Shorter Oxford Textbook of Psychiatry, 6th Ed.

CHAPTER 22. Child psychiatry


Normal development

Classification of psychiatric disorders in children and adolescents



Psychiatric assessment of children and their families

Physical examination

Psychiatric treatment for children and their families

Review of syndromes

Ethical and legal problems in child and adolescent psychiatry

Appendix: History taking and examination in child psychiatry


The practice of child psychiatry differs from that of adult psychiatry in four important ways.

• Children seldom initiate the consultation. Instead, they are brought by a parent, or another adult, who thinks that some aspect of the child’s behaviour or development is abnormal. Whether a referral is sought depends on the attitudes and tolerance of these adults, and how they perceive the child’s behaviour. Healthy children may be brought to the doctor by over-anxious and solicitous parents or teachers, while in other circumstances severely disturbed children may be left to themselves.

• The child’s problems may reflect the problems of other people—for example, illness in the mother. Also, when a child’s problems have previously been contained within the family or school, that child may be referred when another problem, which reduces their capacity to cope with the child, arises in the family or school.

• The child’s stage of development must be considered when deciding what is abnormal. For example, repeated bed-wetting may be normal in a 3-year-old child but is abnormal in a 7-year-old. In addition, the child’s response to life events changes with age. Thus separation from the parents is more likely to affect a younger child than an older one.

• Children are generally less able to express themselves in words. For this reason, evidence of disturbance often comes from observations of behaviour made by parents, teachers, and others. These informants may give differing accounts, in part because children’s behaviour varies with their circumstances, and in part because the various informants have different ideas of what is abnormal. For this reason, informants should be asked for specific examples of any problem they describe, and asked about the circumstances in which it has been observed.

• The emphasis of treatment is different. Medication is used less in the treatment of children than in the treatment of adults, and is usually started by a specialist rather than the family doctor. Instead, there is more emphasis on working with the parents and the whole family, reassuring and retraining children, and coordinating the efforts of others who can help children, especially at school. Thus multidisciplinary working is even more important in child psychiatry than in adult psychiatry. Consequently, treatment is usually provided by a team that includes at least a psychiatrist, psychiatric nurses, a psychologist, and a therapist.

The first part of this chapter is concerned with a number of general issues concerning psychiatric disorder in childhood, including its frequency, causes, assessment, and management. The second part of the chapter contains information about the principal syndromes encountered in the practice of child psychiatry. The chapter does not provide a comprehensive account of child psychiatry. It is an introduction to the main themes for psychiatrists who are undertaking specialist general training. It is expected that they will follow it by reading a specialist text such as one of those listed under further reading on p. 680. Learning disability among children is considered in Chapter 23. Although this is a convenient arrangement, the reader should remember that many aspects of the study and care of children with learning disability are similar to those described in this chapter.

Normal development

The practice of child psychiatry calls for some knowledge of the normal process of development from a helpless infant into an independent adult. In order to judge whether any observed emotional, social, or intellectual functioning is abnormal, it has to be compared with the corresponding normal range for the age group. This section gives a brief and simplified account of the main aspects of development that concern the psychiatrist. A textbook of paediatrics should be consulted for details of these developmental phases (for example, see Hopkins et al., 2004).

The first year of life

This is a period of rapid development of motor and social functioning. Three weeks after birth, the baby smiles at faces, selective smiling appears by 6 months, fear of strangers by 8 months, and anxiety on separation from the mother shortly thereafter.

Bowlby (1980) emphasized the importance in the early years of life of a general process of attachment of the infant to the parents, and of more selective emotional bonding. Although bonding to the mother is most significant, important attachments are also made to the father and other people who are close to the infant. Other studies have shown the reciprocal nature of this process and the role of early contacts between the mother (or other carers) and the newborn infant in initiating bonding. Attachment and bonding are discussed further on p. 634.

By the end of the first year, the child should have formed a close and secure relationship with the mother or other close carer. There should be an ordered pattern of sleeping and feeding, and weaning has usually been accomplished. Children will have begun to learn about objects outside themselves, simple causal relationships, and spatial relationships. By the end of the first year, they enjoy making sounds and may say ‘mama’, ‘dada’, and perhaps one or two other words.

Year two

This too is a period of rapid development. Children begin to wish to please their parents, and appear anxious when they disapprove. They begin to learn to control their behaviour. By now, attachment behaviour should be well established. Temper tantrums occur, particularly if exploratory wishes are frustrated. These tantrums do not last long, and should lessen as the child learns to accept constraints. By the end of the second year the child should be able to put two or three words together as a simple sentence.

Preschool years (2–5 years)

This phase brings a further increase in intellectual abilities, especially in the complexity of language. Social development occurs as children learn to live within the family. They begin to identify with the parents and adopt their standards in matters of conscience. Social life develops rapidly as they learn to interact with siblings, other children, and adults. Temper tantrums continue, but diminish and should disappear before the child starts school. Attention span and concentration increase steadily. At this age, children are very curious about the environment and ask many questions.

In children aged 2–5 years, fantasy life is rich and vivid. It can form a temporary substitute for the real world, enabling desires to be fulfilled regardless of reality. Special objects such as teddy bears or pieces of blanket become important to the child. They appear to comfort and reassure the child, and help them to sleep. They have been called transitional objects.

Children begin to learn about their own identity. They realize the differences between males and females in appearance, clothes, behaviour, and anatomy. Sexual play and exploration are common at this stage. According to psychodynamic theory, at this stage defence mechanisms develop to enable the child to cope with anxiety arising from unacceptable emotions. These defence mechanisms have been described on p. 154.

Common problems in early childhood

In children from birth to the beginning of the fifth year, common problems include difficulties in feeding and sleeping, as well as clinging to the parents (separation anxiety), temper tantrums, oppositional behaviour, and minor degrees of aggression.

Middle childhood

By the age of 5 years, children should understand their identity as boys or girls, and their position in the family. They learn to cope with school, and to read, write, and begin to acquire numerical concepts. The teacher becomes an important person in children’s lives. At this stage children gradually learn what they can achieve and what their limitations are. Conscience and standards of social behaviour develop further. According to psychoanalytical theory, defence mechanisms develop further while psychosexual development is quiescent (the latent period). The latter notion has been questioned, and it now seems that in children aged 5–10 years sexual interest and activities are present, although they may be concealed from adults.

Common problems in middle childhood.

The common problems in this age group include fears, nightmares, minor difficulties in relationships with peers, disobedience, and fighting.


Adolescence is the growing-up period between childhood and maturity. Among the most obvious features are the physical changes of puberty. The age at which these changes occur is quite variable, usually between 11 and 13 years in girls, and between 13 and 17 years in boys. The production of sex hormones precedes these changes, starting in both sexes between the ages of 8 and 10 years. Adolescence is a time of increased awareness of personal identity and individual characteristics. At this age, young people become self-aware, are concerned to know who they are, and begin to consider where they want to go in life. They can look ahead, consider alternatives for the future, and feel hope and despair. Some experience emotional turmoil and feel alienated from their family, but such experiences are not universal.

Peer group relationships are important and close friendships often develop, especially among girls. Membership of a group is common, and this can help the adolescent in moving towards autonomy. Adolescence brings a marked increase in sexual interest and activity. At first, tentative approaches are made to the opposite sex. Gradually these become more direct and confident. In late adolescence, there is a capacity for affection towards the opposite sex as well as sexual feelings. How far and in what way sexual feelings are expressed depends greatly on the standards of society, the behaviour of the peer group, and the attitudes of the family.

Common problems in later childhood and early adolescence

Common problems among children aged from 12 to 16 years include moodiness, anxiety, minor problems of school refusal, difficulties in relationships with peers, disobedience and rebellion including truancy, experimenting with illicit drugs, fighting, and stealing. Schizophrenia and bipolar disorder may have their onset in this age group, but are uncommon.

Developmental psychopathology

In child psychiatry it is important to adopt a developmental approach for three reasons.

• The stage of development determines whether behaviour is normal or pathological. For example, as noted above bed-wetting is normal at the age of 3 years but abnormal at the age of 7 years.

• The effects of life events differ as the child develops. For example, infants aged under 6 months can move to a new carer with little disturbance, but children aged 6 months to 3 years of age show great distress when separated from the original carer, because an attachment relationship has been formed. After the age of 3 years, attachment bonds are still strong but the child’s ability to understand and to use language can reduce the effect of a change of caretaker provided that it is arranged sensitively.

• Psychopathology may change as the child grows older. Anxiety disorders in childhood tend to improve as the child develops, depressive disorders often recur and continue into adult life, conduct disorders frequently continue into adolescence as aggressive and delinquent behaviour, and also commonly as substance abuse—a problem that is less common in younger children. These changes and continuities are sometimes related as much to changes in the environment as to developmental changes in the child.

The psychopathology of individual disorders will be discussed later in the chapter. Here some general issues regarding childhood psychopathology are summarized.

The influence of genes. Susceptibility genes have been identified for a few disorders, namely autism, attention deficit disorder, and specific reading disorder. However, the amount of variance explained by the identified genes is not great, suggesting multiple genes of small effect, and interactions between genes and the environment. Also, genes may indirectly cause stressful life events–for example, they may control personality traits of irritability and impulsiveness that lead to the repeated breakdown of relationships.

The influence of the environment. Factors in the environment may predispose to or precipitate disorders, and they may maintain them. They may also protect from the effects of other causative agents—for example, the risk of depression in adulthood as a consequence of poor parental care in childhood is reduced by the experience of a caring relationship with another person. This experience is not protective against all causative factors—for example, a caring relationship does not reduce the risk of depression in adulthood following child abuse (Hill et al., 2001).

The dividing line between normal and abnormal. Many childhood disorders are at the extreme end of a continuum with normal behaviour. Despite this, we study them using a categorical system because decisions about treatment require ‘yes or no’ answers. To form a category, a cut-off point has to be decided, and this is usually rather arbitrary. Children who fall just below the cut-off point—and hence outside the diagnostic group—can nevertheless have problems and need help. For example, some adolescents who fall below the threshold for depressive disorder have significant psychosocial impairment, and even mild depressive symptomatology can be associated with poor academic performance (Pickles et al., 2001).

Continuities and discontinuities. Some symptoms and behaviour problems in childhood are associated with problems in adulthood. For example, overactivity and difficulties in management at the age of 3 years are associated with offending in adult life. In contrast, anxiety in childhood is less likely to persist into adulthood (see p. 661).

Parent–child interactions. Maternal behaviour affects the child, but children elicit behaviours from their parents. For example, a mother is likely to be less responsive to an infant who does not respond to cuddling and play than to a sibling who is responsive to her. Her lack of responsiveness may then affect the infant, thus increasing the difficulty with attachment.

Of course, most of these issues are also relevant to psychopathology in adults, and are considered further in Chapter 5.

Classification of psychiatric disorders in children and adolescents

Both DSM-IV and ICD-10 contain a scheme for classifying the psychiatric disorders of childhood. Disorders of adolescence are classified partly with this scheme, and partly with the categories used in adult psychiatry.

Seven main groups of childhood psychiatric disorders are generally recognized by clinicians. The terms used in this book for the seven groups are listed below, with some alternatives shown in parentheses:

• adjustment reactions

• pervasive developmental disorders

• specific developmental disorders

• conduct (antisocial or externalizing) disorders

• attention-deficit hyperactivity disorders

• emotional (neurotic or internalizing) disorders

• symptomatic disorders.

Many child psychiatric disorders cannot be classified in a satisfactory way by allocating them to a single category. Therefore multi-axial systems have been proposed. ICD-10 has six axes:

1. clinical psychiatric syndromes

2. specific delays in development

3. intellectual level

4. medical conditions

5. abnormal social situations

6. level of adaptive functioning.

In DSM-IV, psychiatric syndromes and specific delays in development are both on axis 1, and the other axes are the same as in ICD-10, so that there are five in total. The scheme is easy to use, and allows clinicians to record systematically the different kinds of information required when categorizing children’s problems.

The DSM-IV and ICD-10 classifications for child psychiatric disorders are shown in Table 22.1. (In this book, learning disability (mental retardation) in childhood is considered in Chapter 23.) Both schemes are complicated, so only the main categories are shown in the table. In most ways the two schemes are similar. Both have categories for the following:

Table 22.1 Classification of childhood psychiatric disorders


• specific and pervasive developmental disorders, with the former divided into disorders affecting motor skills, speech and language (communication), and scholastic skills (learning)

• disorders of behaviour, which are divided into conduct (or disruptive behaviour) disorder and attention deficit (hyperkinetic) disorder

• anxiety (emotional) disorders (DSM-IV does not provide a separate category for childhood anxiety disorder, but uses the adult category instead)

• tic disorders.

DSM-IV has a category for eating and elimination disorders, whereas in ICD-10 these conditions are classified with sleep disorders under ‘other behavioural and emotional disorders.’ For further information about classification in child psychiatry, see Scott (2009a).


Behavioural and emotional disorders are common in childhood. Estimates vary according to the diagnostic criteria and other methods used, but it appears that rates in different developed countries are similar. Moreover, the rates of emotional and behavioural disorders in developing countries are similar to those in developed ones. In the UK, the prevalence of child psychiatric disorder in ethnic-minority groups has usually been found to be similar to that in the rest of the population. The few exceptions are mentioned later in this chapter.

A landmark study was carried out more than 40 years ago on the Isle of Wight in the UK. The study was concerned with the health, intelligence, education, and psychological difficulties in all the 10- and 11-year-olds attending state schools on the island—a total of 2193 children (Rutter et al., 1970a). In the first stage of the enquiry, screening questionnaires were completed by parents and teachers. Children identified in this way were given psychological and educational tests and their parents were interviewed. The 1-year prevalence rate of psychiatric disorder was about 7%, with the rate in boys being twice that in girls. The rate of emotional disorders was 2.5%, and the combined rate of conduct disorders together with mixed conduct and emotional disorders was 4%. Conduct disorders were four times more frequent among boys than girls, whereas emotional disorders were more frequent in girls, in a ratio of almost 1.5:1. There was no correlation between psychiatric disorder and social class, but prevalence increased as intelligence decreased. It was also associated with physical handicap and especially with evidence of organic brain damage. In addition, there was a strong association between reading retardation and conduct disorder. A subsequent study using the same methods was conducted in an inner London borough (Rutter et al., 1975a,b). Here the rates of all types of disorder were twice those in the Isle of Wight.

Overall rates of psychiatric disorder. The results of these early studies have been broadly confirmed in subsequent investigations, using standard diagnostic criteria, conducted in the UK (Meltzer et al., 2000), in New Zealand (Fergusson et al., 1993), and in the Great Smoky Mountains in the USA (Costello et al., 1996). For example, in the study in the UK, ICD-10 disorders were present in about 10% of more than 10 000 children aged 5–15 years. The most common problem was conduct disorder (5%), closely followed by emotional disorders (4%), while 1% were rated as hyperactive. The less common disorders (autism, tic disorders, and eating disorders) were present in about 0.5% of the population.

Rates in adolescence. Evidence about adolescence was provided originally by a 4-year follow-up of the Isle of Wight study (Rutter et al., 1976b). At the age of 14 years, the 1-year prevalence rate of significant psychiatric disorder was about 20%. Prevalence estimates in subsequent studies vary somewhat, but most broadly confirm the original findings, with rates between 15% and 20%. However, there are some minor differences over time. In the UK, for example, there is evidence of a substantial increase in conduct and emotional problems in adolescents over the past two decades (Collishaw et al., 2004, 2010).

Variations with gender and age. Before puberty, disorders are more frequent overall among males than among females; after puberty, disorders are more frequent among females. Particular disorders also vary in frequency according to gender and age (see Table 22.2).

Table 22.2 Comparative frequency of psychiatric disorders in males and females


Other sources of variation. In the middle years of childhood, rates of psychiatric problems differ between areas of residence, being twice as high in urban areas (about 25%) as in rural areas (about 12%) (Rutter et al., 1975a). Subsequent studies have confirmed these earlier findings of a higher frequency of psychiatric disorder in association with:

• breakdown of the parental relationship, parental illness, and parental criminality

• residence in urban areas with social disadvantage

• attendence at a school with a high turnover of teachers.

Comorbidity. Studies using DSM criteria find high rates of comorbidity between childhood disorders. This could be because one disorder predisposes to another, or because they have common predisposing factors, or because the classification system has gone too far in identifying, as distinct disorders, patterns of behaviour that can occur in more than one condition.

For a review of the psychiatric epidemiology of childhood and adolescence, see Costello and Angold (2009).

Follow-up studies

Mild symptoms and behavioural or developmental problems are usually short-lived. However, conditions severe enough to be diagnosed as psychiatric disorders often persist for years. Thus in the Isle of Wight study, 75% of children with conduct disorder and 50% of those with emotional disorders at age 10 years were still handicapped by these problems 4 years later (Rutter et al., 1976b).

The prognosis for adult life of psychiatric disorder in childhood has been investigated in long-term studies. Robins (1966) followed up people who had attended a child guidance clinic 30 years previously, and compared them with a control group who had attended the same schools but had not been referred to the clinic. Children with conduct disorder had a poor outcome. As adults they were likely to develop antisocial personality disorder or alcoholism, have problems with employment or relationships, or commit offences. Some emotional problems that start in childhood also have a poor prognosis. Fombonne et al. (2001a,b) studied 149 people who, 20 years earlier, had attended a child and adolescent psychiatry clinic with a major depressive disorder, and some of whom also had a conduct disorder. Depression on the first occasion carried a high risk of adult depression, and comorbid conduct disorder on the first occasion carried an increased risk of drug dependence, alcoholism, and antisocial personality disorder in adulthood.

These examples show that continuities in psychiatric disorder between childhood and adulthood can be homotypic (where the same disorder persists over time) or heterotypic (where later disorders are of an apparently different kind). Although it is true that many young adult patients with psychiatric disorder have also had a diagnosable disorder in childhood or adolescence, childhood disorders do not invariably result in adult illness. For example, the majority of children with anxiety and depression do not have mood disorders in adulthood (Maughan and Kim-Cohen, 2005).


In discussing the causes of child psychiatric disorders, the principles are similar to those described in Chapter 5 on the aetiology of adult disorders. Most childhood disorders have multiple causes, with both genetic and environmental components. There is also a developmental aspect; children mature psychologically and socially as they grow up, and their disorders reflect this maturation. In the following paragraphs, four interacting groups of factors will be considered briefly. These are heredity, temperament, physical impairment with special reference to brain damage, and environmental factors in the family and society. All aspects of aetiology are discussed more fully in the textbook edited by Rutter et al. (2008).

Genetic factors

Children with psychiatric problems often have parents who suffer from a psychiatric disorder. Environmental factors account for a substantial part of this association. However, population-based twin studies indicate that there is a significant genetic contribution to some psychiatric disorders, especially to hyperactivity and anxiety disorders. Genetic studies are considered further when the aetiology of particular childhood psychiatric disorders is described later in this chapter. Here, we consider some general matters concerned with the interpretation of the results of genetic studies. Methods of genetic investigation are discussed on p. 98. Examples of genetic findings in child psychiatry are given at relevant points in this chapter, especially in the sections on autism and hyperactivity, where genetic research has produced some relevant findings (see p. 651). Some general points about the interpretation of genetic studies are shown in Box 22.1.

Temperament and individual differences

Many years ago, Thomas et al. (1968) conducted an influential longitudinal study in New York. They found that certain temperamental factors detected before the age of 2 years predisposed to later psychiatric disorder. In the first 2 years, one group of children (‘difficult children’) tended to respond to new environmental stimuli by withdrawal, slow adaptation, and an intense behavioural response. Another group (‘easy children’) responded to new stimuli with a positive approach, rapid adaptation, and a mild behavioural response. This second group was less likely than the first to develop behavioural disorders later in childhood. The investigators thought that these early temperamental differences were determined both genetically and by environmental factors.

More recent studies have confirmed an association between temperamental styles and subsequent childhood psychopathology. For example, behavioural inhibition predicts later anxiety disorders, while low positive affect has been linked to the development of depression. Perhaps unsurprisingly, children characterized by difficulties with self-control and high levels of irritability are more likely to be diagnosed subsequently with disruptive behaviour problems, including conduct disorder (Caspi and Shiner, 2008).

Box 22.1 Interpreting the results of genetic studies of childhood psychiatric disorders

Phenotype and genotype. Although standard diagnostic criteria are valuable in research, the results of this research show that not all diagnostic categories relate closely to the genetic factors in aetiology. For example, the phenotype of autism seems to include some kinds of developmental language disorder, while the phenotype of Gilles de la Tourette syndrome seems to extend to obsessional thinking and behaviour.

Comorbidity. Many children have symptoms that qualify for more than one psychiatric diagnosis—for example, depressive disorder and conduct disorder, or depressive disorder and eating disorder. As noted above (under epidemiology), it is not clear whether this overlap arises because one disorder predisposes to another, or because they are two manifestations of the same genetic predisposition.

Polygenic effects and environmental influences. The hereditary factors of importance in child psychiatry are largely polygenic. As in adulthood, these multiple genes interact with psychosocial factors, and genetic investigations may include estimates of environmental factors.

Indirect effects. Genes may exert their effects through factors such as intelligence and temperament, which in part determine whether certain situations are experienced as stressful.

Non-shared environment. Analyses of population genetic data divide the variance into genetic, shared environmental, and non-shared environmental factors (see p. 98). Many psychological traits have a heritability of about 50% (see p. 99 for an explanation of heritability), and for most, shared environmental effects are rather small. It might be concluded therefore that non-shared environment accounts for the remaining 50% of the variance. However, this remaining variance also includes an unknown but possibly quite large component of error. Moreover, a shared environment can affect different children in different ways, and this effect appears in the analyses as ‘non-shared environment.’ For these reasons, analyses that suggest large non-shared environmental effects should be supplemented with studies designed to measure such effects directly. Clinicians should not conclude, from estimates of variance, that shared family influences are unimportant.

Brain disorder

Although serious physical disease of any kind can predispose to psychological problems in childhood, brain diseases are the most important. In the Isle of Wight study (see above), about 7% of physically healthy children aged 10–11 years were classified as having psychiatric problems, compared with about 12% of physically ill children of the same age, and 34% of children with brain disorders (Rutter et al., 1976b). The high prevalence in the latter group was not explained by the adverse social factors known to be associated with the risk of brain disorder. Nor is it likely to have been due to physical disability as such, because rates of psychiatric disorder are lower in children equally disabled by muscular disorders. Instead, the rate of psychiatric disorder among children with brain damage is related to the severity of the damage, although it is not closely related to the site. It is as common among brain-injured girls as boys, a finding which contrasts with the higher rate of psychiatric disorder among boys in the general population.

Children with brain injury are more likely to develop psychiatric disorder if they encounter adverse psychosocial influences of the kind that provoke psychiatric disorder in children without brain damage.

Maturational changes and delayed effects

The effects of brain lesions are more complex in childhood than in adult life because the brain is still developing. This has two consequences:

• Greater capacity to compensate. The immature brain is more able than the adult brain to compensate for localized damage. For example, even complete destruction of the left hemisphere in early childhood can be followed by normal development of language.

• Delayed effects. Early damage may not be manifested as a disorder until a later stage of development when the damaged area takes up some key function. It is well established that brain injury at birth may not result in seizures until many years later. It has been suggested that there may be similar delays in the behavioural consequences of brain injury.

Lateralized damage and psychopathology

The behavioural effects of lateralized brain damage are less specific in childhood than in adult life. Therefore attempts to explain, for example, educational problems in terms of left- and right-sided damage or dysfunction are of doubtful value, and attempts to use left-brain or right-brain training are unlikely to be helpful.

