Shorter Oxford Textbook of Psychiatry, 6th Ed.

CHAPTER 14. Eating, sleep, and sexual disorders

Eating disorders

Eating disorder not otherwise specified (EDNOS; atypical eating disorders)

Sleep disorders

Disorders of sexual function, preference, and gender identity

The drives to eat, sleep, and have sex can all be impaired or become otherwise dysfunctional in many psychiatric and medical disorders. They can also all be primary disorders, and it is the latter which are the focus of this chapter.

Eating disorders

Eating disorders are characterized by abnormalities in the pattern of eating which are determined primarily by the patient’s attitude to their weight and shape. The disorders share a distinctive core psychopathology which is best described as an over-evaluation of weight and shape, such that patients judge their self-worth in terms of their shape and weight and their ability to control these. The disorders covered in this chapter include anorexia nervosa and bulimia nervosa as well as a number of related conditions. Obesity is not covered, as it is not a psychiatric disorder (Marcus and Wildes, 2009), although it is associated with an increased risk of various psychiatric disorders (Petry et al., 2008).

Until the late 1970s, eating disorders were believed to be uncommon. Following the description of bulimia nervosa, they have increasingly been seen as conspicuous and disabling. It remains uncertain whether the rapid rise in presentation and diagnosis reflects a true increase in incidence or an increase in detection and diagnosis (Curren et al., 2005). Many eating disorders go clinically unrecognized, and it is estimated that only about 50% of the cases of anorexia nervosa in the general population are detected in primary care; for bulimia nervosa the figure is substantially less, and most individuals with bulimia are untreated.

Anorexia nervosa and bulimia nervosa are part of a larger range of eating disorders which include those that are clinically very similar to the two main diagnoses, but fail to meet their precise diagnostic criteria. These disorders are classified within DSM-IV as ‘Eating disorders not otherwise specified (EDNOS).’ They are also known as atypical eating disorders, and in community samples are actually more frequent than either anorexia nervosa or bulimia nervosa. In fact, many patients with EDNOS have previous histories of anorexia nervosa or bulimia nervosa, and some patients with EDNOS will eventually meet the criteria for one of the two latter diagnoses. The fact that patients with eating disorders tend to ‘migrate’ between these various diagnoses (see Figure 14.1) suggests that they all share a common pathophysiology, and that the boundaries between them are largely arbitrary (Fairburn and Harrison, 2003; see also Eddy et al., 2008). Despite these reservations about the current classification of eating disorders, this chapter will follow the standard terminology as used in DSM-IV and ICD-10 (see Table 14.1).


Figure 14.1 Schematic representation of temporal movement between the eating disorders. The size of the arrow indicates the likelihood of movement in the direction shown. Arrows that point outside of the circle indicate recovery. Reprinted from The Lancet, 361 (9355), Christopher G Fairburn and Paul J Harrison, Eating disorders, pp. 407–16, Copyright (2011) with permission from Elsevier.

Anorexia nervosa

Although there were many previous case histories, anorexia nervosa was first named in 1868 by the English physician William Gull, who emphasized the psychological causes of the condition, the need to restore weight, and the role of the family. The other key description at this time was by Charles Lasegue in Paris. The main features are very low body weight (defined as being 15% below the standard weight, or body mass index (BMI) of less than 17.5 kg/m2), an extreme concern about weight and shape characterized by an intense fear of gaining weight and becoming fat, a strong desire to be thin and, in women, amenorrhoea (for DSM-IV criteria, see Box 14.1).

Most patients are young women (see the section on epidemiology below). The condition usually begins in adolescence, although childhood-onset and older-onset cases are encountered. It generally begins with ordinary efforts at dieting, which then get out of control. The central psychological features are the characteristic overvalued ideas about body shape and weight (Fairburn et al., 1999a). The patient may have a distorted image of her body, believing herself to be too fat even when she is severely underweight. This belief explains why most patients do not want to be helped to gain weight.

The pursuit of thinness may take several forms. Patients generally eat little and set themselves very low daily calorie limits (often between 600 and 1000 kcal). Some try to achieve weight loss by inducing vomiting, exercising excessively, and misusing laxatives. Patients are often preoccupied with thoughts of food, and sometimes enjoy cooking elaborate meals for other people. Some patients with anorexia nervosa admit to stealing food, either by shoplifting or in other ways.

Table 14.1 Diagnostic classification of eating disorders


Binge eating. A subgroup of patients have repeated episodes of binge eating (uncontrollable overeating). This behaviour becomes more frequent with chronicity and increasing age. During binges, the patient typically eats foods that are usually avoided. After overeating they feel bloated and may induce vomiting. Binges are followed by remorse and intensified efforts to lose weight. If other people encourage them to eat, the patient is often resentful; they may hide food or vomit in private as soon as the meal is over. In DSM-IV, anorexia nervosa with binge eating and purging (self-induced vomiting or the misuse of laxatives or diuretics) is recognized as a distinct type, which differs from the restricting type.

Amenorrhoea is one of several physical abnormalities that have traditionally been incorporated in diagnostic criteria. It occurs early in the development of the condition, and in about 20% of cases it precedes obvious weight loss, although careful history taking generally reveals that these patients had already started dieting. Some cases first come to medical attention with amenorrhoea rather than disordered eating.

Other symptoms. Depressive, anxiety, and obsessional symptoms, lability of mood, and social withdrawal are all common. Lack of sexual interest is usual. For a review of anorexia nervosa, see Russell (2009).

Physical consequences

A number of important symptoms and signs of anorexia nervosa are secondary to starvation. Several body systems can be affected (see Table 14.2). For a review of the physical consequences, see Mitchell and Crow (2006).

Box 14.1 DSM-IV criteria for anorexia nervosa

• Refusal to maintain body weight at or above a minimally normal weight for age and height (e.g. weight loss leading to maintenance of body weight at less than 85% of that expected, or failure to make expected weight gain during period of growth, leading to body weight less than 85% of that expected).

• Intense fear of gaining weight or becoming fat, even though underweight.

• Disturbance in the way in which one’s body weight or shape is experienced, undue influence of body weight or shape on self-evaluation, or denial of the seriousness of the current low body weight.

• In post-menarchal females, amenorrhoea—that is, the absence of at least three consecutive menstrual cycles. (A woman is considered to have amenorrhoea if her periods occur only following hormone administration, e.g. oestrogen treatment.)


Restricting type. During the current episode of anorexia nervosa, the person has not regularly engaged in binge-eating or purging behaviour (i.e. self-induced vomiting or the misuse of laxatives, diuretics, or enemas).

Binge-eating/purging type. During the current episode of anorexia nervosa, the person has regularly engaged in binge-eating or purging behaviour (i.e. self-induced vomiting or the misuse of laxatives, diuretics, or enemas).


Incidence. Estimates of the incidence of anorexia nervosa which are based on case registers in the UK and the USA range from 0.4 to 4 per 100 000 members of the population per year, but are much higher among young women (Keski-Rahkonen et al., 2007). Reported incidence rates increased from the beginning of the twentieth century up to the 1970s, but have probably remained fairly stable since then (Curren et al., 2005).

Prevalence. A large US community survey found a lifetime prevalence of 0.9% among women and 0.3% in men (Hudson et al., 2007), and a comparable Finnish study found a 2.2% prevalence in women, with 50% of cases unknown to services (Keski-Rahkonen et al., 2007). The gender difference is greater in clinical samples. The condition is more common in the upper than the lower social classes, and is reported to be rare in non-Western countries and in the non-white population of Western countries (Hoek and van Hoeken, 2003).


Genetics and neurobiology

Among the female siblings of patients with established anorexia nervosa, 5–10% suffer from the condition, compared with 0.5–1.0% of the general population of the same age. Siblings also have an increased risk of other eating disorders, suggesting a common familial liability. This increase in risk might be due to family environment or to genetic influences. A surprisingly large heritability (up to 75%) is reported in twin studies (Bulik et al., 2007a), although the figures have been questioned (Fair-burn et al., 1999b). Several genes have been implicated, including the 5-HT transporter (Lee and Lin, 2010), but most of the findings have not been well replicated, and a recent genome-wide study had no unequivocal positive findings (Wang et al., 2011). Neurobiologically, the focus is on alterations in neurotransmitter systems, such as changes in 5-HT1A and 5-HT2A receptors (Kaye, 2008). However, it is often difficult to determine whether abnormalities are causal, or are the result of starvation and weight loss.

Social factors

The fact that anorexia nervosa is more common in certain societies suggests that cultural factors play a part in its development. Important among such factors is likely to be the notion that thinness is desirable and attractive. Surveys in more affluent societies show that most schoolgirls and female college students diet at one time or another. However, when other risk factors are taken into account, people who develop anorexia nervosa have no greater exposure to factors that increase the risk of dieting. This suggests that the problem is more due to how an individual reacts to dieting than to dieting itself.

Individual psychological causes

Bruch (1974) was one of the first writers to discuss the psychological antecedents of anorexia nervosa. She suggested that these patients are engaged in ‘a struggle for control, for a sense of identity and effectiveness, with the relentless pursuit of thinness as a final step in this effort.’ These clinical observations are supported by epidemiological studies, which implicate low self-esteem and perfectionism in the development of the disorder (Fairburn, 1999). It has been suggested that these premorbid personality traits can make it particularly difficult for an individual to negotiate the demands of adolescence.

Table 14.2 Main physical features of anorexia nervosa

Physical symptoms

• Heightened sensitivity to cold

• Gastrointestinal symptoms—constipation, fullness after eating, bloating

• Dizziness and syncope

• Amenorrhoea (in females not taking an oral contraceptive), lack of sexual interest, infertility

• Poor sleep, with early-morning wakening

Physical signs

• Emaciation; stunted growth and failure of breast development (if onset is preprubertal)

• Dry skin; fine downy hair (lanugo) on the back, forearms, and sides of face; orange discolouration of the skin of the palms and soles

• Swelling of the parotid and submandibular glands (especially in bulimic patients)

• Erosion of the inner surface of the front teeth (perimylolysis) in those who vomit frequently

• Cold hands and feet; hypothermia

• Bradycardia; hypotension; cardiac arrhythmias (especially in underweight patients and those with electrolyte abnormalities)

• Dependent oedema (complicating assessment of body weight)

• Weak proximal muscles (elicited as difficulty in rising from a squatting position)

Abnormalities on physical investigation

Endocrine abnormalities

• Low concentrations of luteinizing hormone, follicle-stimulating hormone, and oestradiol

• Low T3 with T4 in low normal range, normal concentrations of thyroid-stimulating hormone (low T3 syndrome)

• Increase in plasma cortisol and dexamethasone non-suppression

• Raised growth hormone concentration

• Severe hypoglycaemia (rare)

Cardiovascular abnormalities

• ECG abnormalities (especially in those with electrolyte disturbance)

• Conduction defects, especially prolongation of the QT interval

Gastrointestinal abnormalities

• Delayed gastric emptying

• Decreased colonic motility (secondary to chronic laxative misuse)

• Acute gastric dilatation (rare, secondary to binge eating or excessive re-feeding)

Haematological abnormalities

• Moderate normocytic normochromic anaemia

• Mild leucopenia with relative lymphocytosis

• Thrombocytopenia

Other metabolic abnormalities

• Hypercholesterolaemia

• Raised serum carotene

• Hypophosphataemia (exaggerated during re-feeding)

• Dehydration

• Electrolyte disturbance (varied in form; present in those who vomit frequently or misuse large quantities of laxatives or diuretics)

• Possible hypokalaemia distinct from electrolyte disturbances

Other abnormalities

• Osteopenia and osteoporosis (with heightened fracture risk)

Reproduced with permission from Fairburn and Harrison (2003).

