Clinical review: Eating disorders

The diagnosis and management of eating disorders in primary care, covering validated screening tools, differential diagnoses and clinical features.

Section 1: Epidemiology and aetiology
Section 2: Making the diagnosis
Section 3: Managing the condition
Section 4: Prognosis
Section 5: Case study
Section 6: Evidence base

Section 1: Epidemiology and aetiology

Eating disorders are a group of mental illnesses characterised by severe disturbances in eating habits and distress, and excessive preoccupation about weight and shape. They include anorexia nervosa (AN), bulimia nervosa (BN), binge eating disorder (BED) and atypical variants.

Eating disorders can occur at any weight. Obesity is not classified as an eating disorder, since it is defined by a weight metric rather than by thoughts and behaviours. However, obesity is a risk factor for onset of eating disorders because it is associated with high levels of body image dissatisfaction and dieting behaviour, and can also be a marker of disordered eating, in particular binge eating.

Eating disorders have the highest mortality rates among psychiatric disorders and are increasingly recognised as an important cause of morbidity and mortality in young individuals. Due to the nature of these illnesses and their impact on both psychological and physical well-being, treatment and recovery is often a long and complex process requiring a robust multidisciplinary team approach.

A report commissioned by the charity Beat in 2015 found that more than 725,000 people in the UK are affected by an eating disorder with an upward trend in prevalence over time (approximately 7% each year). The increase was largely accounted for by increases in atypical presentations, and presentations in boys and men.

The onset of AN is typically around 15-24 years of age with the highest incidence occurring in adolescent girls between 15-19 years (2 per 1000). BN has a slightly later peak of incidence, and BED even later - often into adult maturity. Eating disorders affect women approximately 7-10 times more than men, although this ratio is changing. 


Causation is bio-psychosocial and multifactorial, involving the following:

  • Genetic factors: first-degree female relatives and monozygotic twin offspring of patients with eating disorders have higher rates of AN and BN. A child is 10 times more likely to suffer from an eating disorder if they have a family history of eating disorders and bulimia is 4 times more likely to occur in relatives of patients suffering with the condition.
  • Neurobiological factors: characteristic neuropsychological profiles have been found in adults with restrictive AN on recovery. Body and brain changes at puberty are thought to play a part.
  • Social factors: eating disorders are associated with perfectionism, and are therefore found in high achieving individuals, although in most studies IQ is not significantly above average. Certain professions, such as ballet dancers and gymnasts, have a higher incidence of eating disorders. The influence of the media and in particular the rise of social media, along with the increasing popularity of what is termed `clean eating' and diets excluding entire food groups may also contribute towards the recent rise in eating disorders.
  • Individual factors: childhood temperament and personality traits, such as perfectionism, rigidity, drive for thinness and obsessive-compulsive traits, may be predisposing factors. Puberty and significant life events (school changes, bereavement, or trauma) may further increase the risk and act as a trigger for the onset of illness. High personal expectations, low self-esteem and early traumatic experiences are risk factors.
  • Family factors: studies suggest that non-shared factors (patient specific) are more important than shared (within the family) in terms of risk. Family factors and relationship dynamics within the home may be more important in maintaining eating disorders than in their causation.


The ICD-10 and DSM-5 are the mostly widely used classification systems used to diagnose eating disorders. 

Anorexia nervosa is characterised by intentional weight loss, low body weight (less than 15% of that expected or BMI <17.5), body image distortion, excessive weight or shape concerns, avoidance of fattening foods and other behaviours contributing to weight loss and endocrine disturbances (possible sexual dysfunction, amenorrhoea, poor growth, and delay or arrest of pubertal changes).

Bulimia nervosa is characterised by recurrent episodes of binge eating (uncontrolled consumption of large amounts of food in a short space of time) followed by compensatory behaviours (fasting, self-induced vomiting, excessive exercise, or laxative/diuretic abuse) occurring at least once a week for 3 months, and an undue influence of body weight and shape in self-evaluation. 

The remaining eating disorders differ depending on the classification system used. They include binge eating disorder (recurrent binge episodes without associated compensatory behaviours) and atypical variants (typically one defining feature is absent, or the frequency of behaviours is below the diagnostic threshold). All feeding and eating disorders included in the ICD-10 and DSM-5 are listed in table 1 (below).