The consequences of head injury in childhood

Head injury is a common cause of neurological damage in childhood. The form of the consequent disorder is not very specific, partly because the effects of head injury are seldom localized to one area of the brain. Common consequences of severe injury are intellectual impairment and behaviour disorder. The former is proportional to the severity of the injury, but the relationship of behaviour disorder to the injury is less direct.

Epilepsy as a cause of childhood psychiatric disorder

The relationship between epilepsy and psychiatric disorder in adults is considered on pp. 338–43. In childhood there is a strong association between recurrent seizures and psychiatric disorder. As in adult life, the causal relationship may be of four kinds:

• the brain lesion causing the epilepsy may also cause the psychiatric disorder

• the psychological and social consequences of recurrent seizures may cause the disorder

• the effects of epilepsy on school performance may cause the disorder

• the drugs used to treat epilepsy may cause the disorder through their side-effects.

The site and the type of epilepsy appear to be generally less important, apart from the fact that temporal lobe epilepsy seems to be more likely to be associated with psychological disorder. The age of onset of seizures also determines the child’s response to epilepsy. For a review of the influence of brain injury and epilepsy on psycho-pathology, see Harris (2008).

Environmental factors

The effect of life events

The concept of life events (see p. 94) is useful in child psychiatry as well as adult psychiatry. Life events may predispose to or provoke a disorder, or protect against it. Events can be classified by their severity, their social characteristics (e.g. family problems, or the death of a parent), or their general significance—for example, exit events (i.e. separations) or entrance events (i.e. additions to a family by the birth of a sibling). The way in which stressful life events contribute to childhood psychiatric disorders is not well understood, but as in adults the psychological impact of such events is influenced by factors such as temperament, cognitive style, and previous life experience.

It is important to note that life events often occur in a setting of chronic stress which may itself play a causal role in the precipitation of psychiatric disorder. Also, in children as in adults, there are important differences in the liability of individuals to experience negative life events. This may be a reflection of societal factors such as poverty and discrimination, but the behaviour of individuals can also shape life experience in important ways. For example, children with conduct disorder have a substantially increased risk of experiencing negative life events (Sandberg and Rutter, 2008).

Family influences

As children progress from complete dependence on others to independence, they need a stable and secure family background with a consistent pattern of emotional warmth, acceptance, help, and constructive discipline. Prolonged separation from or loss of parents can have a profound effect on psychological development in infancy and childhood. Poor relationships in the family may have similar adverse effects, and overt conflict between the parents seems to be especially important.

Maternal deprivation and attachment

The much quoted work of Bowlby (1951) led to widespread concern about the effects of ‘maternal deprivation.’ Bowlby suggested that prolonged separation from the mother was a major cause of juvenile delinquency. Although this suggestion has not been confirmed, it was an important step towards his idea that attachment is a crucial stage of early psychological development. Attachments are formed in the second 6 months of life, and adaptive attachments are characterized by an appropriate balance between security and exploration. Bowlby proposed that infants derive, from their experiences of attachment, internal models of themselves, of other people, and of their relationships with others. He proposed further that these internal models persist through later childhood into adult life, and that they influence self-concept and relationships. Thus children whose caregivers are loving, sensitive to their child’s needs, and consistent in their responses are likely to grow up with self-esteem and able to form loving and trusting relationships.

Bowlby’s ideas have been generally supported by studies in which attachment has been measured in infancy and in later years. Attachment has usually been measured at around 6–18 months of age using the Strange Situations Procedure (Ainsworth et al., 1978), which evaluates the infant’s response to separation from, and subsequent reunion with, the mother or other attachment figure. In this test, the response to the initial separation appears to be determined more by temperament than by attachment, and it is the response to reunion that is used to measure attachment. Other measures of attachment have been developed for older children. Using such tests, attachments have been divided into four types:

• secure: present in about 60% of infants, and associated with caregiving that is sensitive and responsive to the child’s needs

• avoidant: present in about 15%, and associated with caregiving that is rejecting or intrusive

• disorganized: present in about 15%, and associated with caregiving that is unpredictable or frightening

• resistant–ambivalent: present in about 10%, and associated with caregiving that is inconsistent or lacking.

Studies of older children indicate that the type of attachment develops early in life and tends to be stable thereafter, although it is modified to some extent by changing life circumstances. Securely attached children generally do better than those with insecure attachments. Insecure attachments are associated not only with the factors mentioned above, but also with maternal depression, maternal alcoholism, and child abuse. Of the three types of insecure attachment, the disorganized type is the best predictor of future problems, especially externalizing problems.

The effect of separation. Bowlby’s original studies of the effects of separation on young infants have been broadly confirmed, but it has become clear that the effects depend on many factors, including:

• the age of the child at the time of separation

• the previous relationship with the carers

• the reasons for the separation

• how the separation was managed

• the quality of care during the separation.

For further information about attachment, see Goodman and Scott (2005), on which the above account is partly based.

Family risk factors

Family risk factors for psychiatric disorder in childhood are multiple and cumulative. The risk increases in children of families with severe marital or other relationship conflict, low social status, large size or overcrowding, paternal criminality, and parental psychiatric disorder. The risk is also increased in children who are placed in care away from the family.

Protective factors

Protective factors reduce the rate of psychiatric disorder associated with a given level of risk factors. Protective factors include good parenting, strong affectionate ties within the family, including good sibling relationships, sociability, and the capacity for problem solving in the child, and support outside the family from individuals, or from the school or church (Jenkins, 2008).

Child-rearing practices

Some patterns of child rearing are clearly related to psychiatric disturbance in the child, particularly those that involve verbal or physical abuse and scapegoating. Sexual abuse is another important risk factor (Glaser, 2008).

Effects of alternative childcare

Working mothers entrust part of the care of their children to other people. In general, the use of alternative childcare does not appear to be a major risk factor for psychiatric disorder, although multiple features of the nature of care need to be assessed. For example, high-quality care with sufficient individual attention can result in improved development of cognitive, social, and language skills. Generally, however, exposure to a large amount of alternative care correlates with increased levels of behavioural problems reported subsequently by teachers (Belsky et al., 2007).

Effects of parental mental disorder

Rates of psychological problems are higher in the children of parents with mental illness than in the children of healthy parents. These problems usually involve poor adjustment at home or at school, and often include disruptive behaviours. The causes are complex and are likely to be both genetic and environmental. However, an effect of illness in reducing the effectiveness of parenting has been demonstrated for depressed mothers, where attachment to the infant is impaired by lower levels of warmth and responsiveness. It also seems likely that maternal depression and anxiety during pregnancy can be associated with later childhood disturbance, independent of any postnatal effects of maternal illness (Jenkins, 2008).

Mothers who misuse alcohol or drugs. Children of such mothers suffer from a series of disadvantages which are related not only to the effects of drug taking on the development of the fetus and the mother’s care of the child, but also to features of the mother’s personality that led her to take drugs, and to social problems associated with the drug taking. For a review, see Greenfield et al. (2010).

Effects of divorce

The children of divorced parents have more psychological problems than the children of parents who are not divorced. It is not certain how far these problems precede the divorce and are related to conflict between the parents, the behaviour of one or both parents that contributed to the decision to divorce, and the changes in family relationships that accompany remarriage or cohabitation following the divorce. Distress and dysfunction in the children are greatest in the year after the divorce; after 2 years these problems are still present but are generally less severe than those of children who remain in conflictual marriages. For a review of the long-term effects of divorce on children, see Hetherington (2005).

Death of a parent

In children, the response to the death of a parent varies with age. Children aged below 4–5 years do not have a complete concept of death as causing permanent separation. Such children react with despair, anxiety, and regression to separation, however caused, and their reaction to bereavement is no different.

Children aged 5–11 years have an increasing understanding of death. They usually become depressed and overactive, and may show disorders of conduct. Some have suicidal thoughts and ideas that death would unite them with the lost parent. Suicidal actions are infrequent. In children over the age of 11 years, the response is increasingly like that of adults (see p. 171).

Bereavement may have long-term effects on development, especially if the child was young at the time of the parent’s death, and if the death was sudden or violent. Outcome probably depends largely on the effects of the bereavement on the surviving parent. Most studies of bereavement in children have been concerned with the death of a parent; few have concerned the death of a sibling. For a review, see Black and Trickey (2009).

Social and cultural factors

Although the family is undoubtedly the part of the child’s environment that has most effect on his development, wider social influences are also important, particularly as a cause of conduct disorder. In the early years of childhood, these social factors act indirectly through their influence on the patterns of family life. As the child grows older and spends more time outside the family, they have a direct effect as well. These factors have been studied by examining the associations between psychiatric disorder and type of neighbourhood and school.

Effects of neighbourhood

Rates of childhood psychiatric disorder are higher in areas of social disadvantage. For example, as already noted above (see p. 631), the rates of both emotional and conduct disorder were found to be higher in a poor inner-London borough than on the Isle of Wight. The important features of inner-city life may include lack of play space, inadequate social amenities for older children and teenagers, overcrowded living conditions, and lack of community involvement.

Effects of school

It has long been known that rates of child psychiatric referral and delinquency vary between schools. These differences persist when allowance is made for differences in the neighbourhoods in which the children live. It seems that children are less likely to develop psychiatric problems in a school in which teachers praise, encourage, and give responsibility to their pupils, set high standards, and organize their teaching well. Factors that do not seem to affect rates of psychiatric disorder include the size of the school and the age of its buildings.

Bullying is one of the stressful events that children may encounter at school. Goodman and Scott (2005, p. 243) have defined bullying as the repeated and deliberate use of physical or psychological means to hurt another child, without adequate provocation and in the knowledge that the victim is unlikely to retaliate effectively. It seems that around 2–8% of children are bullied once or more a week. They may be divided into a passive group, who are insecure and anxious and withdraw when attacked, and a provocative group, who are themselves aggressive either directly or by getting others into trouble. The bullies are more often boys than girls; boys are more likely to be physically aggressive, whereas girls are more likely to exclude their victims or campaign against them. Children who were bullied around the age of 5 years were found to exhibit a greater risk of emotional problems and disruptive behaviours at 2-year follow-up (Arseneault et al., 2006). Bullying in older children has important associations with self-harm, violent behaviour, and psychotic symptoms (Arseneault et al., 2010).

Psychiatric assessment of children and their families

The aims of assessment are to obtain a clear account of the presenting problem, to find out how this problem is related to the child’s past development and present life in its psychological and social context, and to plan treatment for the child and the family.

The psychiatric assessment of children differs in several ways from that of adults:

• A more flexible approach. With children it is often difficult to follow a set routine, so a flexible approach to interviewing is required, although it is still important that information and observations are recorded systematically.

• Interviewing of family members. Both parents or other carers should be asked to attend the assessment interview, and it is often helpful to have other siblings present.

• Information from schools. Time can be saved by asking permission to obtain information from teachers before the child attends the clinic. This information should be concerned with the child’s behaviour in school and their educational attainments.

Child psychiatrists vary in their methods of assessment. All agree that it is important to see the family together at some stage to observe how they interact. Some psychiatrists do this before seeing the patient alone, while others do it afterwards. It is generally better to see adolescent patients on their own before seeing the parents. With younger children the main informants are usually the parents, but children over the age of 6 years should usually be seen on their own at some stage. In the special case of suspected child abuse, the interview with the child is particularly important. Whatever the problem, the parents should be made to feel that the interview is supportive and does not undermine their confidence.

Interviewing the parents

Parents are likely to be anxious, and some fear that they may be blamed for their child’s problems. Time should be taken to put them at ease and explain the purpose of the interview. They should then be encouraged to talk spontaneously about the problems before systematic questions are asked. The methods of interviewing are similar to those used in adult psychiatry (see Chapter 3). The items to be included in the history are listed in the Appendix to this chapter. As in adult psychiatry, an experienced interviewer will keep the complete list in mind, while focusing on those items that are relevant to the particular case. It is important to obtain specific examples of general problems, to elicit factual information, and to assess feelings and attitudes.

Interviewing and observing the child

Because younger children may not be able or willing to express their ideas and feelings in words, observations of their behaviour and interaction with the interviewer are especially important. With very young children, drawing and the use of toys may be helpful. With older children, it may be possible to follow a procedure similar to that used with adults, provided that care is taken to use words and concepts appropriate to the child’s age and background. Standardized methods of observation and interviewing have been developed mainly for research purposes.

Starting the interview. It is essential to begin by establishing a friendly atmosphere and winning the child’s confidence, and asking what they like to be called. It is usually appropriate to begin with a discussion of neutral topics such as pets, favourite games, or birthdays, before turning to the presenting problem.

Techniques of interviewing. When a friendly relationship has been established, the child can be asked about the problem, their likes and dislikes, and their hopes for the future. It is often informative to ask what they would request if they were given three wishes. Younger children may be given the opportunity to express their concerns and feelings in paintings or play. Children can generally recall events accurately, although not always in the correct sequence. They are more suggestible than adults, and when asked a question are used to trying to give the answer that the adult has in mind—as they might, for example, in school. Therefore it is particularly important not to use leading questions when interviewing, and not to suggest actions or interpretations to a child who is being observed while painting or at play.

Behaviour and mental state. The items to be noted are listed in the Appendix to this chapter (see p. 680). Children who are brought to see a psychiatrist may appear silent and withdrawn at the first meeting; this behaviour should not be misinterpreted as evidence of depression.

Developmental assessments. By the end of the interview an assessment should have been made of the child’s stage of development relative to other children of the same age.

For further information and advice about interviewing and communicating with children, see Bostic and Martin (2009).

Interviewing the family

A family interview can contribute to the assessment of the interactions between family members, but it is not a good way to obtain factual information. The latter is generally more effectively obtained in interviews with the parents or other family members on their own. Of the various aspects of family interaction, the psychiatrist will usually be most interested in discord and disorganization, which are the features most closely associated with the development of psychiatric disorder. Patterns of communication between family members are also important.

It is usually best to see the family at the first assessment or soon after this, before the interviewer has formed a close relationship with the young patient, or with one of the parents, as this may make it more difficult to interview the other family members.

The interviewer could begin by asking ‘Who would be the best person to tell me about the problems?’ If one family member monopolizes the interview, another member should be asked to comment on what has been said. A useful question to stimulate discussion is ‘How do you think that your partner (daughter, son) would see the problem?’ The interviewer can then ask the partner (daughter, son) how they in fact see the problem. As an alternative, family members who are present can be asked what they think an absent member would think about the issues.

While observing the family’s ways of responding to these and other questions, the interviewer should consider the following:

• Who is the spokesperson for the family?

• Who seems most worried about the problem?

• What are the alliances within the family?

• What is the hierarchy in the family? For example, who is most dominant?

• How well do the family members communicate with one another?

• How do they seem to deal with conflict?

Psychological assessment

Measures of intelligence and educational achievement are often valuable. If mental development and achievement are inconsistent with chronological or mental age, or with the expectations of parents or teachers, this may indicate a generalized or specific disorder of development or may indicate a source of stress in disorders of other kinds. Some of the more commonly used procedures are listed in Table 22.3. For further information, see Charman et al. (2008).

Other information

The most important additional informants are the child’s teachers. They can describe classroom behaviour, educational achievements, and relationships with other children. They may also make useful comments about the child’s family and home circumstances. It is often helpful for a member of the psychiatric team to visit the home. This visit can provide useful information about material circumstances in the home, the relationship of family members, and the pattern of their life together.

Table 22.3 Notes on some psychological measures in use with children


Physical examination

If the child has not been examined recently by the general practitioner, an appropriate physical examination may be needed to complete the assessment. What is appropriate depends on the nature of the problem, but it will often be concerned with evidence of conditions that might affect the brain. Therefore the first step is to observe the child’s appearance, coordination, and gait, at rest and during play. A basic physical examination may follow, with emphasis on the nervous system. If abnormalities are found or suspected, the opinion of a paediatrician or paediatric neurologist may be needed.

Ending the assessment

At the end of the assessment the psychiatrist should explain to the parents—and to the child, in terms appropriate to their age—the result of the assessment and the plan of management. He should explain how he proposes to inform and work with the general practitioner, and seek permission to contact other people involved with the child, such as teachers or social workers. Throughout, the psychiatrist should encourage questions and discussion.


As in adult psychiatry, a formulation is helpful when summarizing key issues. This has the same format as that used with adult patients (see p. 64). It starts with a brief statement of the current problem. The diagnosis and differential diagnosis are discussed next. Aetiology is then considered, with attention to predisposing, precipitating, and perpetuating factors. The developmental stages of the child should be noted, as well as any particular strengths and achievements. An assessment of the problems and the strengths of the family is also recorded. Any further assessments should be specified, a treatment plandrawn up, and the expected outcome recorded.

Court reports

Psychiatrists may be asked to prepare court reports in relation to children. These reports are usually undertaken by specialists in child psychiatry; therefore only an outline will be given here. If general psychiatrists are required to prepare such a report, they should ensure that they are thoroughly aware of the relevant legislation, ask advice from a colleague with relevant special experience, and read a more detailed account of the requirements for court reports.

Courts concerned with children obtain evidence from several sources, including social workers, probation officers, community nurses, psychologists, and psychiatrists. In their reports, psychiatrists should focus on matters within their expertise, including:

• the child’s age, stage of development, and temperament, and the relevance of these to the case

• whether the child has a psychiatric disorder

• the child’s own wishes about their future, considered in relation to their age and understanding

• the parenting skills of the carers, how far they can meet the child’s needs, and other relevant aspects of the family.

While focusing on these matters, the psychiatrist should also be prepared to provide information about the following issues:

• the child’s physical, emotional, and educational needs

• the likely effect on the child of any possible change of circumstances (e.g. removal from home, or living with one or other parent after a divorce)

• any harm that the child has suffered or is likely to suffer.

The wishes of the child should be considered in relation to their age and ability to understand the present situation and possible future arrangements, and to relevant factors in the present situation—for example, some abused children maintain strong attachments to the abuser.

Parenting skills are judged partly on the basis of the history and reports of other people. They are also judged partly on direct observations of the interactions between the parents and the child, including the parents’ attachment to the child, their sensitivity to cues from the child, and their ability to meet the child’s needs.

The report is similar in structure to a court report for an adult, and is presented under the headings shown in Box 22.2.

Box 22.2 Topics to be addressed in a court report about a child

• The qualifications of the writer

• Who commissioned the report, and what questions were asked

• What written information was available, and who was interviewed

• A summary of the findings from the interview (it is not necessary to repeat information contained in social enquiry reports)

• The writer’s interpretation of the information from the interviews and written material

• In the light of the findings, comments on the options before the court, remembering always that it is for the court and not the psychiatrist to determine which option is selected

Thompson and Pearce (2009) suggest the use of the following principles when evidence is obtained from a child:

• Allow the child to talk freely, asking as few questions as possible.

• Obtain the evidence as soon as possible after the event, and whenever possible before any counselling has taken place.

• Try to complete the account on the first occasion, as subsequent accounts are likely to be less accurate.

• Be aware that the greater the pressure to remember, the less accurate the account is likely to be.

Children as witnesses

A child’s evidence is important in cases of suspected abuse, and occasionally in other kinds of case, such as an assault on the mother. When seeking such evidence, the interviewer should be aware of certain points about children’s memory and the factors that can influence their accounts.

Memory varies with age, as do the verbal skills required to describe what has been remembered. Children below the age of 3 years seldom have the cognitive and other capacities to produce an account that can be presented in a court of law. Children above the age of 3 years can produce detailed and accurate memories, although they may not be able to describe them clearly without some prompting.

Most children above the age of 6 years can use straightforward grammar and syntax adequately, but their vocabulary is limited and they may be confused by complicated questions. Also, children of this age may expect that an adult who asks a question already knows the right answer, as this is how they learn (e.g. ‘How many flowers did you see in the picture?’) Consequently, they may agree with leading questions asked by an adult or, when they cannot remember, make up an answer in the hope that it will be the one that is wanted by the questioner. A further problem is that young children do not have an accurate sense of the sequencing and timing of events. Finally, the events that children are asked to recall when they are questioned as witnesses are usually frightening and experienced in a state of emotional arousal. Memories of such events are often incomplete, although the recalled fragments may be detailed and vivid.

Psychiatric treatment for children and their families

This section is concerned with the general features of treatment for children. Aspects of treatment that are specific to individual disorders are discussed later in the sections concerned with those disorders.

The role of the primary care team

General practitioners and other members of the primary care team spend much of their time in advising parents about children, but they refer only a small proportion of these children to a child psychiatrist or a paediatrician. General practitioners are more likely to refer children with the following to a paediatrician:

• developmental difficulties

• physical symptoms with a probable psychological cause

• psychological complications of physical illness.

They are more likely to refer emotional and conduct disorders to a child psychiatrist. Many of the disorders that are referred are no more severe than those which the general practitioner manages, because the decision to refer depends in part on the wishes of the parents or the situation of the family.

The psychiatric team

Way of working. Although psychiatrists, psychologists, nurses, and social workers have special skills, when working in a child psychiatry team they do not confine themselves to their traditional professional role. Instead, they adopt whatever role is most likely to be helpful in the particular case, as they do when working in a community team caring for adults.

The team usually adopts a family-centred approach, and it liaises closely with other professionals involved with the child or the family. These professionals include paediatricians, members of child health and social services, teachers, and educational psychologists. Since many childhood problems are evident at school, or lead to educational difficulties, the child’s teachers usually need to be involved in some way. They may need advice about the best way to manage disturbed behaviour in the classroom, changes that may be needed in the child’s school timetable, or remedial teaching that may be required. Occasionally a change of school is indicated.

In the following sections, brief accounts are given of the main kinds of treatment. In the second part of the chapter, information is given about the management of specific disorders. Additional information about treatment in child psychiatry can be obtained from one of the textbooks listed at the end of this chapter.


Drugs have a limited place in child psychiatry. The main indications are in the treatment of epilepsy, psychosis, some depressive disorders, obsessional disorders, overactivity syndromes, and Gilles de la Tourette syndrome, and in some children with autism. In all cases, dosages should be checked carefully in the manufacturer’s literature or a standard reference book, making sure that the dose is correct for the child’s age and body weight.

Psychological treatment

Psychological aspects of management

Whatever the plan of management, children benefit from a warm, secure, and accepting relationship with the therapist. The security of this relationship helps the child to express their feelings and to find alternative ways of behaving. For younger children, play with toys can help to establish the relationship and provide a medium through which they can express their problems and feelings more effectively than they can with words.

At first, children often perceive the therapist as an agent of the parents, and expect the therapist to share their attitudes. Children should feel accepted in their own right, and not criticized. However, they should not be allowed to feel that anything they do will be approved. It is often advisable to delay discussion of the presenting problems until the child’s confidence has been gained by talking about neutral things that interest them.

Two forms of more elaborate psychological treatment are in general use—cognitive–behaviour therapy and psychodynamic psychotherapy.

Cognitive–behaviour therapy

Behavioural methods have several applications in child psychiatry. They can be used to encourage new behaviour by positive reinforcement and modelling. This is often done by first rewarding behaviour that approximates to the desired behaviour (shaping), and then giving reinforcement in a more discriminating way. For example, with autistic children, shaping has been used for behavioural problems such as temper tantrums and refusal to go to bed. Punishment is not used in shaping behaviour because its effects are temporary and it is ethically unacceptable. Instead, efforts are made to identify and remove any factors in the child’s environment that are reinforcing unwanted behaviour. It is often found that undesired behaviour is being reinforced unwittingly by extra attention given to the child when the behaviour occurs. If the child is ignored at such times and attended to when their behaviour is more normal, beneficial changes often take place. More specific forms of behaviour therapy are available for enuresis (see p. 668) and phobias (see p. 581). Social skills training in a group or in individual sessions may be used for children who have difficulty in relationships with other children and adults. The methods generally resemble those used with adults (see Chapter 20).