Causes within the family

Disturbed relationships are often found in the families of patients with anorexia nervosa, and some authors have suggested that they have an important causal role. Minuchin et al. (1978) held that a specific pattern of relationships could be identified, consisting of ‘enmeshment, overprotectiveness, rigidity and lack of conflict resolution.’ They also suggested that the development of anorexia nervosa in the patient served to prevent dissent within the family.

Epidemiological studies suggest that people who develop anorexia nervosa are more likely than healthy controls to be exposed to a range of childhood adversities, including poor relationships with parents and parental psychiatric disorder, particularly depression. However, these risk factors are not specific to anorexia nervosa, but are found with equal frequency among people who subsequently develop other psychiatric disorders (Fairburn, 1999). It seems likely that these general risk factors interact with specific factors within the individual, such as perfectionism and low self-esteem, to increase the risk of developing anorexia nervosa.

Course and prognosis

In its early stages, anorexia nervosa often runs a fluctuating course, with exacerbations and periods of partial remission, and full recovery is not uncommon in cases with a short history. The long-term prognosis is difficult to judge due to incomplete follow-up, or because there may be normalization in weight or menstrual function but persistent abnormalities of eating habits and attitudes to weight and shape. One large population study found that about two-thirds of women with anorexia nervosa had largely or fully recovered at 5 years (Keski-Rahkonen et al., 2007).

There is a sixfold increase in mortality, due both to suicide and to natural causes (Papadopoulos et al., 2009), the latter reflecting the many medical complications of anorexia nervosa (see Table 14.2). The elevated mortality rate remains 20 years after first hospitalization. Reported poor prognostic factors include older age at onset, long history, premorbid personality problems, comorbid substance misuse, and childhood obesity.


Box 14.2 Assessment of eating: some topics to be included

What is a typical day’s eating? To what degree is the patient attempting restraint?

Is there a pattern? Does it vary? Is eating ritualized?

Does the patient avoid particular foods? And if so, why?

Does she restrict fluids?

What is the patient’s experience of hunger or of any urge to eat?

Does she binge? Are these objectively large binges? Does she feel out of control?

Are the binges planned? How do they begin? How do they end? How often do they occur?

Does she make herself vomit? If so, how? Does she vomit blood? Does she wash out with copious fluids afterwards?

Does she take laxatives, diuretics, emetics, or appetite suppressants? If so, with what effects?

Does she chew and spit? Does she fast for a day or longer?

Can she eat in front of others?

Does she exercise? Is this to ‘burn off calories’?

Reproduced with permission from Palmer B (ed.) (2000). Helping People with Eating Disorders. A clinical guide to assessment and treatment. John Wiley & Sons, Chichester.

Most patients with anorexia nervosa are reluctant to see a psychiatrist, so it is important to try to establish a good relationship. This means listening to the patient’s views, explaining the treatment alternatives, and being willing to consider compromises. A thorough history should be taken of the development of the disorder, the present pattern of eating and weight control, and the patient’s ideas about body weight (see Boxes 14.2 and 14.3). In the mental state examination, particular attention should be given to depressive symptoms, as well as to the characteristic psychopathology of anorexia nervosa itself. More than one interview may be needed to obtain this information and gain the patient’s confidence. The parents or other informants should be interviewed whenever possible. It is essential to perform a physical examination, with particular attention to the degree of emaciation, cardiovascular status, and signs of vitamin deficiency. Other wasting disorders, such as malabsorption, endocrine disorder, or cancer, should be excluded. Electrolytes should be measured if there is any possibility that the patient has been inducing vomiting or abusing laxatives.

Box 14.3 Assessment of psychological issues: some topics to be included

What does the patient feel about her body and her weight?

If she is restraining her eating, what is her motivation?

Does she feel fat? Does she dislike her body? If so, in what way?

Does she have a distorted body image? If so, in what way?

What does she feel would happen if she did not control her weight or her eating?

Does she fear loss of control? Is she able to say what she means by this?

Does she feel guilt or self-disgust? If so, what leads her to feel this?

Does anything about her disorder lead her to feel good?

If she binges, what are her feelings before, during, and after bingeing?

What has she told others about her eating disorder—if anything?

How does she think about her disorder? What does she make of it?

Reproduced with permission from Palmer B (ed.) (2000). Helping People with Eating Disorders. A clinical guide to assessment and treatment. John Wiley & Sons, Chichester.

Evidence about treatment effectiveness

There is a lack of good evidence about treatment and management. In part, this reflects the wide range in severity and the difficulty of evaluating complex interventions. This means that many current views and treatment guidelines (National Institute for Clinical Excellence, 2004a) depend upon clinical experience and opinion. Preventive treatments are also under investigation (Stice et al., 2007).


Family therapy has been advocated, reflecting the belief that family factors are important in the origins of anorexia nervosa. Various kinds of family therapy have been used, and it is unclear which of them is most effective. Systematic reviews suggest that there are some benefits in adolescents, but not in adults (Bulik et al., 2007b), with no clear advantages over individual treatments (Fisher et al., 2010).

Cognitive–behaviour therapy (CBT) has largely replaced psychodynamically oriented psychological treatments. Overall, there is modest evidence for its efficacy in reducing relapse rates in anorexia nervosa (Bulik et al., 2007b). The efficacy of conventional CBT is unproven with regard to initial weight gain (Bulik et al., 2007b), but a specifically tailored form of CBT has efficacy in weight restoration as well as in maintenance in eating disorders, including anorexia nervosa (Fairburn et al., 2009a).

Guidelines also suggest interpersonal therapy and cognitive analytic therapy for anorexia nervosa, but the evidence is limited (National Institute for Clinical Excellence, 2004a).


Many patients have significant symptoms of depression, and SSRIs are sometimes used for this indication. However, medication is not a routine part of the treatment of anorexia nervosa, because of the lack of efficacy (Walsh et al., 2006; Bulik et al., 2007b), although one trial reported that olanzapine may aid initial weight gain (Bissada et al., 2008). The increased risks of medication side-effects or toxicity because of the low weight and medical complications of the disorder must also be borne in mind.


Starting treatment

Success largely depends on establishing a good relationship with the patient. It should be made clear that achieving an adequate weight is essential in order to reverse the physical and psychological effects of starvation. It is important to agree a specific dietary plan, while emphasizing that weight control is only one aspect of the problem, and help should be offered with the accompanying psychological problems. Educating the patient and their family about the disorder and its treatment is important.

Place of treatment

Most cases may be treated on a day-patient or outpatient basis, ideally within a specialist eating disorder service. Admission to hospital is now unusual, but may be indicated if the patient’s weight is dangerously low, if there is a comorbid disorder (e.g. severe depression), if there is a high and acute suicide risk, or if outpatient care has failed.

Admission to a medical ward is appropriate if the main reason for admission is life-threatening consequences of weight loss, such as electrolyte disturbance, hypoglycaemia, or severe infection.

Compulsory treatment for anorexia nervosa is controversial both legally and ethically (see Chapter 4), and is rarely used.

Restoring weight

A reasonable aim is an increase of 0.5 kg a week, which will usually require an extra 500–1000 calories a day. The target weight usually has to be a compromise between a healthy weight (a BMI above 20 kg/m2) and the patient’s idea of what her weight should be. A balanced diet of about 3000 kcal should be taken as three or four meals a day. It is good practice to monitor the patient’s physical state regularly, and to prescribe vitamin supplements if indicated. It is also important to assess and modify other weight-reducing strategies that the patient may employ, such as over-exercising and laxative misuse.

For those who require inpatient treatment, there should be an understanding that the patient will stay in hospital until her agreed target weight has been reached and maintained and there is a comprehensive, mutually agreed treatment plan for subsequent outpatient care. Eating must be supervised by a nurse, who has three important roles—to reassure the patient that she will not lose control over her weight, to be clear about the agreed targets, and to ensure that the patient does not induce vomiting or take laxatives. Inpatient weight restoration usually takes between 8 and 12 weeks. Some patients demand to leave hospital before their treatment is finished, but with patience the staff can usually persuade them to stay. In the past, strict behavioural regimens were used, but these had no proven advantage and often appeared punitive.

For guidelines on the treatment of anorexia nervosa, see the National Institute for Clinical Excellence (2004a), and see also Box 14.4.

Bulimia nervosa

The term bulimia refers to episodes of uncontrolled excessive eating, sometimes called ‘binges.’ As mentioned above, the symptom of bulimia occurs in some cases of anorexia nervosa. The syndrome of bulimia nervosa was first described by Russell (1979) in a highly influential paper in which he named the condition and described the key clinical features in 30 patients who were seen between 1972 and 1978. However, Russell’s cases also suffered from concomitant anorexia nervosa and were therefore unlike the cases that we now think of as bulimia nervosa. Thereafter the syndrome ‘bulimia’ was included in DSM-III, and it soon became evident that bulimia nervosa was common in the general population among people who did not have anorexia nervosa.

Box 14.4 Guidelines for management of anorexia nervosa

• Most people with anorexia nervosa should be managed on an outpatient basis, with psychological treatment and monitoring of their physical condition.

• Psychological therapies for anorexia nervosa in clude cognitive analytic therapy (CAT), cognitive–behaviour therapy (CBT), interpersonal psychotherapy (IPT), focal psychodynamic therapy, and family interventions focused explicitly on eating disorders.

• Outpatient psychological treatment for anorexia nervosa should normally be of at least 6 months’ duration. Failure to improve or deterioration should lead to more intensive forms of treatment (e.g. a move from individual therapy to combined individual and family work, or day care or inpatient care). Dietary counselling should not be provided as the sole treatment for anorexia nervosa.

• For inpatients with anorexia nervosa it is important to monitor the patient’s physical status during re-feeding. Psychological treatment should be provided which has a focus both on eating behaviour and attitudes to body weight and shape, and on wider psychosocial issues with the expectation of weight gain. Rigid inpatient behaviour modification programmes should not be used in the management of anorexia nervosa.

• Following inpatient weight restoration, people with anorexia nervosa should be offered outpatient psychological treatment that focuses both on eating behaviour and attitudes to body weight and shape, and on wider psychosocial issues, with regular monitoring of both physical and psychological risk.

Adapted from the National Institute for Clinical Excellence (2004a).

For a review, see Fairburn et al. (2009b).

The central features of bulimia nervosa are as follows:

1. an irresistible and recurrent urge to overeat

2. the use of extreme measures to control body weight

3. overvalued ideas concerning shape and weight of the type seen in anorexia nervosa.

