Introduction

Although obsessivecompulsive disorder (OCD) has long been considered as an adult disorder, the child psychiatric literature contains early descriptions of typical cases. At the beginning of the twentieth century, Janet reported on a 5-year-old with classical obsessivecompulsive symptoms, and Freud described in his adult patients obsessional behaviours dating back from childhood, while speculating on the strong constitutional influence in the choice of these symptoms. In 1957 Kanner(1) noted the resemblance and sometimes the association between compulsive movements and tics, and in 1955 Despert(2) described the first large series of obsessivecompulsive children (N = 68), noting the preponderance of males and the children's perception of the abnormality and undesirability of their behaviours. Tics have been described since antiquity, but the first systematic reports are those of Itard, in 1825, and Gilles de la Tourette, in 1885, who noted the association between tic disorders and obsessivecompulsive symptoms, and speculated on the hereditary nature of the syndrome. Over the past two decades, there has been a tremendous growth of interest and research on OCD and tic disorders. Significant advances have occurred regarding the phenomenology, epidemiology, genetics, neurophysiology, pathogenesis, and treatment of both disorders. The frequent association of OCD with tic disorders, as well as with other neuropsychiatric disorders, has led to a fascinating aspect of current neurobiological researchthe possible localization of brain circuits mediating the abnormal behaviours. OCD and tic disorders now appear as model neurobiological disorders to investigate the role of genetic, neurobiological, and environmental mechanisms that interact to produce clinical syndromes of varying severity.

Clinical features
Obsessivecompulsive disorder A number of systematic studies have been conducted over the past 10 years on children and adolescents with OCD, both in clinical settings and in the community. They have greatly increased our knowledge of the disorder in its early stage and shown that, in contrast with other forms of psychopathology, the specific features of OCD are essentially identical in children, adolescents, and adults. Obsessions are persistently recurring thoughts, impulses, or images that are experienced as intrusive, inappropriate, and distressing, and that are not simply excessive worries about realistic problems. Compulsions are repetitive behaviours or mental acts that a person feels driven to perform according to a rigidly applied rule, in order to reduce distress or to prevent some dreaded outcome. Obsessions and compulsions are egodystonic, considered by the subject himor herself as irrational or unrealistic, and at least partly resisted. They may be kept secret for a long time, or will only appear at home or in the presence of close family members. They are always source of psychological distress, and interfere with personal and occupational functioning, social life, and relationships to others.

The nature and frequency of obsessivecompulsive symptoms during childhood and adolescent years have been documented in various countries and various cultures. In a series of 70 young patients studied at the National Institute of Mental Health (NIMH) in the United States,(3) obsessions dealt primarily with fear of dirt or germs (40 per cent), danger to self or a loved one (24 per cent), symmetry (17 per cent), or scrupulous religiosity (13 per cent); the major presenting ritual symptoms included washing (85 per cent), repeating (51 per cent), checking (46 per cent), touching (20 per cent), ordering (17 per cent), counting (18 per cent), and hoarding (11 per cent). Toro et al.(4) described a series of 72 children and adolescents with OCD in Barcelona, for whom the most common compulsions were repeating (74 per cent) and cleaning rituals (56 per cent). For Khanna and Srinath,(5) in India, obsessions were less frequently reported by obsessivecompulsive children than by adults, with fear of harm being the single theme which occurred most often. In a group of 61 obsessivecompulsive children in Japan, the most common obsession was dirt phobia and the most common compulsion was washing.(6) Typically, children and adolescents with OCD experience multiple obsessions and compulsions, whose content may change over time.(3,7) However, in some community-based samples of adolescents, there are high proportions of individuals with either obsessions only or compulsions only.(8,9) Tics Tics are defined as sudden rapid recurrent non-rhythmic stereotyped movements, gestures, or utterances, which may affect any part of the body, and typically mimic some aspects or fragments of normal behaviour. They vary greatly in nature, location, number, intensity, forcefulness, and frequency. Tics may be simple or complex. Simple motor tics are fast, darting, and meaningless muscular events, while complex motor tics are slower and resemble purposeful actions. Comings and Comings(10) have studied the nature and frequency of motor and phonic tics in 250 subjects with Tourette's disorder: the most common simple motor tics were neck jerking (63 per cent), eye blinking (56 per cent), elevation of shoulders (32 per cent), mouth movements (38 per cent); the most common vocal tics were throat clearing (56 per cent) and sniffing (21 per cent). Other simple vocal tics may involve sucking air, grunting, snorting, humming, barking. Frequent complex motor tics include facial movements, jumping, gyrating, touching, kicking, grooming behaviours, and echokinesis. In a small fraction of cases, the complex motor tics are self-injuring behaviours, which may be potentially dangerous. Complex vocal tics are coprolalia (explosive utterance of dirty words), echolalia (repeating the last sound heard), palilalia (repeating one's own words or sounds), or repeating phrases out of context. Initially, tics have been decribed as irresistible, but they can actually be suppressed for varying lengths of time (some children can suppress their tics at school or during specific activities); however, this suppression usually produces a secondary exacerbation of tics (at home or when the child is alone). Tics are markedly diminished during sleep. They are classically exacerbated by stress and tiredness, and attenuated during absorbing activities such as reading or counting.

Many young children are completely oblivious of their tics or experience them as wholly involuntary movements or sounds. By the age of 10 years, most are aware of premonitory urges, which may be experienced as a sensory perception or a mental awareness in a specific region of the body where the tic is about to occur; this unpleasant phenomenon is momentarily relieved by the performance of the tic.(11) Children and adolescents with tic disorders may present a broad array of behavioural difficulties including obsessivecompulsive symptoms, disinhibited speech or conduct, impulsivity, distractiblity, and motor hyperactivity.(11) The presence of motor and/or phonic tics is associated with difficulties in self-esteem, self-definition, family life, peer acceptance and relationships, and sometimes school performance. Age of onset Reports of the mean age at onset of OCD in children and adolescents have ranged from 9.0 years in referred subjects(12) to 12.8 years in a community sample.(7) In the NIMH study, boys tended to have an earlier (prepubertal) onset, whereas girls were more likely to have a later (pubertal) onset.(3) The onset of tics usually occurs between 2 and 15 years of age, with a mean at about 7 years.(13) Children between the ages of 7 and 11 years appear to have the highest rate of tics. Phonic tics usually begin 1 or 2 years after motor symptoms. Sex ratio In community-based samples of adolescents with OCD, there are approximately equal numbers of males and females, while in most studies of referred children and adolescents with OCD, males outnumber females by 2:1 or 3:1;(14) in addition, boys are more likely than girls to have a comorbid tic disorder.(15) Most studies show a preponderance of tic disorders in males, with a sex ratio around 3:1;(9) for Tourette's disorder, the male-to-female ratio, estimated in the community, may be as high as 9:1.(16,17) Comorbidity In referred children and adolescents with OCD, the frequency of a diagnosis of any tic disorder ranges from 17 per cent to 40 per cent, and that of Tourette's disorder from 11 per cent to 15 per cent.(18) Conversely, Kano et al.(13) found, among 157 Tourette's disorder patients, 29 per cent with obsessivecompulsive behaviours. In longitudinal studies, about 50 per cent of children and adolescents with Tourette's disorder develop obsessivecompulsive symptoms or OCD by adulthood,(19) whereas, in a follow-up study of children and adolescents initially treated for OCD, nearly 60 per cent were found to have a lifetime history of tics that ranged from simple, mild, and transient tics to Tourette's disorder, for which the rate was 11 per cent.(15) On the basis of personal or family history of tics, a distinction has been proposed between ticrelated OCD' and non-tic-related OCD', under the assumption that the two forms might differ in terms of clinical phenomenology, neurobiological concomitants, and responsiveness to pharmacological interventions.(20) Tic-related OCD appears to have an earlier onset, and to occur more frequently in boys than in girls. The need to touch or rub, blinking and staring rituals,