Box 1: Eating disorders according to the main classification systems
ICD-10*  Eating disorders DSM-5  Feeding and eating disorders
F50.0 Anorexia nervosa 307.1 Anorexia nervosa
F50.1 Atypical anorexia nervosa 307.51 Bulimia nervosa
F50.2 Bulimia nervosa 307.51 Binge eating disorder
F50.3 Atypical bulimia nervosa 307.59 Avoidant/restrictive food intake disorder (ARFID)
F50.4 Overeating associated with other psychological disturbances 307.53 Rumination disorder
F50.5 Vomiting associated with other psychological disturbances 307.52 Pica
F50.8 Other eating disorders 307.59 Other specified feeding or eating disorder
F50.9 Eating disorder, unspecified 307.50 Unspecified feeding or eating disorder

* Note: In ICD 10 feeding disorders appear in a separate section of disorders first diagnosed in childhood and early infancy. DSM 5 has removed the diagnosis ‘feeding disorder’, which is now subsumed within ARFID. ICD 11 is likely to follow a similar model.

Section 2: Making the diagnosis

Patients with eating disorders are often taken to see their family GP or another healthcare professional by concerned relatives/friends, sometimes against their will and when the illness has already become established.

If the sufferer themselves is the one seeking help, it may be covert - in other words they may seek help for other symptoms due to shame or fear of stigma. It is therefore vital to identify signs early and to undertake steps towards treatment with the help of a experienced GP, practice nurse, dietitian and counsellor. 

Once an eating disorder is suspected, referral to secondary care for specialist advice and management is recommended in most cases, with a view to transfer back to the primary care once stable or in remission. Occasionally, treatment may be delivered in a primary care setting, under the overall supervision of a specialist community eating disorders team. The urgency of the referral will depend on an initial risk assessment. 

The pathway for referral to treatment for a young person (<18 years="" with="" an="" eating="" disorder="" within="" england="" is="" now="" subject="" to="" a="" href="" target="_blank" data-mce-href="">access and waiting times standards, with the aim that all patients receive an empirically-supported treatment within four weeks from the point of identification, or sooner if urgent. Treatment is delivered in the context of specialist community eating disorders services for young people (CYP-EDS). 

The guidance includes the requirement of GPs to contact the CYP-EDS via telephone or electronically following discussion with the child or young person and their parents or carers, as soon as an eating disorder is first identified.

In a minority of cases, where the child or young person is extremely physically compromised or where there is very high psychiatric risk (for example, suicidality), the GP should arrange immediate paediatric and/or psychiatric care to manage risk via local protocols and should inform the CYP-EDS accordingly. These standards are good practice for care across the age range, although as yet these are not mandated for adult care.

When making a diagnosis and assessing risk to establish the urgency of referral, it is important to consider the following:

  • A detailed history of eating difficulties and weight/shape concerns, taking into account parent/carer concerns as well as those of the young person
  • Methods used by the patient to control weight such as dietary restriction, laxative abuse, self-induced vomiting, excessive exercise, inappropriate use of diuretics and so on
  • Presence of any subjective/objective binge episodes
  • Current weight expressed as BMI (kg/m2) adjusted for age and gender by plotting on a BMI centile chart or calculating %BMI/median BMI for age and gender. BMI <15 or="" 9th="" centile="" 85="" median="" bmi="" a="" history="" of="" very="" rapid="" weight="" loss="">1 kg/ week) suggest further physical examination is needed to establish the degree of medical compromise (see box 2)
  • Presence of other comorbid psychiatric illnesses (depression and anxiety disorders are common)
  • Mental state examination
  • Menstrual history
  • General medical history
  • Medication history
  • Family history including any history of eating difficulties within the family

Behavioural indicators of an eating disorder include avoidance or reluctance to seek professional help, seeking help for physical symptoms, resisting weighing and examination, covering the body, being secretive or evasive when symptoms are being explored, increased energy (and in some cases agitation), and appearing angry or distressed when asked about eating problems. 

Clinicians should be aware of physical features, such as emaciation, dry skin, hair changes, amenorrhoea, constipation, delayed gastric emptying, syncope, myopathy, enamel erosion (from repeated vomiting), callused knuckles (from scraping against incisors when inducing vomiting) and swollen salivary glands.

The patient's beliefs and fears are best elicited through a face to face assessment, although disordered behaviours may be better reported by family members. The SCOFF questionnaire, although only validated in adults, can provide a framework for screening. The SCOFF questions are:

  • Do you make yourself Sick because you feel uncomfortably full?
  • Do you worry that you have lost Control over how much you eat?
  • Have you recently lost more than One stone (14 lb) in a 3-month period?
  • Do you believe yourself to be Fat when others say you are too thin?
  • Would you say that Food dominates your life?