Two subtypes of bulimia nervosa are recognized in DSM-IV (see Box 14.5). The purging type is characterized by the use of self-induced vomiting, laxatives, and diuretics to prevent weight gain. In the non-purging type, ‘purging’ symptoms do not occur regularly but the person uses other behaviours to avoid weight gain, such as fasting and excessive exercise. Patients with bulimia nervosa are usually of normal weight. Patients who are substantially underweight usually qualify for a diagnosis of anorexia nervosa, which takes precedence. Most patients are female, and they often have normal menses.

Patients have a profound loss of control over eating. Episodes of bulimia may be precipitated by stress or by the breaking of self-imposed dietary rules, or may occasionally be planned. During the episodes large amounts of food are consumed, on average over 2000 kcal per episode (e.g. a loaf of bread, a whole pot of jam, a cake, and biscuits). This voracious eating usually takes place when the patient is alone. At first it brings relief from tension, but this is soon followed by guilt and disgust. The patient may then induce vomiting or take laxatives. There can be many episodes of bulimia and purging each day.

Box 14.5 DSM-IV criteria for bulimia nervosa

Recurrent episodes of binge eating. An episode of binge eating is characterized by both of the following:

• eating, in a discrete period of time (e.g. within any 2-hour period), an amount of food that is definitely larger than most people would eat during a similar period of time and similar circumstances

• a sense of lack of control over eating during the episode (e.g. a feeling that one cannot stop eating or control what or how much one is eating).

Recurrent inappropriate compensatory behaviour in order to prevent weight gain, such as self-induced vomiting, misuse of laxatives, diuretics, enemas, or other medication, fasting, or excessive exercise.

The binge eating and inappropriate compensatory behaviour both occur, on average, at least twice a week for 3 months.

The disturbance does not occur exclusively during episodes of anorexia nervosa.


Purging type. During the current episode of bulimia nervosa the person has regularly engaged in self-induced vomiting or the misuse of laxatives, diuretics, or enemas.

Non-purging type. During the current episode of bulimia nervosa the person has used other inappropriate compensatory behaviours, such as fasting or excessive exercise, but has not regularly engaged in self-induced vomiting or the misuse of laxatives, diuretics, or enemas.

Depressive symptoms are more prominent than in anorexia nervosa, and are probably secondary to the eating disorder. A high proportion of patients meet the criteria for major depression. The depressive symptoms usually remit as the eating disorder improves. A few patients appear to suffer from a depressive disorder of sufficient severity and persistence to require treatment with antidepressant medication.

Physical consequences

Repeated vomiting leads to several complications. Potassium depletion is particularly serious, resulting in weakness, cardiac arrhythmia, and renal damage. Rarely, urinary infections, tetany, and epileptic fits may occur. The teeth become pitted by the acidic gastric contents in a way that dentists can recognize as characteristic.


The prevalence of bulimia nervosa is around 1% among women aged between 16 and 40 years in Western societies (Hoek and van Hoeken, 2003). It is at least ten times less common in men. The dramatic increase in presentation and diagnosis seen in the UK in the early 1990s has been followed by stability or a modest decline. It is possible that the earlier increase represented increased rates of detection of relatively long-standing cases (Curren et al., 2005).

Onset and development of the disorder

Bulimia nervosa usually has an onset in late adolescence (i.e. several years later than anorexia nervosa). It often follows a period of concern about body shape and weight, and 25% of patients have a history of a previous episode of anorexia nervosa. There is usually an initial period of dietary restriction which, after a variable length of time, but usually within 3 years, breaks down, with increasingly frequent episodes of overeating. As the overeating becomes more frequent, the body weight returns to a more normal level. At some stage, self-induced vomiting and laxative misuse are adopted to compensate for the overeating. However, this may result in even less control of eating.


Like anorexia nervosa, bulimia nervosa appears to be the result of exposure to general risk factors for psychiatric disorder, including a family history of psychiatric disorder, especially depression and substance misuse, and a range of adverse childhood experiences. It used to be thought that sexual abuse was especially common, but the evidence now suggests that the rate is no higher than among those who develop other types of psychiatric disorder (Fairburn, 1999). Epidemiological studies also suggest that, unlike those with anorexia nervosa, patients with bulimia nervosa have increased exposure to factors that specifically promote dieting, such as childhood obesity, parental obesity, and early menarche. Perfectionism appears to be less of a risk factor than in anorexia nervosa (Fairburn, 1999).

The neurobiological and genetic contributions appear to be broadly similar to those described above for anorexia nervosa (Kaye, 2008; Fairburn et al., 2009b).

Course and prognosis

This is uncertain, as there have been few long-term studies. However, despite the original assertion by Russell (1979) that bulimia nervosa was an ‘ominous variant’, its outcome is clearly better in terms of both recovery and mortality (Keel et al., 2003). Nevertheless, even 5 to 10 years later between one-third and a half of individuals still have a clinical eating disorder, although in many of these cases it will take an atypical form (Fairburn et al., 2000).

No convincing predictors of course or outcome have been identified, although childhood obesity and low self-esteem may be associated with a worse prognosis (Fairburn et al., 2000).

Evidence about treatment effectiveness

There has been more research into the treatment of bulimia nervosa than into that of anorexia nervosa, and more evidence for effective psychological and pharmacological treatments (Shapiro et al., 2007). For some patients, guided self-help according to cognitive–behavioural principles may be sufficient, but for most patients formal treatment is indicated. For a review, see Fairburn et al. (2009b).


Both CBT and interpersonal therapy are effective in bulimia nervosa. Of those who complete treatment (about 20% drop out), 60% will have stopped binge eating, and there is an 80% reduction overall. Psychological aspects improve in parallel. Early response is a strong predictor of outcome. The effect is maintained, with low relapse rates seen over 12 months.

The most striking evidence in this field comes from a trial of a specifically tailored CBT-based approach (‘CBT-E’). In a relatively large study, CBT-E reduced or abolished core behavioural and psychological symptoms of eating disorders, and showed sustained efficacy, with over 50% of patients remaining essentially recovered at 5-year follow-up (Fairburn et al., 2009a). The treatment was delivered in two forms—a ‘simple’ intervention which focused solely on the eating disorder, and a ‘complex’ intervention in which personality, mood, and interpersonal issues were also addressed. The authors suggested that the simple treatment is the default version and widely applicable to bulimia nervosa (and other eating disorders), whereas the latter could be limited to individuals with additional psychopathology.


Antidepressants are effective, producing a reduction of about 50% in the frequency of binge eating, and cessation in 20% of cases. The onset is more rapid than in depression, but a higher dose may be needed (e.g. fluoxetine 60 mg daily). However, long-term data are less encouraging and show poor compliance. Anti-depressants should be viewed as second-line treatment if effective psychological treatment is available. The combination of the two treatments does not appear to offer any advantages.


The management of bulimia nervosa is easier than that of anorexia nervosa because the patient is more likely to wish to recover, and a good working relationship can often be established. Furthermore, there is no need for weight restoration. It is necessary to assess the patient’s physical state and to measure electrolyte status in those who are vomiting frequently or misusing laxatives.

As with many common disorders, a ‘stepped-care’ approach appears to be the best way of providing appropriate care for large numbers of people with varying degrees of severity of disorder (National Institute for Clinical Excellence, 2004a).

Step 1. Identify the small minority (less than 5%) of individuals who need specialist treatment because of severe depression, physical complications, or substance abuse that requires treatment in its own right.

Box 14.6 Questions patients and their families ask about eating disorders and their treatment


1. What kind of eating problem do I have?

2. What does it mean for my health, daily life, work, or schooling?

3. Why have you decided to offer me this kind of treatment?

4. Are there other treatments that might suit me better?


1. How can we help with the problem?

2. Should we be more directly involved in supervising how the person eats?

3. Can you let us know how treatment is progressing?

4. Might we benefit from support as a family?

See National Institute for Clinical Excellence (2004a).

Step 2. Offer guided cognitive–behavioural self-help as appropriate, using a self-help book and with the guidance of a non-specialist facilitator. Treatment usually takes about 4 months and requires eight to ten meetings with the facilitator. Such treatment is appropriate for primary care, and appears to lead to good progress in about one-third of patients.

Step 3. Patients who do not show benefit within around 6 weeks of commencing Step 2 require full CBT. In a minority of cases, where concomitant depressive symptoms are severe, it is worthwhile adding an antidepressant drug such as fluoxetine in doses of up to 60 mg daily. However, most depressive symptoms will resolve with successful psychotherapy.

Step 4. Patients who do not improve with CBT require comprehensive specialist reassessment. In some cases, measures to provide more intensive cognitive therapy or an antidepressant drug may be useful. It is important to review the initial treatment with the patient with the aim of agreeing a treatment approach that the patient finds acceptable (see Box 14.6).

Eating disorder not otherwise specified (EDNOS; atypical eating disorders)

These DSM-IV and ICD-10 categories are for disorders of eating that do not meet the criteria for anorexia nervosa or bulimia nervosa, but which are of clinical severity. As noted above, they are frequent, and are diagnosed in at least one-third of referrals when current diagnostic criteria are applied strictly.

Binge eating disorder

A subgroup within EDNOS, binge-eating disorder, is recognized in DSM-IV as ‘requiring further study.’ It is characterized by recurrent bulimic episodes in the absence of the other diagnostic features of bulimia nervosa, particularly counter-regulatory measures such as vomiting and purging. Patients may have depressive symptoms and some dissatisfaction with their body weight and shape; however, this is usually less severe than in bulimia nervosa. Binge eating disorder is of comparable chronicity and stability to other eating disorders (Pope et al., 2006).

The risk factors for binge eating disorder are similar to those for psychiatric disorder in general, and for obesity. About 25% of patients who present for treatment for obesity have features of binge eating disorder. The condition generally affects an older age group than bulimia nervosa, and up to 25% of those who present for treatment are men.

Unlike the other eating disorders described here, binge eating disorder has a high spontaneous remission rate and larger effect sizes for response to CBT, interpersonal therapy, and antidepressants (Vocks et al., 2010).

Psychogenic vomiting

Psychogenic vomiting is chronic and episodic vomiting without an organic cause, which commonly occurs after meals and in the absence of nausea. It should be distinguished from the more common syndrome of bulimia nervosa, in which self-induced vomiting follows episodes of binge eating (uncontrolled overeating). Psychogenic vomiting appears to be more common in women than in men, and usually presents in early or middle adult life. It is reported that both psychotherapeutic and behavioural treatments can be helpful.

For a review of idiopathic vomiting disorders, see Olden and Chepyala (2008).

Sleep disorders

Psychiatrists may be asked to see patients whose main problem is either difficulty in sleeping or, less often, excessive sleep. Many other patients also complain about sleep problems as one of their symptoms. However, sleep disturbances are often overlooked or misdiagnosed (Stores, 2007).

Sleep problems are important for several reasons.

• They may represent primary sleep disorders (which are the focus of this section).

• They may be causes of psychological symptoms.

• They may be symptoms of psychiatric disorder, especially mood disorders (see Chapter 10).

• Sleep problems can also occur in a range of medical disorders.

Many patients who sleep badly complain of tiredness and mood disturbance during the day. Although prolonged sleep deprivation leads to some impairment of intellectual performance and disturbance of mood, loss of sleep on occasional nights is of little significance except in those whose responsibilities or activities require maximum alertness. The daytime symptoms of people who sleep badly are probably related more to the cause of their insomnia than to insomnia itself.

