

A number of brief screening methods have been developed that have some utility in detecting eating disorders. The SCOFF has been shown to be capable of determining cases of eating disorders in adult women in primary care. The place of longer questionnaires (e.g. EAT, EDI, BITE, EDE-Q) may be in further assessment, once index of suspicion has been raised. They may also be useful to facilitate decisions regarding referral to secondary care or other specialist services. Certain clinical presentations should also raise the index suspicion, for example, adolescent girls with concerns about weight, and women consulting with menstrual disturbances, gastrointestinal or psychological symptoms.

5.2.4. Identification

The most important factor in the identification of eating disorders in generalist settings is for the practitioner to consider the possibility of an eating disorder and to be prepared to inquire further in an empathic and non-judgmental manner. The history is paramount and special investigations are not normally required to make a diagnosis.

5.2.4.1. Anorexia nervosa

The first contact with health care services is often made by a worried family member, friend or schoolteacher rather than the patient. Concerns expressed may be related to weight loss, food-related behaviours such as skipping meals, hiding food or adopting a restrictive diet. There may be a change in mood, sleep patterns and increased activity. Typical psychopathological features are fear of gaining weight or becoming fat despite being underweight, disturbance in evaluating or experiencing body weight or shape, undue influence of eating or changes in body weight on self-evaluation and preoccupation with shape or weight-related matters. These features may not all be present, easy to elicit or they may be denied. However, denial of the seriousness of the weight loss or consequences, both physical and psychological is usually present.

Established anorexia nervosa with signs of emaciation is usually obvious. However, patients may present initially in primary care with non-specific physical symptoms such as abdominal pain, bloating, constipation, cold intolerance, light headedness, hair, nail or skin changes. Amenorrhoea, combined with unexplained weight loss, in the population at risk should always prompt further enquiry. Apparent food allergy/intolerance and chronic fatigue syndrome sometimes precede the development of an eating disorder and may cause diagnostic confusion. In children, growth failure may be a presenting feature.

In practice, typical cases should cause little difficulty when the time is taken to explore the history including corroborative information and the patient’s attitude to the weight loss. Indeed, diagnosis is often delayed when doctors inadvertently collude by over-investigating and referring to other specialties rather than confronting the possibility of an eating disorder.

Diagnostic criteria

These are outlined in the introduction (see Section 2.1.2).

In children and adolescents under 18, the use of BMI centile charts should be encouraged (Cole, Bellizi et al., 2000) with a cut off less than the 2.4th centile of the reference population indicating underweight.

Centile charts for weight and height are also helpful in showing failure to progress over time.

The list of potential differential diagnoses of weight loss or amenorrhoea is large, but in practice, typical cases should cause little difficulty when the time is taken to explore the history including corroborative information and the patient’s attitude to the weight loss. The following factors need to be considered:

  • Risk factors – family history of eating disorder, Type 1 diabetes, previously overweight, occupation (e.g. athlete, dancer, model). Although adolescent girls and young women constitute the principal population at risk, it should be remembered that eating disorders also occur in ethnic minorities, men and children.
  • Differential diagnosis of weight loss – includes malabsorbtion (e.g. coeliac disease, inflammatory bowel disease), neoplasm, illicit drug use, infection (e.g. TB), autoimmune disease, endocrine disorders (e.g. hyperthyroidism).
  • Differential diagnosis of amenorrhoea – includes pregnancy, primary ovarian failure, poly cystic ovary syndrome, pituitary prolactinoma, uterine problems and other hypothalmic causes.
  • Psychiatric differential diagnosis – includes depression, obsessive-compulsive disorder, somatisation and, rarely, psychosis.
Initial physical assessment

The rationale for physical assessment is more to determine the presence and severity of emaciation and secondary physical consequences of the anorexia nervosa than to ascertain the primary diagnosis.

It should include as a minimum:

  • Height weight and BMI
  • Centile charts for age less than 18
  • Pulse and blood pressure.

The following may also be helpful to assess the risk of physical instability:

  • Core temperature (this is easily done by ear thermometer)
  • Examination of peripheries (circulation and oedema)
  • Cardiovascular examination including postural hypotension
  • Situp/squat test (a test of muscle power) (Robinson, 2003).
Laboratory investigations

Extensive laboratory investigation is not usually required in the diagnosis or assessment of anorexia nervosa in a primary care setting. Many tests remain normal even with extreme weight loss and are a poor guide to physical risk. The diagnosis is always made on the basis of the clinical history.