The ICD-10 and DSM-5 criteria can be used to help establish a diagnosis. The Eating Disorders Examination Interview or Questionnaire (EDE/EDE-Q) has been validated for use in children (>8 years) and adults, and is often used by specialist eating disorder teams for diagnosis and research purposes. It also provides a useful springboard to explore eating disorder psychopathology in detail with patients during the initial assessment process.

Box 2: Examination and investigations
  • Check weight and height, temperature, hydration, peripheral circulation, BP (lying and standing), pulse, and look for muscle weakness
  • For young people, plot BMI on centile chart (weight (kg)/height (m2)) or calculate percentage median BMI for age and gender (BMI/median BMI for age and gender)
  • Blood tests: these rarely inform management but can reflect risk. FBC, U&Es, LFTs, TFTs (T3 is often low), ESR (normal or low), growth hormone, LH, FSH (low), cortisol (normal or high)
  • ECG essential if bradycardia, electrolyte imbalance (secondary to vomiting or laxative use), very low BMI
  • Bone mineral density if there is persistent amenorrhoea

Differential diagnosis

The common differential diagnoses can either be physical or psychiatric in origin. A good psychiatric history, physical examination and investigation screen are essential in making the correct diagnosis. However, the relatively common occurrence of eating disorders (affecting up to 13% of adolescents) should be borne in mind when considering alternative diagnoses.

Physical disorders:

  • Chronic wasting diseases: malignancy, peptic ulcer disease, coeliac disease, inflammatory bowel disease, chronic infection.
  • Endocrine disorders: Addison's disease, hyperthyroidism, anterior pituitary insufficiency, diabetes mellitus.

Psychiatric disorders:

  • Mood disorders - depression 
  • Anxiety disorders - obsessive compulsive disorder, generalised anxiety disorder
  • Autistic spectrum disorder
  • Pervasive refusal syndrome, an oppositional defiant disorder

Other causes of amenorrhoea:

  • Pregnancy, primary ovarian failure, prolactinoma, polycystic ovary syndrome, uterine problems and other hypothalamic causes.

Section 3: Managing the condition

Once an eating disorder is identified, the findings of examination and risk assessment should be fed back to the parents and young person together with information about eating disorders or direction to relevant resources, together with actions agreed regarding onward referral.

Referral criteria

The needs of the patient should be the primary basis for decisions on referral. Referrals should be accompanied by a full referral letter explaining why a particular route has been chosen. The letter should include current weight and height as well as other information relevant to assessing risk such as physical observations and the results of any medical investigations including recent blood test results. Include any previous measures of weight or height, for a sense of how severe and long standing the problem is.

Risk assessment will guide how urgently and where treatment should take place, and should be communicated to the patient and carer(s) in a clear and non-judgemental manner. For young people, the risk assessment framework in the Junior MARSIPAN guidelines are useful in making referrals, determining the level of urgency for intervention  and informing future management.

The most severe cases should be directed straight to emergency care. The MARSIPAN checklist aims to identify those at risk of refeeding complications and can be applied across the age range.

Refeeding syndrome

Refeeding syndrome is rare, and occurs when patients who have undergone a period of insufficient nutrition rapidly increase their dietary intake over a short period of time.This can occur in hospital where acutely unwell patients with eating disorders are admitted to be fed, or in the community where patients may binge following a period of fasting/severe dietary restriction. It is most often observed in patients with rapid weight loss and very low BMI, who have eaten little or nothing in the past week or have abnormal biochemical parameters.

Patients and parents should be advised not to increase dietary intake rapidly without advice and close monitoring from an experienced clinical team. Although dieticians/clinicians vary in what is an acceptable starting energy requirement for an individual and how quickly this should be increased, it is agreed that if the risk of refeeding syndrome is high, regular monitoring of physical observations (BP, pulse, presence of oedema) and electrolytes (calcium, phosphate and magnesium) during the initial phase is essential. It is important for clinicians to note that insufficient nutrition (underfeeding) is more likely to cause complications than overfeeding.

If the risk of refeeding is deemed to be high enough to require daily blood tests, the patient should be referred for paediatric or medical admission.