worries over symmetry and exactness, a sense of incompleteness, and intrusive aggressive thoughts and images, are significantly more common in tic-related OCD, whereas contamination worries and cleaning compulsions are more frequent in patients with non-tic-related OCD. Overall lifetime psychiatric comorbidity in children and adolescents with OCD is about 75 per cent, both in referred and in community cases. In the NIMH sample,(3) depression and anxiety disorders were the most common comorbid conditions, occurring in 35 per cent and 40 per cent of the group, respectively; disruptive behaviour disorder or substance abuse were seen in 33 per cent of the sample. In the study by Toro et al.,(4) the most frequent associated diagnosis was an anxiety disorder (42 per cent) or an affective disorder (37 per cent), and anorexia nervosa was seen in 8 per cent. A wide range of impulsive and attentional symptoms are frequently observed in patients with Tourette's disorder. However, there are divergent findings regarding comorbidity of Tourette's disorder with attention-deficit hyperactivity disorder (ADHD); some studies report elevated personal or family rates(21) while others do not.(22) Some authors have suggested that individuals with Tourette's disorder may have weaker capacities for attention and impulse control, but at a subthreshold level for an ADHD diagnosis.

Classification
Obsessivecompulsive disorder Both DSM-IV(23) and ICD-10(24) define OCD, regardless of age, by obsessions and/or compulsions (criterion A), which are described, at some point during the course of the disorder, as excessive or unreasonable (criterion B), and are severe enough to cause marked distress or to interfere significantly with the person's normal routine, or usual social activities or relationships (criterion C). The specific content of the obsessions or compulsions cannot be restricted to another Axis I diagnosis, such as an eating disorder, a mood disorder, or schizophrenia (criterion D). DSM-IV adds that the disturbance is not due to the direct physiological effects of a substance or a general medical condition (criterion E). ICD-10 allows subclassification of forms with predominant obsessions, predominant compulsions, or mixed symptoms. In DSM-IV, the only difference in diagnostic criteria between children and adults appears in criterion B; although most children and adolescents actually acknowledge the senselessness of their symptoms, the requirement that insight is preserved is waived for children. Tic disorders In both DSM-IV and ICD-10, tic disorders are divided into four categories, according to duration of the symptoms and presence of vocal tics in addition to motor tics: Tourette's disorder, chronic motor or vocal tic disorder, transient tic disorder, and tic disorder not otherwise specified. By definition, all tic disorders must have onset before age 18 years (criterion D). In all, the disturbance causes marked distress or significant impairment in social, occupational, or other important areas of functioning (criterion C in DSM-IV), and is not due to the direct physiological

effects of a substance or a general medical condition (criterion E in DSM-IV, and criterion C in ICD-10). Transient tic disorder is defined by single or multiple motor and/or vocal tics (criterion A) that occur many times a day, nearly everyday for at least 4 weeks, but for no longer than 12 consecutive months (criterion B). Both Tourette's disorder and chronic motor or vocal tic disorder each have a duration of more than 1 year, with no tic-free period of more than 3 months (criterion B). In Tourette's disorder, both multiple motor and one or more vocal tic have to be present, although not necessarily concurrently (criterion A), while in chronic motor or vocal tic disorder, either motor or vocal tics, but not both, have been present at some time during illness (criterion A). The major difference between ICD-10 and DSM-IV concerns the comorbidity between Tourette's disorder and OCD. In ICD-10, it is stated that OCD cannot be diagnosed if the patient meets Tourette's disorder criteria, while both diagnoses may be given simultaneously in DSMIV.

Diagnosis and differential diagnosis

Symptoms of OCD in children are clearly distinct from developmentally normal childhood rituals.(1) Normal rituals include bedtime rituals, not stepping on cracks, counting, having lucky and unlucky numbers, and wanting things in their right place. They are most intense in 4- to 8year-olds, they stress rules about daily life, help the child master anxiety, and enhance the socializing process. In contrast, obsessivecompulsive rituals are perceived, even by young children, as unwanted and irrational; they are incapacitating and painful, promoting social isolation and regressive behaviour. OCD must be distinguished from other anxiety disorders and, in some cases, from autism or schizophrenia. In phobias, subjects are preoccupied by their fears only when confronted to the phobogen stimuli, and, in separation anxiety disorder, fear of harm to parents or loved ones are part of persistent worries and behaviours which are not criticized by the child. Stereotyped movements and ritualistic behaviours are frequent in mental retardation and autism, but they convey no particular intentionality, and the child does not try to resist them. In schizophrenia, there are erroneous belief systems in several areas, but the subject does not criticize them and does not consider the subsequent behaviours to be abnormal. Tics should be differentiated from other types of abnormal movements, which can occur in numerous different congenital or acquired neurological and neuropsychiatric disorders (Sydenham's chorea, encephalitis, Huntington disease, tuberous sclerosis, mental retardation, autism, etc.). The term of secondary tics or Tourettism has been applied to these disorders, and the abnormal movements can be choreiform movements, dystonic movements, myoclonic movements, spasms, or stereotypies. The distinction between tic disorders and other disorders with abnormal movements is based on anamnesis, family history, observation, and neurological examination. Involuntary vocal utterances are uncommon neurological signs in the absence of a tic disorder. Some medications such as central nervous system stimulants (methylphenidate, amphetamine, pemoline, cocaine), antihistaminic and anticholinergic drugs, antiepileptics (carbamazepine, phenytoin), antipsychotics, and opioids may also produce or exacerbate tics.(26)

Sometimes, complex motor tics can be confused with other complex repetitive behaviours, such as stereotypies or compulsive rituals, and differentiation between them may be difficult. Unlike tics, stereotyped movements are often rhythmic and repeated endlessly over a discrete period of time. Complex motor tics are rare in the absence of simple tics. Typically, compulsions are quite complex, they are aimed at neutralizing the anxiety resulting from an obsession, and/or they are performed according to rules that must be applied rigidly. Both compulsive rituals and complex tics may be preceded by premonitory urges, which persist until the action is completed. In individuals with both Tourette's disorder and OCD, these symptoms are sometimes so closely intertwined that efforts to distinguish them would be futile.