For a review of the biology of sleep and its disorders, see Siegel (2009) and Stores (2009a).


Table 14.3 shows the DSM-IV classification of sleep disorders, which is compatible with the more elaborate International Classification of Sleep Disorders. Classification in ICD-10 is rather different, with sleep disorders occurring in three different parts of the classification.


Sleep disorders are frequent, but there is a wide range of variation in estimates, depending on the definition and the population studied; for example, 22% of adults meet the DSM-IV criteria for insomnia, but only 4% meet the ICD-10 criteria (Roth et al., 2011). Excessive sleepiness occurs in 5% of adults, and possibly 15% of adolescents and 14% of the adult population have some form of chronic sleep–wake disorder.

In the Epidemiologic Catchment Area Study (Ford and Kamerow, 1989), 10.2% of the community sample reported insomnia and 3.2% reported hypersomnia. In total, 40% of those with insomnia and 46.5% of those with hypersomnia had a psychiatric disorder, compared with 16.4% of those with no sleep complaints. Groups at particular risk of persistent sleep problems include young children, adolescents, the physically ill, those with learning disability, and those with dementia. Women have a higher rate of insomnia than men, across all conditions (Zhang and Wing, 2006). Ageing is associated with fragmentation and other changes in sleep patterns, and in the causes of sleep disorder (Wolkove et al., 2007).


Assessment requires a full psychiatric and medical history, together with detailed enquiries about the sleep complaint (see Box 14.7). In some cases specialist investigation, including polysomnography, is necessary.

Table 14.3 Classification of primary sleep disorders in DSM-IV


Primary insomnia

Primary hypersomnia


Breathing-related sleep disorder

Circadian-rhythm sleep disorder

Dyssomnia not otherwise specified


Nightmare disorder (dream anxiety disorder)

Sleep terror disorder

Sleepwalking disorder

Parasomnia not otherwise specified

Sleep disorder related to another mental disorder



Other sleep disorders

Secondary sleep disorder due to a general medical condition

Substance-induced sleep disorder


Transient insomnia occurs at times of stress or as ‘jet lag.’ Short-term insomnia is often associated with personal problems—for example, illness, bereavement, relationship difficulties, or stress at work. Insomnia in clinical practice is usually secondary to other disorders, notably painful physical conditions, depressive disorders, and anxiety disorders; it also occurs with excessive use of alcohol or caffeine, and in dementia. Sleep may be disturbed for several weeks after stopping heavy drinking. Sleep problems are also common in association with any medical illness that results in significant pain or discomfort or is associated with metabolic disturbances. They may also be provoked by prescribed drugs. In about 15% of cases of insomnia, no cause can be found (primary insomnia). People vary in the amount of sleep that they require, and some of those who complain of insomnia may be having enough sleep without realizing it. For a review of insomnia, see Sateia and Nowell (2004).

Box 14.7 Assessment of sleep disturbance

Screening questions

Do you sleep well enough and long enough?

Are you very sleepy during the day?

Is your sleep disturbed at night?

Sleep history

Detailed history of the sleep complaint

Pattern of occurrence, factors that improve or worsen sleep

Consequences for mood, everyday life, and family

Past and present treatment

Typical 24-hour sleep–wake schedule

Sleep diary

Systematic 2-week or longer record

History from bed partner


Video recording

Wrist scintigraphy (monitoring body movement)

Polysomnography (EEG, EMG)

HLA typing

CSF hypocretin (orexin) levels

See Stores (2009a).

Assessment of insomnia

Usually the diagnosis of insomnia is made on the basis of the account given by the patient, focusing both on the nature of the sleep problem and its daytime consequences, as well as screening for psychiatric and medical disorders and use of alcohol and drugs. Sleep studies are rarely used.

Treatment of insomnia

Both pharmacological and non-pharmacological approaches can be used to treat insomnia. Drugs were widely used in the past, but non-pharmacological approaches are now recommended as first-line treatment.

Box 14.8 Principles of sleep education (sleep hygiene)

Sleep environment

Familiar and comfortable




Bedtime routines

Consistent time for going to bed and waking up

Going to bed only when tired

Thinking about problems before going to bed

Regular exercise


Overexcitement before going to bed

Late-evening exercise

Caffeine-containing drinks late in the day

Excessive alcohol and smoking

Excessive daytime sleep

Large late meals

Too much time in bed lying awake

Sleep hygiene refers to a series of steps which are useful for all patients with insomnia and can be applied in primary care (see Box 14.8). For more serious or persistent insomnia, a range of specific non-pharmacological treatments have been shown to be effective (see Table 14.4). Cognitive therapy (also called cognitive restructuring) identifies and changes distorted cognitions about insomnia. Stimulus control therapy focuses on the principle that one only goes to bed when sleepy, and if not asleep within 20 minutes, one should get up and engage in a relaxing activity before returning to bed. Although it is sometimes justifiable and useful to give a hypnotic for a few nights, demands for prolonged medication should be resisted. This is because withdrawal of hypnotics may lead to insomnia that is as distressing as the original sleep disturbance. Continuation of hypnotics may be associated with impaired performance during the day, tolerance of the sedative effects, and dependency. The use of hypnotic drugs is described further on p. 515. Sedative antidepressants, such as trazodone, are another alternative for short-term treatment of insomnia (e.g. in the early stages of SSRI treatment).

Table 14.4 Treatments for insomnia

Non-pharmacological treatments

Sleep hygiene

Cognitive therapy

Stimulus control therapy

Sleep restriction

Progressive muscle relaxation

Pharmacological treatments

Short-acting benzodiazepines (e.g. temazepam)

‘Z drugs’ (zolpidem, zalepon, zopiclone)

Low-dose sedative antidepressants (e.g. trazodone)


For a review, see Espie and Bartlett (2009). For guidelines on the use of medication to treat insomnia, see Wilson et al. (2010).

Hypersomnia and excessive daytime sleepiness

Excessive daytime sleepiness is common (with a reported prevalence of 3–5%) and underdiagnosed. Many cases are secondary to loss of night-time sleep. Table 14.5 lists the principal causes. For a review, see Zeman et al. (2004). For a discussion of the treatment of excessive daytime sleepiness, see Billiard (2009).


Narcolepsy usually begins between the ages of 10 and 20 years, although it may start earlier. Onset is rare after middle age. The prevalence is about 5 per 10 000 members of the population. The key features are cataplexy (sudden brief episodes of paralysis, with loss of muscle tone), which occurs in most cases, and sleep paralysis and hypnagogic hallucinations, which occur in only about 25% of patients. Narcolepsy usually presents with excessive sleepiness or the consequences of cataplexy (e.g. collapse), and has major effects on lifestyle. For a review, see Dauvilliers et al. (2007).


Many aetiological theories have been advanced. Evidence favours a complex genetic predisposition, although occasional families with apparent autosomal-dominant transmission are known. The strongest genetic association is with HLA DR2 tissue type, and specifically the HLA DQB1*0602, which is present in over 85% of cases, compared with 18–35% in the general population. Other HLA haplotypes are strongly protective against narcolepsy.

The hypocretins (also called orexins) are hypothalamic neuropeptide transmitters which regulate the sleep–wake cycle. Recent evidence strongly implicates them in narcolepsy, especially cases with cataplexy. In these individuals, concentrations of hypocretin-1 and hypocretin-2 are decreased in the brain and cerebrospinal fluid (CSF), and there are fewer hypocretin-positive cells in the hypothalamus. The relationship between hypocretins and the HLA findings in narcolepsy is unknown, but an autoimmune mechanism is possible.


Narcolepsy usually presents to neurologists. The differential diagnosis is from other causes of excessive daytime sleepiness, and occasionally epilepsy, schizophrenia, or chronic fatigue syndrome. Psychiatric referral may occur if the latter are suspected. A full history, especially a sleep history, is the main assessment tool. The Epworth Sleepiness Scale is often used. Sleep laboratory studies, if available, are valuable. Measurement of HLA status or hypocretin levels in CSF is sometimes undertaken, but is not routine.

Table 14.5 Causes of excessive daytime sleepiness

Insufficient night-time sleep

Unsatisfactory sleep routines or circumstances

Circadian-rhythm sleep disorders

Frequent parasomnias

Chronic physical illness

Psychiatric disorders

Pathological sleep


Obstructive sleep apnoea

Other central nervous system disease

Drug effects

Kleine–Levin syndrome

Depressive illness

Psychiatric aspects of narcolepsy

Strong emotions sometimes precipitate cataplexy. Patients with narcolepsy often have secondary emotional and social difficulties, and their difficulties are increased by other people’s lack of understanding.

Treatment of narcolepsy

Patients need considerable help in adjusting to a disabling chronic illness. They should be encouraged to follow a regular routine, with planned naps during the day. If stressful events or other factors (e.g. fatigue) appear to provoke cataplexy, efforts should be made to avoid them or to arrange the patient’s lifestyle so as to minimize their impact.

Most patients require treatment with stimulant drugs, such as dexamphetamine, which decrease sleepiness and moderately reduce the frequency of sleep attacks. Modafinil, a non-amphetamine stimulant, also reduces daytime sleepiness and has fewer side-effects; it has not been compared directly with amphetamines. Clomipramine and other antidepressants can also be used to decrease cataplexy.

For guidelines on the treatment of narcolepsy, see Billiard et al. (2006).

Breathing-related sleep disorder

This syndrome consists of daytime drowsiness together with periodic respiration, recurrent apnoeas, and excessive snoring at night. It is usually associated with upper airways obstruction, hence the term obstructive sleep apnoea syndrome. The prevalence is about 4% in the male population. The typical patient is a middle-aged overweight man who snores loudly. Treatment consists of relieving the cause of the respiratory obstruction and encouraging weight loss. Continuous positive pressure ventilation using a face mask is often effective. Compliance with advice is often poor. Obstructive sleep apnoea is a risk factor for stroke.

Kleine–Levin syndrome

This very rare secondary sleep disorder consists of episodes of somnolence, increased appetite, and hypersexuality, often lasting for days or weeks, with long intervals of normality between them. It usually affects adolescent boys. The symptoms suggest a hypothalamic disorder, but its aetiology is unknown. There is no established treatment, but stimulants, and prophylaxis with mood stabilizers, are used (Billiard, 2009).

Circadian-rhythm sleep disorder (sleep–wake schedule disorders)

There are several forms of circadian-rhythm sleep disorder, of which jet lag is the most familiar. Shift-work sleep disorder is a common and increasing problem, the consequences of which are widely underestimated. Fatigue and transient difficulties in sleeping accompany regular changes of shift, or the irregular alternation of night work and days off may lead to chronic problems of poor sleep, fatigue, impaired concentration, and an increased risk of accidents as well as adverse effects on family life.

Circadian rhythms, their disorders, and their involvement in various psychiatric disorders have become of greater neurobiological and clinical interest recently (Woolf et al., 2010). This has been in part because of the availability of modafinil and other agents such as melatonin which have some efficacy. For a review of the assessment and treatment of circadian-rhythm sleep disorders, see Morgenthaler et al. (2007).


