Chapter 19 Eating disorders
Learning outcomes
Introduction
The disorders discussed in this chapter are those classified by the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, DSM-IV-TR (American Psychiatric Association (APA) 2000) as eating disorders. They include anorexia nervosa, bulimia nervosa and eating disorder not otherwise specified (EDNOS). This last category, which includes atypical eating disorders and binge eating disorder, allows diagnosis of people whose disordered eating behaviours and symptoms are not entirely consistent with the criteria for anorexia nervosa or bulimia nervosa.
Eating disorders are characterised by one or more serious disturbances of eating behaviours such as food avoidance, self-induced vomiting, excessive exercising, recurrent episodes of uncontrolled eating and misuse of laxatives or diuretics in an effort to control weight. People with anorexia nervosa or bulimia nervosa are preoccupied with their weight, and judge their self-worth largely, or even exclusively, by their shape and weight and their ability to control them (Fairburn & Harrison 2003). Although DSM-IV-TR describes the criteria for diagnosis of specific eating disorders, symptoms have been known to occur on a continuum between these disorders (Tozzi et al 2005). While clients cannot be diagnosed with anorexia nervosa, bulimia nervosa or EDNOS at the same time, their disorderedeating behaviours can cross over between these disorders over time. Body sculpting and cosmetic procedures to remove weight are also becoming more widely used at the beginning of the twenty-first century.
Historically, descriptions of people with illnesses that might have been anorexia nervosa appear as early as the seventeenth century (Cartwright 2004). Sometime during the late nineteenth and early twentieth centuries, Western societies began to focus on thinness as a mark of beauty, and Russell (1995) argues that the incidence of anorexia nervosa has risen significantly since the 1950s. Bulimia nervosa was first documented in the 1900s and was identified in 1979 as a variant of anorexia nervosa (Russell 1979). EDNOS is a more recently identified and defined category of eating disorder.
Characteristics of eating disorders
Anorexia nervosa
Anorexia nervosa is a complex and usually chronic psychiatric illness with potentially fatal medical complications. It is characterised by determined efforts to lose weight or avoid weight gain (Box 19.1). If the onset of anorexia nervosa is prepubertal, the sequence of pubertal events is delayed or even arrested (growth ceases—in girls the breasts do not develop and there is primary amenorrhoea, and in boys the genitals remain juvenile) (Royal Australian and New Zealand College of Psychiatrists (RANZCP) 2004).
Box 19.1 Diagnostic criteria for anorexia nervosa
Reprinted with permission from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (Copyright 2000) American Psychiatric Association.
The four essential criteria defined in the DSM-IV-TR (APA 2000) for anorexia nervosa and its two subtypes are listed in Box 19.1.
Bulimia nervosa
Bulimia nervosa is characterised by overwhelming urges to over-eat (binge), followed by compensatory behaviour to avoid weight gain, such as self-induced vomiting, excessive exercise, food avoidance or laxative misuse. A cycle is often seen where self-induced nutritional restriction leads to an increased hunger for food, both physiologically and psychologically, which in turn encourages bingeing behaviour. For the person concerned with weight and body shape, there then arises an overwhelming desire to prevent weight gain through behaviours such as vomiting, laxative abuse, over-exercising or restricting food intake again. The use of these compensatory behaviours leads to feelings of disgust and act to reinforce the person’s sense of poor self-worth, and the bulimic cycle begins again.
The main feature that distinguishes bulimia nervosa from anorexia nervosa is that attempts to restrict food intake are punctuated by repeated binge eating, or episodes of eating during which there is an aversive sense of loss of control and an unusually large amount of food is eaten (Fairburn & Harrison 2003). Clients with bulimia nervosa are likely to have normal or near normal body weight.
The five essential criteria defined in the DSM-IV-TR (APA 2000) for bulimia nervosa and its two subtypes are listed in Box 19.2.
Box 19.2 Diagnostic criteria for bulimia nervosa
Reprinted with permission from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (Copyright 2000) American Psychiatric Association.