Cognitive therapy is useful for older children who can describe and learn to control the ways of thinking that give rise to symptoms and problem behaviours. Older children and adolescents with anxiety disorders, depressive disorders, and eating disorders can be treated with the techniques originally devised for adults. Special techniques have been developed for children with aggressive behaviour (see p. 658).

For a review of the efficacy of cognitive–behaviour therapy in childhood and adolescent anxiety and depression, see Graham (2009).

Psychodynamic therapies

Most psychodynamic approaches for children are based on the methods of Anna Freud and Melanie Klein. Reviews have generally concluded that, although many outcome studies have methodological flaws, the better investigations indicate that dynamic psychotherapy is rather more effective than no treatment. However, psychodynamic therapy has not been shown to produce better results than those of counselling or cognitive–behaviour therapy. Consequently, long-term, analytically based psychodynamic therapies are now used infrequently in the UK, although group therapy and creative therapies (e.g. play therapy and art therapy) continue to be provided for selected children with severe and enduring difficulties. For a more positive assessment of the value of psychodynamic therapies for children, see Fonagy and Target (2009).

Parent training

Parent training is used to improve the skills of parents with deficient parenting skills, including those who abuse or neglect their children and those with low intelligence. It is also used to assist parents of children with behaviour problems that require special parenting skills—for example, the parents of children with conduct disorder or hyperactivity. Most research has been with parents of children who are oppositional or defiant or have conduct disorders. Parent training can be carried out with an educational approach, in which skills of general importance are taught, or with a behavioural approach, in which the specific problems of the particular parent and child are analysed and corrected. In the behavioural approach, use may be made of video-recorded vignettes showing desirable and undesirable parental responses to children’s behaviour. These responses are discussed with the parents of a particular child, or with a group of parents. Whatever approach is adopted, it is important to take account of the stage of development of the child, and of the changing needs of children of different ages.

Studies of the behavioural training of parents have now firmly established the effectiveness of this approach in improving parenting skills and parent–child relationships, and reducing antisocial behaviour in children. Research is now focusing on which families can benefit, and the long-term effects. For a review of parent training programmes, see Scott (2008).

Family therapy

This is a specific form of treatment to be distinguished from the general family approach to treatment described above. In family therapy, the symptoms of the child or adolescent are viewed as an expression of malfunctioning of the family, which is the primary focus of treatment. Several approaches have been used, based on behavioural or psychoanalytical, interpersonal or structural theories. These kinds of therapy are described on pp. 596–7. In practice, most therapists adopt an eclectic approach.

The indications for family therapy are still debated, but there is general agreement that it can be used appropriately when:

• the child’s symptoms are judged to be part of a disturbance of the whole family

• individual therapy is not proving effective

• family difficulties have arisen during another kind of treatment.

The contraindications to family therapy are that the parents’ relationship is breaking up, or the child’s problems do not seem to be closely related to family function. It is important that a therapist’s interest in family therapy should not prevent a thorough evaluation of the case and the use of other treatments when indicated.

The best evidence for effectiveness is for structural and behavioural forms of family therapy, where benefits have been shown in adolescents with behaviour problems and in young people with anorexia nervosa (Eisler and Lask, 2008).

Social work

Social workers play an important role in the care of children with psychiatric disorders, and of their families. In the UK they have statutory duties with regard to the protection of children who are at risk within the family, and who require special care or special supervision. They help parents to improve their skills in caring for their children, and to solve problems with finances or accommodation. Social workers carry out family assessments and family therapy, and may also provide individual counselling for the child and members of the family.

Occupational therapy

Occupational therapists can play a valuable part in assessment of the child’s development, in psychological treatment, and in devising measures to improve parent–child interaction. They work both in day-patient and inpatient units, and in the community. They work closely with teachers in both assessing and providing therapeutic activities for children.

Special education

Children who are attending outpatient clinics, and the smaller number who are day patients or inpatients, often benefit from additional educational arrangements. Special teaching may be needed to restore confidence, and to remedy backwardness in writing, reading, and arithmetic, which is common among children with conduct disorders as well as those with specific developmental disorders. For further information about special education and the issue of integration with ‘mainstream’ teaching, see Howlin (2008).

Substitute care

Residential care

Residential care can be valuable for children with symptoms that result from or are maintained by a severely unstable home environment or extreme parental rejection. Children who are considered for residential placement often have conduct disorders and severe educational problems. Removal of a child from home should be considered only after every practical effort has been made to improve the circumstances of the family. Residential care may be provided in a foster home, a children’s home (in which a group of about 10 children live in circumstances as close as possible to those of a large family), or a residential school.

Residential care, other than fostering, is seldom arranged for children under 5 years of age, because they have a special need for attachment to parental figures. In general, children who have been in residential care have high rates of psychosocial problems in later childhood and in adult life. As men they most often have problems with the law, while as women they most often have problems with unwanted pregnancy and single parenting problems. It is not clear how far these problems relate to the experience of residential care, or to previous adverse experiences that led to the residential placement. Reports of the abuse of children placed within care are reminders of the need to ensure good training and supervision of the staff of children’s homes and residential schools.

For a review of residential care, see Rushton and Minnis (2008).


Foster care may be of three kinds:

• short-term emergency care—for example, when a care-giver is ill or when the parents of an autistic child need respite

• medium-term care, which may be followed by a return home—for example, if the caregiver is receiving treatment for problems that led to neglect or abuse of the child

• long-term care, in which the child remains until they are able to live independently.

Children in long-term foster care have more problems than children who have been adopted, but it is difficult to determine to what extent these problems are related to experiences before fostering, and to what extent they are due to the lesser security of fostering as opposed to adoption. Problems seem to be greater when the fostered child is older and when children in the fostering family are of the same age as the fostered child. Children in foster care usually retain some contact with their biological parents, but it is not helpful to the child to have sporadic and distressing contacts. For a review of fostering, see Rushton and Minnis (2008).

Inpatient and day-patient care

Child psychiatric inpatient units require easy access to paediatric advice, adequate space for play, easy access to schooling, and an informal design that still allows close observation. There should be some provision for mothers to stay with their children.

Admission for inpatient treatment is usually arranged for any of three reasons:

• severity: when the disorder is too severe to treat in any other way—for example, extreme hyperactivity, severe pervasive developmental disorder, life-threatening anorexia nervosa, or school refusal that is resistant to outpatient treatment

• observation: when the diagnosis is uncertain

• separation: inpatient treatment can provide a necessary period away from a severely disturbing home environment—for example, when there is child abuse or gross overprotection.

Sometimes the mother is admitted as well as the child, thus helping to maintain the bonds between the two. This arrangement also allows close observation of the ways in which the mother responds to the child (e.g. in cases of child abuse). In addition, it provides an opportunity for the mother to learn new parenting skills by taking an increasing part in the child’s care while both of them remain in hospital.

Day-hospital treatment for children provides many of the advantages of inpatient care without removing them from home. Unless there is any danger that the child may be abused, remaining at home has the advantage that relationships with other family members are maintained. Day care can relieve the family from some of the stressful effects of managing an overactive or autistic child. For a review of inpatient and day-patient care, see Green and Worrall-Davies (2008).

Intensive home-based care and outreach services

Increasingly, intensive community-based support is provided for children and adolescents with severe problems. Such support requires a well-staffed and experienced team of professionals with an ‘assertive-outreach’ approach (see p. 617). Although such services are valuable, they do not remove the need for inpatient provision. For a review of outreach services for children, see Green and Worall Davies (2008).

Review of syndromes

The review of syndromes begins with the problems encountered in preschool children. Specific and pervasive developmental disorders are described next. An account is then given of the main psychiatric disorders of childhood, in the order in which they appear in the major systems of classification. Other psychiatric disorders of childhood are described next, before a brief account of the disorders of adolescence (which are generally similar to those of either childhood or adulthood). The chapter ends with a discussion of child abuse.

Problems of preschool children and their families

It has already been noted that in the preschool years children are learning several kinds of social behaviour. They are acquiring sphincter control. They are learning how to behave at mealtimes, to go to bed at an appropriate time, and to control angry feelings. They are also becoming less dependent. All of these behaviours are learned within the family. The psychiatric problems of preschool children centre around these behaviours, and they often reflect factors in the family as well as factors in the child. Many psychological problems at this age are brief, and can be thought of as delays in normal development. Most are treated by general practitioners and paediatricians. The more serious problems may be referred to child psychiatrists.


In a much quoted study, Richman et al. (1982) studied 705 families with a 3-year-old child in a London borough. The most frequent abnormalities of behaviour in these children were bed-wetting at least three times a week (present in 37%), wetting by day at least once a week (17%), overactivity (14%), soiling at least once a week (13%), difficulty in settling at night (13%), fears (13%), disobedience (11%), attention seeking (10%), and temper tantrums (5%).

Whether these behaviours are reported as problems depends on the attitudes of the parents as well as on the nature, severity, and frequency of the behaviour. Richman et al. overcame this difficulty by making their own ratings of the extent of problems. They based this assessment on the effects on the child’s well-being and the consequences for the other members of the family. They used common-sense criteria to decide whether the problems were mild, moderate, or severe. In total, 7% of 3-year-olds in their survey had behaviour problems of marked severity and 15% had mild behaviour problems. The behaviours most often rated as problems were temper tantrums, attention seeking, and disobedience.

Subsequent epidemiological surveys employing more modern diagnostic criteria have also found significant rates of psychological problems in preschool children. Pooling a number of epidemiological surveys, Egger and Angold (2006) concluded that about 15% of preschool children had a psychiatric problem. The most common difficulties were emotional disorders, anxiety disorders, and oppositional disorders. The latter included temper tantrums and disobedience. About 50% of the children had more than one psychological problem.

Some common problems of preschool children

Temper tantrums and disruptive behaviours

Occasional temper tantrums are normal in toddlers, and only persistent or very severe tantrums are abnormal. The immediate cause is often unwitting reinforcement by excessive attention and inconsistent discipline on the part of the parents. When this arises it is often because the parents have emotional problems of their own, or because the relationship between them is unsatisfactory.

Temper tantrums usually respond to kind but firm and consistent setting of limits. In treatment it is first necessary to discover why the parents have been unable to set limits in this way. They should be helped with any problems of their own and advised how to respond to the tantrums.

Sleep problems

The most common sleep difficulty is wakefulness at night, which is most frequent between the ages of 1 and 4 years. About 20% of children of this age take at least an hour to get to sleep or are wakeful for long periods during the night. When wakefulness is an isolated problem and not very distressing to the family, it is enough to reassure the parents that it is likely to improve. If sleep disturbances are severe or persistent, two possible causes should be considered. First, the problems may have been made worse by physical illness or an emotional disorder. Secondly, they may have been exacerbated by the parents’ excessive concern and inability to reassure the child.

If no medical or psychiatric disorder is detected, the reasons for the parents’ concerns should be sought and dealt with as far as possible. Some parents overstimulate their child in the evening, or unwittingly reinforce crying in the night by taking the child into their own bed. A behavioural approach to these problems is generally helpful. Other sleep problems such as nightmares and night terrors are common among healthy toddlers, but they seldom persist for long. They are discussed on p. 364. For a review of sleep problems, see Stores (2009a,b).

Feeding problems

Minor food fads and food refusal are common in preschool children, but do not usually last long. In a minority, however, the behaviour is severe or persistent, although not accompanied by signs of poor nourishment. When this happens it is commonly because the parents, who are often over-attentive and perfectionistic, are unwittingly reinforcing the child’s behaviour.

Treatment is directed to the parents’ management of the problem. They should be encouraged to ignore the feeding problem and refrain from offering the child special foods or otherwise attempting to do anything unusual to persuade him or her to eat. Instead, the child should be offered a normal meal and left to decide whether to eat it or not.


Pica is the eating of items that are generally regarded as inedible, such as soil, paint, or paper. It is often associated with other behaviour problems. Cases should be investigated carefully because some are due to brain damage, autism, or mental retardation. Others are associated with emotional distress, which should be reduced if possible. Otherwise, treatment consists of common-sense precautions to keep the abnormal items of diet away from the child. Pica usually diminishes as the child grows older.

Reactive attachment disorders of infancy and early childhood

As explained above, children’s attachments to their parents vary in their security, and they may vary between caregivers—for example, being insecure with one parent but secure with the other. Attachment disorders are more extreme variations from the norm, and do not correspond exactly to any of the types of insecure attachment described earlier (see p. 634). They are pervasive, affecting all relationships. They start before the age of 5 years and are associated with grossly abnormal caregiving. In DSM-IV, the diagnosis requires that the disturbance of relationships appears to be a direct result of abnormal caregiving. ICD-10 does not use this criterion, but requires that the behaviour is present in several situations. In addition, clinicians are cautioned about making the diagnosis unless maltreatment has occurred. Two clinical subtypes are usually recognized:

• Disinhibited. These children seek comfort but do so indiscriminately, seeking it as much from strangers as from caregivers. As infants, they relate to these people with clinging behaviour, and in early childhood with attention seeking. Such behaviour has been described most clearly in children raised in institutions, or experiencing repeated changes in foster care.

• Inhibited. This type is less clearly defined. These children show a combination of behavioural inhibition, vigilance, and fearfulness, which is sometimes called frozen watchfulness. They show a mixture of approach and avoidance behaviours. They are miserable and difficult to console, and sometimes aggressive. Some fail to thrive. This behaviour is seen among children who have been abused and those raised in institutions.

Aetiology. It seems that these syndromes are related more to the type of caregiving (abusive or institutional) than to the child.

Prognosis. Insecure attachment in infancy is often followed by conflicts with caregivers and behaviour problems later in childhood. It seems that improvement can occur if the child experiences a secure attachment to a caregiver—for example, as a result of fostering or adoption. However, some institutionalized children continue to show significant attachment disturbances despite apparently adequate foster parenting. For a review of attachment disorders, see Zeanah and Smyke (2008).

Assessment and treatment of the problems of preschool children

Assessment. Usually the information is largely obtained from the parents. The assessor seeks to discover whether the problem is primarily in the child or related to difficulties in the mother or the entire family. The problem behaviour is assessed, together with the child’s general level of development, and the functioning of the family.

Treatment. Apart from the particular points already mentioned under the specific disorders, treatment includes advice for the parents (and if necessary for other family members) about relevant aspects of child rearing. There is little systematic evidence about the value of treatment, although parenting interventions are the best established approaches, and have some benefit even where the underlying condition has a strong genetic basis—for example, in some forms of hyperactivity (Gardner and Shaw, 2008). It may be helpful to arrange for the child to spend part of the day away from the family in a playgroup or nursery school, provided that the care offered is of high quality. Specially adapted cognitive approaches can be helpful for emotional and anxiety disorders. For further information about the assessment and treatment of preschool children, see Gardner and Shaw (2008).


As explained above, many psychological problems of preschool children are brief. However, Richman et al. (1982) found that certain problems detected in 3-year-old children were still present at the age of 8 years. These problems included overactivity, difficulty in controlling the child, speech difficulty, and autism. About 50% of children who show high rates of disruptive behaviour at age 3–4 years continue to show this behaviour at school age (Gardner and Shaw, 2008).

Specific developmental disorders

Both DSM-IV and ICD-10 contain categories for specific developmental disorders, which are circumscribed developmental delays that are not attributable to another disorder or to lack of opportunity to learn (see Table 22.4). It is debatable whether these conditions should be classified as mental disorders at all, since many children who meet the criteria have no other signs of psychopathology. In ICD-10, these developmental disorders are divided into specific developmental disorders of scholastic skills, speech and language, and motor function. In DSM-IV, the same disorders are called learning disorder, communication disorders, and motor skill disorder, respectively.

Specific developmental disorders of scholastic skills are divided further into specific reading disorder, specific spelling disorder, and specific arithmetic disorder. In DSM-IV, these conditions are called reading disorder, disorder of written expression, and mathematics disorder, respectively.

Specific reading disorder (dyslexia)

In DSM-IV, this condition is called reading disorder. It is defined by a reading age well below (usually defined as 2 standard deviations below) the level expected from the child’s age and IQ. In a much quoted study, the disorder was found in about 4% of 10- to 11-year-olds on the Isle of Wight, and about twice that percentage in London (Yule and Rutter, 1985), and subsequent investigations have confirmed these estimates.

Clinical features

Specific reading disorder is to be distinguished from general backwardness in scholastic achievement resulting from low intelligence, lack of opportunity to learn at home or at school, or poor visual acuity. The child has serious delay in learning to read, evident from the early years of schooling, and sometimes preceded by delayed acquisition of speech and language. Writing and spelling are impaired, and in older children these problems may be more obvious than the reading problems. Errors in reading and spelling include omissions, substitutions, or distortions of words, slow reading, long hesitations, and reversals of words or letters. There may also be poor comprehension. There may be associated emotional problems, but development in other areas is not affected. Compared with children with general backwardness at school, those with specific reading retardation are much more often boys. They are also more likely to have minor neurological abnormalities, and are likely to come from socially disadvantaged homes.

Table 22.4 Specific disorders of psychological development


Specific reading retardation is associated with conduct disorder more often than would be expected through chance. The association may arise in part because the two conditions have common neurodevelopmental or temperamental origins, in part because reading retardation leads to conduct problems at school when the child is frustrated by failures, and in part because conduct disorder gives rise to problems in learning to read (Remschmidt and Schulte-Körne, 2009).


Reading is a complex skill which depends on more than one psychological process and is learned in several stages. It is not surprising, therefore, that no single cause has been identified for specific reading disorder. A widely held theory of the learning of reading is that children first use visual methods; they learn the appearance of whole words, and cannot decipher new words. The next stage of learning is alphabetical; children become able to decode new words from the sounds associated with the letters. In the final stage, reading becomes automatic and flexible in combining visual and alphabetical methods. (This model of reading, although not accepted by all, is useful in clinical practice.)

Genetic causes. The frequent occurrence of reading disorder in family members suggests a genetic cause, and the family patterning suggests that there is not a single mode of inheritance. Several quantitative trait loci have been identified by linkage analysis, and many of the implicated genes play a role in neuronal migration, supporting a role for abnormal neurodevelopment in aetiology (Scerri and Schulte-Körne, 2010).

Neurological causes. Children with cerebral palsy and epilepsy have increased rates of specific reading disorder. It has been suggested that children who have specific reading disorder, but no obvious neurological disease, may have minor and less obvious neurological abnormalities. Another suggestion is that there is a disorder of brain maturation affecting one or more of the skills required in reading. This explanation is consistent with the following findings: difficulty in visual scanning, confusion between right and left, and general improvement with age.

Social factors. It seems likely that difficulties that would be insufficient in themselves to cause reading retardation sufficient to be diagnosed as specific reading disorder, may do so when the child is brought up in an illiterate or otherwise disadvantaged family, receives little attention at school, or changes school frequently. For a review of aetiology, see Remschmidt and Schulte-Körne (2009).

Assessment and treatment

It is important to identify the disorder early. Assessment is carried out by an educational or clinical psychologist using an individually administered standardized test of reading accuracy and comprehension. Treatment is educational unless there are additional medical or behavioural problems requiring separate intervention. Treatment should be started as early as possible, before the child has a sense of failure. Several educational approaches are used, but it is most important to reawaken the interest of a child with a long experience of failure. Parental interest and continued extra teaching seem to be helpful, but there is no evidence that any one method of teaching is better than others. If there are behavioural problems secondary to frustration caused by the reading difficulty, they may lessen as reading improves; others may need separate attention.


The prognosis varies with the severity of the condition. Among children with a mild problem in mid-childhood, only about 25% achieve normal reading skills by adolescence. Very few with severe problems in mid-childhood overcome them by adolescence. Although there is insufficient evidence to be certain what happens to these people as adults, those with substantial difficulties in adolescence seem likely to retain them. Specific reading difficulties are common in adults with antisocial behaviour. For a review of reading disorders, see Snowling and Hulme (2008).

Mathematics disorder (specific arithmetic disorder)

The first of these terms is used in DSM-IV; the term in parentheses is used in ICD-10. Difficulty with arithmetic is probably the second most common specific disorder. Problems include failure to understand simple mathematical concepts, failure to recognize numerical symbols or mathematical signs, difficulty in carrying out arithmetic manipulations, and inability to learn mathematical tables. These problems are not due simply to lack of opportunities to learn, and are evident from the time of the child’s first attempts to learn mathematics. Although it causes less severe handicap in everyday life than reading difficulties, mathematics disorder can lead to secondary emotional difficulties when the child is at school.

Epidemiology. The incidence reported in several studies lies between 3% and 6%, and the disorder frequently co-occurs with specific reading difficulties (Remschmidt and Schulte-Körne, 2009).

Aetiology. This is uncertain, but cognitive deficits involving working memory have been implicated. Dyscalculia occurs in some adults with parietal lobe lesions, and structural imaging investigations have suggested that individuals with mathematics disorder have a reduced volume of parietal structures (Rotzer et al., 2008). Twin studies have also implicated genetic factors in mathematics disorder.

Assessment. This is usually based on the arithmetic subtests of the Wechsler Intelligence Scale for Children (WISC) and the Wechsler Adult Intelligence Scale (WAIS), and on specific tests.

Treatment. This involves remedial teaching similar to the approach used for specific reading difficulties, but it is not known whether it is effective. Without treatment the condition appears to persist over several years of follow-up. For a review, see Snowling and Hulme (2008).

Communication disorders (developmental disorders of speech and language)

Children vary widely in their achievement of speech and language. Half of all children use words with meanings by about 12 months of age, and 97% do so by 21 months. Half of all children form words into simple sentences by 23 months. Vocabulary and complexity of language develop rapidly during the preschool years. However, when children start school, 1% are seriously retarded in speech and 5% have difficulty in making themselves understood by strangers. The process by which language is acquired is complex, and is still not fully understood. Language disorders are associated with psychiatric problems because both may be an expression of a common brain abnormality, and also because language disorder impairs social interaction and education.

Causes of speech and language disorder

No cause can be found in the majority of children with speech and language disorders. These cases are said to have specific developmental speech and language disorder. It is most important to detect the primary conditions that are present in the minority. The most common of these causes is learning disability. Other important causes are deafness, cerebral palsy, and autism and autistic spectrum disordersSocial deprivation can cause mild delays in speaking or add to the effects of the other causes.


The classification differs in some ways between ICD-10 and DSM-IV. ICD-10 uses the title ‘specific developmental disorders of speech and language’, whereas DSM-IV has the wider title ‘communication disorders.’ Three disorders appear in both classifications, although with some differences in nomenclature:

• phonological disorder (in DSM-IV) or specific speech articulation disorder (in ICD-10)

• specific developmental expressive language disorder (this term is used in both classifications)

• mixed receptive–expressive disorder (in DSM-IV); here ICD-10 has the narrower term ‘specific developmental receptive language disorder. ‘(The reason for the difference is explained below.)

ICD-10 (but not DSM-IV) has a fourth category of acquired aphasia with epilepsy. In DSM-IV, the wider title of the group allows the inclusion of stuttering; the narrower title in ICD-10 does not include stuttering, which is coded instead under behavioural disorders of childhood (see Table 22.4).

Phonological disorder (specific developmental speech articulation disorder)

In this condition, accuracy in the use of speech sound is below the level appropriate for the child’s mental age, but language skills are normal. Errors in making speech sounds are normal in children up to the age of about 4 years, but by the age of 7 years most speech sounds should be normal. By age 12 years nearly all speech sounds should be made normally. Children with specific speech articulation disorder make errors of articulation so severe that it is difficult for others to understand their speech. Speech sounds may be omitted or distorted, or other sounds substituted. When assessing speech production, appropriate allowance should be made for regional accents and dialects. The sounds most often affected are those that develop later in the normal sequence of development (l, r, s, z, th, and ch for English speakers).

