Eating disorders affect 2-3% of people and 90% of sufferers are women. Only a minority of sufferers present for treatment. Effective treatments are available and the outcome can be very good. General practitioners have an important role in the diagnosis and treatment of eating disorders. Anorexia nervosa is preferably treated in a specialist centre with access to inpatient facilities. Bulimia nervosa and related eating disorders, such as binge eating disorder, respond to a number of psychotherapeutic approaches, particularly cognitive behaviour therapy. A selective serotonin reuptake inhibitor may be effective in the short term.
Introduction and clinical features
Case reports of anorexia nervosa date from the late 17th century and definitive descriptions appeared in the 1870s. Reports of bulimia nervosa only emerged this century, with recognition of the disorder in the late 1970s.
In Australia, the most widely used diagnostic criteria for eating disorders are those of the 4th edition of the American Psychiatric Association Diagnostic and Statistical Manual.1
Patients with anorexia nervosa are characterised by a relentless pursuit of thinness, resulting in weight loss and a refusal to maintain a normal body weight. The essential features of bulimia nervosa are the presence of regular episodes of uncontrolled overeating of large amounts of food, namely binge eating, associated with weight control methods to `counteract' the perceived and feared effects of overeating, and an intense preoccupation with weight and shape issues as expressions of self-worth. Using present criteria, if features of both anorexia and bulimia are present, the diagnosis of anorexia nervosa takes precedence.
Many patients who present with an eating disorder do not meet these diagnostic criteria and are classified as having eating disorder not otherwise specified (EDNOS). An example is binge eating disorder, where there are recurrent episodes of binge eating in the absence of the regular use of inappropriate compensatory behaviours characteristic of bulimia nervosa.
In clinical practice, the boundaries between binge eating disorder and the non-purging form of bulimia nervosa may be indistinct.
Epidemiology and risk factors
Anorexia nervosa is relatively uncommon, but occurs worldwide. The point prevalence is no more than 0.5% of women over 15 years old. The community prevalence of bulimia nervosa is higher, between 0.5 and 1.0%, with an even social class distribution. The estimated community prevalence of EDNOS brings the combined prevalence of eating disorder syndromes in the community to around 2-3%. The majority of sufferers, up to 90%, are women. The evidence for changes in the incidence of anorexia nervosa over time is controversial, and there is certainly no 'epidemic'. There is better evidence for a true increase in the incidence of bulimia nervosa in the past two decades.
Many of the risk factors for eating disorders are not specific to eating disorders. The elucidation of the relative importance of risk factors for different eating disorder syndromes and other psychological disorders, such as depression, is ongoing. The pathway by which, for example, low self-esteem leads to the development of eating disorders varies between individuals. This reflects a complex interplay between biological, psychological and social factors.
The diagnosis is based on clinical assessment and no special investigations are necessary for diagnosis. The assessment includes:
- attitudes to weight, shape and body image
- detailed dietary intake and weight history
- menstrual function history
- presence of binge eating, extreme restrictive dieting, excessive exercise and/or purging (vomiting and/or laxative or diuretic misuse).
The risk factors include:
- female gender
- restrictive dieting
- substance abuse
- personal and family histories of obesity and/or mood disorders
- low self-esteem and perfectionism.
In anorexia nervosa, self-imposed weight loss, with body weight less than 85% expected (or a failure to gain weight with developmental maturation in younger patients), occurs. Bulimia nervosa can be more difficult to identify because of the extreme secretiveness of patients about binge eating and purgative behaviours. Weight may be relatively normal, but patients often describe a past history of anorexia nervosa or restrictive dieting and some alternate between the two disorders. Assessment should include an evaluation of the quality of interpersonal relationships and family environment. These are often disrupted, and can be important factors in the persistence of the disorder.