Epidemiology
Tics might be one of the most common behavioural problems in childhood, but frequency estimates vary greatly. Community surveys indicate that between 1 and 13 per cent of boys, and between 1 and 11 per cent of girls manifest frequent tics, twitches, mannerisms or habit spasms'(27) Race and socio-economic status do not seem to influence the frequency of tics, while urban living may be associated with elevated rates. There are virtually no general population studies of transient tic disorder or chronic motor or vocal tic disorder. For Tourette's disorder, several studies have been conducted among school-age children in the United States, yielding widely divergent prevalence estimates, ranging from 2.87 per 10 000 in Monroe County, New York,(28) and 5.2 per 10 000 in the North Dakota Survey,(16) to 105 per 10 000 boys and 13.2 per 10 000 girls in a California school district.(17) These data indicate that children are five to 12 times more likely than adults to have a tic disorder. In adolescents, epidemiological studies using strict diagnostic criteria and structured clinical interviews have been conducted in several parts of the world, estimating the prevalence of OCD between 1 and 4 per cent, and the prevalence of Tourette's disorder at 4.3 per 10 000 (Table 1). In the largest epidemiological study on juvenile OCD, conducted in a North American population of 5596 high-school students, (7) the current prevalence rate of OCD in adolescents was estimated to 1 ( 0.5) per cent, and its lifetime prevalence rate to 1.9 (0.7) per cent. The study showed that the disorder was clearly underdiagnosed and undertreated in this age group, none of the cases identified having been previously diagnosed, and only 20 per cent ever treated for (comorbid) psychological problems. In a later study,(8) examining 562 consecutive inductees into the Israeli Army, the point prevalence of OCD was 3.6 (0.7) per cent. Of note was the high proportion of subjects with obsessions only (50 per cent), potentially less disruptive of everyday functioning. If the prevalence of OCD was estimated excluding those individuals with only obsessions, the point prevalence dropped to 1.8 per cent. Among the OCD cases, there was a significant elevation of tic disorders (Tourette's disorder 5 per cent, chronic multiple tics 10 per cent, transient tics 10 per cent, versus 0 per cent, 0.9 per cent, and 1.1 per cent respectively in individuals with no diagnosis of OCD). In two longitudinal studies following cohorts of children in the community up to the age of 18 years, one from the United States found a lifetime prevalence rate for OCD of 2.1 per cent,(29) and the other from New Zealand found an overall 1year prevalence rate of 4 per cent, but only 1.2 per cent when subjects with obsessions only were excluded.(9) Th us, it appears that OCD might be as frequent in adolescents as it is in adults (see Chapter 4.8).

Table 1 Epidemiological studies of OCD and Tourette's disorder in adolescence (studies using structured diagnostic interviews)

A study conducted amongst the total population of inductees into the Israeli Defence Force over a 1-year period estimated the point prevalence of Tourette's disorder at 4.9 per 10 000 males and 3.1 per 10 000 females.(30) The point prevalence of OCD among subjects with Tourette's disorder was significantly elevated (41.7 per cent) compared with those without (3.4 per cent).

Aetiology
Psychological factors Psychological theories of OCD have encompassed psychoanalytic as well as more general nonpsychodynamic aetiological approaches, focusing alternatively on volitional, intellectual, and/or emotional impairment. Freud's famous patient, the Rat Man, has been seen as a paradigm of a psychologically determined illness, illustrating the central role of anal sadistic concerns with control, ambivalence, magical thinking, and the salience of defences such as reaction formation, intellectualization, isolation, and undoing.(31) Freud went on to formulate a theory of pregenital organization of the libido, determined by constitutional rather than experimental factors, and crucial to the obsessional neurosis. He also provided fascinating speculations on the similarity between obsessivecompulsive phenomena, children's games, and religious rites. Later, Anna Freud(32) stated that: obsessional outcomes are promoted by a constitutional increase in the intensity of the anal-sadistic tendencies probably as the result of inheritance combined with parental handling'. However, despite the beautifully described dynamics of obsessional symptoms, most illustrative of unconscious processes, the psychoanalysts have also pointed out the extreme difficulty in treating severe OCD with classical analytic treatment. Even though psychological factors are insufficient to cause Tourette's disorder, tic behaviours have long been identified as stress-sensitive conditions, and an intimate association has been noted between the content and timing of tics and dynamically important events in the lives of children. Biochemical factors Although a variety of biological aetiologies have been proposed in OCD since 1860,(33) modern neurobiological theories began with the clinical studies showing that clomipramine and other serotonin reuptake inhibitors (SRIs) had a unique efficacy in treating the disorder. This inspired a serotoninergic hypothesis' of OCD (see Chapter 4.8). In children, the involvement of the serotonin system in the pathophysiology of OCD is supported by one study in which improvement of obsessivecompulsive symptoms during clomipramine treatment was closely correlated with pretreatment platelet serotonin concentration,(34) and reports of decreased density

of the platelet serotonin transporter in children and adolescents with OCD but not in those with Tourette's disorder.(35) However, the delayed and incomplete action of serotonergic drugs, suggesting multiple effects on other neurotransmitters as well, and numerous biochemical studies of OCD patients and controls have not yet indicated a single biochemical abnormality as a primary aetiological mechanism in OCD. In Tourette's disorder, multiple neurochemical systems have been implicated by pharmacological and metabolic studies, but dopaminergic involvement is a prime candidate, especially in view of the dramatic improvement of tics with dopamine receptor antagonists. Furthermore, the frontal subcortical circuits seem to represent the anatomic region implicated in the disorder, and dopamine plays an important role in the neurotransmission of those areas.(36) Genetic factors Several studies have shown that OCD is much more common among relatives of individuals with OCD than would be expected from estimated occurrence rates for the general population.(37) Lenane et al.(38) investigating 147 first-degree relatives of children and adolescents with OCD found that 44 per cent of the families had a positive history of tics in at least one first-, second-, or third-degree relative. Pauls et al.(39) reported that the prevalence rates of OCD and tic disorders were significantly greater among the first-degree relatives of 100 probands with OCD (10.3 per cent and 4.6 per cent respectively) than among relatives of psychiatrically unaffected subjects (1.9 per cent and 1.0 per cent). It has been suggested that at least some forms of OCD could be genetically related to Tourette's disorder. The pattern of vertical transmission among family members in Tourette's disorder has led to specific genetic hypotheses favouring models of autosomal dominant transmission.(40) However, no genetic linkage studies have yet shown the role of a specific gene in the expression of the disorder, even though more than 60 per cent of the genome has been examined so far. Dysfunction of the frontalsubcortical circuits It has been known for a long time that obsessivecompulsive symptoms could be associated with neurological disorders of motor control, including Tourette's disorder, Huntington's disease, Parkinson's disease, as well as traumatic or infectious lesions of the basal ganglia.(33) Advances in neuroimaging have allowed the study of brain functioning in OCD patients and controls. Studies using single-photon electroencephalography, positron emission tomography, and functional magnetic resonance imaging have generally demonstrated metabolic abnormalities in the circuits involving orbitofrontal/cingulate cortex and the basal gangliamost particularly the caudate nucleiin obsessivecompulsive patients. Studies done at rest and during symptom provocation demonstrated selective increases in regional blood flow in the caudate and orbitofrontal cortex, which correlated with symptom intensity.(41) In Tourette's disorder, morphological studies have provided evidence of volume differences within the basal ganglia, and functional imaging studies have identified decreased regional metabolic activity in frontal, cingulate, and insular cortices.(36) Autoimmune factors

Recently, a strong association has been demonstrated between OCD and Sydenham's chorea, a childhood movement disorder associated with rheumatic fever which is thought to be a result of an antineuronal antibody-mediated response to group A -haemolytic streptococcus (GABHS), directed at portions of the basal ganglia.(41) OCD or some of its symptoms are seen in 70 per cent of Sydenham's chorea cases. Swedo et al.(42) have also documented post-streptococcal cases of OCD and/or tics in children and adolescents, without the neurological symptoms of Sydenham's chorea, giving them the acronym of PANDAS (paediatric autoimmune neuropsychiatric disorders associated with streptococcal infections). Therapeutically, this finding of a probable autoimmune-caused OCD raises the clinical possibility that immunosuppressant and even antibiotic treatments will be effective in treating or preventing some cases of OCD. An antigen labelled D8/17, on the surface of peripheral blood mononuclear cells has been shown to be a marker for the genetic tendency to generate abnormal antibodies to GABHS. Two independent groups of researchers have found a greater expression of the D8/17 antigen in the B lymphocytes of patients with childhood-onset OCD or Tourette's disorder compared with healthy controls, indicating that the presence of the D8/17 antigen may serve as a marker of susceptibility for OCD.(39)

Course and prognosis

Several retrospective or prospective follow-up studies of subjects treated for OCD during childhood or adolescence have looked at the outcome of the disorder in early adulthood.(14) The design and results of the main studies, in Europe and the United States, are summarized in Table 2. In the first study,(43) where no long-term treatment was specified, only 13 per cent of subjects had recovered at follow-up. In two studies(44,45) in which subjects received non-specific psychological and/or drug treatment, the recovery rate was still poor (2830 per cent). By the time patients had access to specific treatment with SRIs and/or cognitivebehavioural therapy, recovery rates increased to 57 to 65 per cent,(46,47 and 48) although, in one study, many of the symptom-free subjects were still taking medication at follow-up.