Parasomnias are characterized by abnormal behaviour or physiological events occurring in association with sleep, specific sleep stages, or sleep–wake transitions. For a review, see Schenk and Mahowald (2009).

Nightmares (dream anxiety disorder)

A nightmare is an awakening from REM sleep to full consciousness with detailed dream recall. Children experience nightmares with a peak frequency around the age of 5 or 6 years. Nightmares may be stimulated by frightening experiences during the day, and frequent nightmares usually occur during a period of anxiety. Other causes include post-traumatic stress disorder, fever, psychotropic drugs, and alcohol detoxification.

Night terror disorder

Night terrors are much less common than nightmares. They are sometimes familial. The condition begins and usually ends in childhood, but occasionally persists into adult life. A few hours after going to sleep, the child, while in stage 3–4 non-REM sleep, sits up and appears terrified. They may scream, and they usually appear confused. There are marked increases in heart rate and respiratory rate.

After a few minutes the child slowly settles and returns to normal calm sleep. There is little or no dream recall. A regular bedtime routine and improved sleep hygiene have been shown to be helpful. Benzodiazepines and imipramine have been shown to be effective in preventing night terrors, but their prolonged use should be avoided.

Sleepwalking disorder

Sleepwalking is an automatism that occurs during deep non-REM sleep, usually in the early part of the night. It is most common between the ages of 5 and 12 years, and 15% of children in this age group walk in their sleep at least once. Occasionally, the disorder persists into adult life. Sleepwalking may be familial. Most children do not actually walk, but sit up and make repetitive movements. Some walk around, usually with their eyes open. They do not respond to questions, and are very difficult to wake. They can usually be led back to bed. Most episodes last for a few seconds or minutes.

As sleepwalkers can occasionally harm themselves, they need to be protected from injury. Doors and windows should be locked and dangerous objects removed. Adults with severe problems should be given advice about safety, avoidance of sleep deprivation, and any other circumstances that might make them excessively sleepy (e.g. drinking alcohol before going to bed).

Other parasomnias

Rapid eye movement (REM) sleep behaviour disorder is a parasomnia which should be considered when behavioural problems, particularly agitation or aggression, occur during the night. It is thought to occur when the normal atonia of REM sleep is lost, so that dreams are acted out. It is more common in the elderly, particularly men, and is associated with neurological disorders, particularly Parkinsonism. Clonazepam and donepezil may be effective.

Sleep paralysis is an inability to perform voluntary movements during the transitions between sleep and wakefulness, either at sleep onset (hypnagogic) or during awakening (hypnopompic). The episodes are often accompanied by extreme fear.

Restless legs syndrome is a distressing and painful condition which can result in severe insomnia and periodic limb movements during sleep. It is common, affecting up to 10% of the population, especially women. It can be mistaken for a psychiatric symptom or disorder (e.g. anxiety, akathisia).

Disorders of sexual function, preference, and gender identity

In the past, as reflected in earlier editions of this book, sexual matters had greater prominence than is common in present-day psychiatric practice, for several reasons.

• Sexual factors were considered to be important in psychodynamic theories of causation in psychiatry and, perhaps related to this historical view, more attention was paid to the patient’s sexual history, attitudes, and behaviour than is now usual. (However, set against this trend, there is increasing evidence about the frequency and psychiatric implications of childhood sexual abuse.)

• Homosexuality was a category of psychiatric disorder until the 1970s (Mendelson, 2003).

• Sex therapy by psychiatrists was more widely practised (Althof, 2010).

Here we discuss the main features of the three categories of disorders affecting sex and gender that are recognized in ICD-10 and DSM-IV (see Table 14.6).

• Sexual dysfunction denotes impaired or unsatisfying sexual enjoyment or performance. Such conditions are common. They are subdivided, according to the stage of the sexual response that is mainly affected, into disorders of sexual desire, disorders of sexual arousal, and disorders of orgasm. There are also categories for the painful conditions vaginismus and dyspareunia.

• Abnormalities of sexual preference (paraphilias) are uncommon, but they take many forms and have forensic implications.

• Gender identity refers to one’s sense of being male or female. When this sense of identity is at variance with an individual’s anatomical sex, that person is said to have a gender identity disorder.

The epidemiology, physiology, and psychology of sexual behaviour are not covered here. For reviews of these aspects, see Levin (2009). Similarly, the reader is directed elsewhere for information about sexual orientation (Byrne and Parsons, 1993; Mutanski et al., 2002; Swaab and Garcia-Falgueras, 2009) and its mental health aspects (Herek and Garnets, 2007).

Sexual dysfunction

In men, sexual dysfunction refers to repeated impairment of normal sexual performance, and less often to impairment of sexual interest and pleasure. In women it more often refers to a repeated unsatisfactory quality to the experience, or to impaired desire for it. Note also that there is no agreed definition of ‘dysfunction’ in this context. What is regarded as ‘normal’ (e.g. in terms of the frequency or duration of sexual activity) depends in part on the expectations of the individuals concerned.

Table 14.6 Classification of sexual and gender identity disorders


As explained above, problems of sexual dysfunction are classified into those that affect

• sexual desire and sexual enjoyment

• the genital response (erectile impotence in men, and lack of arousal in women)

• orgasm (premature or retarded ejaculation in men, and orgasmic dysfunction in women).

To this list can be added problems that result in pain during sex (vaginismus and dyspareunia in women, and painful ejaculation in men).

It is important to remember that sexual function may not be disclosed directly, but only revealed during enquiries about another complaint, such as depression or poor sleep, or gynaecological symptoms.

Prevalence of sexual dysfunction

In the UK, a stratified probability sample of 11 161 people aged 16–44 years were asked about sexual problems that had lasted for more than 1 month and for more than 6 months (Mercer at al, 2003). The response rate was 65%. It is not known whether the self-reported problems would have met the diagnostic criteria for the corresponding sexual dysfunction. The reported rates are shown in Table 14.7. More women than men reported problems, and among both the majority of problems lasted for less than 6 months.

Table 14.7 Reported frequency of problems related to sexual function among people aged 16–44 years with a sexual partner (respondents could report more than one disorder)*


Two UK studies have surveyed general practice patients. Dunn et al. (1998) conducted a postal survey, which one-third of the men who responded and two-fifths of the women reported a current sexual problem. Erectile dysfunction and premature ejaculation were the most common problems among the men. Vaginal dryness and infrequent orgasm were the most common problems among the women. Half of those with a problem reported that they would have liked help, but only one in ten had received it. Nazareth et al. (2003) surveyed patients from 13 general practices in London, with a 70% response rate to the questionnaire. Around 22% of the men and 40% of the women met the criteria for an ICD diagnosis of a sexual problem. The most frequent problems were inhibited female orgasm (19%), lack or loss of sexual desire (17% of women and 7% of men), male erectile dysfunction (9%), and female sexual arousal dysfunction (3.6%).

Most studies show an increasing prevalence of sexual dysfunction with age, although concern about such dysfunction declines (Lindau et al., 2007). For a review of the epidemiology of sexual dysfunction, see DeRogatis and Burnett (2008).

Causes of sexual dysfunction

Sexual dysfunction arises from varying combinations of a number of factors, including a poor general relationship with the partner, low sexual drive, ignorance about sexual technique, and anxiety about sexual performance. Other important factors include physical illness, depressive and anxiety disorders, medication, and alcohol or drug misuse. Some of these factors will now be considered. When assessing aetiology in an individual, it is important to recognize that both psychological and physical factors are often present. For a review of sexual dysfunction in women, see Raina et al. (2007), and for a review of such dysfunction in men, see Beutel et al. (2006).

Low sexual drive

Sexual drive varies from one person to another, but the reason for this is not known. It is not directly related to androgen levels.


Anxiety is an important cause of sexual dysfunction. Sometimes anxiety is an understandable consequence of earlier experiences, such as a man’s failure during his first attempt at intercourse, or a woman’s experience of sexual abuse. There is no evidence to support psychoanalytic theories that anxiety about sexual relationships originates from failure to resolve the Oedipal complex in boys, or the corresponding attachment to the father in girls.

Physical illness and surgical treatments

Sexual dysfunction sometimes dates from a period of abstinence associated with pregnancy or childbirth, or from the debilitating effects of physical illness (see Table 14.8). Of the diseases that have a direct effect on sexual performance, diabetes mellitus is particularly important, due both to autonomic neuropathy and vascular disease. For a review of the pathophysiology of endocrine-related sexual dysfunction, see Bhasin et al. (2007). Other important organic causes of sexual dysfunction include neurological disorders (Rees et al., 2007), as well as a range of general medical conditions (Basson and Schulz, 2007) and, in women, pelvic surgery (Raina et al., 2007). In addition to a direct physical contribution, chronic disease and its treatment are associated with sexual dysfunction because of indirect and psychological effects (e.g. due to fatigue, anxiety about harm being caused by sexual exertion, or lowered self-esteem or self-confidence).

Sexual dysfunction in psychiatric disorders

Many psychiatric disorders may impair sexual function, as discussed in other chapters. There is a strong relationship with depression, in which rates of sexual dysfunction are estimated to be up to 75% (Williams and Reynolds, 2006).

Table 14.8 Medical and surgical conditions commonly associated with sexual dysfunction

Medical conditions


• Diabetes, hyperthyroidism, myxoedema

• Addison’s disease, hyperprolactinaemia


• Vaginitis, endometriosis, pelvic infections


• Angina pectoris, myocardial infarction, peripheral vascular disease


• Asthma, obstructive airways disease


• Arthritis

• Renal failure

• Pelvic autonomic neuropathy, spinal cord lesions, stroke

Surgical conditions


Colostomy, ileostomy


Episiotomy, operations for prolapse

Sexual dysfunction and physical handicap

Physically handicapped people may experience sexual problems for a number of different reasons:

• direct effects of the physical handicap on sexual function

• non-specific effects, such as tiredness or pain

• fears about the effects of intercourse on their condition

• lack of information about the sexual behaviour of other people with the same disability.

Social attitudes towards sexual activity in the physically handicapped (and, indeed, in people with learning disabilities) have changed markedly in recent years. There is less prudery and a wider acceptance of the sexual needs of these groups. Much can be done to help disabled people by overcoming their fears and discussing the forms of sexual activity that are possible despite their disability, and, if appropriate, using the relevant parts of the methods described below for treating sexual dysfunction. Several informal support groups exist.

Effects of medication

Several commonly used classes of drugs—psychiatric, medical, and recreational—have side-effects that involve sexual function (see Table 14.9). Antidepressants are of particular importance (see Chapter 19), with sexual side-effects reported in 25–80% of patients (Serretti and Chiesa, 2009), which add to the dysfunction associated with depression itself (Kennedy and Rizvi, 2009). SSRIs are commonly associated with decreased desire and anorgasmia, especially in women. Antipsychotics (and tricyclic antidepressants) can cause erectile dysfunction and ejaculatory failure in men, and loss of libido and anorgasmia, especially with agents that produce hyperprolactinaemia, in women (Dossenbach et al., 2006). Anxiolytics, sedatives, and hormones have more effect on sexual activity in men than in women. Intoxication from or the excessive use of alcohol and street drugs can also impair sexual performance.