Prevalence. This depends on the criteria used to determine when speech production is abnormal; a rate of 2–3% has been cited among 6- to 7-year-olds (Remschmidt and Schulte-Körne, 2009).

Aetiology and treatment. Because the disorder tends to run in families, a genetic influence has been assumed. Speech therapy can be helpful if the articulation disorder problem is an isolated problem without, for example, associated hearing loss or cognitive impairment.

Specific developmental expressive language disorder

In this disorder, the ability to use expressive spoken language is markedly below the level appropriate for the child’s mental age. Language comprehension is within normal limits, but there may also be abnormalities in articulation. Language development varies considerably among normal children, but the absence of single words by 2 years of age, and of two-word phrases by 3 years of age signifies abnormality. Signs at later ages include restricted vocabulary, difficulties in selecting appropriate words, and immature grammatical usage. Non-verbal communication, if impaired, is not affected as severely as spoken language, and the child makes efforts to communicate. Disorders of behaviour are often present.

Cluttering. Some children speak rapidly and with an erratic rhythm such that the grouping of words does not reflect the grammatical structure of their speech. This abnormality, which is known as cluttering, is classified as an associated feature of expressive language disorder in DSM-IV, but in ICD-10 it is classified (with stammering) among other behavioural disorders of childhood.

Prevalence. The prevalence of expressive language disorder depends on the method of assessment; a rate of 3–5% of children has been proposed (Remschmidt and Schulte-Körne, 2009).

Prognosis. It is reported that about 50% of the children who meet the DSM-IV criteria develop normal speech by adulthood, while the rest have long-lasting difficulties. The prognosis is worse when the language disorder is severe, and when there is a comorbid condition, such as conduct disorder.

Treatment. This mainly involves special education. Psychiatrists may be involved when there is a comorbid disorder, and may need to advise the parents about the child’s rights for special education. For a review of expressive language disorder, see Bishop and Norbury (2008).

Receptive–expressive (or receptive) developmental language disorder

In this disorder the understanding of language is below the level appropriate to the child’s mental age. In almost all cases, expressive language is also disturbed (a fact recognized in DSM-IV by the term receptive–expressive language disorder).

The development of receptive language ability varies considerably among normal children. However, failure to respond to familiar names, in the absence of non-verbal cues, by the beginning of the second year of age, or failure to respond to simple instructions by the end of the second year, are significant signs suggesting receptive language disorder—provided that deafness, learning disability, and pervasive developmental disorder have been excluded. Associated social and behavioural problems are particularly frequent in this form of language disorder.

Prevalence. This depends on the criteria for diagnosis, but a frequency of up to 3% of school-age children has been suggested (Remschmidt and Schulte-Körne, 2009).

Prognosis and treatment. The prognosis is poor, with only about 50% of these children having normal conversational skills in adulthood. The prognosis is worse when the language disorder is severe, or when there is a comorbid condition, such as conduct disorder.

Treatment. This involves special education. The psychiatrist’s role is the same as in expressive language disorder (see above). For a review of receptive language disorder, see Bishop and Norbury (2008).

Acquired epileptic aphasia (Landau–Kleffner syndrome)

In this rare disorder, a child whose language has so far developed normally loses both receptive and expressive language but retains general intelligence. There are associated EEG abnormalities, nearly always bilateral and temporal, and often with more widespread disturbances. Most of the affected children develop seizures either before or after the change in expressive language. The disorder generally starts between 3 and 9 years of age, usually over a period of several months but sometimes more rapidly.

The aetiology is unknown. The prognosis is variable; about two-thirds of children are left with a receptive language deficit, but the other third recover completely. Treatment of the seizures does not always lead to improvement in language.

Assessment of speech and language disorders

Early investigation is essential both to determine the nature and severity of the speech and language disorder and to exclude mental retardation, deafness, cerebral palsy, and pervasive developmental disorder. The speech-producing organs should be examined. It is particularly important to detect deafness at an early stage.

Parents can give some indication of the child’s speech and language skills, especially if they complete a standardized inventory. With younger children it may be necessary to rely on this information, but children from the age of about 3 years can be tested by a standard test of language appropriate to the child’s age. If possible, such a test should be carried out by a speech therapist or a psychologist specializing in the subject.


Treatment depends partly on the cause, but usually includes a programme of speech training carried out through play and social interaction. In milder cases this help is best provided at home by the parents, who are given information on what to do. More severe difficulties are likely to require specialized help in a remedial class or a special school. Treatment should start early.

For a review of the development of speech and language disorders, see Bishop and Norbury (2008).

Motor skills disorder

Some children have delayed motor development, which results in clumsiness in school work or play. In ICD-10, this condition is called specific developmental disorder of motor function. It is also known as clumsy child syndrome or specific motor dyspraxia. The child can carry out all normal movements, but their coordination is poor. They are late in developing skills such as dressing, walking, and feeding. They tend to break things and are poor at handicrafts and organized games. They may also have difficulty in writing, drawing, and copying. IQ testing usually shows good verbal but poor performance scores.

These children are sometimes referred to a psychiatrist because of secondary emotional disorder. An explanation of the nature of the problem should be given to the child, the family, and the teachers. Special teaching may improve confidence. It may be necessary to exempt the child from organized games or other school activities that involve motor coordination. There is usually some improvement with time. For further information, see Snowling and Hulme (2008).

Pervasive developmental disorders

The term pervasive developmental disorder refers to a group of disorders characterized by abnormalities in communication and social interaction and by restricted repetitive activities and interests. These abnormalities occur in a wide range of situations. Usually development is abnormal from infancy, and most cases are manifested before the age of 5 years.

Six conditions are included under this rubric in ICD-10 (see Table 22.5), two of which do not appear in DSM-IV, namely atypical autism and overactive disorder with mental retardation and stereotyped movements. The incidence of all forms of pervasive developmental disorder is currently thought to be about 60–70 per 10 000, and the frequency has increased over recent decades. Most of the increase appears to be due to changes in diagnostic criteria and improved awareness of the condition. Diagnostically, the most frequent forms are pervasive developmental disorders that do not meet the full criteria for a specific syndrome (Fombonne, 2009) (see Box 22.3).

Autistic spectrum disorder (ASD). This term is often used to denote the whole range of pervasive developmental disorders, including those that do not precisely meet the formal diagnostic criteria. It recognizes that the boundaries between the different syndromes are difficult to decide, and that it can be more helpful to parents to think of a continuum of disability.

Childhood autism (autistic disorder)

This condition was described by Kanner (1943), who suggested the name ‘infantile autism.’ The term childhood autism is used in ICD-10, but autistic disorder is the term used in DSM-IV.

Table 22.5 Pervasive developmental disorders


Clinical features

In his original description, Kanner (1943) identified the main features, which are still used to make the diagnosis. In both DSM-IV and ICD-10, three kinds of abnormality are required to make the diagnosis of autism:

• abnormalities of social development

• abnormalities of communication

• restriction of interests and behaviour.

Box 22.3 Epidemiology of autism spectrum disorders


Of these, the abnormalities of social development are the most specific to autism. The clinical picture is variable both between individuals, and in the same individual at different ages. Abnormal development is usually apparent before the age of 3 years. There are reports that early signs of autism can be detected in infancy—for example, absence of babbling and pointing by 12 months, lack of imitation and oversensitivity to sounds, and dislike of change (for further details, see van Engeland and Buitelaar, 2008).

Abnormalities of social development. The child is unable to form warm emotional relationships with people (autistic aloneness). Autistic children do not respond to their parents’ affectionate behaviour by smiling or cuddling. Instead, they appear to dislike being picked up or kissed. They are no more responsive to their parents than to strangers, and do not show interest in other children. There is little difference in their behaviour towards people and inanimate objects. A characteristic sign is gaze avoidance—that is, the absence of eye-to-eye contact.

Abnormalities of communication. Speech may develop late or never appear. Occasionally, it develops normally until about the age of 2 years and then disappears in part or completely. This lack of speech is a manifestation of a severe cognitive defect. As autistic children grow up, about 50% acquire some useful speech, although serious impairments usually remain, such as the misuse of pronouns and the inappropriate repeating of words spoken by other people (echolalia). Some autistic children are talkative, but their speech is a repetitive monologue rather than a conversation with another person.

The cognitive defect also affects non-verbal communication and play. Autistic children do not take part in the imitative games of the first year of life, and later they do not use toys in an appropriate way. They show little imagination or creative play.

Restriction of interests and behaviourObsessive desire for sameness is a term applied to the autistic child’s stereotyped behaviour, and their distress if there is a change in the environment. For example, some autistic children insist on eating the same food repeatedly, on wearing the same clothes, or on engaging in repetitive games. Some are fascinated by spinning toys. Odd behaviour and mannerisms are common. Some autistic children carry out odd motor behaviours such as whirling round and round, twiddling their fingers repeatedly, flapping their hands, or rocking. Others do not differ obviously in motor behaviour from normal children.

Other features. Autistic children may suddenly show anger or fear without apparent reason. They may be over-active and distractible, sleep badly, or soil or wet themselves. Some injure themselves deliberately. About 25% of autistic children develop seizures, usually about the time of adolescence.

Intelligence level. Kanner originally believed that the intelligence of autistic children was normal, but later research has shown that this is true only of a minority. Severe to profound intellectual impairment is found in about 40%, of autistic children, while a further 30% show mild to moderate impairment. Some autistic children show areas of ability despite impairment of other intellectual functions, and in some cases they have exceptional but restricted powers of memory or mathematical skill (van Engeland and Buitelaar, 2008).


The prevalence of autism is probably about 20 per 10 000 children. It is four times as common in boys as in girls (Fombonne, 2009).


The cause of childhood autism is unknown, but the social, language, and behavioural problems in autism suggest abnormalities in not one but several neural systems.

Genetic influences are of great importance. The condition is 50 times more frequent in the siblings of affected individuals than in the general population. Several twin studies have shown a much higher concordance between monozygotic than between dizygotic twins, with a calculated heritability of 80–90%. Mild social and language problems are more frequent among the siblings of autistic probands than in the general population, suggesting that the phenotype may be wider than the syndrome of autism as currently defined. Linkage investigations in autism have identified several replicated susceptibility loci on chromosomes 2, 7, and 17, while other analyses have implicated synaptic genes such as NRXN1 and NLGN3. Copy number variations (see p. 101), including 15q11–13 duplications, are also more frequent in people with autism. The findings suggest that the clinical syndrome of autism has a number of different genetic aetiologies (Kumar and Christian, 2009).

Brain abnormalities. Structural brain studies have established that autistic children have larger brain volumes than normal children, although these apparently normalize during adolescence. Regional abnormalities in grey and white matter have also been frequently reported, but the nature of these has been somewhat inconsistent. Functional MRI (fMRI) scanning has provided evidence for aberrant functional connectivity in the neural circuitry that supports social cognition and theory of mind (Stigler et al., 2011).

Other evidence for biological causes. Autism is associated with fragile X syndrome (see p. 695), Rett’s syndrome, and tuberose sclerosis, but only in a small proportion of cases. About 10% of autistic children have a concomitant medical condition of some kind (Volkmar and Klin, 2009). Claims of an association with MMR vaccine have not been confirmed in careful epidemiological studies.

Abnormal parenting. In a much quoted paper, Kanner (1943) suggested that autism was a response to abnormal parents, who were characterized as cold, detached, and obsessive. Kanner’s idea has not been substantiated. It is now thought that any psychological abnormalities in the parents are likely to be either a response to the problems of bringing up the autistic child, or (more likely) a manifestation in the parents of the genes that have produced autism in the child. Although unsupported by evidence, persistent lay beliefs about the role of parenting still cause distress among the parents who hear of it.

Relationship to schizophrenia. Clinically, the distinction between schizophrenia and autism in terms of symptomatology is not difficult, and the two have usually been regarded as separate disorders. However, there is growing evidence for a degree of overlap in genetic mechanisms and perhaps in abnormalities in functional neural connectivity (King and Lord, 2010). Whether future neurobiological findings will establish a pathophysiological relationship between the two disorders remains to be seen.

For a review of the aetiology of autism, see van Engeland and Buitelaar (2008).


Theory of mind in autism. This theory attempts to identify a basic psychological disorder in autism. By the age of 4 years, normal children are able to form an idea of what others are thinking. As an example, consider a normal child who watches while another normal child is first shown the location of a hidden object and then sent out of the room while the object is moved to a new hiding place. The child who has remained in the room will conclude that the child who left temporarily will expect the object to be in the original position when he returns to the room. An autistic child tends to lack this appreciation of what another child is likely to be thinking. In the example, an autistic child is likely to say that the child who left the room will think that the object has been moved to its new place. It is not certain how specific to autism is this difficulty in appreciating what others know and expect, nor how central it is to the psychopathology. In any case, its cause is not known.

Other possible ‘core’ psychological disorders in autism. These include impairment of frontal lobe executive functions involved in planning and organization, resulting in perseveration and poor self-regulation, and impaired ability to extract high-level meaning from diverse sources of information. None of these proposed core psychopathologies can account for more than a part of the clinical picture of autism. For a review of cognitive phenotypes in autism, see Charman et al. (2010).

Differential diagnosis

It is more usual to encounter partial syndromes than the full syndrome of childhood autism. These partial syndromes must be distinguished from the following:

• deafness, which can be excluded by appropriate tests of hearing

• communication disorder (see p. 648), which differs from autism in that the child usually responds normally to people and has good non-verbal communication

• learning disability, in which responses to other people are more normal than those of an autistic child. Also, an autistic child has more impairment of language relative to other skills than is found in a learning-disabled child of the same age

• Asperger’s syndrome (see p. 653)

• childhood disintegrative disorder (see p. 653).


Around 10–20% of children with childhood autism begin to improve between the ages of about 4 and 6 years, and are eventually able to attend an ordinary school and obtain work. A further 10–20% can live at home, but cannot work and need to attend a special school or training centre and remain very dependent on their families and/ or support services. The remainder, at least 60%, improve little and are unable to lead an independent life; many need long-term residential care. Those who improve may continue to show language problems, emotional coldness, and odd behaviour. As noted already, a substantial minority develop epilepsy in adolescence. Factors associated with a better prognosis are communicative speech by the age of about 6 years, and higher IQ, although outcome is highly variable even within the normal IQ range (Volkmar and Klin, 2009).


Assessment should be concerned with more than the diagnosis of autism. The following additional factors need to be considered:

• cognitive level

• language ability

• communication skills, social skills, and play

• repetitive or otherwise abnormal behaviour

• stage of social development in relation to age, mental age, and stage of language development

• associated medical conditions (e.g. comorbid epilepsy)

• psychosocial factors, including the needs of the family.

For a review of the diagnosis and assessment of autism, see van Engeland and Buitelaar (2008).


In the absence of any specific treatment, management has three main aspects—management of the abnormal behaviour, education and social services, and help for the family.

Management of abnormal behaviourContingency management (see p. 591) may control some of the abnormal behaviour of autistic children. Such treatment is often provided at home by the parents, instructed and supervised by a clinical psychologist. It is not known whether this treatment has any lasting benefit, but in autism even temporary changes are often worthwhile for the patient and the family.

Education and social services. Most autistic children require special schooling. It is generally thought better for them to live at home and to attend special day schools. If the condition is so severe that the child cannot stay in the family, residential schooling is necessary. Special care is needed to avoid an institutional atmosphere, as this can increase social withdrawal. In some cases, the educational and residential needs of autistic children can be provided through the services for the learning disabled. Older adolescents may need vocational training.

For a review of psychological and educational treatments, see van Engeland and Buitelaar (2008).

Help for the family. The family of an autistic child need considerable help to cope with the child’s behaviour, which is often difficult to understand and distressing. They need prompt assessment of their child’s needs and easy access to appropriate educational and other provisions. Although doctors may be able to do little specifically to help the patient, they must not withdraw from the family, who need continuing support as well as support for their efforts to help the child themselves, and obtain help from educational and social services. Some parents request genetic counseling, and it seems that the risk of a further autistic child is about 3%. Many parents find it helpful to join a voluntary organization in which they can meet other parents of autistic children and discuss common problems.

Other suggested treatments. Individual psychotherapy has been used in the hope of effecting more fundamental changes, but there is no evidence that it succeeds in doing so. There is evidence of short-term (up to 6 months) benefit for treatment with the antipsychotic drug, risperidone, in terms of significant reductions in irritability, aggression, temper tantrums, and self-injurious behaviours. Treatment with SSRIs may also have a place in lowering compulsive and repetitive behaviours as well as stereotypies and rituals. For a general review of treatment, see van Engeland and Buitelaar (2008).

Rett’s disorder

Rett’s disorder (or Rett’s syndrome) is a rare X-linked condition which occurs almost exclusively in girls. The reported prevalence is about 1 per 10 000 girls. After a period of normal development in the first months of life, head growth slows and over the next 2 years there is arrest of cognitive development and loss of purposive skilled hand movements. Stereotyped movements develop with hand-clapping and hand-wringing movements. Ataxia of the legs and trunk may develop. Interest in the social environment diminishes in the first few years of the disorder, but may increase again later. Expressive and receptive language development is severely impaired and there is psychomotor retardation. Some patients develop severe learning disability. The disorder is associated with sporadic mutations in the MeCP2 gene, located on the X chromosome; aberrant imprinting (p. 693) of the gene may also occur. MeCP2 regulates the expression of genes involved in brain development. For a review, see Volkmar and Klin (2009).

Childhood disintegrative disorder

Childhood disintegrative disorder (also known as Heller’s disease) is a rare condition which begins after a period of normal development usually lasting for more than 2 years. It is unclear how far the childhood disintegrative disorder is distinct from childhood autism. It resembles childhood autism in the marked loss of cognitive functions, abnormalities of social behaviour and communication, and unfavourable outcome. It differs from childhood autism in the loss of motor skills and of bowel or bladder control. The condition may arrest after a time, or progress to a severe neurological condition.

Asperger’s syndrome

This condition was first described by Asperger (1944), and his original paper, in German, has been translated into English by Frith (1991) The condition has also been referred to as autistic psychopathy. The condition is more common in boys than in girls. These children develop normally at first, but by the third year begin to lack warmth in their relationships. They go on to show marked abnormalities of social behaviour similar to those of childhood autism. There may be stereotyped and repetitive activities, often in the form of intense but narrow interests rather than motor mannerisms. The condition differs from autism in that there is no general delay or retardation of cognitive development or language, although speech may be stilted and intonation unusual. Conversation may consist more of repeated monologues than of social interchange. These children are eccentric, solitary, and may spend much time pursuing narrow interests. Many of these individuals are clumsy. They are more interested in others than are children with autism, but they do not share interests or pleasures with others, and are without friends.

Epidemiology. The prevalence of the syndrome is uncertain, and reported rates depend on the definition used in the research. Fombonne (2009) estimated that it is about a third as common as autism, with a prevalence of about 6 in 10 000; however, the confidence limits of this estimate are wide, Whatever the true frequency, the condition is increasingly recognized as important and these children are recognized as requiring educational and other help.

Aetiology. The cause of Asperger’s syndrome is unknown. It is uncertain whether the condition is a milder variant of childhood autism or a separate disorder. There appears to be heritability of social difficulties in families.

Prognosis. The abnormalities usually persist into adulthood. Most people with the disorder can work, but few form successful relationships and marry. Recognition and diagnosis of the disorder have become increasingly common in adult psychiatric practice, where it has been linked with various kinds of psychiatric comorbidity, such as depression and obsessive–compulsive disorder as well as a possible increased risk of criminal behaviour (Dein and Woodbury-Smith, 2010).

For a review of Asperger’s syndrome, see Volkmar and Klin (2009).

Atypical autism, and pervasive developmental disorder not otherwise specified

The terms atypical autism (in ICD-10) and pervasive developmental disorder not otherwise specified (NOS) (in DSM-IV) denote a residual category for pervasive developmental disorders that resemble autism but do not meet the diagnostic criteria for any of the syndromes within this group. The prevalence of these cases varies according to the criteria adopted, but most investigations show that they are more common than autism itself (Fombonne, 2009). The relationship between these cases and those which meet the criteria for the other syndromes within the group of pervasive developmental disorders is poorly understood.

Hyperkinetic disorder

About one-third of children are described by their parents as overactive, and 5–20% of school children are so described by teachers. These reports encompass a continuum of behaviour ranging from normal high spirits to a severe and persistent disorder. This overactivity often varies in different situations. Hyperkinetic disorders are severe forms of overactivity associated with marked inattention—hence the widely used term attention-deficit hyperactivity disorder (ADHD) adopted in DSM-IV. In ICD-10 the term is hyperkinetic disorder, and because of slight differences in the criteria for diagnosis (see below), cases diagnosed as hyperkinetic disorder are rather more severe than those that meet the criteria for attention-deficit hyperactivity disorder.

Clinical features

The cardinal features of this disorder are as follows:

• extreme and persistent restlessness

• sustained and prolonged motor activity

• difficulty in maintaining attention

• impulsiveness and difficulty in withholding responses.

These features are pervasive, occurring across situations, although they can vary somewhat in different circumstances, so that parents and teachers may give rather different accounts of the child’s behaviour.

Children with the disorder are often reckless, and prone to accidents. They may have learning difficulties, which result in part from poor attention and lack of persistence with tasks. Many develop minor forms of antisocial behaviour as the condition continues, particularly disobedience, temper tantrums, and aggression. These children are often socially disinhibited and unpopular with other children. Mood fluctuates, but low self-esteem and depressive mood are common.

Restlessness, overactivity, and related symptoms often start before school age. Sometimes the child was overactive as a baby, but more often significant problems begin when the child starts to walk; they are constantly on the move, interfering with objects and exhausting their parents.

Diagnostic criteria

In both ICD-10 and DSM-IV the cardinal features for the diagnosis of the disorder are impaired attention, hyper-activity, and impulsiveness starting in childhood and lasting for at least 6 months to a degree that is maladaptive and inconsistent with the developmental level of the child. However, the two systems differ in the details of the criteria for diagnosis.

• ICD-10 requires that the symptoms started before 6 years of age, whereas DSM-IV specifies that they started before 7 years of age.

• ICD-10 requires both hyperactivity and impaired attention, whereas DSM-IV requires either inattention, or hyperactivity with impulsiveness.

• ICD-10 requires that the criteria are met both at home and at school, whereas DSM-IV requires only that they be present in one situation, with impairment (which does not have strict criteria) in the other.

The result of these differences is that children who meet the ICD-10 criteria are more severely impaired than those who meet the DSM-IV criteria.

In ICD-10 the disorder can be further classified as:

1. disturbance of activity and attention

2. hyperkinetic conduct disorder.

The latter term is used when criteria for both hyperkinetic disorder and conduct disorder are met. (The provision is made because the presence of associated aggression, delinquency, or antisocial behaviour is associated with a poorer outcome; see below.)

Comorbidity. About 50% of children with the disorder meet the diagnostic criteria for other conditions as well, principally conduct disorder, depressive disorder, or anxiety disorder. Learning disability and language impairment are also commonly present. Autistic children are often hyperactive and inattentive, but these features are regarded as part of the syndrome of childhood autism; hyperactivity disorder is not diagnosed in addition to autism.


Estimates of the prevalence of hyperkinetic disorder vary according to the criteria for diagnosis. Using DSM-IV criteria, a prevalence of about 5% is proposed among the school-age population; however, ICD-10 diagnosis yields lower rates, at around 1–2% (Taylor, 2009). Rates of the disorder in epidemiological studies do not seem to have increased over the last two decades, although the frequency of treatment with stimulants has done. This suggests increased medical recognition and a lower threshold for using pharmacological treatment, rather than a true increase in incidence. Rates are about three times higher in boys than in girls. The disorder is more frequent in areas of social deprivation and among children raised in institutions.