Table 2 Main follow-up studies of children and adolescents with OCD

All studies clearly demonstrate the continuity of the diagnosis of OCD from childhood to adulthood: when subjects are still symptomatic, the main diagnosis is almost invariably OCD, although comorbid disorders, especially mood and/or anxiety disorders, are frequent. Evolution towards psychosis is exceptional (and possibly the result of initial misdiagnosis). Spontaneous course is most often marked by a waxing and waning severity of the disorder, whereas remissions under treatment may be followed by relapses, even after long periods of time. Several studies, both on the natural course of OCD and on clinical response to treatment interventions, have attempted to identify demographic or clinical features that may influence course of illness or outcome in OCD, with mainly negative or inconsistent results.(14) In short-

term pharmacological treatment, only one study found that males responded significantly better than females.(49) In the long-term study by Leonard et al.(46), a worse OCD outcome at follow-up was predicted by more severe OCD symptoms after 5 weeks of initial drug therapy, the presence of parental Axis I psychiatric diagnosis, and a lifetime history of tics at baseline. In the study by Bolton et al.,(47) good treatment outcome in adolescence predicted medium-term prognosis (between 1 and 4 years) fairly well, but it failed to predict long-term prognosis (between 9 and 14 years). The course and outcome of childhood tic disorders have been less formally studied. It is known that, in the majority of cases, the tic disorder will be transient (by definition, no longer than 12 consecutive months) or will tend to improve with age. Tourette's disorder is considered as a chronic disorder. Symptoms are often at their most severe during the first decade of the disorder, but tend to improve in late adolescence and early adulthood;(50) motor tics may be reduced in frequency and number, with most patients having a more or less stable repertoire that wax and wane over a reduced range of severity, and vocal tics may disappear. Chronic motor or vocal tic disorders usually have a similar course. Although the prognosis for tic disorders is generally good, the course in adulthood may vary:(51) in some cases the disorder disappears entirely, a significant minority (1530 per cent) continue to have severe symptoms into adulthood, and in rare instances the most severe expression of the disorder can be seen in adulthood. In tic disorders, poorer prognoses are associated with comorbid developmental and mental disorders, chronic physical illness, unstable and unsupportive family environments, or exposure to psychoactive drugs such as cocaine.(11)

Treatment
Evidence The treatment of paediatric OCD has changed dramatically over the past 15 years, with two modalities being empirically shown to ameliorate the core symptoms of the disorder: cognitive behavioural therapy and pharmacological treatment with SRIs. In Tourette's disorder, D2 dopamine antagonists have been used with relative success since the 1960s, and cognitive behavioural therapy techniques have been employed without formal assessment of efficacy. Psychopharmacological treatment Several randomized controlled clinical trials of pharmacological agents have been conducted in children and adolescents with OCD, demonstrating, as many more similar studies with adult patients, the selective and unique efficacy of SRIs (clomipramine, fluoxetine, sertraline) in the short-term treatment of the disorder; a few open studies have given preliminary positive results for other selective serotonin reuptake inhibitors (SSRIs) (fluvoxamine, citalopram, paroxetine). The design and main results of these studies are summarized in Table 3. Overall, it appears that:

Table 3 Percentage improvement from baseline on obsessivecompulsive symptoms measures during short-term pharmacological treatment of children and adolescents with

OCD

1. the anti-obsessional effect is independent of the presence of depressive symptoms at baseline; 2. it takes longer to appear than the antidepressant effect; 3. the therapeutic response occurs gradually over a few weeks or a few months; 4. the final response is most often incomplete, with a mean reduction from baseline on scales measuring intensity of obsessive compulsive symptoms between 20 and 44 per cent across measures and across studies. A commonly used rating scale for measuring severity and change in OCD is the Children's Yale Brown ObsessiveCompulsive Scale.(52) Practice parameters for the assessment and treatment of children and adolescents with OCD(61) can be summarized as follows. 1. An adequate therapeutic trial of clomipramine generally consists of dosages up to 3 mg/kg/day for 3 months. Anticholinergic side-effects, typical of tricyclic antidepressants, might be dry mouth, somnolence, dizziness, tremor, headache, constipation, stomach ache, sweating, and insomnia. Risks of toxicity include seizures and ECG changes (ECG monitoring is recommended). 2. Initial studies suggest that the SSRIs are safe, effective, and well tolerated in children and adolescents, with a side-effect profile similar to that seen in adults. Although there are differences between them, the most common described side-effects include nausea, headache, tremor, gastrointestinal complaints, drowsiness, insomnia, akathisia, disinhibition, agitation, and hypomania. 3. Systematic doseresponse data are not available for children, but side-effects generally appear dose-dependent. It is therefore recommended to start with a low dosage that is increased slowly up to the minimum dose found effective in adult patients (fluoxetine, 20 mg/day; sertraline, 50 mg/day; citalopram, 20 mg/day; paroxetine, 40 mg/day; no fixeddose studies for fluvoxamine), although much higher doses have been used in published studies (Table 3). 4. Despite differences in potency and selectivity, it is not known whether one SRI is more effective than another for treating childhood OCD. If there is no clinical response after 10 to 12 weeks of treatment with one SRI, switching to another is reasonable. 5. The optimal duration of maintenance treatment is unclear, since relapses are frequent when discontinuing medication. Anti-obsessional medication should be maintained for at least 12 to 18 months after a satisfactory response has been obtained. Once the decision is made to attempt reduction or discontinuation, the tapering should be gradual. In Tourette's disorder, the relatively selective D2-receptor antagonists have consistently been shown to be effective in 80 to 90 per cent of patients, with at least 50 per cent reduction of tic