Eating disorder not otherwise specified
The diagnosis of eating disorder not otherwise specified (EDNOS) is given when the disordered eating behaviour is not entirely consistent with the essential diagnostic criteria for either anorexia or bulimia nervosa. For example, a woman may have all the symptoms of anorexia nervosa except that she may still have a regular menstrual cycle. Binge eating disorder also fits into the category of EDNOS. Binge eating disorder is characterised by recurrent episodes of binge eating without compensatory behaviours to avoid weight gain, a lack of self-control during the binge eating, and marked distress after a binge (APA 2000). As binge eating disorder does not involve avoidance of weight gain, it has a strong association with obesity.
Eating disorders in children and adolescents
Special medical consideration, including early and more aggressive re-feeding to a healthy weight range, is required for children and adolescents with eating disorders as they can more rapidly become medically compromised and permanently lose growth potential (Hamilton 2007). Emaciation occurs more rapidly as younger clients have lower energy stores than adults, and children also dehydrate more quickly than adults do (Royal College of Psychiatrists (RCP) 2005).
In order to prevent potentially irreversible physical growth and developmental complications, paediatricians, child psychiatrists and nurses who are aware of the physiological differences and psychosocial developmental needs of these younger clients, treat children and adolescents with eating disorders. Unlike adults, children should continue to grow in height during the course of their treatment. Therefore their height measurements and healthy target weights need to be revised every three months. When nursing adolescents with eating disorders, the philosophy should be consistent with nursing care aimed at promoting the achievement of adolescent developmental tasks as well as management of the eating disorder (Anderson 1996). Bulimia nervosa is rarely seen in children.
Eating disorders in males
Behaviours such as excessive exercise to increase bulk, bingeing and purging and steroid use are not uncommon among young males and there is concern that eating disorders are increasing among men. There is an increased prevalence in certain subgroups of males who are vulnerable to weight and shape concerns, such as wrestlers and homosexual men (Clarke-Stone & Joyce 2003) and athletes (Muise, Stein & Arbess 2003). The clinical features of eating disorders specific to males include loss of libido, decrease in spontaneous early morning erections, nocturnal emissions and desire to masturbate associated with malnutrition and lower testosterone levels and higher rates of comorbid depression and substance use.
The increased incidence of eating disorders in men may be due to an increasing obsession among men in Western societies with their appearance. In contrast to the popular thin ideal of feminine beauty, young men experience two conflicting social pressures. One is the pressure to ‘bulk up’ or become more muscular. A condition termed muscle dysmorphia that is most prevalent in males has been described (see Ch 21). In contrast to people with eating disorders who see themselves as fat when they are thin, people with muscle dysmorphia feel ashamed of looking small when they are actually well built (Grieve 2007; Pope, Phillips & Olivardia 2000). The other recent social pressure on young men is highlighted by the current popularity of some male role models in the media spotlight, such as musicians and singers, who have been diagnosed with anorexia nervosa or look particularly underweight. The impact of these underweight role models on the prevalence of disordered eating among men is yet to be studied.
Incidence and prevalence
The eating disorders are encountered predominantly in Western industrialised countries but appear to be increasing in non-Western countries. Incidence and prevalence rates are difficult to report accurately as many people with eating disorders go undetected. For example, one Australian study (O’Dea & Abraham 2002) investigated eating, weight, shape and exercise behaviours in young men, and found that one in five men worried about their weight and shape, followed rules about eating, and limited their food intake, and while 9% reported disordered eating, none of the population studied had sought treatment.
Incidence rates refer to the number of new cases in the population over a specified time. Incidence rates for anorexia nervosa have been estimated to be 8 per 100,000 people per year, and 13 per 100,000 people per year for bulimia nervosa (Hoek 2006). The incidence rate for anorexia nervosa in males has been reported to be below 1 per 100,000 people per year (Hoek 2006).
Prevalence rates refer to the actual number of cases in the population at a particular point in time. Eating disorders are more common among adolescent girls and young women, with a 0.3% average prevalence rate reported for anorexia nervosa and 1% for bulimia nervosa (Hoek 2006). Studies of lifetime prevalence have reported various rates, from 0.3% to 3.7% for anorexia nervosa and 1% to 4.2% for bulimia nervosa (APA 2006). Machado et al (2007) reported a 2.7% prevalence rate for EDNOS. One Australian study (Hay 1998) reported a 1% prevalence of binge eating disorder using DSM-IV criteria and a 2.5% prevalence using a broader definition. Binge eating disorder has been described as almost as common among men as women, with approximately 40% of sufferers being male (Pope et al 2000).