Table 14.9 Some drugs that may impair sexual function

Therapeutic agents

Antihypertensive agents

• Diuretics, spironolactone, α-blockers, β-blockers


• Tricyclic antidepressants, monoamine oxidase inhibitors, SSRIs

Mood stabilizers

• Lithium

Anxiolytics and hypnotics

• Benzodiazepines


Hormonal agents

• Anabolic steroids, corticosteroids, oestrogens

Misused substances

Alcohol, cocaine, marijuana

Assessment of patients with sexual dysfunction

History taking

The interviewer needs to be particularly sensitive when enquiring about sexual function and dysfunction, and when detecting and dealing with embarrassment experienced by the patient. Whenever possible, the patient and their sexual partner should be interviewed, both separately and together. The major areas of enquiry are listed in Table 14.10. In addition to full characterization of the dysfunction itself, the assessment should search for evidence of any of the medical and psychiatric disorders noted earlier which can lead to sexual dysfunction.

Physical examination and special investigations

If the general practitioner or another specialist has not already done so, a physical examination should be performed, especially in men (see Table 14.11). Laboratory tests or other investigations should be arranged in appropriate cases—for example, fasting blood sugar, testosterone, and other hormones in men with erectile dysfunction.

Table 14.10 Assessment of sexual dysfunction

Define the problem (ask both partners)

Origin and course

Prior (baseline) sexual function

With other partners?

Sexual drive

Knowledge and fears

Social relationships generally

Relationship between the partners

Psychiatric disorder

Substance misuse

Medical illness; medical or surgical treatment

Why seek help now?

Physical examination

Laboratory tests

Table 14.11 Physical examination of male patients presenting with sexual dysfunction

General examination (directed especially to evidence of diabetes mellitus, thyroid disorder, and adrenal disorder)

Hair distribution


Blood pressure

Peripheral pulses

Ocular fundi


Peripheral sensation

Genital examination

Penis: congenital abnormalities, foreskin, pulses, tenderness, plaques, infection, urethral discharge

Testicles: size, symmetry, texture, sensation


The treatment of sexual dysfunction

The main treatments for sexual dysfunction are shown in Table 14.12. The first step, after a detailed assessment, is to provide advice and reassurance, as the problem is often long-standing by the time the patient presents, and it may have led to adverse secondary effects—for example, by increasing anxiety or impairing the quality of the relationship generally. Where possible, the underlying causes should be treated.

Specific interventions include a range of psychological and behavioural treatments, including sex therapy. There are also physical treatments, of which drug therapy with phosphodiesterase type 5 (PDE-5) inhibitors is now the most widely used. The roles of the various treatment modalities, and the evidence for their efficacy, vary according to the specific type of sexual dysfunction, as discussed below.

Table 14.12 Treatments for sexual dysfunction

Advice, information, and reassurance

Treatment of underlying cause

Psychological methods

Behavioural techniques

Sex therapy

Drug treatments

PDE-5 inhibitors

Other physical treatments

Vacuum devices


Sex therapy

As noted above, many patients benefit from simple advice and reassurance. Other cases resolve with treatment of the cause of the sexual dysfunction, or the use of medication. For some people, a formal type of behavioural therapy, known as sex therapy, may be appropriate. In the UK, its availability within the National Health Service is very limited, although it is also provided by the charity Relate.

Sex therapy owes much to the work of Masters and Johnson (1970), and has four characteristic features.

1. The partners are treated together.

2. They are helped to communicate better about their sexual relationship.

3. They receive education about the anatomy and physiology of sexual intercourse.

4. They complete a series of graded tasks, which focus as much on what must not yet be attempted as what is to be done. This prohibition often significantly reduces ‘performance anxiety’, resulting in increased confidence and subsequent success.

There have been few adequately controlled studies of sex therapy, and most of them were conducted before 1990. Heiman and Meston (1997) concluded that sex therapy was effective for the treatment of orgasmic dys-function in women, and erectile dysfunction in men.

Types of sexual dysfunction

In this section we discuss the features and treatment of the more commonly encountered categories of sexual dysfunction.

Lack or loss of sexual desire (hypoactive sexual desire disorder)

Complaints of diminished sexual desire are much more common among women than among men. There is controversy about the extent to which lack of sexual desire, especially in women, should be considered a psychiatric disorder (Angel, 2010). Lack of desire can be categorized as either a lack or a loss, and as either global or situational.

global lack of desire, in either sex, suggests either a constitutionally low sex drive, or occasionally a sexual orientation that differs from the person’s current sexual relationship. A global loss of desire requires consideration of a medical or psychiatric cause, or a side-effect of medication, as outlined in Table 14.9. Among women, sexual desire may diminish premenstrually, or in relation to pregnancy and the postnatal period, or after hysterectomy or breast removal. Women who experience loss of desire in these circumstances should be reassured that this problem is common and likely to be short-lived. Situational loss of desire often reflects general problems in the relationship between the sexual partners.

Hypoactive sexual desire disorder has a poor prognosis relative to other types of sexual dysfunction. There is little evidence for any effective treatment (Clayton and Hamilton, 2010). PDE-5 inhibitors (see below) do not show clear evidence of benefit, except perhaps in women with antidepressant-induced dysfunction (Nurnberg et al., 2008). Lack of desire following ovariectomy may respond to testosterone, which can be administered either by intramuscular injection or transdermally. Testosterone may also be used in men with hypogonadism.

Sexual aversion disorder

Sexual enjoyment is sometimes replaced by a positive aversion to genital contact. When this aversion is persistent or severe and is accompanied by avoidance of almost all genital sexual contact with a sexual partner, the condition is classified as sexual aversion disorder. The causes of the condition are not well understood; they seem to be similar to the psychological causes of hypoactive sexual desire disorder. Assessment and treatment are similar to that for hypoactive sexual desire disorder.

Female sexual arousal disorder

Lack of sexual arousal in the female is manifested as reduced vaginal lubrication. This may be due to inadequate foreplay, lack of sexual interest, anxiety, or low oestrogen levels (typically related to the menopause). As with low sexual desire, there is little evidence for any effective treatment (Clayton and Hamilton, 2010), although yohimbine (an alpha-2 antagonist) and PDE-5 inhibitors are sometimes tried.

Male erectile disorder

This condition is the inability to achieve an erection or to sustain it for long enough for satisfactory coitus. It may be present from the first attempt at intercourse (primary male erectile disorder) or develop after a period of normal function (secondary male erectile disorder). It is more common among older than younger men (in contrast with premature ejaculation; see below).

Primary cases may occur due to a combination of low sexual drive and anxiety about sexual performance. Secondary cases may arise from diminishing sexual drive in the middle-aged or elderly, loss of interest in the sexual partner, anxiety, depressive disorder, or organic disease and its treatment. It is thought that abnormalities of the vascular supply to the penile erectile tissue are important factors in erectile failure associated with physical disease, including diabetes and peripheral vascular disease.

The following aspects of the assessment are particularly important.

• Has there been a previous period of normal function?

• Does erection occur during foreplay?

• Does erection occur on waking or in response to masturbation? Erection in these circumstances suggests psychological causes for the failure of erection at other times.

• Is there evidence of alcohol or drug abuse? (Ask the partner as well as the patient.)

• Are there possible effects of any medication?

If the aetiology is uncertain after history taking, physical examination, and blood tests, several special investigations may be considered, including measurements of tumescence and blood flow using ultrasound.

Any reversible causes should be treated. The first-line specific therapy is now the use of PDE-5 inhibitors—sidenafil, vardenafil, and tadalafil—which are effective in about 70% of cases (Tsertsvadze et al., 2009). These inhibit the breakdown of cyclic GMP by a phosphodiesterase located in the vascular smooth muscle of the penis. They do not initiate arousal but enhance the effects of sexual stimulation. Care should be taken to observe the manufacturer’s advice about contraindications and interactions. The most common side-effects are headache, flushing, dyspepsia, and nasal congestion. PDE-5 inhibitors have largely supplanted earlier physical treatments for erectile dysfunction (i.e. intracavernosal injections, vacuum devices, and surgical methods).

Female orgasmic disorder

Many women do not regularly reach orgasm during intercourse, but do so in response to clitoral stimulation. Whether failure to reach orgasm regularly should be regarded as a disorder has been questioned (e.g. Heiman, 2002). Nevertheless, some women regard it as a problem and ask for advice and help. It may arise due to normal variations in sexual drive, poor sexual technique by the partner, lack of affection for the partner, tiredness, depressive disorder, physical illness (including gynaecological disorders), long-term effects of sexual abuse, and the effects of medication.

Educational and behavioural methods can help the woman and her partner to increase arousal and thus orgasm—for example, via graduated self-stimulation or the use of a vibrator. Sex therapy may also be effective. Whichever approach is chosen, help should be given for any relationship or other problems.

Male orgasmic disorder

This term refers to serious delay in, or absence of, ejaculation. It is usually associated with general inhibition about sexual relationships, but it may be caused by drugs, including SSRIs, antipsychotics, and monoamine oxidase inhibitors.

Premature ejaculation

This term refers to habitual ejaculation either before penetration or so soon afterwards that the woman has not experienced pleasure. It is more common among younger than older men, especially during their first sexual relationships. When it persists, it is often caused by fear of failure.

Psychological causes should be treated initially with advice and reassurance. If the partner is willing, the ‘squeeze technique’ can be useful. When the man indicates that he will soon have an orgasm, his partner grips the penis for a few seconds and then releases it suddenly. Intercourse is then continued. There are also other methods designed to regulate and reduce the amount of sexual stimulation in order to delay orgasm, which can be implemented as part of sex therapy. Sometimes SSRIs, which delay ejaculation, are used in treatment, but continued use is required.

Sexual pain disorders

Dyspareunia. This refers to pain on intercourse, and implicitly relates to women, although it can occur in men. Such pain can have many different causes. Pain that is experienced after partial penetration may result from impaired lubrication of the vagina, from scars or other painful lesions, or from the muscle spasm of vaginismus. Pain on deep penetration strongly suggests pelvic pathology such as endometriosis, ovarian cysts or tumours, or pelvic infection, although it can be caused by impaired lubrication associated with low sexual arousal. For a review, see Steege and Zolnoun (2009).

Vaginismus. This refers to spasm of the vaginal muscles which causes pain when intercourse is attempted, in the absence of a physical cause. The spasm is usually part of a phobic response to penetration, and may be made worse by an inexperienced partner. Spasms often begin as soon as the man attempts to enter the vagina; in severe cases they occur even when the woman attempts to introduce her own finger. Points of special relevance in the sexual history include the circumstances and objects that provoke spasm, the partner’s sexual technique, and any history of traumatic sexual experience.

There have been very few trials of treatment for dyspareunia, so therapy has to be based on clinical experience. Treatment employs the general sex therapy techniques described above, with emphasis on an initial ban on attempts at intercourse. Fears are treated by using psychological approaches, and the woman is helped to desensitize herself gradually by inserting first her finger and then dilators of increasing size. For a review, see Crowley et al. (2009).