severity.(62) Haloperidol has been used since the 1960s, but most patients discontinue their treatment because of emergent side-effects. In a placebo-controlled crossover study, Sallee et al.(63) evaluated the efficacy and safety of pimozide and haloperidol in 22 subjects aged 7 to 16 years. At equivalent dose, pimozide was superior to haloperidol for controlling tics, and haloperidol exhibited a threefold higher frequency of serious side-effects. In addition to the usual side-effects of neuroleptics, pimozide may cause ECG changes. The usual starting dose is 0.25 mg/day of haloperidol or 1 mg/day of pimozide. Increments (0.5 mg haloperidol or 1 mg pimozide) may be added at 7 to 14 days intervals, and daily dosages of 0.5 to 6 mg haloperidol, or 1 to 10 mg pimozide over a period of 4 to 8 weeks generally lead to adequate control of tic symptoms, which will often continue to wax and wane at a much lower level. The 2-adrenergic receptor agonist clonidine may be effective for a proportion of Tourette's disorder subjects, presumably via acute and chronic downstream effects on dopamine. Clinical trials indicate an average 25 to 35 per cent reduction in symptoms over an 8- to 12-week period.(64) Clonidine seems especially useful in improving attentional problems and ameliorating complex motor tics. Treatment must be started at low dose (0.05 mg in the morning), and slowly increased to 0.15 to 0.3 mg in several doses throughout the day. The major side-effects of clonidine are sedation, hypotension, dizziness, and a decrease of salivatory flow. For children and adolescents with OCD and comorbid tic disorder, SSRIs alone might have little anti-obsessional effect and there are reports suggesting that fluvoxamine and fluoxetine may exacerbate or even induce tics in some patients. In adults, a personal or family history of chronic tic disorder has been associated with a positive response to haloperidol augmentation of fluvoxamine.(65) For children and adolescents with OCD and comorbid tic disorder, combined treatment with an SSRI and a low dose of a dopamine antagonist is recommended. Cognitivebehavioural therapy Cognitivebehavioural therapy is regarded as the psychological treatment of choice for children and adolescents with OCD. In contrast to medication, where relapse is common when treatment is withdrawn, cognitivebehavioural therapy has been shown to be a more durable treatment, although booster sessions may be required from time to time. Treatment generally involves a three-stage approach, consisting of information gathering, therapist-assisted graded exposure and response prevention, and homework assignments. Interventions for children with predominantly internalizing symptoms also include relaxation training and cognitive training. Family needs to be involved in treatment, at varying extent according to individual situations. Cognitive behavioural therapy is usually implemented initially with 13 to 20 weekly individual or family sessions and homework assignments. Partial or non-responders may require more frequent visits, and out-of-office therapist-assisted training. A few open studies have shown beneficial effects of cognitivebehavioural therapy, alone or in addition to pharmacotherapy, in series of 14 to 15 children and adolescents with OCD;(66,67 and 68) post-treatment, symptoms were relieved entirely or reduced to a mildly incapacitating level in 50 to 86 per cent of cases. In tic disorders, the relative suppressibility of symptoms may have implications for cognitive behavioural therapy, and a variety of cognitive and behavioural approaches have been used. A battery of habit-reversal training techniques encompassing awareness training, self-monitoring,

relaxation training, competing response training (where a movement is performed that is opposite to a particular tic), and contingency management have been reported to reduce tic symptoms markedly in one study.(69) Most other reports are case reports. Many families say that some methods, temporarily useful, may lose their effectiveness over time. Management As described above, OCD and tic disorders are frequently chronic, if not lifelong, conditions, and most treatments, notably drug treatment, are suspensive but not curative. Therefore, when defining a treatment plan, clinicians should be aware that they embark on a long-lasting task. For OCD of significant severity, the initial treatment choices are cognitivebehavioural therapy or anti-obsessional medication, either alone or in combination. A recent panel of experts(70) favoured cognitivebehavioural therapy as the initial treatment of choice, especially in milder cases without significant comorbidity, whereas presence of comorbid depression, anxiety, disruptive behaviour, or insufficient cognitive or emotional ability to co-operate in cognitive behavioural therapy are indications for including an SRI in the initial treatment. The combination treatment might be more efficient than either one alone, and cognitivebehavioural therapy may reduce relapse rate in patients withdrawn from medication. If tics are a common childhood problem, only a small minority of cases find their way to clinics. Given the waxing and waning nature of tic disorders, usual therapeutic practice will initially focus on careful clinical observation, along with educational and supportive interventions, and pharmacological treatments are held in reserve. The decision about whether to treat and how to treat will depend on the primary diagnosis, and the degree of interference of tics with the child's development and functioning. Most simple tics occurring in the absence of severe functioning impairment respond to a simple explanation of the mechanisms. When tics are responsible of functional impairment, pharmacotherapy can provide a substantial source of relief. However, in order not to expose the subject to excessive unwanted side-effects, pharmacological treatment should not aim at complete disappearance of tics. Even if they are less efficacious alone, other less specific treatment modalities should not be neglected. Obsessivecompulsive and tic symptoms may have a profound impact on the inner life of subjects affected, and traditional psychotherapeutic approaches may be useful to help children and adolescents to address intrapsychic conflicts that affect or result from their illness. Some families become extensively involved in participating in compulsive rituals or reassuring obsessional worries, or on the contrary they may become mired in angry struggles with their symptomatic child. In all cases work with families on how to manage the child's symptoms and participate effectively in behavioural or pharmacological treatment is essential. The growing availability, in all countries, of patient/family lay associations on OCD and tics may be most useful to alleviate the discouragement and incomprehension created by these disorders and give access to appropriate treatment resources.

Possibilities for prevention

At present, there is no known preventive strategy individually targeted at and given for either OCD or tic disorders. However, early intervention and comprehensive treatment, for as long as needed, are certainly the best ways to prevent severe incapacitation, and possibly achieve complete recovery in some cases. Even when response to successive treatment efforts is less than optimal, the improvement in function and quality of life may be considerable. In families with one or several cases of OCD or tic disorder, clinicians should be attentive to the onset of similar symptoms in children and adolescents, and treat cases of newly occurring disorder early and vigorously. Although this might concern only a fraction of patients, evidence of the onset or exacerbation of OCD or tics associated with streptococcal exposure warrants standard antibiotic treatment and ongoing monitoring for recurrent infection.

Conclusions
Childhood OCD represents the disorder, in child psychiatry, whose clinical picture most closely resembles its adult counterpart. Despite a relative diversity, the symptom pool is remarkably finite and very similar to that seen in older individuals. Prevalence, comorbidity, and response to drug and behavioural treatment also appear similar across the lifespan. For tic disorders, there is continuity between child and adult presentations, but the disease is much more prone to resolve spontaneously or to be less disruptive in adulthood. Both disorders occur more often in males than in females and are likely to be linked to an array of neurobiological abnormalities, many of which remain to be understood. Invaluable benefits can now be obtained from available pharmacological and behavioural treatments, but complete remission remains uncertain and long-term management may be required for childhood-onset OCD and tics. One of the main problems we are still facing is the difficulty in predicting outcome. More research is needed to gain better therapeutic response, and to disentangle the role of several neurotransmitter systems and other pathogenic factors in disorders that may prove to be neurobiologically heterogeneous.

OXFOFRD 2
Obsessivecompulsive (anankastic) personality disorder (JLC) While DSM-IV labels this personality disorder as obsessivecompulsive personality disorder (Table 14), ICD-10 prefers the term anankastic, previously used in European psychiatry to refer to fearful, insecure, and compulsive individuals. The cardinal feature of this disorder is an exaggerated and pervasive attempt to control. Anankastic patients need to control those who are close to them, to control every uncertainty, and to control their own thoughts and emotions. The anankastic lacks an internal sense of security and tries to make the external world totally predictable. The anankastic is afraid of his own internal aggressive drives and avoids free emotional expression. Others perceive this kind of personality as characterized by inflexibility and stubborn inefficiency.