A physical examination, full blood count, urea, electrolytes (including phosphate) and an electrocardiogram are mandatory to assess the severity of physical complications of starvation and purging.2 In moderate to severe cases, a chest x-ray, glucose, B12, folate, transketolase, liver and thyroid function tests should also be considered. Hormonal investigations are not warranted unless major pituitary or ovarian dysfunction is suspected. Bone densitometry is useful to define the degree of bone loss which may persist despite an improvement in weight and resumption of normal menstruation. At very low weights, cognitive impairment (and cerebral atrophy) is found and this may not reverse.
Bulimia nervosa patients usually require fewer investigations, but serum urea, creatinine and electrolyte tests are indicated. A chance finding of an abnormal result, such as hypokalaemia, may suggest the diagnosis of an unsuspected eating disorder if not previously considered.
Treatment - anorexia nervosa
Denial of the severity or even existence of a problem is extremely common in this condition, and the early phases of treatment will often be difficult. The guiding principles of treatment are the restoration of a normal weight for height and age and then the identification and resolution of the contributing family and personal problems. A multidisciplinary approach is necessary, including individual and family therapy, dietary advice and, where appropriate, pharmacotherapy. Formal family therapy has been shown to improve the outcome of adolescent patients still residing with their families, but its status for older patients is less certain.
Anorexia nervosa is a potentially fatal condition with significant mortality levels and a high morbidity. Around 40% of patients will make a good 5-year recovery, 40% will remain symptomatic but function reasonably well, and 20% of patients remain severely symptomatic and chronically disabled. Referral to a specialised service should always be contemplated, even in those individuals whose weight loss is not yet marked. However, access to such services may be particularly difficult for patients in rural or remote areas.
The treatment of anorexia nervosa in many units has evolved from long-term (up to 12 or more months) inpatient programs with outpatient follow-up, to shorter- or medium-term (e.g. 6-12 weeks) hospital admissions for supervised weight restoration. These are followed by more intensive outpatient programs with hospital backup or, less commonly, partial hospitalisation programs. It is now much more common to begin treatment as an outpatient and the primary care practitioner is often involved in assessment. General practitioners are also frequently asked to assist with monitoring of progress after discharge. Inpatient care is reserved for those individuals who fail to make progress, or are at physical risk with, for example, rapid and/or severe weight loss (including young patients, where it is essential to gain weight to allow normal physical development) or who are at risk of suicide. Insisting on admission to hospital is difficult, but can be life-saving. Such admissions to hospital should preferably be in consultation with members of the ongoing treatment team.
Antidepressant medication is not routinely prescribed. It is indicated if there is a co-morbid depressive illness. In the binge eating/purging subtype of anorexia nervosa, a selective serotonin reuptake inhibitor may be a useful adjunct to psychotherapy. If a patient is highly agitated, especially when attempting to eat, then a major tranquilliser that does not have the disinhibiting effects of an anxiolytic drug may be useful. This prescription should probably be part of the formal treatment program. There may also be a role for prokinetic drugs, such as cisapride, in patients who suffer significant distress with meals and where there is delayed gastric emptying. However, these drugs are not commonly indicated. The most appropriate management of osteopenia is restoration of normal menstruation and weight. The efficacy of hormone 'replacement' therapy or bisphosphonates for preventing continued bone loss is not established.3
Treatment - bulimia nervosa
Although bulimia nervosa is a recently recognised disorder, there has been extensive research into its treatment. The choice of treatment influences the outcome. A solely behavioural approach is likely to be less effective than full cognitive behaviour therapy or interpersonal psychotherapy. 4
Cognitive behaviour therapy is an effective approach in bulimia nervosa (Table 1). Therapy helps the eating disorder and other psychological aspects such as self-esteem. It aims to change not only the eating habits and weight control behaviours, but also the preoccupation with shape and weight. Comprehensive descriptions of this form of therapy are available (see Further Reading).
For some patients, cognitive behaviour therapy is unnecessarily intensive, but for others, it is not sufficient, and other psychological therapies may be as effective. Longer-term psychotherapies may be needed in patients with severe personality disturbance or other problems, such as post-traumatic stress disorder, often following sexual or other abuse in their formative years. Patients with bulimia nervosa should be admitted to hospital if they are at risk of suicide, medically unwell, in the first trimester of pregnancy (because of the risk of spontaneous abortion), or if their symptoms are refractory to outpatient care. A short period in hospital can sometimes be useful to help the sufferer obtain control over their symptoms.