Investigations

The following would represent a reasonable initial screen in primary care if there are no other indications or diagnostic concerns:

  • Full Blood Count, ESR, Urea and Electrolytes, Creatinine, Liver Function Tests, Random Blood Glucose, Urinalysis.
  • ECG: This should be considered in all cases and is essential if symptoms/signs of cardiac compromise, bradycardia, electrolyte abnormality or BMI less than 15 kg/m2. (Or equivalent on centile chart.)

Further tests may be required in more severe cases or to assess complications: Calcium, Magnesium, Phosphate, Serum Proteins, Creatine Kinase (CK or CPK).

Tests that may be needed in the differential diagnosis of amenorrhoea and weight loss:

  • Thyroid Function Tests, Follicle Stimulating Hormone, Lutenising Hormone, Prolactin, Chest X-Ray.

A DXA scan may be considered for identification of osteopenia/osteoporosis, which may occur after six to 12 months of amenorrhoea. Although this is not necessarily a primary care level investigation, it has been suggested that it may be helpful in encouraging motivation for change in those not yet ready to accept referral, by demonstrating the real physical consequences of anorexia nervosa.

5.2.4.2. Bulimia nervosa

Identification

The patient with bulimia nervosa is more likely to be older and to consult alone than a patient with anorexia nervosa. There may be a history of previous anorexia nervosa or of unhappiness with previous weight and attempts to diet. Appropriate questioning (see screening section above) may reveal patterns of restriction, binge eating and purging and psychopathology that make the diagnosis clear. Not infrequently, physical symptoms are presented which may be related to or consequences of purging or laxative use. These symptoms, particularly in a young woman should be a ‘red flag’ in prompting the practitioner to consider further enquiry.

Where the patient does not disclose bulimia nervosa, a range of symptoms may present which should raise the index of suspicion. These include requests for help with weight loss, menstrual disturbance and the physical consequences of vomiting and laxative and diuretic use. Non-specific symptoms may include fatigue, lethargy. Gastrointestinal disorders may be present including bloating, fullness, abdominal pain, irritable bowel syndrome type symptoms, constipation, diarrhoea and rectal prolapse as well as oesophagitis and gastrointestinal bleeding. Oropharyngeal symptoms may include a sore throat, parotid swelling and dental enamel erosion.

Physical examination and investigation

In bulimia nervosa and related conditions, characteristic physical signs have been described (for example, parotid enlargement, Russell’s sign (callus formation on the dorsum of the hand) and dental enamel erosion, which are usually manifestations of purging. In practice these are not seen in the majority of patients presenting in primary care with bulimic disorders, although electrolyte abnormalities are reasonably common, so urea and electrolytes should be routinely obtained. These are covered in Section 7.5.2.

5.2.5. Clinical practice recommendations

5.2.5.1.

For people with eating disorders presenting in primary care, GPs should take responsibility for the initial assessment and the initial co-ordination of care. This includes the determination of the need for emergency medical or psychiatric assessment. [C]

5.2.5.2.

Where management is shared between primary and secondary care, there should be clear agreement amongst individual health care professionals on the responsibility for monitoring patients with eating disorders. This agreement should be in writing (where appropriate using the care programme approach) and should be shared with the patient and, where appropriate, his or her family and carers. [C]

5.2.5.3.

Target groups for screening should include young women with low body mass index (BMI) compared with age norms, patients consulting with weight concerns who are not overweight, women with menstrual disturbances or amenorrhoea, people with gastrointestinal symptoms, patients with physical signs of starvation or repeated vomiting, and children with poor growth. [C]

5.2.5.4.

When screening for eating disorders one or two simple questions should be considered for use with specific target groups (for example, ‘Do you think you have an eating problem?’ and ‘Do you worry excessively about your weight?’). [C]

5.2.5.5.

Young people with Type 1 diabetes and poor treatment adherence should be screened and assessed for the presence of an eating disorder. [C]

5.2.5.6.

In anorexia nervosa, although weight and body mass index (BMI) are important indicators of physical risk they should not be considered the sole indicators (as on their own they are unreliable in adults and especially in children). [C]

5.2.5.7.

In assessing whether a person has anorexia nervosa, attention should be paid to the overall clinical assessment (repeated over time), including rate of weight loss, growth rates in children, objective physical signs and appropriate laboratory tests. [C]

5.2.5.8.

Patients with enduring anorexia nervosa not under the care of secondary care services should be offered an annual physical and mental health review by their general practitioner. [C]

Copyright © 2004, The British Psychological Society & The Royal College of Psychiatrists.

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Enquiries in this regard should be directed to the British Psychological Society.

Bookshelf ID: NBK49312

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.1M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...