Table 14 DSM-IV diagnostic criteria for obsessivecompulsive personality disorder

Epidemiology

The prevalence of obsessivecompulsive personality disorders is about 1 per cent in community samples and up to 10 per cent in psychiatric patients, especially those with depressive and anxiety disorders. It is most frequent among males. Some obsessivecompulsive traits are sanctioned in some cultures, and a personality disorder should not be diagnosed unless the traits are markedly beyond the average for the culture.
Aetiology

Biological factors and learning seem to be involved in the aetiology of obsessivecompulsive personality disorder. The personality may be partly inherited.(102) Early psychodynamic theories linked obsessive personality to the anal phase of psychosexual development between the ages of 2 and 4, when libidinal drives come into conflict with parental attempts to socialize the child, especially in sphincter control and toilet training. Later psychoanalytic theory(103) emphasized earlier manifestations of the child's autonomy versus parental wishes. The expression of drives and emotions, including anger, is shaped by parental responses and may evoke shame and criticism. According to this theory, as children, obsessional patients were often praised for what they did as opposed to who they were. Feelings were relegated to the realm of weakness and shame. The child could avoid criticism by focusing on tasks and displacing anger. By adopting moralistic attitudes towards anger, the child gained affection and attention from the parents. This dynamic sequence is reinforced in societies which are strongly influenced by the Protestant work ethic, in families where individual emotions are subordinated to the group, and in societies in which open expression of emotions is discouraged.
Clinical picture

The behaviour of an obsessivecompulsive personality has been consistently described as one of orderliness. The patient is preoccupied with details, and pays attention to rules, procedures, schedules, and punctuality. Patients with obsessional personalities often produce their own detailed lists of symptoms and are annoyed if any item is neglected or misinterpreted. They repeat actions and check for mistakes, despite the inconvenience and annoyance that result from this behaviour. As a consequence, their conduct is frequently inefficient. For example, the combination of unproductive perfectionism and rigidity may lead to difficulty in finishing a written report on time because of excessive correction and rewriting. Since this striving for perfection and order is time consuming, other areas of their lives often appear disorganized. One room or one desk drawer may fall into disarray, or parts of their social or family lives may be disorganized.

People with obsessivecompusive personality focus on work and productivity. It is difficult for them to take vacations or even to have free time. They do not enjoy leisure activity, which they may consider a waste of time. Often, they need to take work home to alleviate their anxiety. Hobbies and leisure pursuits become formally organized activities. They insist on perfect performance of sports or games and transform them into a serious task requiring careful organization and hard work. Leisure activities may be an unpleasant experience for the others involved, owing to the insistence on rules and standards. Stubbornness is another characteristic of these people. They need things to be done in their way, and realistic arguments do not usually make them change their insistence. They need others to submit to their way of doing things, and often believe that no one can do the tasks as perfectly as they can. They give detailed instructions, insisting that their way is the only way of doing things, and are irritated if others suggest alternatives. Therefore, they generally insist in doing everything themselves and are unable to delegate, which increases their inefficiency at work. Paradoxically, their stubbornness is associated with doubt. Indecisiveness is a constant characteristic unless they have structured guidelines. They fear making mistakes or misjudgements, and delay repeatedly until they have enough data to take what they consider the only right decision. When rules do not dictate the correct answer to a problem or when procedures for tasks are not laid down, decision-making or task initiation may become a lengthy and painful process. People with this personality disorder are characterized by excessive conscientiousness and scruples. They are inflexible about matters of morality, ethics, or values. Moral principles and standards of performance have to be followed rigidly, and respect for authority and rules is absolute. Failure to do these things leads to irritation, anger, and self-criticism. These people are stingy and mean, and often live with standards far below their actual socioeconomic status. They dislike spending, believing that money should be saved in case of future difficulties. They have great difficulty in discarding worn-out or worthless objects, believing that they might be useful some day. They may hoard objects such as newspapers or broken appliances, even when they have no sentimental value. These people are humourless and lack spontaneity of emotional expression. Usually they do not express anger directly. However, they are often angry in situations in which they are unable to control the behaviour of themselves or others. Anger is generally manifested by indirect aggressive acts (such as leaving a small tip or not providing minor help when expected). Their management of anger is closely related to their attitude of dominancesubmission toward authority figures. They may be excessively submissive to a person in authority whom they respect, but obstructive with an authority figure whom they do not respect. The affect of the obsessive person is controlled and stilted. It is not flat or blunted, but constricted. They do not laugh or cry, and feel uncomfortable with people who express their feelings. Their mood is usually serious but may appear anxious or depressed. In a clinical interview they may sit in a stiff unnatural posture, and seldom make spontaneous comments about their emotions. They usually relate their history in a pedantic and circumstantial manner. If interrupted by a question from the doctor, they have to finish their monologue before answering.

When asked about feelings, they answer with lists of facts and circumstances. They can label emotions and feelings, but are unable to display them. In summary, obsessive personalities love order, neatness, and sameness, and hates novelty, spontaneity, and change. They need control, security, and certainty, and avoid creativity, art, and excitement. They mitigate anxiety by following strict rules and repress emotional expression by avoiding spontaneity. They fear their inner fragile and aggressive emotional world.
Course

Like other personality disorders, obsessivecompulsive personality disorder is present in early adulthood and tends to be persistent and constant. However, some adolescents with marked obsessive traits become warm, loving, and tender adults. On the other hand, intense obsessional traits in adolescence are occasionally a premorbid stage of schizophrenia (pseudoneurotic schizophrenia'). The developmental relationship between obsessivecompulsive personality disorder and obsessivecompulsive disorder is controversial. In the past it was suggested that most obsessivecompulsive personality disorder evolved to a full obsessivecompulsive disorder, indicating that the two syndromes were expressions of the same basic disorder. More recent investigations(104) indicate that most obsessivecompulsive disorder patients do not have a comorbid obsessivecompulsive personality disorder. A variety of psychiatric disorders may present in a patient with obsessive personality, but depressive and anxiety disorders are the most common, followed by phobic, somatoform, and obsessivecompulsive symptoms. Hypochondriacal syndromes are commonly found in obsessive individuals when they lose control of situations. Persons with this personality disorder may do well in jobs that demand working with detail, order, and structured procedures, and may adjust to interpersonal relationships with submissive spouses. However, they are particularly vulnerable to unexpected changes in their occupational and social environment. Late-onset depression is a common occurrence in obsessivecompulsive personalities.
Differential diagnosis

The main difficulty in diagnosing obsessivecompulsive personality disorder is to differentiate it from obsessivecompulsive disorder. The latter diagnosis is made when occupational and personal functioning is severely impaired as a consequence of doubt, indecisiveness, hoarding, or any other obsessive behaviour. In many, but not all, cases of obsessive personality, the traits and behaviours are egosyntonic and no resistance is present, in contrast with obsessivecompulsive disorder. The perfectionism of obsessive personalities may be present in narcissistic personality disorder. However, narcissistic individuals tend to believe that they have achieved perfection, while obsessive individuals tend to be highly critical of their own achievements.

Social detachment and the lack of empathy and warmth may suggest schizoid personality disorder. However, obsessive individuals constrain their emotional expression to keep control of a situation, while schizoids lack the fundamental capacity for affective display or intimacy. Not all individuals with obsessive traits have obsessivecompulsive personality disorder. Obsessive traits can be adaptive in some situations; it is only when they are maladaptive, inflexible, and persistently cause functional impairment that a personality disorder be diagnosed.
Treatment

Pharmacological treatment may be tried in patients with anxiety and distress due to intense doubts, indecisiveness, and scruples. Benzodiazepines may alleviate tension in these cases. Antidepressants with a serotonergic profile sometimes improve mood and global functioning. Psychological treatment, focusing on perfectionism, rigidity, scrupulousness, and intolerance of failure, is the main therapeutic approach. Repressed aggression, guilt, and dependency needs should be addressed using a psychodynamic approach.