Outcome studies have found that about 50% of patients make a full recovery, about 30% make a partial recovery and 20% continue to be notably symptomatic. A small number of studies have found a good prognosis in bulimia nervosa to be associated with a shorter illness duration and absence of personality disorder. One study has also found a poor prognosis to be associated with premorbid and paternal obesity. 4
Much less is known about the treatment and outcome of those with binge eating disorder or other EDNOS syndromes. However, it is most likely that strategies similar to those used for patients with bulimia nervosa would be appropriate and helpful to sufferers.
The role of the general practitioner (Table 2)
General practitioners can play an important part in management, especially in the recognition of the disorders. Initial education for sufferers and their families about bulimia nervosa and its complications is very important. An empathic and informed response is critical to the treatment alliance and outcome. For those with an interest, further training in psychotherapy for bulimia nervosa and related EDNOS syndromes is now available in several centres. Therapies such as cognitive behaviour therapy lend themselves well to the primary care setting where patients may be more accepting of help.
Families of anorexia nervosa sufferers are commonly disrupted by the behaviour of the anorexic patient, and will need support and advice. The general practitioner can play a useful role in providing ongoing information and in ensuring that families continue to maintain their children's safety, despite entreaties by the children that they will best recover by being left alone. For patients who refuse active treatment, respond poorly to treatment or live in isolated areas, the general practitioner is often asked to provide ongoing monitoring of the patient's physical health.
General practitioners are also in a key position to increase public awareness about the risks of restrictive dieting. This can be done in a number of ways, including the promotion of healthier attitudes towards weight and shape, and provision of sound nutritional advice in waiting rooms and surgeries.
Eating disorders are a notable source of morbidity and it is known that only a minority of sufferers present for treatment. Primary prevention has appeared elusive, although attempts are being made such as the comprehensive educational programs in Norwegian schools. The role of the family practitioner in secondary prevention strategies, such as the early identification of sufferers, is very important as there are now a range of effective treatments and the outcome can be very good.
This paper is based in part on prior work by the author in collaboration with Professor Peter N. Gilchrist, Professor David I. Ben Tovim and Kay Walker, of the Weight Disorder Unit, Flinders Medical Centre, Adelaide.
Fairburn CG, Hay PJ. The treatment of bulimia nervosa. Ann Med 1992;24: 297-302.
Garner DM, Garfinkel PE, editors. Handbook of treatment for eating disorders. 2nd ed. New York: Guilford Press, 1997.
Brownell KD, Fairburn CG, editors. Eating disorders and obesity: a comprehensive handbook. New York: Guilford Press, 1995.
American Psychiatric Association. Practice guidelines for eating disorders. Am J Psychiatry 1993;150:212-28.
Recent advances in the treatment of bulimia nervosa. 8th World Psychiatric Association Conference. Athens, October 1989. J Psychosom Res 1991;35 (1 Suppl):1S-49S. (In particular, pp41-49 - Freeman CP. A practical guide to the treatment of bulimia nervosa.)
- American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: American Psychiatric Association, 1994.
- Sharp CW, Freeman CP. The medical complications of anorexia nervosa. Br J Psychiatry 1993;162:452-62.
- Garfinkel PE, Walsh BT. Drug therapies. In: Garner DM, Garfinkel PE, editors. Handbook of treatment for eating disorders. 2nd ed. New York: Guilford Press, 1997:372-80.
- Fairburn CG, Norman PA, Welch SL, O'Connor ME, Doll HA, Peveler RC. A prospective study of outcome in bulimia nervosa and the long-term effects of three psychological treatments. Arch Gen Psychiatry 1995;52:304-12. (randomised trial)
- Wood A. Pharmacotherapy of bulimia nervosa - experience with fluoxetine. Int Clin Psychopharmacol 1993;8:295-9.