Until they are seen in the specialist clinic, the patient should be seen at least weekly for weight monitoring, blood tests and ECG. Where there is a continued role for GPs in monitoring physical health, regular communication is good practice.

Patients who express or display risk of harm to themselves should be referred to the local child and adolescent or adult eating disorders team. Comorbid mental health problems, such as depression, may resolve with improved nutrition and should be treated as part of a comprehensive eating disorders treatment plan.

New developments

In recent years there has been a move away from inpatient psychiatric care for all but the most severe cases, meaning higher-risk patients are being managed by specialist teams in the community. Effective involvement of families, particularly in children, has proved important in improving outcomes and limiting the burden of these disorders.

Section 4:Prognosis

Eating disorders can be complicated and it can be difficult to predict the outcome. A significant proportion of patients have residual symptoms and relapses throughout their lifetime. The average time to recovery is around five years, depending on age of onset and other prognostic factors.

The medical and psychological consequences of eating disorders may persist throughout life in the patients who remain chronically ill.

Research to date suggests that adolescents have a better prognosis than the adult population, showing long term recovery rates of 70%, compared to 50% in adults. The reason for the better prognosis in adolescents is unknown; it may be due to the support provided by families for the young patient, which may be lacking in adults, or perhaps as a result of the developmental plasticity of adolescence.

With specific regard to anorexia, studies have shown that less than half of patients recover fully, a third improve, and approximately a fifth will remain chronically unwell or deteriorate. 

Mortality rate

The standardised mortality rate is 5.86 for anorexia nervosa and approximately 2 for bulimia and atypical eating disorders.The mortality from anorexia nervosa is estimated to increase by 5% every decade a patient remains ill. 

The mortality of atypical eating disorders is also high, because comorbidity is higher in this group. 

Almost 50% of deaths occurring in eating disorders are due to medical complications of the illness. Of these, cardiac complications are the most common followed by other physiological and biochemical disturbances such as electrolyte abnormalities, or refeeding syndrome. The remainder of deaths have been registered as being due to suicide (25%) and the remaining from unknown causes (25%).

Section 5: Case study

K, a 14-year-old girl with a six-month history of weight loss, was brought to the GP by her mother.

On presentation she appeared cachexic, weak and pale. On examination, her weight was 40kg and her height was 160cm. From her previous records (six months before) she weighed 53kg.

Her mother brought her to see the doctor because she was losing weight and becoming isolated and withdrawn. K was angry about this because she did not feel that anything was wrong with her.

She told the GP that she was in good health until six months ago. Following a move with her family to a different city, she voluntarily restricted her intake and began a regular strenuous exercise programme.

Her mother was concerned because she was no longer eating breakfast and only eating a small salad and a cup of tea for dinner. Her mother recalled herself having some eating problems when she was young. K's mother informed the GP that K had missed her menstrual periods for three months.

Her mother was also concerned about fights around meal times. As K lost increasing amounts of weight, her appetite diminished and she found it easier to diet. She refused to discuss her eating patterns with her GP.


In view of K's rapid decline in weight, her GP examined her physically and ordered blood investigations and an ECG. The examination findings were discussed with K and her mother and the rationale for the investigations explained.

K seemed neither surprised nor concerned when she was told that she had lost more than 25% of her body weight. She was referred to the eating disorders team for management.

Her GP arranged a follow-up appointment for two weeks' time, and gave K's mother advice on when to seek urgent medical assessment.

Section 6: Evidence base

There are a limited number of clinical trials showing evidence of superiority of one treatment over another.

In general, a family-based approach is taken for new onset, young patients, whereas for adults an individual psychological intervention is the usual first step.

Clinical trials

  • A randomised controlled multicentre trial of treatments for adolescent anorexia nervosa including assessment of cost-effectiveness and patient acceptability - the TOuCAN trial. Health Technol Assess 2010; 14(15): 1-98.

Analysis demonstrated no difference in effectiveness for inpatient compared with outpatient treatment, or specialist over generalist treatment, but specialist treatment had advantages increasing over time and was most cost effective.

Outcome from inpatient treatment was poor for those patients who were referred due to failed outpatient treatment.

This study compared the best available treatment for bulimia nervosa, CBT augmented by fluoxetine if indicated, with a stepped-care treatment approach in order to enhance treatment effectiveness.

Therapist-assisted self-help was an effective first-level treatment in the stepped-care sequence, and the full sequence was more effective than CBT at one year, suggesting that treatment of the disorder is enhanced with a more individualised approach.