OXFORD 3
Obsessivecompulsive disorder

Obsessivecompulsive disorder is characterized by intrusive and distressing thoughts, impulses, or images about possible harm coming to oneself or others. Thoughts with a similar content to the intrusions of obsessional patients (e.g. a young mother having an intrusive thought about dropping her baby) are common in the general population.(13) For this reason, it has been suggested that the key cognitive abnormality in obsessivecompulsive disorder is not the content of obsessional thoughts, but rather the way the thoughts are interpreted.(14) In particular, it would appear that obsessional patients interpret recurrent obsessional thoughts and impulses as a sign that something terrible will happen, for which they will be responsible. For example, the young mother mentioned above may think that because she had a thought of dropping her baby, she is very likely to do so, despite finding the idea repugnant. In order to prevent the feared consequences of their obsessional thoughts, patients engage in a wide range of putting right' acts including (when relevant) washing and checking.
Post-traumatic stress disorder

Surveys indicate(15) that unwanted, intrusive, and distressing memories and the other symptoms of post-traumatic stress disorder (avoidance of reminders and hyperarousal/numbing) are common immediately after traumatic events. Over the next few months many people recover but in a subgroup post-traumatic stress disorder becomes chronic. It is the latter group that normally present for treatment. Research indicates that chronic post-traumatic stress disorder is associated with appraising the traumatic event and/or its sequelae in a manner that would produce a sense of serious current threat to one's view of oneself and/or the world.(16) Examples are given in Table 1. There is also evidence that chronic post-traumatic stress disorder tends to be associated with a

fragmented memory for the traumatic event and that recovery is associated with developing a more coherent narrative.(16,17)

Table 1 Some examples of idiosyncratic, negative appraisals leading to a sense of current threat in post-traumatic stress disorder

Why do negative thoughts and beliefs persist? If the world is not as dangerous as anxiety disorder patients assume, why do they not notice this and correct their thinking? For many patients with chronic anxiety disorders, the persistence of their fears can seem strangely irrational, at least at first glance. Consider, for example, panic disorder patients who think during their panic attacks that they are having a heart attack. Before they come for treatment they may have had several thousand panic attacks, in each one of which they thought they were dying, but they are not dead. Despite what might appear to an outsider as stunning disconfirmation of their belief that a panic attack can kill, their thinking has not changed. Several factors that appear to prevent patients from changing their negative thinking are outlined below. Such factors are important because reversing them is likely to be a particularly efficient way of treating anxiety disorders.
Avoidance, escape, and safety-seeking behaviours

Early conditioning theorists identified avoidance of, and escape from, feared stimuli as important factors in the maintenance of anxiety disorders. It is easy to see how avoidance of a feared situation (e.g. a supermarket for an agoraphobic) or escape from the situation before a feared event (e.g. a panic attack) occurs could prevent phobics from disconfirming their fears. However, situational avoidance/escape is not so obviously relevant to non-phobic anxiety and some phobics (especially those with social phobia) regularly endure feared situations without marked improvement in their fears. Salkovskis(18) introduced the concept of in-situation safety behaviours to deal with this problem. In particular, Salkovskis suggested that while in feared situations most patients engage in a variety of (often subtle) behaviours that are intended to prevent, or minimize, a feared outcome. For example, cardiac concerned panic disorder patients may sit down, rest, and slow down their breathing during attacks and believe, erroneously, that performing these safety behaviours is the reason why they did not die. Experimental studies have confirmed that (a) anxious patients engage in safety behaviours(19) while in feared situations, and (b) dropping these behaviours facilitates fear reduction.(20,21) Recent work(22) has highlighted several other important features of safety behaviours. First, although termed behaviours', many are internal mental processes. For example, patients with

social phobia who are worried that what they say may not make sense and will sound stupid, often report memorizing what they have said and comparing it with what they are about to say, whilst speaking. If everything goes well, patients are likely to think It only went well because I did all the memorizing and checking; if I had just been myself people would have realized how stupid I was'. In this way their basic fear persists. Second, it is common for patients to engage in a large number of different safety behaviours while in a feared situation. Table 2 illustrates this point by summarizing the safety behaviours used by a patient who had a fear of blushing, especially while talking to men whom she thought other people would think were attractive. Third, safety behaviours can create some of the symptoms that patients fear. For example, responding to a feeling of breathlessness in panic attacks by breathing more quickly and deeply (hyperventilating) can enhance the feeling of being short of breath. Similarly, post-traumatic stress disorder patients who are concerned that unwanted intrusive recollections of the trauma mean they are going mad often try hard to suppress such recollections. Unfortunately, active suppression increases the probability that the intrusion will occur.(23) Fourth, some safety behaviours can draw other people's attention to problems that patients wish to hide. For example, a secretary who covered her face with her arms whenever she felt she was blushing discovered that colleagues in her office were much more likely to look at her when she did this than when she simply blushed. Finally, some safety behaviours influence other people in a way that tends to maintain the problem. For example, the tendency of social phobics to monitor continually what they have said, and how they think they come across, often makes them appear distant and preoccupied. Other people can interpret this as a sign that the phobic does not like them and, as a consequence, they respond to the phobic in a less warm and friendly fashion.

Table 2 Safety behaviours associated with a fear of blushing

Attentional deployment

Selective attention plays an important role in maintaining some anxiety disorders. Patients with panic disorder or hypochondriasis fear certain bodily sensations and symptoms, believing they indicate the presence of a serious physical disorder (heart attack, cardiac disease, cancer, etc.). Such patients have often had several medical investigations that indicate they do not have the physical illness(es) they fear, but they are not convinced. One reason appears to be that their fears lead them to focus attention on relevant parts of their bodies and, as a consequence of this attentional deployment, they become aware of benign bodily sensations that other people do not notice.(24) The presence of such sensations is then taken by the patient as evidence that a serious physical illness has been missed. (Hypochondriasis is classified as a somatoform disorder in DSM-IV(10) and as a somatization disorder in ICD-10.(25) However, it has many features in common with anxiety disorders and can be conceptualized as such for the purposes of psychological treatment.(26))

Social phobia appears to be associated with two attentional biases. First, when in feared social situations, patients with social phobia report becoming highly self-focused, constantly monitoring how they think and feel they are coming across, and paying less attention to other people. Reduced processing of other people means that social phobics have less chance to observe other people's responses in detail and, therefore, are unlikely to collect from other people's reactions information that would help them to see that they generally come across more positively than they think.(27) Second, there is some evidence that when social phobics do focus on other people, they are particularly good at detecting negative reactions(28) and are poor at detecting positive reactions.
Spontaneously occurring images

Spontaneously occurring images are common in anxiety disorders and also appear to play a role in maintenance. Patients with social phobia often report observer-perspective' images in which they see themselves as if viewed from outside.(29) Unfortunately, in their images they do not see what a true observer would see, but rather their fears visualized. For example, a teacher who was anxious about talking with colleagues in coffee breaks noticed that before speaking she felt tense around her lips. The tension would trigger an image in which she saw herself with a twisted and contorted mouth, looking like the village idiot'. At that moment, she was convinced everyone else thought she was stupid. Negative images are also used as information in other anxiety disorders. For example, obsessional patients who have images of committing a repugnant act (e.g. stabbing one's child) take the occurrence of the image as evidence that they are in danger of performing the act. Similarly, patients with post-traumatic stress disorder report that flashbacks increase the perceived likelihood of a future trauma.
Emotional reasoning

A further source of misleading information that can enhance patients' perception of danger is anxiety itself.(30) For example, social phobics often think they look as anxious as they feel but in general this is not the case. Similarly, generalized anxiety disorder patients often take feeling on edge as a sign that something bad is about to happen.
Memory processes

Some anxiety disorders are associated with a tendency for the selective recall of information that would appear to confirm the patient's worst fears. For example, high socially anxious individuals selectively recall negative information about the way they think they have appeared to others in the past when anticipating a stressful social interaction.(31) Similarly, patients with hypochondriasis selectively recall illness-related information.
Rumination

Anxious patients often spend protracted periods of time ruminating about negative things that could happen in the future and about how bad they would be. They may also ruminate about things that they feel have gone wrong in the past. Studies by Davey and Matchett(32) indicate that such rumination can enhance fear. There are several ways in which rumination might operate.