The aetiology is uncertain, although much of the evidence suggests a disorder of higher cognitive executive function, related to abnormalities of neurotransmission in the prefrontal cortex and associated subcortical structures, and with genetic as well as environmental origins.

Neurological findings. Signs suggesting neurodevelopmental impairment or delay are found in children with hyperkinetic disorder—for example, clumsiness, language delay, and abnormalities of speech. Although these signs are generally associated with birth complications, they could result from factors acting at an earlier stage of development of the brain. The disorder can also occur as a consequence of severe traumatic brain injury.

Neuroimaging studies. These show functional and structural abnormalities in areas involved in self-organization, including the prefrontal and striatal regions, as well as the cerebellum. There is also evidence for white matter disruption and disordered anatomical and functional connectivity between these various brain regions (Emond et al., 2009; Konrad and Eickhoff, 2010), Genetic studies. Investigations of family members, twins and adopted children, and monozygotic and dizygotic twins all suggest that genetic factors are important, with heritability estimates of about 80%. Compared with controls, probands with attention-deficit hyperactivity disorder have more first-degree relatives with the same disorder. Also there is a much higher concordance for monozygotic than for dizygotic pairs. Adoption studies show that the biological parents of children with attention-deficit hyperactivity disorder are more likely to have had the same or a related disorder than are the adoptive parents.

Linkage and association studies have implicated genes involved in the dopaminergic transmitter system, including the dopamine D4 receptor and the dopamine transporter. There may also be a role for the SNAP-25 gene. These findings are consistent with evidence that drugs which affect the dopamine system lead to improvement in the disorder (see below). However, the effect size of these genes in contributing towards the disorder is individually quite small (in the range 1.1–1.5), and it seems likely that hyperactivity is a genetically heterogeneous condition. For a review of the genetic causes, see Faraone and Mick (2010).

Social factors. Social factors increase the child’s innate tendency to hyperactivity. Thus, as noted above, overactive behaviour is more frequent among young children living in poor social conditions and institutions. Also, studies of twins indicate that both shared and non-shared environmental effects contribute to aetiology (Taylor, 2009).

Other suggested causes. In the past, lead intoxication has been suggested as a cause of hyperkinetic syndrome, and more recently zinc deficiency has been suggested, but there is no convincing evidence that either of these is a common cause. Some parents report that certain food additives increase their child’s overactivity, but there is no evidence that this is a general cause (for a review, see Taylor, 2009).


Overactivity usually lessens gradually as the child grows older, especially when it is mild and not present in every situation. It usually ceases by puberty. However, inattention and impulsivity may persist, and about 50% of the cases diagnosed in childhood retain the full diagnosis in adolescence. The prognosis for any associated learning difficulties is less good, whilst antisocial behaviour has the worst outcome. When the overactivity is severe, and is accompanied by learning failure or low intelligence, the prognosis is poor and the condition may persist into adulthood, as antisocial disorder and drug misuse rather than as continued hyperactivity.

By adulthood most patients will no longer meet the full criteria for attention-deficit hyperactivity disorder or hyperkinetic syndrome, but the majority will retain some functional impairment. Although this would suggest that the incidence of the disorder should be low in adulthood, surveys have indicated rates of about 4%, suggesting the presence of additional cases that were not detected during childhood. Adults with attention-deficit hyperactivity disorder may describe procrastination, poor motivation, and mood lability, while irritability, inattention, and poor organization lead to problems with both work and social relationships. Comorbid mood disorders and substance misuse are common. For a review of the assessment and treatment of adults with attention-deficit hyperactivity disorder, see Janakiraman and Benning (2010).


Support and psychological treatment. A hyperactive child exhausts their parents, who need support from the start of treatment, particularly as it may be difficult to reduce the child’s behaviour. Teachers also need support, and special education may be needed. Psychosocial intervention is usually combined with stimulant medication except in the milder cases. Family, group, and behaviour therapy are all used, but are less generally effective than medication in treatment trials. Behavioural treatment is directed towards any conduct problems, often by modifying the response of the carers or teachers to these problems. Modified contingency management may be helpful.

MedicationStimulant drugs should be tried if there is severe restlessness and attention deficit, and where parent training and psychological approaches have not proved effective. These drugs increase dopamine and noradrenaline activity, and it is thought that these actions underlie their therapeutic effects. The most commonly prescribed medication is methylphenidate. Dexamphetamine is also used, and longer-acting preparations are now available. The dosage should be related to body weight, following the manufacturers’ instructions and advice about contraindications.

The potential short-term benefits of stimulant treatment are decreased restlessness, reduced aggressiveness and, sometimes, improved attention. These effects do not usually diminish with time, but it has not been shown that they are associated with better long-term outcome. The side-effects include irritability, depression, insomnia, and poor appetite. High doses may provoke stereotyped behaviour, which disappears when the dose is reduced. With high doses, there may be some slowing of growth, but adult stature and weight do not seem to be affected. Tics may be made worse. The drug may be needed for many months, and some children take it for years; careful monitoring is essential. Some parents are understandably reluctant to agree to long-term drug treatment for their children. In such cases it can be helpful to give the drugs for a trial period so that the benefits and disadvantages can be assessed for a particular child. The drug may be stopped from time to time in an attempt to minimize side-effects and to confirm that medication is still needed.

The noradrenaline reuptake inhibitor, atomoxetine, has also been licensed for the treatment of hyperkinetic syndrome. Common adverse effects of atomoxetine include nausea, abdominal pain, loss of appetite, and sleep disturbance. Rarely (about 1 in 50 000 patients treated), severe liver damage can occur. Although atomoxetine has the theoretical advantage of lacking psychostimulant properties, experience with its use is necessarily limited compared with that for methylphenidate. It may be more suitable in patients with comorbid tic disorders, which are often worsened by stimulant treatment. It can also be used in those who have not responded to stimulant treatment, or who are poorly tolerant of it.

In clinical trials, short-term benefits of stimulants have been shown in about two-thirds of children with hyperkinetic syndrome, but the long-term benefits are uncertain. It therefore seems best to reserve drug treatment for more severe cases that have not responded to other treatment. Surprisingly, there is no report of children treated in this way becoming addicted to the medication. For guidance on the treatment of attention-deficit hyperactivity disorder in both children and adults, see the National Institute for Health and Clinical Excellence (2008).

Conduct disorders

Conduct disorders are characterized by severe and persistent antisocial behaviour. They form the largest single group of psychiatric disorders in older children and adolescents.

Clinical features

The essential feature of conduct disorder is persistent abnormal conduct which is more serious than ordinary childhood mischief. The abnormal behaviours centre around defiance, aggression, and antisocial acts. In the pre-school period, the disorder usually manifests as defiant and aggressive behaviour in the home, often with over-activity. The behaviours include disobedience, temper tantrums, physical aggression towards siblings or adults, and destructiveness. In later childhood, conduct disorder is manifested in the home as stealing, lying, and disobedience, together with verbal or physical aggression. Later, the disturbance often becomes evident outside as well as inside the home, especially at school, or as truanting, delinquency, vandalism, and reckless behaviour, or as alcohol or drug abuse. Antisocial behaviour among teenage girls includes spitefulness, emotional bullying of peers, and running away.

In children older than 7 years, persistent stealing is abnormal. Below that age, children seldom have a real appreciation of other people’s property. Many children steal occasionally, so that minor or isolated instances need not be taken seriously. A small proportion of children with conduct disorder present with sexual behaviour that incurs the disapproval of adults. In younger children, masturbation and sexual curiosity may be frequent and obtrusive. Frequent unprotected sex and pregnancy may be a particular problem in adolescent girls. Fire-setting is rare, but obviously dangerous (see p. 720).

To constitute conduct disorders, these behaviours have to be more persistent than a reaction to changing circumstances, such as adjusting to the arrival in the family of a new step-parent. There is no sharp dividing line between conduct disorder and ordinary bad behaviour; instead there is a continuum on which diagnostic criteria define a cut-off point. The cut-off defines the most severe cases that have the worst outcome and are most in need of help. Much of this help is social or educational, but psychiatrists have an important role in identifying comorbid disorders and arranging multidisciplinary care.


Both ICD-10 and DSM-IV require the presence of three symptoms from a list of 15, and a duration of at least 6 months. The criteria are closely similar in the two systems of classification.

Because conduct disorders vary widely in their clinical features, both systems divide conduct disorders. In DSM-IV they are divided into childhood-onset type (with onset before 10 years of age) and adolescent-onset type (with onset at 10 years of age or later). DSM-IV has an additional category, ‘oppositional defiant disorder’, for persistently hostile, defiant, provocative, and disruptive behaviour outside the normal range, but without aggressive or dyssocial behaviour. This disorder occurs mainly in children below 10 years of age. ICD-10 has four subdivisions of conduct disorder—socialized conduct disorder, unsocialized conduct disorder, conduct disorders confined to the family context, and oppositional defiant disorder.


The prevalence of conduct disorders is difficult to estimate because the dividing line between them and normal rebelliousness is imprecise. Approximate rates were established many years ago, and have not been revised much by more recent evidence. Rutter et al. (1970a) found the prevalence of ‘antisocial disorder’ to be about 4% among 10- to 11-year-olds on the Isle of Wight. In a subsequent study in London, about twice this rate was found (Rutter et al., 1975a, 1976a). Moffitt and Scott (2008) concluded that in Western industrial societies, the prevalence of significant and persistent oppositional, disruptive, or aggressive behaviour in children and adolescents was 5–10%. Conduct disorders are about three times more common in boys than in girls.


Environmental factors. These are important. Conduct disorders are commonly found in children from unstable, insecure, and rejecting families living in deprived areas. Antisocial behaviour is frequent among children from broken homes, those from homes in which family relationships are poor, and those who have been in residential care in their early childhood. Conduct disorders are also related to adverse factors in the wider social environment of the neighbourhood and school (Scott, 2009b).

Genetic factors. Conduct disorder clusters in families. There is a highly heritable trait of liability to externalizing disorders in general (conduct disorder, oppositional defiant disorder, and attention-deficit hyperactivity disorder), but shared environmental factors are also important in the aetiology of conduct disorder. Nevertheless, shared environmental influences on antisocial behaviors seem to decline with age such that persistent antisocial behaviour in late adolescence and adulthood is influenced mainly by genetic factors (Bornovalova et al., 2010). There is evidence that a variant of the monoamine oxidase-A genepredisposes to conduct disorder, but only when combined with adverse factors in the child’s environment (Moffitt and Scott, 2008). If genetic factors are involved, it is not known whether they exert their effect by influencing temperament, or in some other way.

Organic factors. Children with brain damage and epilepsy are more prone to conduct disorder, as they are to other psychiatric disorders.

Other associations. An important finding in the Isle of Wight survey was a strong association between antisocial behaviour and specific reading disorder. It is not known whether antisocial behaviour and reading disorder result from common predisposing factors, or whether one causes the other and, if so, which is primary.

For a review of the aetiology of conduct disorder, see Scott (2009b).


The long-term outcome of conduct disorder varies considerably with the nature and extent of the disorder. In an important study, Robins (1966) found that almost 50% of people who had attended a child guidance clinic for conduct disorder in adolescence showed some form of antisocial behaviour in adulthood. No cases of sociopathic disorder were found in adulthood among those with diagnoses other than conduct disorder in adolescence. Follow-up of conduct-disordered children cared for in children’s homes and of controls led to similar conclusions. About 40% of the conduct-disordered children had antisocial personality disorder in their twenties, and many of the rest had persistent and widespread social difficulties below the threshold for diagnosis of a personality disorder. Where conduct disorders first present in adolescence, the prognosis is better, with about 80% no longer demonstrating significant antisocial behaviour in adulthood (Scott, 2009b) (see Box 22.4).

Box 22.4 Factors that predict poor outcome in children with conduct disorder

In the young person

• Early onset

• Severe, frequent, and varied antisocial behaviours

• Hyperactivity and attention problems

• Low IQ

• Pervasiveness (at home, in school, and elsewhere)

In the family

• Parental criminality and alcoholism

• High hostility/discord focused on the child

• Low income

In the wider environment

• Economically deprived area

• Ineffective schools

From Scott (2009b).


Mild conduct disorders often subside without treatment other than common-sense advice to the parents. For more severe disorders, treatment for the child is often combined with treatment and social support for the family. Any coexisting disorders (e.g. attention-deficit hyperactivity disorder, depression) should be treated. There is no convincing evidence that any treatment affects the overall long-term prognosis. Nevertheless, some short-term benefits can often be achieved, and in some cases adverse family factors can be modified in a way that could improve the prognosis. Some families are difficult to help by any means, especially where there is material deprivation, with chaotic relationships, and poorly educated parents.

Parent training programmes. These programmes use behavioural principles (see p. 573). Parents are taught how the child’s antisocial behaviour may be reinforced unintentionally by their attention to it, and how it may be provoked by interactions with members of the family. Parents are also taught how to reinforce normal behaviour by praise or rewards, and how to set limits on abnormal behaviour—for example, by removing the child’s privileges, such as an hour less time to play a game. As aids to learning, parents are provided with written information and video recordings showing other parents applying behavioural procedures. For a review of parent training programmes, see Scott (2008).

Anger management. Young people who are habitually aggressive have been shown to misperceive hostile intentions in other people who are not in fact hostile. They also tend to underestimate the level of their own aggressive behaviour, and choose inappropriate behaviours rather than more appropriate verbal responses. Anger management programmes seek to correct these ideas by teaching how to inhibit sudden inappropriate responses to angry feelings—for example, the child says to himself ‘Stop! What should I do?’ They also learn how to reappraise the intentions of other people and use socially acceptable forms of self-assertion.

Other methods. Remedial teaching should be arranged if there are associated reading difficulties. Medication is of little value unless there is comorbid disorder which is appropriately treated in this way.

Residential care. Occasionally, residential placement may be necessary in a foster home, group home, or special school. This should be done only for compelling reasons. There is no evidence that institutional care improves the prognosis for conduct disorder.

Juvenile delinquency

A juvenile delinquent is a young person who has been found guilty of an offence that would be categorized as a crime if committed by an adult. In most countries, the term applies only to a young person who has attained the age of criminal responsibility (at present 10 years in the UK, but varying widely in other countries). Thus delinquency is not a psychiatric diagnosis but a legal category. However, juvenile delinquency may be associated with psychiatric disorder, especially conduct disorder. For this reason, it is appropriate to interrupt this review of the syndromes of child psychiatry in order to consider juvenile delinquency.

The majority of adolescent boys, when asked to report their own behaviour, admit to offences against the law, and about 20% are convicted at some time. Most offences are against property. In England in 2009, young people aged between 10 and 17 years committed about 200 000 detected offences of sufficient gravity to result in a court disposal. The largest category (about 42 000 offences) was for theft and handling stolen goods, although violence against the person was a close second (about 38 000 offences). Many fewer girls than boys are delinquent, although the ratio has fallen from 11:1 to 4:1 over the past decades (Youth Justice Board, 2011). About 75% of those with three or more convictions as juveniles go on to offend as adults (Farrington, 2002).

Delinquency is sometimes equated with conduct disorder. This is wrong, for although the two categories overlap, they are not the same. Many delinquents do not have conduct disorder (or any other psychological disorder). Equally, many of those with conduct disorder do not offend. Nevertheless, in an important group, persistent law-breaking is preceded and accompanied by abnormalities of conduct, such as truancy, aggressiveness, and attention seeking, and by poor concentration.


The causes of juvenile delinquency are complex and overlap with the causes of conduct disorder. The causes are reviewed briefly here; for a more detailed account, see Bailey (2009).

Social factors. Delinquency is related to low social class, poverty, poor housing, and poor education. There are marked differences in delinquency rates between adjacent neighbourhoods which differ in these respects. Rates also differ between schools. Many social theories have been put forward to explain the origins of crime, but none offers a completely adequate explanation.

Family factors. Many studies have found that crime runs in families. For example, about 50% of boys with criminal fathers are convicted, compared with 20% of those with fathers who are not criminals. The reasons for this are poorly understood, but they may include poor parenting and shared attitudes to the law (Farrington and Welsh, 2008).

In a much quoted retrospective study, Bowlby (1944) examined the characteristics of ‘juvenile thieves’ and argued that prolonged separation from the mother during childhood was a major cause of their problems. More recent work has not confirmed such a precise link (see p. 144). Although delinquency is particularly common among those who come from broken homes, this seems to be largely because separation often reflects family discord in early and middle childhood. Other family factors that are correlated with delinquency are large family size and child-rearing practices, including erratic discipline and harsh or neglecting care.

Factors in the childGenetic factors appear to be less significant among the causes of delinquency than in the more serious criminal behaviour of adulthood (see p. 710). (The possible role of genetic factors in conduct disorder is considered above.) There are important relationships between delinquency and somewhat below average IQ as well as educational and reading difficulties (Rutter et al., 1976a). There are at least two possible explanations for the latter finding. Temperament or social factors may predispose to both delinquency and reading failure. Alternatively, reading difficulties may result in frustration and loss of self-esteem at school, and these may in turn predispose to antisocial behaviour. Physical abnormalities probably play only a minor role among the causes of delinquency, even though brain damage and epilepsy predispose to conduct disorder.


When the child is seen as part of an ordinary psychiatric referral and the delinquency is accompanied by a psychiatric syndrome, the latter should be assessed in the usual way. Sometimes the child psychiatrist is asked to see a delinquent specifically to prepare a court report. In these circumstances, as well as making enquiries among the parents and teachers, it is essential to consult any social worker or probation officer who has been involved with the child.

Psychological testing of intelligence and educational achievements can be useful. Other factors to be taken into account are listed in Table 22.6. The form of the report is similar to that described in Chapter 24(p. 730). It should include a summary of the history and present mental state together with recommendations about treatment.

Table 22.6 Assessment of young offenders*

The offence

• Nature and seriousness

• Characteristics of victim

• Motive

• Role in the group, if others are involved

• Attitude to the offence and the victim

Other problem behaviours

• Other offences (number, nature, and whether detected/ convicted)

• Violence

• Self-harm

• Cruelty to children or animals

• Fire setting

*Modified from Goodman and Scott (2005), table 7.1.

Violence among adolescents

There is concern about increasing rates of violent offences by adolescents. The causes of violent behaviour among young people are not fully understood, and most violent offenders commit other kinds of offence. In the UK, Bailey and Aulich (1997) studied 50 cases of the most extreme form of violence among juveniles—homicide. Many had pre-existing conduct or emotional disorders, and adverse family and social circumstances. A smaller number had learning difficulties. However, none of the 76 features studied separated these young people from other young offenders. For further information, see Bailey and Scott (2008).


In this section we consider measures intended to reduce the chances of further offending. Many of the children and adolescents who appear before juvenile courts have psychiatric disorders, including conduct disorders, mood disorders, attention-deficit hyperactivity disorder, substance misuse and dependence, learning disability, and epilepsy. These disorders are treated in ways described elsewhere in this chapter.

Psychiatrists who treat delinquent children and adolescents need to understand the legal system in the country in which they work. The legal responses include a warning not to offend again, a fine, the requirement that the parent or guardian take proper control, supervision by a social worker, a period at a special centre, or an order committing the child to the care of the local authority. The exact provisions vary from one country to another. Since delinquent behaviour is common, mainly not serious, and often a temporary phase, it is generally appropriate to treat first offences with minimal intervention coupled with firm disapproval. The same applies to minor offences that are repeated. A more vigorous response is required for more serious, recurrent delinquency. For this purpose a community-based programme is usually preferred, with the main emphasis on improving the family environment, reducing harmful peer group influences, helping the offender to develop better skills for solving problems, and improving educational and vocational accomplishments. In the UK, such a programme has been introduced by the setting up of Young Offender Teams. When this approach fails, custodial care may be considered.

The main aim of the law as it applies to children and adolescents is treatment rather than punishment. There has been extensive criminological research to determine the effectiveness of the measures used. The general conclusions are not encouraging, although not surprising since, as explained above, delinquency is strongly related to factors external to the child, including family disorganization, antisocial behaviour among the parents, and poor living conditions. The risks of reconviction are greater among children who have had any court appearance or period of detention than among children who have committed similar offences without any official action having been taken (Farrington and Welsh, 2008).

Of the many special approaches to the treatment of delinquency, four have been shown to have some benefit.

• Problem-solving skills training. This is one of several related approaches which focus on teaching social skills, anger management, and problem solving. An additional component, attributional retraining, helps to correct cognitive distortions whereby delinquent youths readily perceive threat and hostility even in neutral interpersonal situations.

• Functional family therapy. The therapist works with the family at home for about 3 months with the following aims: to motivate the family; to select a specific problem and find how to change it; and to learn to generalize the problem-solving skills acquired in the second stage.

• Multisystemic therapy. This has six elements—family therapy, helping the young person to find non-delinquent friends, personal development (including assertiveness training), improving family problem-solving skills, liaison with teachers, and coordination of other involved agencies.

• Multidimensional treatment foster care. The young person lives in a foster home, away from delinquent friends, for about 6 months and learns better life skills At the same time, the family are taught the skills needed to respond to the young person more effectively.

Studies of these and other approaches indicate the need to match the type of treatment to the needs of the particular offender. Some seem to respond better to authoritative supervision, others to more permissive counselling. However, it is not yet possible to provide any satisfactory guidelines for matching treatment and offender. See Bailey and Scott (2008) for a review of forensic aspects of child and adolescent psychiatry.

Anxiety disorders

There is no clear dividing line between normal anxiety and anxiety disorders in childhood. In ICD-10, anxiety disorders in childhood are classified as emotional disorders with onset specific to childhood (see Table 22.7). DSM-IV does not contain this category, and with two exceptions classifies childhood anxiety disorders in the same way as anxiety disorders in adult life. The exceptions are separation anxiety disorder and reactive attachment disorder, which are listed under the heading ‘other disorders of infancy, childhood or adolescence.’ ICD-10 has a diagnosis of sibling rivalry disorder. DSM-IV does not have this diagnosis in the main classification, but sibling relationship problems can be coded under ‘other conditions that may be the focus of clinical attention.’ An individual child’s disorder may fulfil the criteria for more than one disorder listed in ICD-10 or DSM-IV—for example, phobic disorder and separation anxiety disorder


The prevalence of anxiety disorders in childhood is uncertain, because epidemiological studies have usually employed the wider category of emotional disorder, or asked about symptoms rather than syndromes of anxiety. In their survey of the Isle of Wight, Rutter et al. (1970a) found a prevalence of emotional disorders of 2.5% in both boys and girls. In a London suburb, the corresponding figure was doubled (Rutter et al., 1975a). In both places, the rate of conduct disorder was about twice that of emotional disorder. More recent surveys of the general population suggest higher rates of anxiety disorders in children, with an overall prevalence of 5–10% (Pine and Klein, 2008).

Anxiety at different ages

Anxiety is common in childhood, but its nature changes as the child grows older. Infants pass through a stage of fear of strangers. During the preschool years, separation anxiety and fears of animals, imaginary creatures, and the dark are common. In early adolescence these fears are replaced by anxiety about social situations and personal adequacy. Anxiety disorders in childhood resemble these normal anxieties and follow the same developmental sequence, although they are more severe and more prolonged. Phobias and separation anxiety disorder usually start in early childhood, and social anxiety disorder and generalized anxiety disorder start in adolescence.

Table 22.7 Anxiety disorders in childhood


Separation anxiety disorder

Separation anxiety disorder is a fear of separation from people to whom the child is attached which is clearly greater than normal separation anxiety of toddlers or preschool children, or persists beyond the usual preschool period, and is associated with significant problems of social functioning. The onset is before the age of 6 years. The diagnosis is not made when there is a generalized disturbance of personality development.