Abnormalities of sexual preference (paraphilias)

The concept of abnormal sexual preference has a social aspect—that is, the behaviour does not conform to some generally accepted view of what is normal. The accepted view is not the same in every society or at every period in history. For centuries, abnormalities of sexual preference were regarded as offences against the laws of religion, rather than conditions that doctors should study and treat. The systematic investigation of these disorders began in the 1870s. Krafft-Ebing (1840–1902), a professor of psychiatry in Vienna, wrote a systematic account of paraphilias in his book Psychopathia Sexualis, which was first published in 1886. In England, Havelock Ellis (1859–1939) wrote extensively about these and other sexual disorders, which at the time included masturbation.

In a current psychiatric context, the important feature that delineates abnormal sexual preference is harm. That is, abnormal sexual preference is considered to be harmful, physically or emotionally, to another person or persons. In addition, we may add the suffering experienced by the person who has the abnormal preference, arising from the conflict between their sexual urges and the resulting shame, embarrassment, and guilt about the resulting harm or distress caused to others.

Currently, the paraphilias are considered to be psychiatric disorders, as classified in Table 14.6, but with major legal implications and interactions. The more serious disorders are thus primarily in the domain of forensic psychiatrists. However, any doctor may encounter abnormalities of sexual preference in their clinical work—for example, as a request for help, or an expression of concern by a third party.

For a review of this subject, see Fedoroff (2009).

Types of abnormality of sexual preference

Abnormalities of sexual preference can be divided into two groups—abnormalities of the object of the person’s sexual interest, and abnormalities in the preference of the sexual act. Abnormalities of the sexual object include paedophilia and fetishism. Abnormalities of preference of the sexual act include exhibitionism, voyeurism, and autoerotic asphyxia.

Aetiology of abnormalities of sexual preference

A range of theories have been proposed to explain abnormalities of sexual preference, none of which are supported by good evidence.

• Behavioural models. According to these models, sexual preference is shaped by events and reinforcements during development. For example, in 1877 Alfred Binet proposed that fetishism arises from the chance co-occurrence of sexual excitement with the object that later becomes the fetish object.

• Psychoanalytical models. For example, Freud argued that fetishism arises when castration anxiety is not resolved in childhood, with each fetish object representing a phallus.

• Biological models. These include reports linking paraphilias with abnormal brain activity or structure, or with genetic predisposition.

• Disease models. Paraphilias that begin in middle age or later may be secondary to dementia or other organic disorders, or to their treatment. For example, dopamine agonists that are used to treat Parkinson’s disease are associated with a range of abnormal sexual behaviours.

Assessment of abnormalities of sexual preference

Assessment often takes place in the context of possible or actual legal proceedings. It is important therefore to explain the relationship of the interview to any such proceedings, to explain the position with regard to confidentiality and its limits, and to obtain the necessary consent. People who are seen for the assessment of abnormality of sexual preference are often reluctant to disclose the true nature of their desires, or of their sexual behaviour. Whenever possible, consent should be obtained for an interview with any regular sexual partner, and with other informants who may be able to assist with, for example, the assessment of personality.

Assessment includes a psychiatric history, and should incorporate the following steps.

• Exclude another psychiatric disorder, especially if the patient first presents with the abnormal sexual preference in middle age or later. Abnormal sexual preference is sometimes secondary to dementia, alcoholism, or mania.

• Document the sexual behaviour. Obtain details of the normal and abnormal sexual behaviour, both currently and in the past. Remember that it is not uncommon for people to have more than one form of abnormal sexual preference. Find out what part the abnormal sexual preference is playing in the patient’s life other than as a source of sexual arousal.

• Assess the patient’ motivation. This will influence decisions about management. Patients’ motives for seeking treatment are often mixed. Individuals who have little wish to change their sexual behaviour may consult a doctor because of pressure from a sexual partner, a relative, or the police. Such people may hope to be told that no treatment can help, so as to justify the continuation of their paraphilia. Others seek help when they are depressed and feel guilty about their behaviour. A strong desire for change, expressed during a period of depression, may fade quickly as the patient’s normal mood is restored.

• Psychophysiological assessment. Some specialists use penile plethysmography or polygraphy to assess sexual interests. These methods are not part of routine assessment.

The management of abnormalities of sexual preference

Some aspects of management apply to all kinds of abnormality of sexual preference. Management specific to particular disorders is described below along with the other aspects of those disorders. The following outline is based on Fedoroff (2009).

• Complete the assessment in order to make a provisional diagnosis.

• Assess the immediate risks—for example, whether the patient should be reported to the authorities, or whether their level of distress requires urgent treatment or admission. Often it is a crisis that first leads the person to present—for example, if they have been accused of a sexual offence or have been caught in a compromising sexual act.

• Decide what treatment is indicated and acceptable to the patient. This includes agreement about the aims of treatment, and the patient’s motivation. It may also include incentives or punishments determined by the legal system.

• Treatments for the paraphilias are of three types:

Images psychotherapy—individual, couple, or group

Images pharmacotherapy—SSRIs, anti-androgens

Images rehabilitation—counselling, education, social skills training.

Of the treatments for paraphilias, there is evidence for the efficacy of surgical castration, anti-androgens, GnRH agonists, and SSRIs (Fedoroff, 2009). The availability and acceptability of these treatments limit their use. It is less clear whether psychological interventions, such as relapse prevention, are effective, and the findings of a meta-analysis were inconclusive (Hanson et al., 2002).

Types of abnormality of preference for the sexual object

These abnormalities involve preferences for an ‘object’ other than another adult for the achievement of sexual excitement. The alternative ‘object’ may be inanimate, as in fetishism, or may be a child (paedophilia) or an animal (zoophilia). A few of the many specific paraphilias of this kind are mentioned here.


In sexual fetishism, the preferred or only means of achieving sexual excitement are inanimate objects or parts of the human body that do not have direct sexual associations. The objects that evoke sexual arousal are many and varied, but for each individual there is usually only a small number of such objects. Contact with the object causes sexual excitement, which may be followed by solitary masturbation or by sexual intercourse that incorporates the fetish. The disorder merges into normal sexual behaviour, but is considered to be abnormal when the behaviour takes precedence over the usual patterns of sexual intercourse. The estimated prevalence is 1–18% in men, and the disorder is very rare in women. There are no good data on natural history or treatment outcome.

Transvestic fetishism (fetishistic transvestism)

Transvestic fetishism ranges from the occasional wearing of a few articles of clothing of the opposite sex to complete cross-dressing. Transvestic fetishism is rare among women, and the description below applies to men. The prevalence is estimated to be around 1%. Onset usually occurs at around the time of puberty.

Transvestic fetishists generally experience sexual arousal when cross-dressing, and the behaviour often terminates with masturbation. Sexual arousal may diminish with ageing, so that the person dresses mainly in order to feel feminine. Sometimes the clothes are worn in public, either underneath male outer garments or in some cases without such precautions against discovery. Transvestic fetishists are in no doubt that their gender conforms with their external sexual characteristics. Most are heterosexual and many have sexual partners, who may or may not know about the behaviour. There have been no reliable follow-up studies or evaluated treatments.


In contrast to the generally benign effects of fetishism on others, paedophilia has serious forensic and clinical implications. Paedophilia is repeated sexual activity (or fantasizing about such activity) with prepubertal children as a preferred or exclusive method of deriving sexual excitement. Like most paraphilias, it is almost exclusively a disorder of men, although female cases are recognized (Gannon and Rose, 2008). Paedophilia has to be distinguished from intercourse with young people who have passed puberty but not yet reached the legal age of consent (which differs between legislations).

Paedophiles usually choose a child aged between 6 years and puberty, but some prefer very young children. The child may be of the opposite sex (heterosexual paedophilia) or the same sex (homosexual paedophilia). Some paedophiles approach children within their extended family, or in their professional care, while others befriend unrelated children. Although most paedophiles who are seen by doctors are men of middle age, the condition is established early in life. With younger children, fondling or masturbation is more likely to occur than full coitus, but sometimes young children are injured by forcible attempts at penetration. There are rare and tragic cases of paedophilia associated with sexual sadism.

There is no reliable information about the prevalence of paedophilia. It is evident from the existence of child prostitution in some countries, the ready sale of pornographic material depicting sex with children, and the number of Internet sites that show such material, that interest in sexual relationships with children is not rare.

There is no established treatment, nor are there any reliable outcome data. Paedophiles are thought to have a higher reoffending rate than other sexual offenders. For a review, see Seto (2009).

Abnormalities in the preference for the sexual act

The second group of abnormalities of sexual preference involves variations in the behaviour that is engaged in to obtain sexual arousal.


Exhibitionism is the repeated exposure of the genitals to unprepared strangers for the purpose of achieving sexual excitement, but without any attempts at further sexual activity with the other person. The name exhibitionism was suggested by Ernest-Charles Lasègue in 1877. (This technical use of the term is clearly different from its everyday meaning.)

Most exhibitionists who are seen by doctors are aged 20–40 years, and two-thirds of them are married (Gayford, 1981). The act of exposure is usually preceded by a feeling of mounting tension. Most exhibitionists choose places from which escape is easy, and suddenly reveal their penis to the other person, from whom the exhibitionist seeks to evoke a strong emotional reaction. This reaction is interpreted by the exhibitionist as sexual interest, although it may in fact be shock, fear, or laughter. This distorted interpretation is accompanied by a state of intense excitement. Some men masturbate during exposure; others do so afterwards.

As a generalization, two groups of exhibitionists can be described. The first group includes men of inhibited temperament who struggle against their urges and feel much guilt after the act; they sometimes expose a flaccid penis. The second group includes men who have aggressive traits, sometimes accompanied by features of antisocial personality disorder; they usually expose an erect penis. They gain pleasure from any distress they cause, and often feel little guilt.

In the UK, if a man is brought to court because of exhibitionism, he is charged with the offence of indecent exposure. About 80% of men who are charged with indecent exposure are exhibitionists. Few go on to commit violent sexual acts. There are no reliable data on treatment or outcome.

Sexual sadism

Sadism is named after the Marquis de Sade (1740–1814), who inflicted extreme cruelty on women for sexual purposes. Sexual sadism refers to the achievement of sexual arousal habitually, and in preference to heterosexual intercourse, by inflicting pain on another person, by bondage, or by humiliation. The acts may be symbolic, causing little actual harm. However, at times serious injuries are caused. Extreme examples are the rare ‘lust murders’, in which the killer inflicts mutilations on the genitalia of his victim. In these rare cases, ejaculation may occur during the sadistic act, or later during intercourse with the dead body (necrophilia).

Sexual sadism as the predominant sexual practice is probably uncommon, but sexual sadism as a part of ‘normal’ sexual activity has been estimated to have a prevalence of 3–20% (Fodoroff, 2009). No treatment has been shown to be effective. Men who have committed serious injury are dealt with by legal means. The risks must not be underestimated when potentially dangerous behaviour has been planned or has taken place.

Sexual masochism

Sexual masochism involves using the experience of suffering or humiliation as a preferred or exclusive practice for achieving sexual excitement. The condition is named after Leopold von Sacher-Masoch (1836–1905), an Austrian novelist, who described sexual gratification from the experience of pain. The suffering may take the form of being beaten, trodden upon, or bound, or the enactment of various symbolic forms of humiliation. Masochism, unlike most other sexual deviations, occurs in women as well as in men.