Aetiology and risk factors
Gender
Being female in a culture where thinness is equated with beauty and popularity is a risk factor for developing an eating disorder. Paxton (2000) argues that evolutionary factors that predispose women to being more conscious of their appearance may be involved. Ninety per cent of clients with anorexia nervosa are female. Patton et al (1999) argue that dieting is the most important predictor of new eating disorders, and that differences in the incidence of eating disorders between the sexes can be largely accounted for by the higher rates of dieting and psychiatric morbidity in females.
Age
Onset of anorexia nervosa generally occurs during adolescence, although eating disorders are occurring more frequently in children aged from 6 to 12 years (Hamilton 2007). Bulimia nervosa and binge eating disorder are more likely to first occur in late adolescence or early adulthood, although clients with these disorders may not present for treatment until much later in life.
History of dieting
Most eating disorders start with dieting behaviours. One study of adolescent girls found that severe dieters were eighteen times more likely to develop an eating disorder than non-dieters, and that moderate dieters were five times more likely to develop these symptoms (Patton et al 1999). Clients seeking treatment for binge eating disorder, however, have been reported to have weight problems prior to dieting and binge eating behaviour (Reas & Grilo 2007).
Social factors
Low self-esteem and concerns about appearance and body image are exacerbated for many young people by social and cultural pressures to conform to a particular thin ideal of beauty. The media and associated influences such as the fashion industry have been criticised for many years for promoting unrealistically thin images of women. In addition, the ideal bodies presented in the media have become thinner over time. Body comparison, or the process of comparing one’s body with others, such as friends or models and film stars, is rec ognised as a factor contributing to body dissatisfaction, dieting and symptoms of disordered eating (Paxton et al 1999; Stormer & Thompson 1996). Recent pressure is being placed on the fashion, media, marketing and advertising industries to encourage a greater diversity of more realistic weight and body shapes. It will be interesting to see if this pressure has an impact on lowering future incidence and prevalence rates of eating disorders. However, it must be remembered that while all young people are exposed to media and marketing industries, they do not all become eating disordered. Low self-esteem is a common factor in those who do develop eating disorders, and interventions aimed at improving the self-esteem of children and adolescents may also help to decrease the number of people affected by eating disorders. Furthermore, the onset of anorexia nervosa can also be associated with other stressful life events such as loss and grief, child abuse or neglect, bul lying or other developmental traumas. Urban living has also been identified as a risk factor for bulimia nervosa but not for anorexia nervosa (Van Son et al 2006).
Psychological factors
Common personality traits found in clients with anorexia nervosa include low self-esteem, perfectionism, obsessionality, alexithymia (difficulty identifying and expressing feelings verbally) and intimacy concerns along with a sense of not feeling in control of one’s life. Young people with bulimia nervosa tend to be more impulsive, self-critical and demonstrate labile moods in response to environmental events (Sigman 2003).
Cognitive behavioural theories have been proposed to account for the development and maintenance of eating disorders. Fairburn & Harrison (2003) describe two main origins of the restriction of food intake characterising the onset of eating disorders as a need to feel in control of life, which gets displaced onto controlling eating, and over-evaluation of shape and weight in those who have been sensitised to their appearance.
In the psychodynamic literature, clients with anorexia nervosa have been described as having difficulties with separation and autonomy (often manifested as overly close or enmeshed relationships with parents), affect regulation (including the direct expression of anger and aggression) and negotiation of psychosexual development (APA 2006).
Clients with bulimia nervosa have been understood in a number of ways, ranging from viewing symptoms as manifestations of impulsivity or problems with emotion regulation and dissociative states to viewing them along a spectrum of self-harming behaviours commonly seen in borderline personality organisation (Westen & Harnden-Fischer 2001).
Familial factors
Despite its limitations, research indicates that family functioning can play a part in the development and maintenance of weight concerns and eating disorders (May et al 2006). Negative paternal comments and dieting have been known to influence self-esteem, body image and the eating behaviours of children (Cartwright 2004). Freeman (2002) suggests that families with a child with anorexia nervosa often have one parent who is over-involved while the other is passive. Anorexia nervosa can also develop in over-controlled and rigid families that have difficulty expressing and resolving conflict, while bulimia nervosa is more likely to develop when family systems are chaotic (Hamilton 2007).