Clinical picture

Children with this disorder are excessively anxious when separated from parents or other attachment figures, and unrealistically concerned that harm may befall these persons or that they will leave the child. They may refuse to sleep away from these persons or, if they agree to separate, may have disturbed sleep with nightmares. They cling to their attachment figures by day, demanding attention. Anxiety is often manifested as physical symptoms of stomach ache, headache, nausea, and vomiting, and may be accompanied by crying, tantrums, or social withdrawal. Separation anxiety disorder is one cause of school refusal (see p. 666).


Community surveys suggest that rates of separation anxiety disorder are about 5% among 7- to 11-year-old boys and girls (Pine, 2009).


Separation anxiety disorder is sometimes precipitated by a frightening experience. This may be brief (e.g. admission to hospital) or prolonged (e.g. conflict between the parents). In some cases separation anxiety disorder develops in children who react with excessive anxiety to a large number of everyday stressors and who are therefore said to have an anxiety-prone temperament. Sometimes the condition appears to be a response to anxious or overprotective parents.


The disorder often improves with time, but may worsen again when there is a change in the child’s routine, such as a move of school. Some cases progress to generalized or other anxiety disorders in adult life.


Account should be taken of the whole range of possible aetiological factors, including stressful events, previous actual separation, an anxiety-prone temperament, and the behaviour of the parents. Stressors should be reduced if possible, and the children should be helped to talk about their worries. It is more important to involve the family, helping them to understand how their own concerns or overprotection affect the child, and to find ways of making the child feel more secure. Anxiolytic drugs may be needed occasionally when anxiety is extremely severe, but they should be used for short periods only. When separation anxiety is worse in particular circumstances, the child may benefit from the behavioural techniques used for phobias, as described in the next section.

Phobic anxiety disorder

This diagnosis for children corresponds to specific phobia for adults (see p. 186). Minor phobic symptoms are common in childhood. They usually concern animals, insects, the dark, school, and death. The prevalence of more severe phobias varies with age. Severe and persistent fears of animals usually begin before the age of 5 years, and nearly all have declined by the early teenage years.


Most improve, but a minority, probably about 10%, persist into adult life.


Most childhood phobias improve without specific treatment provided that the parents adopt a firm and reassuring approach. For phobias that do not improve, simple behavioural treatment can be combined with reassurance and support. The child is encouraged to encounter feared situations in a graded way, as in the treatment of phobias in adult life.

Social anxiety disorder of childhood

This term is used in ICD-10 to describe disorders starting before the age of 6 years in which there is anxiety with strangers greater or more prolonged than the fear of strangers, which normally occurs in the second half of the first year of life. Children with this condition tend to have an inhibited temperament in infancy. These children are markedly anxious in the presence of strangers and avoid them. The fear, which may be mainly of adults or of other children, interferes with social functioning. It is not accompanied by severe anxiety on separation from the parents.

Treatment resembles that of other anxiety disorders of childhood.

Sibling rivalry disorder

This category is listed in ICD-10 for children who show extreme jealousy or other signs of rivalry in relation to a sibling, starting during the months following the birth of that sibling. The signs are clearly greater than the emotional upset and rivalry which is common in such circumstances, and they are persistent and cause social problems. When the disorder is severe there may be hostility and even physical harm to the sibling. The child may regress in behaviour—for example, losing previously learned control of bladder or bowels—or act in a way appropriate for a younger child. There is usually opposition to the parents and behaviour intended to obtain their attention, often with temper tantrums. There may be sleep disturbance and problems at bedtime. In treatment, parents should be helped to divide their attention appropriately between the two children, to set limits for the older child, and to help him or her to feel valued.

Post-traumatic stress disorder

Although not included in ICD-10 among the anxiety disorders with onset usually in childhood, post-traumatic stress disorder (PTSD) can occur in childhood. The clinical picture resembles that of the same disorder in adult life (see p. 159), with disturbed sleep, nightmares, flashbacks, and avoidance of reminders of the traumatic events. Children with post-traumatic stress disorder often have irrational separation anxiety, and young children may show regressive behaviour.


As in adults, the cause is an encounter with exceptionally severe stressors—for example, those encountered by children caught up in war, civil unrest, or natural disasters. Physical and sexual abuse may also provoke post-traumatic stress disorder in children. As with adults, cases after vehicle and other accidents are more common than those after the less frequent major disasters. One study found a rate of about 14% of post-traumatic stress disorder among children involved in road accidents (Keppel-Benson et al., 2002).


The prognosis of the disorder has not been studied systematically in childhood, but severe reactions may last for a year or longer.


Treatment resembles that for adults (see p. 163). As with adults, immediate counselling (debriefing) is often provided for all those involved in a disaster. At least for adults, the value of this procedure is doubtful, unless cognitive therapy procedures are incorporated to help the victim to ‘process’ the memories of the events (see p. 158). It is likely that the same principles, and the same doubts, apply to the treatment of children.

Obsessive–compulsive disorder

Obsessive–compulsive disorders are rare in childhood. However, several related forms of repetitive behaviour are common, particularly between the ages of 4 and 10 years. These repetitive behaviours include preoccupation with numbers and counting, the repeated handling of certain objects, and hoarding. Normal children commonly adopt rituals such as avoiding cracks in the pavement or touching lamp posts. These behaviours cannot be called compulsive because the child does not struggle against them (see p. 12 for the definition of obsessive and compulsive symptoms). The preoccupations and rituals of obsessive–compulsive disorder are more extreme than these behaviours of healthy children, and take up an increasing amount of the child’s time—for example, rechecking schoolwork many times, or frequently repeated hand washing.

Clinical picture

Obsessional disorder rarely appears in full form before late childhood, although the first symptoms may appear earlier. The onset may be rapid or gradual.

Obsessional disorders in childhood generally resemble those in adulthood (see p. 199). The presenting symptoms are more often rituals than obsessional thoughts. Washing rituals are the most frequent, followed by repetitive actions and checking. Obsessional thoughts are most often concerned with contamination, accidents or illness affecting the patient or another person, and concerns about orderliness and symmetry. The content of symptoms often changes as the child grows older. The obsessional symptoms may be provoked by external cues such as unclean objects. Children with obsessional symptoms usually try to conceal them, especially outside the family. Obsessional children often involve their parents by asking them to take part in the rituals or give repeated reassurance about the obsessional thoughts.


Genetic factors are suggested by the observation that obsessive–compulsive disorder is more frequent among the first-degree relatives of children and adolescents with obsessive–compulsive disorder than among the general population. In addition, the concordance rate in monozygotic twins (about 80%) is significantly more than that of dizygotic twins (about 40%). The association of obsessive–compulsive disorder with tics and Tourette’s syndrome (see below) suggests that they may share genetic causes (Flament and Robaey, 2009).

Neurotransmitter disorders. Studies with adults suggest the involvement of serotonin systems in the brain (see p. 202), and the response of children to SSRIs (see below) suggests that they may share this involvement.

Neurological factors. Obsessive–compulsive disorder in childhood is associated with certain conditions thought to arise from dysfunction of the basal ganglia. Some children with obsessive–compulsive disorder have tics or choreiform movements. Conversely, children with Tourette’s syndrome have obsessional and compulsive symptoms, and these symptoms have also been described in children with Sydenham’s chorea (see below).

Autoimmune factors. The association with Sydenham’s chorea, which is thought to be an autoimmune disorder following group A beta-haemolytic streptococcal infection, has led to the description of a subtype of childhood-onset obsessive–compulsive disorder with similar aetiology. The condition is known as paediatric autoimmune neuropsychiatric disorder associated with streptococcal infection (PANDAS). See Rapoport and Shaw (2008) for a general review of aetiology.


The overall prevalence of obsessive–compulsive disorder in young people up to 18 years of age is 1–4%, with equal rates in males and females. Rates are much lower in young children and rise exponentially with age (Flament and Robaey, 2009).

Associated disorders

Severe and persistent obsessional thoughts and compulsive rituals in childhood are often accompanied by other anxiety and depressive symptoms. In some cases there is a formal associated anxiety or depressive disorder. As noted above, tics occur in 17–40% of children with obsessive–compulsive disorder, while Tourette’s syndrome is present in up to 15%. Conversely, about 30% of children with Tourette syndrome have obsessive–compulsive symptoms, and the frequency increases to about 50% by adulthood (Flament and Robaey, 2009).


There have been no large long-term follow-up studies, but the available evidence indicates that obsessive–compulsive disorder in childhood has a generally poor prognosis, often persisting into adulthood, although sometimes with fluctuations in severity.


The first step is to inform the child, the parents and the teachers about the disorder and allow time for discussion of the implications. When obsessional symptoms occur as part of an anxiety or depressive disorder, treatment is directed to the primary disorder. True obsessional disorders of later childhood are treated along similar lines to the same disorder in adults (see p. 203), with cognitive–behavioural methods, medication, or a combination of the two. Cognitive–behavioural methods have to be modified somewhat with younger children to take account of their stage of development. Whatever the treatment, it is important to involve the family.

Both SSRIs and clomipramine are more effective than placebo. (In the UK, both sertraline and fluvoxamine are licensed for the treatment of obsessive–compulsive disorder in people under 18 years of age.) As in adults, the symptoms are reduced but not removed by this treatment. In one trial with children and adolescents with obsessive–compulsive disorder, treatment with sertraline gave a remission rate of 20%, which is modest but worthwhile when compared with the placebo rate of 4%. In the same trial, the remission rate with cognitive–behavioural treatment was about 40%, and that with combined treatment was over 50% (Pediatric OCD Treatment Study Team, 2004).

Somatoform disorders and other unexplained physical symptoms

Children with a psychiatric disorder often complain of somatic symptoms which do not have a physical cause. These complaints include abdominal pain, headache, cough, and limb pains. Most of these children are treated by family doctors. The minority who are referred to specialists are more likely to be sent to paediatricians than to child psychiatrists.

Chronic fatigue syndrome

This condition, which is described on p. 389, occurs in children over the age of 11 years and in adolescents (Farmer et al., 2004). The principles of treatment resemble those for the treatment of the disorder in adults (see p. 390). Cognitive–behaviour therapy, similar to that used for adults, has been found to be effective in adolescents (Stulemeijer et al., 2005). Often it is best for treatment to be carried out jointly by a paediatrician and a psychiatrist or clinical psychologist. Although the longer-term prognosis for fatigue is probably quite good, the short-term effects on school performance can have serious consequences.

Recurrent abdominal pain

Recurrent abdominal pain has been estimated to occur in around 5–17% of all children, and is a common reason for referral to a paediatrician. In most cases abdominal pain is associated with headache, limb pains, and sickness. Physical causes of the abdominal pain are seldom found, and psychological causes are often suspected. Some are related to anxiety, some to ‘masked’ depressive disorder (see p. 665), and some to stressful events. Treatment is similar to that for other emotional disorders. Follow-up suggests that 25% of cases severe enough to require investigation by a paediatrician develop psychiatric problems or unexplained physical symptoms in adult life.

Conversion disorders

Conversion disorders (or conversion and dissociative disorder in ICD terminology) are described on p. 394. They are more common in adolescence than in childhood, both in individual patients and in the rare epidemic form of the disorders (see p. 397). In childhood, symptoms are usually mild and they seldom last long. The most frequent symptoms include paralyses, abnormalities of gait, and inability to see or hear normally. As in adults, conversion symptoms can occur in the course of organic illness as well as in a primary conversion disorder. As with adults, organically determined physical symptoms are sometimes misdiagnosed as conversion disorder when the causative physical pathology is difficult to detect and stressful events coincide with the onset of the symptom. For this reason, the diagnosis of conversion disorder should be made only after the most careful search for organic disease.


Conversion disorders are encountered rarely in community surveys, including the original Isle of Wight study (Rutter et al., 1970a). Among children referred to paediatricians, these disorders have been reported in 3–13% of cases. In a survey of prepubertal children referred to a psychiatric hospital, Caplan (1970) found that conversion disorder was diagnosed in about 2% of cases. In almost half of those children, organic illness was eventually detected either near the time or during the 4- to 11-year follow-up. Amblyopia was the symptom of organic disorder most likely to be originally diagnosed as psychogenic.


Conversion and other somatoform disorders should be treated as early as possible, before secondary gains (see p. 396) accumulate. The psychiatrist and paediatrician should work closely together. Thorough physical investigation is required before the psychiatric diagnosis is made, but unnecessary physical investigation should be avoided. Treatment is directed mainly at reducing any stressful circumstances and encouraging the child to talk about the problem. Symptoms may subside with these measures, or may need management comparable to that used for conversion disorder in adults (see p. 396), with physiotherapy and behavioural methods for motor symptoms.

For a review of child psychiatric syndromes with somatic presentation, see Gledhill and Garralda (2009).

Mood disorders

Mania is generally thought to be extremely uncommon before puberty, although there have been claims that it is more frequent than this. Particularly in the USA, some workers have suggested that paediatric bipolar disorder presents in an atypical way, with marked irritability, distractibility, and hyperactivity. In addition, the abnormal mood states often have a mixed presentation and cycle rapidly. This gives rise to estimates of bipolar disorder in paediatric populations of about 1%. The diagnostic issue is unresolved, but clearly caution is required, particularly in the use of drug treatments whose efficacy and long-term safety are not established in this population (Leibenluft and Dickstein, 2008).

It is normal for healthy children to feel depressed in response to distressing circumstances—for example when a parent is seriously ill or a grandparent has died. Some, including those experiencing grief, may also lose interest, concentrate poorly, and eat and sleep badly. This section is not concerned with these normal forms of unhappiness. Neither is it concerned with depressive symptoms that are part of another psychiatric disorder, such as anxiety disorder or conduct disorder. Mood disorders in adolescence are considered on p. 672.

Clinical picture

The clinical picture of major depression in childhood is in most ways similar to that in adults. The differences are that young children do not express guilt in an adult way, and may have difficulty in describing feelings of sadness. Also their sleep may not be disturbed in the ways found among adults with depressive disorder (see p. 207).

Masked depression. Some psychiatrists have suggested that in children, depressive disorders can present with little or no depressed mood but instead with symptoms such as unexplained abdominal pains, headache, anorexia, and enuresis. However, while these latter symptoms can be the first to be brought to medical attention, their investigation should include the possibility of a depressive disorder. When this is done, history taking and mental state examination will reveal sadness, anhedonia, irritability, and other typical symptoms of depressive disorder that were not volunteered initially (Luby et al., 2003).


Depressive disorders are infrequent before puberty (they are more common thereafter; see p. 672). Estimates of the prevalence of major depressive disorder give figures of 1% for children in middle childhood. Before puberty, rates of depressive disorders are higher among boys. They are less likely than depression in adolescence to lead to adult depression, and are more often associated with family dysfunction. Many children who meet the diagnostic criteria for depressive disorder also meet the criteria for another diagnosis, especially an anxiety disorder. For more information, see Brent and Birmaher (2009).


The causes of depressive disorder in childhood appear to be similar to those of depressive disorder in adulthood (see p. 219).

Genetic factors. As noted above, bipolar disorders seem to be rare in childhood, so the available information concerns depressive disorders. The rates of depressive disorder among first-degree relatives of children with depressive disorder are higher than the rates in the general population, suggesting genetic factors. It is likely that genetic factors are less important in depression in childhood than they are in adolescence. Genetic liability to anxiety and depression may be transmitted together (Brent and Birmaher, 2009).

Cortisol. Cortisol hypersecretion in non-depressed adolescents predicts the onset of depression (Goodyer et al., 2003). Such a relationship has not been shown in children, but it suggests the general possibility that hypersecretion of cortisol could have some direct or indirect causative role.

Other causes. Negative life events often precede the onset of depressive disorder in children, as they do in adults. Temperament also seems to be important, especially the tendency to react intensely to environmental stimuli. Family environment plays an important part in causing depression in adolescents, especially when a parent has depression (Hammen et al., 2004), and it is likely that the same is true in children.

The aetiology of depression is discussed further, in relation to adolescents, on p. 672.


Childhood depression tends to be chronic and recurrent, particularly when childhood onset is associated with a parental history of depression. There is an increased risk of depression in adulthood as well as an increased risk of bipolar disorder. Chronic depression in childhood is also associated with educational underachievement and interpersonal difficulties (Brent and Birmaher, 2009).


General measures. Any distressing circumstances should be reduced as far as possible, while the child is helped to talk about their feelings. The possibility of depression in the parents should be considered, and treated if necessary. With a school-age child, the management plan should involve the teachers to help them to understand the effect of depression on the child’s performance, and to help to identify and reduce any stressors at school, including bullying. The nature of the disorder is explained to the parents and, in terms appropriate to their age, to the child. This explanation, and the approach to possible stressors, should take full account of culture and ethnicity.

Psychological treatment. If depression does not improve with the above measures, a specific psychological treatment should be considered. Cognitive therapy, interpersonal therapy, and brief family therapy can be used, although most of the evidence for their efficacy is from trials with older children and adolescents.

Medication. When a child has failed to respond to psychosocial measures and psychological treatment, or if the depressive disorder is severe at the outset, drug treatment can be considered. Clinical trials of tricyclic antidepressant drugs for depressed children have not shown significant benefits. There is also uncertainty about the possible benefits of SSRIs in depressed children, where a meta-analysis found specific improvement only for fluoxetine (number needed to treat is approximately 10). In the same meta-analysis, relative to placebo, SSRIs produced a greater risk of self-harm and suicidal ideation (3% vs. 2%), giving a number needed to harm of about 112 (Bridge et al., 2007). There were no completed suicides. These findings suggest that medication should be reserved for children with significant depressive symptomatology that has not responded to psychological approaches. Prescribing decisions in these circumstances are best made by a specialist in child psychiatry. The potential benefits and adverse effects are examined for each child, and discussed with the carers and, in appropriate terms, with the child. The information should include warning about the delay between starting medication and improvement, and possible side-effects, including worsening ideas of self-harm. If an antidepressant is prescribed, it should always be combined with some form of psychological treatment. Of the available drugs, most evidence favours fluoxetine, which is the only medication approved for this purpose in the UK.

In view of the possibility of relapse, the child’s progress should be followed for up to a year after improvement. Very occasionally, when depression is particularly severe and unresponsive to care in the community, it may be necessary to treat the child in hospital. For guidelines about the care of children with depressive disorders, see National Institute for Health and Clinical Excellence (2005d).

Suicide in childhood

Suicide in childhood is considered on p. 426.

School refusal

School refusal is not a psychiatric disorder but a pattern of behaviour that can have many causes. It is convenient to consider it at this point in the chapter because of its association with anxiety and depressive disorder. School refusal is one of several reasons for repeated absence from school. Physical illness is the most common. A small number of children are deliberately and repeatedly kept at home by parents to help in the home or for other reasons. Some are truants who could go to school but choose not to, often as a form of rebellion. Finally, school refusers stay away from school because they are anxious or miserable when there. In an important study, Hersov (1960) compared 50 school refusers and 50 truants, all referred to a child psychiatric clinic. Compared with the truants, the school refusers came from more neurotic families, were more depressed, passive, and overprotected, and had better records of schoolwork and behaviour.


Temporary absences from school are extremely common, but the prevalence of school refusal is uncertain. In the Isle of Wight study, school refusal was reported in rather less than 3% of 10- and 11-year-olds with psychiatric disorder (Rutter et al., 1970a). It is most common at three periods of school life—between 5 and 7 years, at 11 years with the change of school, and especially at 14 years or older.

Clinical picture

At times, the first sign to the parents that something is wrong is the child’s sudden and complete refusal to attend school. More often there is an increasing reluctance to set out, with signs of unhappiness and anxiety when it is time to go. These children often complain of somatic symptoms of anxiety such as headache, abdominal pain, diarrhoea, sickness, or vague complaints of feeling ill. These complaints occur on school days but not at other times. Some children appear to want to go to school, but become increasingly distressed as they approach it. The final refusal can arise in several ways. It may follow a period of gradually increasing difficulty of the kind just described. It may appear after an enforced absence for another reason, such as a respiratory tract infection. It may follow an event at school such as a change of class. It may occur when there is a problem in the family such as the illness of a grandparent to whom the child is attached. Whatever the sequence of events, the child is extremely resistant to efforts to return them to school, and their evident distress makes it hard for the parents to insist that they go.


Several causes have been suggested. Separation anxiety is particularly important in younger children. In older children there may be a true school phobia, i.e. a specific fear of certain aspects of school life, including travel to school. Some fear bullying, or failure to do well in class. Other children have no specific concerns, but feel generally inadequate and depressed. Some older children have a depressive disorder.


Clinical experience suggests that most younger children eventually return to school. However, a proportion of the most severely affected adolescents do not return before the time when their compulsory school attendance ceases. There have been few studies of the longer prognosis of school refusal. It seems that of those with the more severe problems, between a third and a half have emotional problems over the next 10 years.


Except in the most severe cases, arrangements should be made for an early return to school. There should be discussion with the schoolteachers, who should be asked about the child’s problems, asked how they can help the child to catch up with missed education, and advised how to manage any difficulties that may arise when the child returns. By the time that help is sought, parents often have difficulty in pressuring the child to go to school. It is then more satisfactory for someone other than a parent to accompany the child to school at first. Sometimes the parents need help with their own problems before they can help the child. In severe cases, a more formal graded behavioural programme is necessary. In the most severe cases, admission to hospital may be required to reduce anxiety or depression before the child can return to school. Occasionally a change of school is appropriate.

Any depressive disorder should be treated. Otherwise, treatment is psychosocial. In all cases the child should be encouraged to talk about their feelings and the parents given support. For a review of psychosocial interventions, see Pina et al. (2009).

Other childhood psychiatric disorders

Functional enuresis

Functional enuresis is the repeated involuntary voiding of urine occurring after an age at which continence is usual (see below), in the absence of any identified physical disorder. Enuresis may be nocturnal (bed-wetting) or diurnal (daytime wetting), or both. Most children achieve daytime and night-time continence by 3 or 4 years of age. Nocturnal enuresis is often referred to as primary if there has been no preceding period of urinary continence. It is called secondary if there has been a preceding period of urinary continence.

Nocturnal enuresis can cause great unhappiness and distress, particularly if the parents blame or punish the child. This unhappiness may be made worse by limitations imposed by enuresis on activities such as staying with friends or going on holiday.


Estimates of prevalence vary, depending on the definition and method of assessment. In the UK, the prevalence of nocturnal enuresis occurring once a week or more is about 10% at 5 years of age, 4% at 8 years, and 1% at 14 years. Similar figures have been reported from the USA. Nocturnal enuresis occurs more frequently in boys. Daytime enuresis has a lower prevalence and is more common in girls than in boys. More than 50% of daytime wetters also wet their beds at night (Butler, 2008).


Nocturnal enuresis occasionally results from physical conditions, but more often appears to be caused by delay in the maturation of the nervous system, either alone or in combination with environmental stressors. There is some evidence for a genetic cause; about 70% of children with enuresis have a first-degree relative who has been enuretic. Also, concordance rates for enuresis are twice as high in monozygotic as in dizygotic twins (Butler, 2008).

Although most enuretic children are free from psychiatric disorder, the proportion with psychiatric disorder is greater than that of other children. Enuresis is more frequent in large families living in overcrowded conditions. Stressful events are associated with the onset of secondary enuresis. Rigid or other particular kinds of training have not been proved to be important.


A careful history and appropriate physical examination are required to exclude undetected physical disorder, particularly urinary infection, diabetes, or epilepsy, and to assess possible precipitating factors and the child’s motivation. A question should be asked about faecal soiling.