Key texts

More online

Beating Eating Disorders. 

Academy for Eating Disorders.

Royal College of Psychiatrists Faculty of Eating Disorders

  • Dr Dasha Nicholls is consultant child and adolescent psychiatrist, Great Ormond Street Hospital, London; and chair, Eating Disorders Faculty, Royal College of Psychiatrists
  • Dr Farhana Ahmed is specialist registrar in child and adolescent psychiatry, Great Ormond Street Hospital, London

Take a test on this article and claim your certificate on MIMS Learning

This is an updated version of an article that was first published in January 2012.


  1. Klump KL, Bulik CM, Kaye WH et al. Academy for eating disorders position paper: eating disorders are serious mental illnesses Int J Eat Disord 2009; 42(2): 97-103.
  2. Arcelus J, Mitchell AJ, Wales J et al. Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry 2011; 68(7): 724-31.
  3. American Psychiatric Association. (fourth ed). Washington, DC, American Psychiatric Association, 2000.
  4. Fairburn C, Harrison PJ, Brownell K. Eating Disorders. Lancet 2003, 361: 407-16.
  5. Nicholls DE, Lynn R, Viner RM. Childhood eating disorders: British national surveillance study. Br J Psychiatry 2011; 198: 295-301.
  6. Machado PP, Machado BC, Goncalves S et al. The prevalence of eating disorders not otherwise specified. Int J Eat Disord 2007; 40(3): 212-17.
  7. Morgan JF, Reid F, Lacey JH. The SCOFF questionnaire: assessment of a new screening tool for eating disorders BMJ 1999; 319(7223): 1467-8.
  8. Strober M, Freeman R, Morrell W et al. The long-term course of severe anorexia nervosa in adolescents: survival analysis of recovery, relapse, and outcome predictors over 10-15 years in a prospective study. Int J Eat Disord 1997; 22(4): 339-60.
  9. Steinhausen HC. The outcome of anorexia nervosa in the 20th century. Am J Psychiatry 2002; 159(8): 1284-93.
  10. Gowers SG, Clark AF, Roberts C et al. A randomised controlled multicentre trial of treatments for adolescent anorexia nervosa including assessment of cost-effectiveness and patient acceptability - the TOuCAN trial. Health Technol Assess 2010; 14(15): 1-98.
  11. Mitchell JE, Agras S, Crow S et al.Stepped care and cognitive-behavioural therapy for bulimia nervosa: randomised trial. Br J Psychiatry 2011; 198(5): 391-7.
  12. The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Description and Diagnostic Guidelines, 1992. World Health Organisation, Geneva
  13. The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition,2013. American Psychiatric Association

Suggested further CPD activity

These further action points may allow you to earn more credits.

  • Make a register of all of your patients with eating disorders and ensure they are reviewed on a regular basis.
  • Make a template on your computer system to ensure patients are appropriately followed up.
  • Make a poster of the SCOFF questionnaire for your clinic to remind your colleagues how to screen for eating disorders.

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins


Already registered?

Sign in

Follow Us:

Just published

Under 40s to be offered alternative to AstraZeneca COVID-19 vaccine

Under 40s to be offered alternative to AstraZeneca COVID-19 vaccine

Patients aged 30-39 will be offered an alternative to the Oxford/AstraZeneca COVID-19...

GPs per patient down 10% in just five years, NHS data reveal

GPs per patient down 10% in just five years, NHS data reveal

The number of fully-qualified, full-time equivalent (FTE) GPs per patient in England...

Supportive model must replace harmful CQC inspections, BMA chair warns

Supportive model must replace harmful CQC inspections, BMA chair warns

CQC inspections that ‘divert time away from patient care’ and ‘instil fear’ among...

UK COVID-19 vaccination programme tracker

UK COVID-19 vaccination programme tracker

GPs across the UK are playing a leading role in the largest-ever NHS vaccination...

A 'tsunami' of work is drowning general practice: GPs speak out about a profession at its limit

A 'tsunami' of work is drowning general practice: GPs speak out about a profession at its limit

Rocketing workloads created by huge backlogs of cases and continuing COVID-19 disruption...

Slight increase in GP workforce during past year, but number of partners continues to fall

Slight increase in GP workforce during past year, but number of partners continues to fall

The GP workforce grew by 0.4% in the year to March 2021, but the number of GP partners...