McIntosh et al (2000) suggest that the families of individuals with restricting anorexia nervosa are more positive than the families of individuals with bulimia nervosa or binge eating/purging anorexia nervosa. In a discussion of bulimia nervosa, Zerbe (1993) identified family discord, lack of consistency, lack of warmth and emotional connection to the child, and failure to protect the child from sexual and physical abuse or perpetuating such abuse as influential. Families of clients with bulimia nervosa have also been shown to have higher rates of substance abuse, affective disorders and certain personality traits such as perfectionism (Lilenfield et al 2000).
Genetic and biological factors
Biological factors known to predispose people to developing eating disorders include genetics and gene–environment interactions (Strober et al 2000). Women who have had anorexia nervosa often have anorexic daughters and the risk of a first-degree relative of an affected person developing the disease is ten times that of the general population (Beumont 2000, p 81). Molecular genetic studies are currently attempting to identify specific genes that contribute to this vulnerability for eating disorders.
Alterations of central nervous system serotonin activity may also directly affect eating behaviours as well as other psychiatric symptoms such as depression and obsessive compulsive symptoms. Low levels of serotonin activity are associated with impulsivity and may predispose to bulimia nervosa, whereas high levels are associated with rigidity and constraint and may predispose to anorexia nervosa. Furthermore, the neurotransmitters serotonin and noradrenaline are regulated by leptin, which acts on the hypothalamus to regulate appetite and weight. It may be that individuals with bulimia have a normal response to leptin but as they try to reduce their weight past their body’s ‘set-point’ for weight, they fail, because of the body’s overwhelming drive to maintain this set point (Abraham & Llewellyn-Jones 2001).
The risk of developing anorexia nervosa has also been shown to increase with the number of perinatal complications (Favaro, Tenconi & Santonastaso 2006).
Medical complications
The protein–calorie malnutrition seen in anorexia nervosa affects every organ in the body, and nurses must be aware of the potential medical complications as these can be life threatening and require urgent resuscitation. Acute complications of anorexia nervosa include bradycardia and cardiac compromise, hypothermia, dehydration, electrolyte disturbance (with purging), gastrointestinal motility disturbances, renal problems, infertility and perinatal complications (Miller et al 2005).
Cardiovascular effects
Fainting and collapse are common reasons for presentation at emergency departments and often indicate serious cardiovascular complications (Cartwright 2004). Cardiac irregularities caused by protein–calorie malnutrition include bradycardia (heart rate less than 50 beats per minute), hypotension (blood pressure less than 80/50 mmHg) and cardiac arrhythmias. ECG abnormalities, including a prolonged QTc interval and non-specific ST segment depression or T wave changes can be associated with electrolyte disturbances as well as malnutrition. The abuse of the emetic ipecac can lead to cardiomyopathy. As cardiac arrest can result from arrhythmias, cardiac monitoring is recommended for clients with acute medical compromise such as bradycardia or a prolonged QTc interval on ECG.
Gastrointestinal effects
Clients who severely restrict their dietary intake describe feeling bloated or full even after eating small amounts of food, which can indicate shrinking of the stomach, or delayed gastric emptying. Binge eating, on the other hand, can lead to gastric dilation and, in rare cases, stomach rupture or death (RANZCP 2004). In assessing bowel activity, diarrhoea can be a sign of laxative abuse, while constipation may result from inadequate food intake including lack of fibre bulk, dehydration or decrease in gastric motility. It should be noted that laxatives used by clients can include those found in com mon household food supplies, such as artificial sweeteners, chewing gum and diet drinks.
Musculoskeletal effects
Osteopenia, osteoporosis and associated risk of frac tures are common in longstanding and severe cases of anorexia nervosa (Grinspoon et al 2000). Irreversible decreased bone mineral density is associated with pro longed malnutrition, low oestrogen levels and amenorrhoea for longer than six months, and decreased muscle mass. A dual-energy x-ray absorbtiometry (DEXA) scan is generally ordered to assess bone mineral density when clients with an eating disorder have experienced amenorrhoea for longer than six consecutive months. Growth retardation can occur in children when the onset of the disorder occurs before closure of the epiphyses.