Voyeurism refers to observing the sexual activity of others repeatedly as a preferred means of sexual arousal. The voyeur also spies on women who are undressing or without clothes, but does not attempt to engage in sexual activity with them. Voyeurism is usually accompanied or followed by masturbation. Some voyeurs use video cameras to record women undressing or engaging in sexual activities. Although the voyeur usually takes great care to hide from the women whom he is watching, he often takes considerable risks of discovery.

Voyeurism is usually a disorder of heterosexual men whose sexual relationships are limited or inadequate.

Other abnormalities of preference of the sexual act

Auto-erotic asphyxia. Auto-erotic asphyxia is the practice of inducing cerebral anoxia in order to heighten sexual arousal while masturbating. Asphyxia is usually induced by partial strangulation with cords, or by a plastic bag placed over the head. The practice occurs almost exclusively in men. It is hazardous and may lead to death. The act may be accompanied by the use of objects to produce anal stimulation, by the use of fetish objects, or by cross-dressing. The person may look at himself in a mirror or photograph himself, or he may apply bondage—for example by tying the ankles. Few people who practise auto-erotic asphyxia seek help. For further information, see Blanchard and Hucker (1991).

Frotteurism. In frotteurism, the preferred form of sexual excitement is by rubbing the male genitalia against another person, or by fondling the breasts of an unwilling participant, who is usually a stranger, generally in a crowded place.

Abnormalities of gender identity


Transsexual people are convinced that they are of the opposite gender to that indicated by their chromosomes. They have an overpowering wish to live as a member of the opposite gender to their anatomical sex, and seek to alter their bodily appearance and genitalia.

In the past, the condition was called eonism, because it was exemplified by the Chevalier d’Eon de Beaumont. In the psychiatric literature, the condition was mentioned by Esquirol in 1838 and described in more detail by Krafft-Ebing in 1886. In the 1960s, the attention of doctors and the public was directed to the condition by a number of high-profile reports, and by a book by Harry Benjamin (Benjamin, 1966).


Transsexual people generally prefer the usage in which transsexual female refers to people who were identified at birth as male, but experience themselves as female, and transsexual male refers to people who were identified at birth as female, but experience themselves as male. In the past in the medical literature, the term transsexual male has more often been used to denote people identified at birth as male but experiencing themselves as female. In the following account, we generally use the terms male-to-female transsexual and female-to-male transsexual, as they describe the sequence of events in the person’s life. Thus male-to-female transsexuals are people who were identified at birth as male, but who later experience themselves as female.

Some transsexual people object to the use of the word transsexual as a noun. When the word transsexual is used in this way in the following account, it denotes a trans-sexual person, just as the word heterosexual is widely used as a noun to denote a heterosexual person.

Many transsexual people consider that their condition should not be listed as a psychiatric disorder. However, it is listed in both ICD-10 and DSM-IV. This listing may assist transsexual people who request treatment, as in many countries treatment under a state health scheme or private insurance is restricted to disorders that are listed in one of the classifications of diseases.


A survey in the Netherlands found a prevalence of gender identity disorder in adults of 1 per 10 000 born male and 1 per 30 000 born female (Kesteren et al., 1996). Among those seeking help, male-to-female transsexuals generally exceed female-to-male transsexuals by a ratio of 3:1 (Green, 2009).


Transsexual people have a strong conviction, which usually starts before puberty, of belonging to the gender opposite to that to which they were assigned (‘I am a woman trapped in a man’s body’). Parents sometimes report that, as children, these people preferred the company and pursuits of children of the opposite anatomical sex, although such a history is not invariable. However, follow-up studies of effeminate boys have found that they more often grow up as homosexual than as trans-sexual adults.

By the time that medical help is requested, most transsexual people have started to cross-dress at times. In contrast with transvestic fetishists, male-to-female transsexuals cross-dress to feel more like a woman, not to achieve sexual arousal. (They should also be contrasted with homosexual men who dress as women in order to attract other male homosexuals.) Make-up may be worn, the hair is arranged in a feminine style, and facial and body hair may be gradually removed by electrolysis. These physical changes may be added to by cultivating feminine gestures, raising the pitch of the voice, and seeking changes in social role.

Male-to-female transsexuals often ask for help in altering the appearance of the breasts and external genitalia. Usually the first requests are for oestrogens to enlarge the breasts. Later requests may be for surgery to the breasts, castration and removal of the penis, and operations to create an artificial vagina. Many transsexual people are greatly distressed by their predicament. Depression is common, and some threaten self-harm or mutilation if they cannot have surgery. These threats should be taken seriously because they indicate extreme distress, and are sometimes carried out.

Female-to-male transsexuals adopt masculine dress, voice, gestures, and social behaviour. They wish to have intercourse in the role of a man with a heterosexual woman (and not with a homosexual woman). Some request mastectomy or hysterectomy, and a few ask for plastic surgery to create an artificial penis.


The causes are unknown (Green, 2009). The following factors have been investigated:

Genetic causes. Transsexual people have normal sex chromosomes, and there is no convincing evidence of a genetic predisposition.

Early upbringing. There is no convincing evidence that transsexual people have been brought up in the gender role opposite to their anatomical sex. As noted above, most boys with gender identity disorder grow up as homosexual rather than transsexual adults.

Endocrine causes. No specific hormonal profile has been found in adult transsexuals. However, it has been suggested that the condition might result from endocrine abnormalities during early life. For example, when pregnant rhesus monkeys are given large doses of androgens, their female infants behave more like males during play (Young et al., 1964), and girls with adrenogenital syndrome show boyish behaviour in childhood (Ehrhardt et al., 1968). Also, a high rate of polycystic ovarian disease has been reported in female-to-male transsexuals (Bosinski et al., 1997).

Differences in brain structure. The volume of the central subdivision of the bed nucleus of the stria terminalis is larger in men than in women. One well-known study of postmortem brains from male-to-female transsexuals found the size of the nucleus to be similar to that in described in women (Zhou et al., 1995). However, the sample included only six subjects, and the results have been criticized on technical and interpretational grounds (Green, 2009). For a review, see Swaab and Garcia-Falgueras (2009).

Miscellaneous factors. As summarized by Green (2009), male-to-female transsexuals (like homosexuals) are reported to have an increased likelihood and number of older brothers, and a higher ratio of maternal aunts to uncles. They are also more likely to be non-right-handed.


There is no reliable information about the course of untreated transsexual people. Some transsexual people appear to change their aspirations; some adapt to their desired gender identities without medical help, and others abandon their wish to change during the early stages of treatment. It is not known what proportion of trans-sexual people fall into each of these groups.


First stages. There have been no randomized trials in this area. Current treatment generally follows the Harry Benjamin International Gender Dysphoria Association (2001) standards of care document. The following account summarizes the steps.

Agreeing a plan. Transsexual people have often already decided what treatment they want, and may be impatient to obtain it. They may regard psychiatric assessment as an unnecessary obstacle on the path to hormonal treatment and surgery and, as noted above, may become distressed or threaten suicide or self-mutilation if their own timetable is not met. However, it should be explained that certain stages have to be passed through. The first is to demonstrate the psychological stability that is essential if the person is to succeed in the transition that is sought. To help to establish this stability, the person needs to pass through the ‘real-life experience’ (see below).

Counselling should deal with the full range of options available. It emphasizes the need to set realistic goals and to consider the full consequences of any contemplated changes, both for the person concerned and for their family, including any children. The counsellor should also discuss the likelihood that no therapy can permanently eradicate all vestiges of the person’s original gender assignment. There are also practical and legal implications, which differ between countries (e.g. concerning marriage, birth certificate, and passport).

The ‘real-life experience.’ A male-to-female transsexual person has to become accustomed to appearing, speaking, and behaving as a woman. Facial hair is often removed and voice training may be required. The experience requires the person to live in the new gender role continuously for at least a year, and to meet certain criteria. Appropriate hormone treatment (see below) is prescribed. At the end of the year, an individual who can demonstrate to an assessment panel that they are better adjusted in the new gender role than in the old one may be considered for surgery.

Hormonal treatment. Oestrogens may be prescribed to produce breast enlargement and increase fat around the hips and buttocks, giving a more female appearance. Androgens may be prescribed to female-to-male transsexual people, as a result of which the voice deepens, hair increases on the face and body, menstruation ceases, the clitoris enlarges, and the sex drive increases. These changes are generally more pronounced than the changes in male-to-female transsexual patients who are taking oestrogen. Prolonged use of androgens is associated with the risk of liver damage.

Surgery. People who have successfully passed through the previous stages of treatment may be considered for surgery. Opinions differ about the merits of sex reassignment, and decisions should be taken jointly by the psychiatrist and the surgeon, in consultation with the primary care physician. Surgery should be followed by intensive and long-term support.

For a male-to-female transsexual person, possible surgical procedures include mammoplasty, penectomy, orchidectomy, and the creation of a vagina-like structure. For a female-to-male transsexual person, surgical procedures include mastectomy, ovariectomy, and phalloplasty.

Help for the family. Many transsexual people are married, and some have children. Partners, and especially children, require help in coming to terms with the effects on their own lives of the changes in the transsexual person. Concerns may be expressed about the possibility of gender role problems in the children. However, Green (1998) studied 34 children with a transsexual parent and found no instances of gender identity disorder.

Outcome. Follow-up studies show that outcomes are generally favourable, but no studies have compared patients who were allocated randomly to surgery or no surgery. A review found that 90% of male-to-female transsexual people were reported to have had a successful outcome of sex reassignment (Green and Fleming, 1991). However, this figure must be viewed cautiously because of possible bias in selection and loss from follow-up. In one study (Mate-Kole et al., 1990), 40 patients were selected for surgery, and then allocated alternately to treatment within 3 months, or treatment after 2 years. Two years after their operation the first group had lower scores for neurotic symptoms and somewhat better work adjustment than the group that was still waiting for the operation. However, Meyer and Reter (1979) found similar rates of improvement between operated and unoperated groups.

Dual-role transvestism

This term is used in ICD-10 to describe people who wear clothes of the opposite sex but are neither transvestic fetishists (seeking sexual excitement) nor transsexuals (wishing for a change of gender and sexual role). Instead they enjoy cross-dressing in order to gain temporary membership of the opposite sex.

Gender identity disorders in children

Parents more often seek advice about effeminate behaviour in boys than about masculine behaviour in small girls (it is not clear whether such behaviour in girls is less frequent or more socially acceptable). Effeminate boys prefer girlish games and enjoy wearing female clothing. The outcome among these boys is variable; some develop normal male interests and activities, although most become homosexual men. Parents should be discouraged from making attempts to force them into stereotyped ‘masculine’ behaviours and roles, as this has no demonstrated benefits and can cause discord and distress. For further information, see Zucker (2005).

Further reading

Fairburn CG (2008). Cognitive Behavior Therapy and Eating Disorders. Guilford Press, New York.

Kryger MH, Roth T and Dement WC (2005). Principles and Practice of Sleep Medicine. Elsevier Saunders, Philadelphia, PA.

Bancroft J (2008). Human Sexuality and its Problems, 3rd edn. Churchill Livingstone Elsevier, Edinburgh.