Psychiatric disorder should be sought. If none is found, an assessment should be made of any distressing circumstances affecting the child. The attitudes of the parents and siblings to the bed-wetting are evaluated. The parents are asked how they have tried to help the child, and their motivation to do more is assessed.


Any physical disorder should be treated. If the enuresis is functional, an explanation should be given to the child and the parents that the condition is common and the child is not to blame. It should be explained to the parents that punishment and disapproval are inappropriate and unlikely to be effective. The parents should be encouraged to reward success without drawing attention to failure, and not to focus attention on the problem. Many younger enuretic children improve spontaneously soon after being given an explanation of this kind, but those over 6 years of age are likely to need more active measures.

The next step is usually advice about restricting fluid before bedtime, lifting the child during the night, and the use of star charts to reward success.

Enuresis alarms. Children who do not improve with these simple measures may be treated with an enuresis alarm. Modern alarms consist of a detector pad attached to the night clothes, and an alarm buzzer carried in a pocket or on the wrist. When the child begins to pass urine the detector is activated and the alarm sounds. The child turns off the alarm, gets up to complete the emptying of the bladder, and changes their pyjamas and sheets as necessary, with help from the parents if needed. The method requires about 6–8 weeks of treatment, and some families break off before this has been completed.

The enuresis alarm seldom succeeds with children under the age of 6 years, or those who are uncooperative. For the rest, the alarm method, carried to completion, is effective within a month in about 70–80% of cases, although about a third of cases relapse within a year. It seems that children with associated psychiatric disorder do less well than the rest.

Medication. The synthetic antidiuretic hormone des-amino-D-arginine vasopressin (desmopressin) has a more prolonged action than natural vasopressin. It is used in the treatment of nocturnal enuresis in children over 5 years of age. It can be administered as a tablet or in a nasal spray. In one clinical trial, about 50% of the enuretic children treated with intranasal hormone became dry, and good results have been reported for an oral preparation. However, patients relapse when treatment is stopped. Side-effects of the oral preparation include rhinitis and nasal pain; other side-effects are nausea and abdominal pain. For this reason it is often used for temporary relief at important times—for example, during an overnight stay with friends. It is also possible to use desmopressin in conjunction with the enuresis alarm to speed the acquisition of bladder control. For a review of the treatment of enuresis, see Brown et al. (2011).

Faecal soiling

At the age of 3 years, 6% of children are still soiling themselves with faeces at least once a week; at 7 years the figure is 1.5%. By the age of 11 years, the figure is only 1% once a month or more. Soiling is three times more frequent in boys than in girls (Butler, 2008).

The term encopresis is used, but in two senses. In its wider sense it is a synonym for faecal soiling. In its narrower sense it denotes the repeated deposition of formed faeces in inappropriate places, including the underclothes. Because of this ambiguity, the term faecal soiling is used here.

Children who soil their clothes for any reason may feel ashamed, deny what has happened, and try to hide the dirty clothing.


Faecal soiling has several causes.

• Constipation with overflow is a common cause.

Constipation has many causes, but common ones are a low-fibre diet, pain on defecation (due, for example, to an anal fissure), or refusal to pass faeces as a form of rebellion. Hirschsprung’s disease is an uncommon but important cause. Soiling results when, after prolonged constipation, liquid faeces leak round the plug of hard faeces in the rectum.

• Fear of using the toilet. Occasionally children who have no pain on passing faeces fear sitting on the toilet for other reasons—for example, because they believe that some harmful creature lives there. Shy or bullied children may fear going to the toilet at school.

• Failure to learn bowel control. This can occur in children with learning disability or children of normal intelligence whose training has been inconsistent or inadequate.

• Stress-induced regression. Children who have recently learned control may lose it as a result of a highly stressful experience, such as sexual abuse.

• Rebellion. Some children appear to defecate deliberately in inappropriate places, and some children smear faeces on walls or elsewhere. Usually the family has many social problems, and often the child has other emotional or behavioural difficulties. The act appears to be a form of aggression towards the parents, although this intention is usually denied by the child.


Treatment depends on the cause. The first step is to check for chronic constipation, and if it is present, to treat the cause. For this, joint assessment with a paediatrician may be needed. Even when constipation is not the main cause, it may require treatment as a secondary problem. A child who is fearful of the toilet should be reassured sympathetically. Inadequate toilet training may be improved using behavioural techniques, including achievable targets, and star charts or other rewards, together with help for the parents. Stress-induced regression usually disappears when the child has been helped to overcome the trauma. Soiling as rebellion is more difficult to treat, as it is generally part of wider social and psychological difficulties which may require intensive and prolonged help. If outpatient treatment fails in these cases, or in those due to inadequate or unsuitable training, the child may respond to behavioural management in hospital. If the child is admitted, the parents need to be closely involved in the treatment in order to avoid relapse when the child returns home.


Whatever the cause, it is unusual for encopresis to persist beyond the middle teenage years, although associated problems may continue. When treated, most cases improve within a year.

For a review of faecal soiling, see Butler (2008).

Selective mutism

In this condition, sometimes called elective mutism, a child refuses to speak in certain circumstances, although they do so normally in others. Usually speech is normal in the home but lacking at school. There is usually no defect of speech or language, although some children have one. Often there is other negative behaviour, such as refusing to sit down or to play when invited to do so. These children often have a comorbid anxiety disorder. The condition usually begins between 3 and 5 years of age, after normal speech has been acquired. Although reluctance to speak is not uncommon among children starting school, clinically significant elective mutism is rare, probably occurring in about 1 per 1000 children.

Assessment is difficult because the child often refuses to speak to the psychiatrist, so that diagnosis depends to a large extent on the parents’ account. When questioning the parents, it is important to ask whether speech and comprehension are normal at home. Although psychotherapy, behaviour modification, and speech therapy have been tried, there is no evidence that any treatment is generally effective. In some cases, elective mutism lasts for months or years. Generally, symptoms of selective mutism improve considerably over time, but rates of associated psychiatric disorder, especially social phobia, remain high (Bishop and Norbury, 2008).


Stammering (or stuttering) is a disturbance of the rhythm and fluency of speech. It may take the form of repetitions of syllables or words, or of blocks in the production of speech. Stammering is four times more frequent in boys than in girls. It is usually a brief problem in the early stages of language development. However, 1% of children suffer from stammering after they have entered school.

The cause of stammering is not known, although many theories exist. It seems unlikely that all cases have the same causes. Genetic factors, brain damage, and anxiety may all play a part in certain cases, but do not seem to be general causes. Stammering is not usually associated with a psychiatric disorder, even though it can cause embarrassment and distress. Most children improve whether treated or not. Many kinds of psychiatric treatment have been tried, including psychotherapy and behaviour therapy, but none has been shown to be effective. The usual treatment is speech therapy.

Tic disorders

Tic disorders, including Gilles de la Tourette syndrome, are considered on pp. 337–8.


Dementing disorders are rare in childhood. They result from organic brain diseases such as lipidosis, leucodystrophy, or subacute sclerosing panencephalitis. Some of the causes are genetically determined and may affect other children in the family. The prognosis is variable. Many cases are fatal, while others progress to profound learning difficulty.


Schizophrenia is almost unknown before 7 years of age, and seldom begins before late adolescence. When it occurs in childhood, the onset may be acute or insidious. The whole range of symptoms that characterize schizophrenia in adulthood may occur (see Chapter 11), and in both DSM-IV and ICD-10 the criteria for diagnosis in children are the same as those used in adults; there is no separate category of childhood schizophrenia. Before symptoms of schizophrenia appear, many of these children are odd, timid, or sensitive, and their speech development is delayed. Early diagnosis is difficult, particularly when these non-specific abnormalities precede the characteristic symptoms.

Treatment is with antipsychotic drugs, as in the management of schizophrenia in adults, although with appropriate reductions in dosage. The child’s educational needs should be met, and support given to the family. See Hollis (2008) for a review of schizophrenia in childhood.

Gender identity disorders

Effeminacy in boys

Some boys prefer to dress in girls’ clothes and to play with girls rather than boys. Some have an effeminate manner and say that they want to be girls. The cause of this condition is unknown, and there is no evidence of any endocrine cause. Various family influences have been suggested, including the encouragement of feminine behaviour by the parents, a lack of boys as companions in play, a girlish appearance, and a lack of an older male with whom the child can identify. However, many children experience these influences without being effeminate.

With regard to treatment it is difficult to know how far intervention is appropriate. Associated emotional disturbance in the child may require help, and it may be useful to investigate and discuss any family behaviours that seem to be contributing to or maintaining the child’s behaviour. The prognosis is uncertain. Adult males with transvestism and transsexualism frequently recall enjoying feminine play as children, but follow-up studies of effeminate behaviour in early childhood show that the condition is more likely to be associated with homosexuality or bisexuality in adulthood than with transsexualism or transvestism (Green, 2009).

Tomboyishness in girls

In girls, the significance of marked tomboyishness for future sexual orientation is not known. It is usually possible to reassure the parents, and sometimes necessary to discuss their attitudes to the child and their responses to her behaviour. For a review of gender identity disorders, see Green (2009), and also see p. 374.

Suicide and deliberate self-harm

Both deliberate self-harm and suicide are rare among children under 12 years of age (they are more common in adolescence). These problems are discussed in Chapter 16 (see p. 426 and p. 436). For a full account, see Shaffer et al. (2009).

Psychiatric aspects of physical illness in childhood

The associations between physical and psychiatric disorders in children resemble those in adults (see Chapter 13 and Chapter 15). There are three main groups of association which are encountered at least as frequently in paediatric as in child psychiatric practice:

• psychological and social consequences of physical illness

• psychiatric disorders presenting with physical symptoms without a physical cause—for example, abdominal pain

• physical complications of psychiatric disorders—for example, eating disorders and faecal soiling.

Most medical disorders that may affect children are discussed in Chapter 15. In this section we consider only some problems specific to childhood.

The consequences of childhood physical illness

Psychiatric disorder provoked by physical illness

Delirium. When they are physically ill, children are more likely than adults (except the elderly) to develop delirium. A familiar example is delirium caused by febrile illness.

Other psychiatric disorder. In the Isle of Wight study of children (Rutter et al., 1970a), the prevalence of psychiatric disorder was only slightly increased with physical illnesses that do not affect the brain, but considerably increased with organic brain disorder or epilepsy.

Other effects. Chronic illness may lead to poor reading ability and affect general intellectual development. It may affect self-esteem and the ability to form relationships. These consequences may persist into adult life.

Effect on parents

The effects on parents are greater when the child’s physical illness is chronic or disabling. Their response depends on factors such as the nature of the physical disorder, the temperament of the child, the parents’ emotional resources, and the circumstances of the family. The parents may experience a sequence of emotional reactions like those of bereavement, and their marital and social lives may be affected. Most parents eventually develop a warm, loving relationship with a handicapped child and cope successfully with the difficulties. A few manage less well; they may have unrealistic expectations, or they may be rejecting or overprotective.

Effect on siblings

The brothers and sisters of children with physical problems may feel neglected, irritated by restrictions on their social activities, or resentful of having to spend so much time helping in the case of a handicapped child. Although some studies have shown more emotional and behavioural disturbances in siblings than would be expected by chance, most siblings manage well, and some even benefit through increased abilities to cope with stress and to show compassion for others.


Everyone involved in the care of physically disabled children should be aware of the psychological difficulties commonly experienced by these children and their families. When giving distressing information to families, it is particularly important to allow plenty of time. It may be necessary to see the family several times, and to provide continuing advice and support. The paediatrician and the child psychiatrist should collaborate closely, both with each other and with teachers and any social workers or other professionals involved with the child. Short periods of respite care may be needed to enable the family to continue to care for a handicapped child.

Children in hospital

Bowlby (1951) identified successive stages of protest, despair, and detachment in children who were isolated from the parents during admission to hospital. To avoid these problems, it is important to prepare children for admission by explaining in appropriate terms what will happen, and by introducing to them the members of staff who will care for them. Also, the family should be able to visit frequently and, if possible, take part in the care of the child. If the child is young, one of the parents should be able to sleep in the hospital if family circumstances allow this. These arrangements are especially important when hospital admission is long or repeated.

Psychiatric problems of adolescence

There are no specific disorders of adolescence. Nevertheless, special experience and skill are required to apply the general principles of psychiatric diagnosis and treatment to patients at this time of transition between childhood and adult life. It is often particularly difficult to distinguish psychiatric disorder from the normal emotional reactions of the teenage years. For this reason, this section begins by discussing how far emotional disorder is an inevitable part of adolescence. For a general review of problems in adolescence and their treatment, see Goodman and Scott (2005), chapter 22.

Psychological changes in adolescence

Considerable changes—physical, psychosexual, emotional, and social—take place in adolescence. During this time the young person becomes physically and sexually mature, and develops cognitive abilities comparable to those of adults, although they have yet to acquire experience. Adolescents strive to find an identity that reflects their own aspirations as well as those of their family, and which is compatible with their abilities and their circumstances. In this process, the influence of the peer group is important, as well as that of parents and teachers.

In many less developed countries, the adolescent starts work and gradually takes on more responsibility. In developed countries, education continues into mid or late adolescence and responsibilities come later. Adolescents are expected both to conform to the rules of society and to become more independent and develop restraints on their own behaviour. Many of the problems of adolescence relate to conflicts between these two expectations, or to the rejection of the rules of society. Although such problems are sometimes conspicuous, they are not inevitable.

Rutter et al. (1976a) concluded that rebellion and parental alienation are uncommon in mid-adolescence. However, inner turmoil, as indicated by reports of misery, self-depreciation, and ideas of reference, is present in about 50% of all adolescents. This turmoil seldom lasts for long and often goes unnoticed by adults. There may also be problems in relationships, and sexual difficulties. Among older adolescents, rebellious behaviour increases and some become estranged from school during their final years of compulsory attendance. These problems may be expressed, for example, as excessive drinking of alcohol, the misuse of drugs and solvents (see Chapter 17), and irresponsible behaviour when driving cars or motorcycles.

Epidemiology of psychiatric disorder in adolescence

Although psychiatric disorders are only slightly more common in adolescence than in the middle years of childhood, the pattern of disorder is closer to that of adults. In a UK survey, Meltzer et al. (2000) found that between the ages of 11 and 15 years, 13% of boys and 10% of girls had a psychiatric disorder. Both emotional disorders and conduct disorder were more frequent than in childhood, and occurred with approximately the same frequency. An epidemiological survey in Italy found that about 8% of adolescents currently met the DSM-IV criteria for a psychiatric disorder (Frigerio et al., 2009). Anxiety and depressive disorders were the most common.

Clinical features of psychiatric disorders of adolescence

Anxiety disorders

School refusal is common between 14 years of age and the end of compulsory schooling, and at this age is often associated with other psychiatric disorders. Social phobias begin to appear in early adolescence; agoraphobia appears in the later teenage years. Anxiety disorders become more common in girls than in boys during the teenage years.

Conduct disorders

About 50% of the cases of conduct disorder seen in adolescents have started in childhood. Those that begin in adolescence differ in being less strongly associated with reading retardation and family pathology. Among younger children, aggressive behaviour is generally more evident in the home or at school. Among adolescents, it is more likely to appear outside these settings as offences against property. Truancy also forms part of the conduct disorders occurring at this age.

Mood disorders

In depressive disorder of adolescence, depressive mood may be less immediately obvious than anger, alienation from parents, withdrawal from social contact with peers, underachievement at school, substance abuse, and suicide attempts. Bipolar disorder occurs in adolescence, and it too may appear first as abnormal behaviour.

Epidemiology. Depressive symptoms are more common in adolescence than in childhood. In the Isle of Wight study they were 10 times more frequent among 14-year-olds than among 10-year-olds. The prevalence of major depression in adolescence is around 3–8%; it is during adolescence that the female preponderance in depression becomes apparent (Brent and Weersing, 2008).

Aetiology. The heritability of major depression in adolescents is greater than that in children. As in children, genes could act indirectly—for example, through behaviours that increase negative life events (see p. 222). Independent life events, in the family or elsewhere, interact with the genetic predisposition. Negative cognitive biases and increased cortisol secretion have also been identified as risk factors. For an account of aetiology, see Goodyer et al. (2009).


As with depressed children, a stepped-care model is appropriate, starting with the general measures described above for childhood depressive disorder (see p. 665).

Psychological treatments are considered next. There is more evidence about their effectiveness for adolescents than there is for children, and cognitive–behaviour therapy has been shown to be effective for moderate to severe depression, although the effect size of treatment appears to be modest (0.3–0.4). There is also some evidence that interpersonal therapy is effective (Weisz et al., 2006; Brent and Weersing, 2008).

Medication. For bipolar disorder, lithium usually reduces recurrence in adolescents as it does in adults. There is also some use of other mood-stabilizing drugs, such as valproate (see p. 554). Particular care is required when deciding on the dosage to be used.

For depressive disorders, tricyclic antidepressant drugs have not shown significant benefit over placebo in depressed adolescents (National Institute for Health and Clinical Excellence, 2005d), and fluoxetine is the only antidepressant currently considered to have a favourable risk–benefi ratio.

In the Treatment for Adolescents with Depression (TADS) study sponsored by the US National Institute for Mental Health (NIMH), 439 depressed adolescents were randomly allocated to fluoxetine, pill placebo, cognitive–behaviour therapy, or a combination of fluoxetine and cognitive–behaviour therapy. After 12 weeks of treatment, response rates for pill placebo (34%) and cognitive–behaviour therapy (43%) were significantly less than for fluoxetine alone (61%) and the combination treatment (71%). By 18 weeks, the response to cognitive–behaviour therapy (65%) was similar to that found with fluoxetine alone (69%), while the response to combination treatment was now 85%. As noted above, fluoxetine treatment was associated with an increased risk of suicidal ideation and self-harm relative to placebo, but this increase was not seen in patients in whom fluoxetine was combined with cognitive–behaviour therapy (March and Vitiello, 2009). The TADS study supports the view that the best treatment for moderate to severe depression in adolescence is a combination of fluoxetine and cognitive–behaviour therapy.

ECT. In rare cases, where there is extremely severe, life-threatening depression that is unresponsive to medication, ECT may be considered for older adolescents (National Institute for Health and Clinical Excellence, 2005d).

Prognosis. Depression in adolescence may recur, and is often followed by depression in adulthood, and by a long-term increase in suicide rate (see Goodman and Scott, 2005, p. 95). For more detailed information about depressive disorder in adolescence, see Brent and Weersing (2008).


Schizophrenia in adolescence is more common in boys than in girls. Usually the diagnosis presents little difficulty. When there is difficulty it is usually in detecting characteristic symptoms, especially in patients whose main features are gradual deterioration of personality, social withdrawal, and decline in social performance. The prognosis may be good for a single acute episode with florid symptoms, but is poor when the onset is insidious.

Eating disorders

Problems with eating and weight are common in adolescence. They are discussed in Chapter 14, to which the reader is referred, as they closely resemble the same conditions in adulthood. It is particularly important to involve the parents and perhaps other family members in the treatment of an adolescent patient with eating disorder. Formal family therapy has been shown to be of value for anorexia nervosa in adolescents. For a review, see Fairburn and Gowers (2008).

Suicide and deliberate self-harm

In recent years there has been a marked increase in suicide and deliberate self-harm among adolescents. These subjects are discussed in Chapter 16 (see p. 426 and p. 436).

Alcohol and substance abuse

Problems of substance abuse in adolescence are similar to those in adulthood, as described in Chapter 17Excessive drinking is common among adolescents, especially among those with conduct disorder. Most adolescent heavy drinkers seem to reduce their drinking as they grow older, but a few progress to more serious drinking problems in adulthood. Prevention programmes have been developed, but there is little evidence for their effectiveness (Heath et al., 2008).

Occasional drug taking is common in adolescence, and is often a group activity. Cigarette smoking and the use of cannabis and ‘Ecstasy’ are especially frequent. Solvent abuse is largely confined to adolescence, and is usually of short duration (see p. 481). Abuse of drugs such as amphetamines, barbiturates, opiates, and cocaine is less common but more serious, as most drug-dependent adults will have experimented with these drugs during adolescence. There is a strong association between conduct disorder in childhood and drug taking in adolescence.

Most adolescents experiment with drugs for short periods and do not become regular users. Those who persist in taking drugs are more likely to come from discordant families or broken homes, to have failed at school, and to be members of a group of persistent drug users. Feelings of alienation and low self-esteem may also be important.

Since regular drug taking starts less often in adulthood than in adolescence, the limitation of drug taking among adolescents is an important aim. One approach is the application of appropriate psychosocial interventions in young people at high risk of drug misuse (see p. 469 and National Institute for Health and Clinical Excellence, 2007a). See Heath et al. (2008) for an account of substance misuse by adolescents.

Sexual problems

Concern about sexuality is normal in adolescence. Sexual abuse is increasingly a cause of referral to psychiatrists (see p. 671). Teenage pregnancy is also an important problem. There is a raised incidence of prenatal complications as compared with older mothers. Teenage mothers may experience the same postpartum disorders as other mothers (see pp. 416–18). Very young mothers frequently have difficulties as parents, and there is a poor outlook for many teenage marriages. Pregnant teenagers need access to continuing medical and social services during and after pregnancy.

Assessment of adolescents

Special skills are needed when interviewing adolescents. In general, young adolescents require an approach similar to that used for children, while for older adolescents it is more appropriate to employ the approach used with adults. It must always be remembered that a large proportion of adolescents attending a psychiatrist do so somewhat unwillingly, and also that most have difficulty in expressing their feelings in adult terms. Therefore psychiatrists must be willing to spend considerable time establishing a relationship with their adolescent patients. To do this, they must show interest in the adolescents, respect their point of view, and talk in terms that they can understand. As in adult psychiatry, it is important to collect systematic information and describe symptoms in detail, but with adolescents the interviewer must be prepared to adopt a more flexible approach to the interview.

It is usually better to see the adolescent before interviewing the parents. In this way, it is made clear that the adolescent is being seen as an independent person. Later, other members of the family may be interviewed, and sometimes the family may be seen as a whole. As well as the usual psychiatric history, particular attention should be paid to information about the adolescent’s functioning at home, in school, or at work, and about their relationships with peers. Relevant physical examination should be carried out unless this has already been done by the doctor who made the referral.

Such an assessment should allow allocation of the problem to one of three categories. In the first category, no psychiatric diagnosis can be made and reassurance is all that is required. In the second, there is no psychiatric diagnosis, but the adolescent, anxious parents or disturbed family need help. In the third category there is a psychiatric disorder that requires treatment.

Treatment of adolescents

Treatment methods are intermediate between those employed in child and adult psychiatry. As in the former, it is important to work with the patient’s relatives and teachers. It is necessary to help, reassure, and support the parents and sometimes to extend this to other members of the family. This is especially important when the referral reflects the anxiety of the family about minor behavioural problems rather than the presence of a definite psychiatric disorder. However, it is also important to treat the adolescent as an individual who is gradually becoming independent of the family. In these circumstances, family therapy as practised in child psychiatry is usually inappropriate, and may sometimes be harmful.

Services for adolescents

The proportion of adolescents in the population who are seen in psychiatric clinics is less than the proportion of other age groups. Of those referred, some of the less mature adolescents can be helped more in a child psychiatry clinic. Some of the older and more mature adolescents are better treated in a clinic for adults. Nevertheless, for the majority the care can be provided most appropriately by a specialized adolescent service, provided that close links are maintained with child and adult psychiatry services and with paediatricians.