Dental and oral effects
Dental erosion and caries can occur with recurrent self-induced vomiting. Riboflavin deficiency may cause fissures of the lips, especially in the corners of the mouth, and iron and zinc deficiencies cause glossitis and loss of taste sensation (RCP 2005).
Skin/integument effects
Protein–calorie malnutrition leads to loss of subcutaneous fat, and lanugo, a fine, downy hair that grows on the face and body is often seen. Lanugo is believed to be an adaptation to loss of body fat, and functions to help preserve body temperature. Cool hands and feet with bluish discolouration (peripheral cyanosis), callous on the dorsum of the dominant hand due to repeated self-induced vomiting, brittle nails and dry skin are commonly seen. Carotenaemia, a yellow/orange discolouration of the skin caused by vitamin A overload, usually from eating carrots (seen best on the hands or soles of the feet), is occasionally seen.
Neurological effects
Structural changes in the brain including loss of brain volume, cerebral atrophy and ventricular dilatation have been reported (Fairburn & Harrison 2003; Rome & Ammerman 2003) and there are concerns that some of these changes persist even after re-feeding. Severe electrolyte imbalances can lead to abnormal electrical discharges in the brain and seizures.
Psychiatric comorbidity
Anxiety disorders may precede eating disorders and anxiety may also develop or worsen as weight is restored and treatment progresses. Obsessive-compulsive behaviours (see Ch 17) are observed in clients with eating disorders, particularly anorexia, and usually take the form of repetitive counting and ritualistic eating patterns such as chewing food a certain number of times. In most cases the obsessive-compulsive symptoms tend to resolve as the starvation resolves, unless the condition was premorbid. As well as obsessive-compulsive disorder, social phobia and panic disorders have been identified (Godart et al 2006). Clients with anorexia nervosa and comorbid body dysmorphic disorder have been shown to have a high rate of attempted suicide (Grant, Won Kim & Eckert 2002).
Substance misuse occurs particularly in clients with bulimia nervosa, and includes the use of legal and illicit drugs, such as alcohol, tobacco and amphetamines and the typical drugs used for weight loss, such as caffeine, emetics, diuretics and laxatives. Children and adolescents with anorexia nervosa are less likely to substance abuse although some young people with bulimia nervosa consume alcohol. Some studies support a familial relationship between substance-use disorders and eating disorders, and monitoring for substance misuse throughout treatment has been recommended (Herzog et al 2006; Piran & Gadalla 2007).
Assessment
Physical assessment
A full medical examination includes weight and height measures, vital signs, cardiovascular and peripheral vascular function, metabolic status, dermatological manifestations and evidence of self-harm (APA 2006). The physical assessment, including blood chemistry, urinalysis and ECG will detect any of the medical complications previously described.
In adults, height and weight are used to calculate the body mass index (see Box 19.3), which helps determine the degree of starvation. Children and adolescents are assessed on percentage of ideal body weight or genderspecific standardised growth charts. Plotting the child’s previous growth patterns helps to identify any crossover of centiles above or below that expected on the child’s growth curve trajectories.
Box 19.3 Calculation of body mass index
55 divided by 2.65 equals a BMI of 20.75 kg/m2.
BMI < 16 | requires specialist management |
BMI < 18.5 | underweight |
BMI 19–25 | normal weight |
BMI > 25 | overweight |
BMI > 30 | obese |
Note: These ranges relate to Caucasian populations. Norms for Asian populations may be lower.
Rather than undressing the child or adolescent, sexual or pubertal development can be assessed by asking them to point to the diagrammatic picture most closely matching their own body development on the Tanner stages rating scales that accompany most standardised child growth charts.
Psychiatric assessment
A psychiatric assessment will confirm the specific diagnosis, identify any comorbid psychiatric conditions and exclude other primary diagnosis such as depression which can present as loss of appetite without the body image disturbance and fear of weight gain. Other aspects of the client’s psychiatric status that greatly influence clinical course and outcome include mood, anxiety and substance use disorders, as well as motivational status, personality traits and personality disorders (APA 2006). Denial and minimisation are common in adolescents with anorexia nervosa and can complicate the assessment process (Couturier & Lock 2006). Parents or carers of young clients are interviewed to help validate assessment findings.