There are variations in the organization of these units and the treatment that they provide, but most combine individual and family psychological treatment with the possibility of drug treatment for severe disorders. Most units accept outpatient referrals not only from doctors but also from senior teachers, social workers, and the courts. When the referral is non-medical, the general practitioner should be informed and the case discussed. All adolescent units work with schools and social services. Inpatient facilities are usually limited in extent, so that it is important to agree with social services what kinds of problems need admission to a health service unit, and which should be cared for in residential facilities provided by social services (in the UK). Reasons for admission to a health service inpatient unit include the following:

• severe or very unusual mental symptoms requiring that the person’s mental state be observed carefully, investigations carried out, or treatment monitored closely

• behaviour that is dangerous to the self or to others, and that is due to psychiatric disorder.

When dangerous behaviour relates to personality and circumstances and not to illness, a hospital unit is not more effective than secure residential accommodation, and the behaviour of such adolescents may be stressful for others with mental disorders. For a review of the organization of mental health services for children and adolescents, see Wolpert (2009).

Child abuse

In recent years, the concept of child abuse has been widened to include the overlapping categories of physical abuse (non-accidental injury), emotional abuse, sexual abuse, and neglect. Most of the literature on child abuse refers to developed countries, rather than to developing countries in which children commonly face poor nutrition, and other hardships such as severe physical punishment, abandonment, and employment as beggars and prostitutes.

The term fetal abuse is sometimes applied to behaviours that are detrimental to the fetus, including physical assault and the taking by the mother of substances that are likely to cause fetal damage. Factitious disorder (or Munchausen syndrome by proxy) is the name given to apparent illness in children which has been fabricated by the parents, and to conditions induced by parents—for example, by partly smothering the child (see p. 392).

For a general review of child abuse and neglect, see Jones (2009).

Physical abuse (non-accidental injury)

Estimates of the prevalence of physical abuse of children vary according to the criteria used. Most studies report prevalence rates of physical abuse in childhood in the range 5–10%, depending on how this is defined. The rate of significant violence is somewhat less, with a community survey suggesting a rate in the UK of about 90 per 1000, rather similar to that reported in Italy (about 80 per 1000) (Jones, 2008).

Clinical features

Parents may bring an abused child to the doctor with an injury said to have been caused accidentally. Alternatively, relatives, neighbours, or other people may become concerned and report the problem to police, social workers, or voluntary agencies. The most common forms of injury are multiple bruising, burns, abrasions, bites, torn upper lip, bone fractures, subdural haemorrhage, and retinal haemorrhage. Some infants are smothered, usually with a pillow, and the parents report an apnoeic attack. Suspicion of physical abuse should be aroused by the pattern of the injuries, a previous history of suspicious injury, unconvincing explanations, delay in seeking help, and incongruous parental reactions. The psychological characteristics of abused children vary, but include fearful responses to the parents, other evidence of anxiety or unhappiness, and social withdrawal. Such children often have low self-esteem, may avoid adults and children who make friendly approaches, and may be aggressive.


There are many interacting causes (see Table 22.8).

The environment. Child abuse is more frequent in neighbourhoods in which family violence is common, schools, housing, and employment are unsatisfactory, and there is little feeling of community.

The parents. Factors associated with child abuse include young age, abnormal personality, psychiatric disorder, lower social class, social isolation, marital conflict and breakdown, and a criminal record. If a parent has a psychiatric disorder, it is most often a personality disorder; only a few parents have disorders such as schizophrenia or affective disorder. Many parents give a history of having themselves suffered abuse or deprivation in childhood, and the parental relationship is marked by conflict and sometimes violence. Although child abuse is much more common in families with other forms of social pathology, it is certainly not limited to such families.

The children. Risk factors include premature birth, early separation, the need for special care in the neonatal period, congenital malformations, chronic illness, and a difficult temperament.

Assessment and management

Doctors and others involved in the care of children should always be alert to the possibility of child abuse. They need to be particularly aware of the risks to children who have some of the characteristics described above, or who are being cared for by parents with the predisposing factors listed.

Points that may raise suspicion include the following:

• delay in seeking help

• a vague or inconsistent account of the way in which injuries came about

• an account that is inconsistent with the nature and extent of injury

• an apparent lack of concern for the child, or a defensive or suspicious response to questions.

Table 22.8 Risk factors for child abuse

The parent(s)

Young age

Low intelligence

Social isolation/no one to provide help

Breakdown of relationship with partner

Poor parenting skills: lack of awareness of child’s needs, harsh punishment, little reward

Experience as a child: abused or neglected

Criminal record

Psychiatric problems: depression, substance abuse, personality disorder

The child

Factors leading to weak attachment to the parents:

Premature, health problems

Separation from mother during early life (e.g. in neonatal unit)

Difficult temperament, cries a lot

The environment

Problem neighbourhood: high levels of family violence, problem schools, high unemployment

Little feeling of community

Doctors who suspect abuse should refer the child to hospital and inform a paediatrician or casualty officer of their suspicions. The paediatrician will perform a physical examination, noting any of the physical consequences described above. (For further information about the physical examination, see a textbook of paediatrics.) In the hospital emergency department, inpatient admission should be arranged for all children in whom non-accidental injury is suspected. If possible, the doctor’s concerns should be discussed with the parents, and in any case they should be told that admission is necessary to allow further investigations. If the parents refuse admission, it may be necessary in England and Wales to apply to a magistrate for a Place of Safety Order; similar action may be appropriate in other countries. During admission, assessment must be thorough and include photographs of injuries and skeletal radiography. Radiological examination may show evidence of previous injury or, occasionally, of bone abnormalities such as osteogenesis imperfecta. A CT scan may be needed if subdural haemorrhage is suspected. All findings must be fully documented.

Once it has been decided that non-accidental injury is probable, the first priority is to establish the child’s safety. The procedures involved will vary according to the administrative arrangements in different countries. In the UK, the Social Services Department is responsible for child protection and should be notified. It may be decided to put the child’s name on a child abuse register, thereby making the Social Services Department responsible for visiting the home and checking the problem regularly.

In some cases, the risk of returning the child to their parents is too great, and separation is required. If the parents do not agree to separation, a care order can be sought by the Social Services Department. If the abuse is severe, prolonged, or permanent, separation may be necessary and the parents may face criminal charges. Because there are known cases of injury or death in children who have been returned to their parents, it is vitally important that the most careful assessment is undertaken before physically abused children are returned to their family. Countries vary in the requirements and procedures for reporting and monitoring possible physical abuse in children, and readers should familiarize themselves with the arrangements that exist in the area in which they work.

For further information about the assessment of children who may have been abused, see Jones (2009).


Children who have been subjected to physical abuse are at high risk of further problems. For example, the risk of further severe injury is probably around 10–30%, and sometimes the injuries are fatal. Abused children are likely to have subsequent high rates of physical disorder, delayed development, and learning difficulties. There are also increased rates of behavioural and emotional problems in later childhood and adulthood, even when there has been earlier therapeutic intervention. As adults, many former victims of abuse have difficulties in rearing their own children. The outcome is better for abused children who can establish a good relationship with an adult (often from outside the family), and who have dispositions that include sociability and academic competence (Jones, 2008).

Emotional abuse

The term emotional abuse usually refers to persistent neglect or rejection sufficient to impair a child’s development. However, the term is sometimes applied to gross degrees of overprotection, verbal abuse, or scapegoating, which impair development. Emotional abuse often accompanies other forms of child abuse.

Emotional abuse has various effects on the child, including failure to thrive physically, impaired psychological development, and emotional and conduct disorders. Diagnosis depends on observations of the parents’ behaviour towards the child, which may include frequent belittling or sarcastic remarks about them during the interview. One or both parents may have a disorder of personality, or occasionally a psychiatric disorder. The parents should be interviewed separately and together to discover any reasons for the abuse of this particular child—for example, he may fail to live up to their expectations, or may remind them of another person who has been abusive to one of them. The parents’ mental state should be assessed.


In treatment, the parents should be offered help with their own emotional problems and with their day-to-day interactions with the child. It is often difficult to persuade parents to accept such help. If they reject help and if the effects of emotional abuse are serious, it may be necessary to involve the social services and to consider the steps described above for the care of children suffering physical abuse. The child is likely to need individual help.

Child neglect

Child neglect is the failure to provide necessary care. It can take several forms, including emotional deprivation, neglect of education, physical neglect, lack of appropriate concern for physical safety, and denial of necessary medical or surgical treatment. These forms of neglect may lead to physical or psychological harm, including poor academic performance and disturbed behaviour (Jones, 2009).

Child neglect is more common than physical abuse, and it may be detected by various people, including relatives, neighbours, teachers, doctors, or social workers. Child neglect is associated with adverse social circumstances, and is a common reason for a child to need foster care.

Non-organic failure to thrive and deprivation dwarfism

Paediatricians recognize that some children fail to thrive when there is no apparent organic cause. In children under 3 years of age, this condition is called non-organic failure to thrive (NOFT); in older children it is called psychosocial short stature syndrome (PSSS) or deprivation dwarfism.

Clinical picture

Non-organic failure to thrive is caused by the deprivation of food and close affection. In the children who are seen in the psychiatric service, there is usually evidence of problems in the parent–child relationship since the child’s early infancy; these include rejection and, in extreme cases, expressed hostility towards the child. There may be physical or sexual abuse as well. The infant who is failing to thrive may present with recent weight loss, persistently low weight, or reduced height. There may be cognitive and developmental delay. The infant may be irritable and unhappy, or, in more severe cases, lethargic and resigned. There is a clinical spectrum ranging from infants with mild feeding problems to those with all of the severe features described above. If food and care are provided, the infant usually grows and develops quickly.

Psychosocial short stature or ‘deprivation dwarfism.’ These children have abnormally short stature, unusual eating patterns, retarded speech development, and temper tantrums. Although short in stature, the child may be of normal weight. Emotional and behavioural disorders occur, and there may be cognitive and developmental delay with impairment of language skills. These children have low self-esteem and are commonly depressed. There is usually a history of deprivation or of psychological maltreatment. Away from the deprived environment, these children eat ravenously.


When treating either syndrome, the first essential is to ensure the child’s safety, which may require admission to hospital. Subsequently, some children can be managed at home, but others need foster care. Some parents can be helped to understand their child’s needs and to plan for them; other parents are too hostile to be helped. If help is feasible, it should be intensive and should probably focus on changing patterns of parenting. It is unusual for the parents to be psychiatrically ill, but some have severe postpartum depression or other psychiatric disorder. Home-based intervention programmes may improve treatment effects, particularly in cognitive and behavioural domains.


With both syndromes, the prognosis for early cases is relatively good (Rudolf and Logan, 2005). However, some children have to be placed permanently in foster care because family patterns are resistant to change. In a positive emotional setting the abnormal behaviour is usually lost quickly and mental development follows physical growth. For a review, see Sandberg and Stevenson (2008).

Factitious disorder by proxy

This condition, in which a parent brings a child for treatment of fabricated symptoms, is discussed on p. 392.

Sexual abuse

The term sexual abuse refers to the involvement of children in sexual activities which they do not fully comprehend and to which they cannot give informed consent, and which violate generally accepted cultural rules. The term covers various forms of sexual contact with or without varying degrees of violence. The term also covers some activities that do not involve physical contact, such as posing for pornographic photographs or films. The abuser is commonly known to the child and is often a member of the family (incest). A minority of children are abused by groups of paedophiles (sex rings).


Rates of sexual abuse are difficult to establish. A UK survey of 18- to 24-year-olds found that 15% of girls and 6% of boys reported non-penetrative contact sexual abuse that had occurred when they were under 16 years of age. Rates of penetrative sexual abuse are lower, perhaps 4% of females and 2% of males (Glaser, 2008). It is agreed that abused children are more often female—probably with a ratio of about 3:1, and in 90% of cases the offender is male. Children with disabilities are more likely to be victims of sexual abuse. Much sexual abuse takes place within the family, and stepfathers are over-represented among abusers. The extent of sexual abuse by women is not known.

Clinical features

The presentation of child sexual abuse depends on the type of sexual act and the relationship of the offender to the child. Children are more likely to report abuse when the offender is a stranger. Sexual abuse may be reported directly by the child or by a relative, or it may present indirectly with unexplained problems in the child, such as physical symptoms in the urogenital or anal area, pregnancy, behavioural or emotional disturbance, or precocious or otherwise inappropriate sexual behaviour. In adolescent girls, running away from home or unexplained suicide attempts should raise the suspicion of sexual abuse. When abuse occurs within the family, marital and other family problems are common.

Effects of sexual abuse

Early emotional consequences of sexual abuse include anxiety, fear, depression, anger, and inappropriate sexual behaviour, as well as reactions to any unwanted pregnancy. A sense of guilt and responsibility is common. Some children show signs of post-traumatic stress disorder (see p. 159). It is not certain how common these reactions are, or how they relate to the nature and circumstances of the abuse.

Long-term consequences include depressed mood, low self-esteem, self-harm, difficulties in relationships, and sexual maladjustment in the form of either hypersensitivity or sexual inhibition. The effects of abuse are generally greater when the abuse has involved physical violence and penetrative intercourse. Some of the long-term effects may be related to the events surrounding the disclosure of the abuse, including any legal proceedings, and to other problems in the family, such as neglect of the children and sexual deviance or substance abuse. Nevertheless, even when these other factors are controlled for, sexual abuse in childhood seems to be associated with psychiatric disorder later in life, especially with depressive disorders, anxiety disorders and personality disorders (Glaser, 2008).


Sexual abuse of children occurs in all socio-economic groups, although it is more frequent among socially deprived families. However, there is not the strong association with low social class found with physical abuse. There are several preconditions which make sexual abuse more likely. In the abuser, these include deviant sexual motivation, impulsivity, a lack of conscience, and a lack of external restraints (e.g. cultural tolerance); in the child, they include a lack of resistance (due to insecurity, ignorance, or other causes of vulnerability).


It is important to be ready to detect sexual abuse and to give serious attention to any complaint by a child of being abused in this way. When abuse has been established, it is important to assess whether it is likely to continue if the child remains at home and, if so, how dangerous it is likely to be. It is also important not to make the diagnosis without adequate evidence, which requires social investigation of the family as well as psychological and physical examination of the child.

The child should be interviewed sympathetically and encouraged to describe what has happened. Drawings or toys may help younger children to give a description, but great care must be taken to ensure that they are not used in a way that suggests to the child events which have not taken place. Young children can recall events accurately, but they are more suggestible than adults. Interviewing is difficult, and whenever possible it should be carried out by a child psychiatrist or social worker with special experience. See Jones (2009) for further advice on interviewing.

In the UK, responsibility for child protection rests with social services, to whom suspicions of abuse and actual abuse should be reported. Multidisciplinary involvement, often including the police and the judicial system, is usually needed to establish whether abuse has taken place and what the appropriate response should be. As with the physical abuse, the first priority is to protect the child and other children who might be at risk. The arrangements vary in different countries, and readers should find out how they apply in the country in which they are working. For further information about assessment, see Glaser (2008).

Box 22.5 Steps in the assessment and management of child sexual abuse

1. Suspicion and recognition; referral to child protection services.

2. Establish whether immediate protection is needed.

3. Plan investigation: inter-agency discussion, interview child, medical examination, family assessment.

4. Initial multi-agency child protection meeting.

5. Draw up protection plan.

6. Plan implementation and review

7. Prosecution.

8. Therapy.

From Glaser (2008).


The initial management and the measures to protect the child are similar to those for physical abuse (see Box 22.5), including a decision about separating the child from the family. There are particular difficulties involved in intervening with families in which sexual abuse has occurred. These include a marked tendency to deny the seriousness of the abuse and of other family problems, and in some cases deviant sexual attitudes and behaviour of other family members, possibly including other children. Individual and group treatment has been used for the offenders, with the general aim of enabling the person to reduce denial and consider the effects of the abuse on the child. If the mother has a history of abuse, this needs to be discussed to help her to understand how this may have affected her response to her child’s abuse.

Some sexually abused children have highly abnormal sexual development for which they require help. They also need counselling to help them to deal with the emotional impact of the abuse, to come to terms with it, and to improve low self-esteem. Help should be in the form of a staged programme of rehabilitation for the whole family, rather than a brief intervention. For a review of the treatment of child sexual abuse, see Glaser (2008).

Ethical and legal problems in child and adolescent psychiatry

As well as the ethical and legal problems related to the treatment of adults and discussed in Chapter 4 and elsewhere, the following issues are particularly likely to arise in the care of children with psychiatric disorders.

Conflicts of interest

In general, the interests of the child take precedence over those of the parents. This principle is most obvious, and most easily followed, in cases of child abuse. In other cases, the decision is more difficult—for example, when a depressed mother is neglecting her child and is likely to become more depressed if substitute care is arranged. Such problems can usually be resolved by discussion with the parents, and between the professionals who are caring for the child and for the parent.


The care of children often involves collaboration between medical and social services, and sometimes with teachers. Different agencies may have different policies about the confidentiality of records, and doctors should take account of these differences when deciding what information to disclose.


In each country, the law decides an age below which parents give consent on behalf of the child. Below this age, the child’s agreement should be obtained, if possible, as this will aid treatment, but if the child refuses the parents can decide. This problem arises, for example, when an adolescent under the legal age of consent refuses treatment for anorexia nervosa. The parents can also refuse treatment; however, their right to do so is linked to their duty to protect the child. In cases in which the parents’ refusal appears not to be in the interests of the child, countries generally have provisions for a decision by a court of law.

Some of the complexities of English law can be mentioned briefly to illustrate the problems that have to be resolved in all legal systems. A fuller account of these and other issues is given by Hale and Fortin (2008). Readers should find out how these issues are dealt with in the country in which they are working before they undertake the care of patients under the age at which adult rules of consent apply.

In English law, the age from which people are judged legally capable of giving or refusing consent is 18 years. However, English law recognizes that most of those aged between 16 and 18 years have the capacity to give consent to treatment, and allows them to do so without the need for consent by a parent. The position is less clear when a 16- or 17-year-old does not consent to treatment, but it is probably the case that the parents’ consent is sufficient. The decision in each case is likely to depend on the consequences of refusal; the more severe these are, the more likely it is that a court would accept that the child’s refusal can be overridden by the parents. Further complexities arise with children under 16 years of age, some of whom are competent to give consent to certain treatments. There is no general assumption of competence at this age, and it has to be established in the individual case, but if it is established, the minor can give consent.

The question arose most notably with regard to the provision of contraception without the additional consent of a parent. In the Gillick case it was ruled that, in English law, a sufficiently competent minor could consent without the need to obtain the consent of the parent (now called Gillick competence). It is probable, however, that with certain more invasive and risky treatments, the consent of a parent could be legally necessary, as well as clinically desirable. If a minor under the age of 16 years refuses treatment, this can be overruled by the parents if refusal is likely to result in harm. A final complexity concerns the definition of a parent. This problem arises, for example, when the person accompanying the child is not the person recognized by the law as having parental responsibility. For example, in English law, a father who is not married to the mother does not automatically have legal responsibility.

Consent for research poses similar problems for patients below the legal age of consent. In most countries, parents consent on behalf of their children, and they may find it difficult to balance the risks to the child against the benefits, which are usually not to their child but to other children who might be treated for the same condition in the future. It is important that they are able to discuss the issues fully, and in general with a person other than the individual who is requesting the consent (e.g. a nurse). For a brief account of this subject, see Larcher (2005).

Further reading

Gelder MG, López-Ibor JJ Jr, Andreasen NC and Geddes JR (eds) (2009). Part 9: Child and adolescent psychiatry. In: New Oxford Textbook of Psychiatry. Oxford University Press, Oxford. (The 33 chapters in this part of the textbook provide a comprehensive account of the subject written for the general psychiatrist.)

Goodman R and Scott S (2005). Child Psychiatry, 2nd edn. Blackwell Science, Oxford. (A concise introduction to the subject.)

Hopkins B, Ronalg RG, Michel GF and Rochat P (2004). The Cambridge Encyclopaedia of Child Development. Cambridge University Press, Cambridge. (A comprehensive source of information about child development.)

Rutter M et al. (eds) (2008). Rutter’s Child and Adolescent Psychiatry, 5th edn. Blackwell, Oxford. (An established, comprehensive work of reference.).

Appendix: History taking and examination in child psychiatry

The format and extent of an assessment will depend on the nature of the presenting problem. The following scheme is taken from the book by Graham (1999), which should be consulted for further information. Graham suggests that clinicians with little time available should concentrate on the items in bold type

1 Nature and severity of presenting problem(s). Frequency. Situations in which it occurs. Provoking and ameliorating factors. Stresses thought by parents to be important.

2 Presence of other current problems or complaints.

(a) Physical. Headaches, stomach ache. Hearing, vision. Seizures, faints, or other types of attacks.

(b) Eating, sleeping, or elimination problems.

(c) Relationship with parents and siblings. Affection, compliance.

(d) Relationships with other children. Special friends.

(e) Level of activity, attention span, concentration.

(f) Mood, energy level, sadness, misery,

(g) Response to frustration. Temper tantrums.

(h) Antisocial behaviour. Aggression, stealing, truancy.

(i) Educational attainments, attitude to school attendance.

(j) Sexual interest and behaviour.

(k) Any other symptoms, tics, etc.

3 Current level of development.

(a) Language: comprehension, complexity of speech.

(b) Spatial ability.

(c) Motor coordination, clumsiness.

4 Family structure.

(a) Parents. Ages, occupations. Current physical and emotional state. History of physical or psychiatric disorder. Whereabouts of grandparents.

(b) Siblings. Ages, presence of problems.

(c) Home circumstances: sleeping arrangements.

Family function.

(a) Quality of parental relationship. Mutual affection. Capacity to communicate about and resolve problems. Sharing of attitudes over child’s problems.

(b) Quality of parent-child relationship. Positive interaction: mutual enjoyment. Parental level of criticism, hostility, rejection.

(c) Sibling relationships.

(d) Overall pattern of family relationships. Alliance, communication. Exclusion, scapegoating. Intergenerational confusion.

6 Personal history.

(a) Pregnancy complications. Medication. Infectious fevers.

(b) Delivery and state at birth. Birth weight and gestation. Need for special care after birth.

(c) Early mother-child relationship. Post-partum maternal depression. Early feeding patterns.

(d) Early temperamental characteristics. Easy or difficult, irregular, restless baby and toddler.

(e) Milestones. Obtain exact details only if outside range of normal.

(f) Past illnesses and injuries. Hospitalizations.

(g) Separations lasting a week or more. Nature of substitute care.

(h) Schooling history. Ease of attendance. Educational progress.

7 Observation of a child’s behaviour and emotional state.

(a) Appearance. Signs of dysmorphism. Nutritional state. Evidence of neglect, bruising, etc.

(b) Activity level. Involuntary movements. Capacity to concentrate.

(c) Mood. Expression of signs of sadness, misery, anxiety, tension.

(d) Rapport, capacity to relate to clinician. Eye contact. Spontaneous talk. Inhibition and disinhibition.

(e) Relationship with parents. Affection shown. Resentment. Ease of separation

(f) Habits and mannerisms.

(g) Presence of delusions, hallucinations, thought disorder.

(h) Level of awareness. Evidence of minor epilepsy.

8 Observation of family relationships.

(a) Patterns of interaction – alliances, scapegoating.

(b) Clarity of boundaries between generations: enmeshment.

(c) Ease of communication between family members.

(d) Emotional atmosphere of family. Mutual warmth. Tension, criticism.

9 Physical examination of child.

10 Screening neurological examination.

(a) Note any facial asymmetry

(b) Eye movements. Ask the child to follow a moving finger and observe eye movement for jerkiness, incoordination.

(c) Finger-thumb apposition. Ask the child to press the tip of each finger against the thumb in rapid succession. Observe clumsiness, weakness.

(d) Copying patterns. Drawing a man.

(e) Observe grip and dexterity in drawing.

(f) Jumping up and down on the spot.

(g) Hopping.

(h) Hearing. Capacity of child to repeat numbers whispered two metres behind him.