First, thinking about an event may directly increase its subjective probability. Second, selectively focusing on past negative events, feelings, and impressions may further enhance the perceived likelihood of future danger. Third, rumination is rarely focused on constructively processing perceived threats, but instead often seems to elaborate the threats or make them more abstract and hence difficult to deal with. For example, patients with post-traumatic stress disorder often ask themselves Could I have done something different?' during their traumatic event without thinking through in detail what their alternative options might have been, and how feasible they would have been at the time.

Treatment
Assessment interview Table 3 summarizes the main topics covered in the assessment interview. The aims of the interview are as follows: (a) to obtain a detailed description of the patient's fears and behaviour; (b) to identify maintaining factors; (c) to normalize the problem; (d) to develop a model of the problem that can be used to guide treatment.

Table 3 Summary of topics to be covered in assessment interview

The interview would start by asking the patient to provide a brief description of the main presenting problem(s). For example, intense anxiety attacks, anxious apprehension, and avoidance of places where the attacks seem particularly likely or would be embarrassing. The interviewer then obtains a detailed description of a recent occasion when the problem occurred or was at its most marked. This would include the situation (Where were you?', What were you doing?'), bodily reactions (What did you notice in your body?', What sensations did you experience?'), thoughts (At the moment you were feeling particularly anxious, what went through your mind? What was the worst that you thought might happen? Did you have an image/mental picture of that? How do you think you looked?'), behaviour (What did you do?'), and the behaviour of others (How did X react?', What did X say/do?'). Having obtained a detailed description of a recent occasion, the interviewer should check whether the occasion was typical. If not, further descriptions of other recent occasions should be elicited to provide a complete picture. Next a list of situations in which the problem is most likely to occur or is most severe is elicited (Are there any situations in which you are particularly likely to have a panic attack?'), together with information about modulators (Are there any things that you notice make the symptoms stronger/more likely to occur?', Are there any things that you've noticed make the symptoms less likely/less severe/more controllable?').

Possible maintaining factors should be identified, including the following:

avoidance of situations or activities (What situations/activities do you avoid because of your fears?') safety behaviours (When you are afraid that X might happen, is there anything you do to try to stop it happening?') attentional deployment (What happens to your attention when you are worried about X? Do you focus more on your body? Do you become self-conscious?') faulty beliefs (e.g. an obsessivecompulsive disorder patient, believing that thinking something can make it happen) attitudes and behaviour of significant others (What does Y think about the problem?' What does Y do when you are particularly anxious?') current medication.

There are several ways in which excessive use of both prescribed and non-prescribed medications can maintain anxiety disorders. For example, painkillers and tranquillizers can cause derealization and sleep disturbance respectively, and drinking before social occasions prevents disconfirmation of one's social fears. It is also important to assess patients' beliefs about the cause of their problems as some beliefs may make it difficult for patients to engage in therapy. For example, patients with post-traumatic stress disorder who think the best way of dealing with a painful memory is to push it out of their mind are unlikely to engage in imaginal reliving of the event until this belief is dealt with. Finally, a brief description of the onset and subsequent course of the problem should be obtained. This description should particularly focus on factors which may have been responsible for initial onset and for fluctuations in the course of the symptoms and is primarily used to make the development of the problem seem understandable to the patient. It is not always possible to obtain all the information needed for a cognitivebehavioural formulation in an assessment interview. Sometimes it is necessary to follow-up the interview with homework assignments in which the patient collects more information to clarify the formulation. For example, a hypochondriacal patient who was concerned that palpitations meant that she had cardiac disease was asked to record what she did each hour and how many palpitations she experienced. To her surprise, palpitations were not associated with exercise, as she expected, but rather were most common when she was sitting quietly, reading, watching television, or studying. This realization helped convince her that her problem may be disease preoccupation rather than a faulty heart.
OXFORD 6

Obsessivecompulsive disorder All the described surgical techniques have been used for treating obsessivecompulsive disorder. Up to 20 per cent of patients with this disorder do not respond to treatment and, because of its unremitting character, the decision to operate is based on chronicity as well as lack of

response.(11) Suicide risk is also high in this population, but it has not been quantified in reference samples. The placebo response is extremely low in obsessivecompulsive disorder. In two casecontrol studies of such patients undergoing four different procedures,(18,19) the operated patients functioned better and had fewer symptoms than control cases of similar severity who had not undergone surgery. Anxiety disorders Stereotactic subcaudate tractotomy and capsulotomy have been used for pure severe anxiety disorders such as panic disorder; however, the total number of procedures has been extremely low.

Legal and ethical considerations

Neurosurgery is not an available option by law in many countries (e.g. Germany) and in some American states (e.g. Oregon). Special legislation applies to many countries including the United Kingdom, where, however, the provisions in Scotland are different from those in England and Wales. In countries where there is no special legislation the selection of patients is regulated by a panel of experts which sits in the hospitals providing the surgery. In England and Wales, under Section 57 of the Mental Health Act, patients must be assessed by a panel of three representatives appointed by the Mental Health Commission. The panel certifies that the patients have given their free and informed consent and are likely to benefit from the procedure. This process can be cumbersome, at times resulting in fatal delays in providing the service.(20) The procedure is different in Scotland, in that Section 97 only applies to patients who are detained. The main ethical issue to be considered is the balance between adequate consent, adequate previous therapeutic attempts, and allowing treatment. This is dealt with differently in different societies, some of which have produced algorithms for the definition of a treatment-resistant population.(8,11) This balance would be affected by the conduct of appropriately controlled studies, especially in affective disorders where the chances of spontaneous remission are much higher than in obsessive compulsive disorder. Such studies could be carried out either using sham techniques (now possible for some procedures that can be carried out with a cobalt gamma knife) or by comparing with aggressive and prolonged pharmacological treatment.

Outcome
Comparisons of outcomes between different centres, and across different diagnostic categories, is made simpler by the fact that the scale traditionally used by those centres reporting on neurosurgery is easily applicable and similar in concept to the Clinical Global Impression scale. On the whole, the five categories can be collapsed into good outcome (I and II), some

improvement (III), and poor outcome (IV and V). Some authors report category III as being of poor outcome(7) and some report it as good outcome.(14) The scale has been modified to also take into account social functioning as an important outcome. It correlates well with self-reported measures of depression and anxiety,(7) with observer-rated scales,(21) and it has good inter-rater reliability.(22) On the other hand, cohort effects, different referral patterns, and different methods of outcome assessment (face-to-face, questionnaires, and hospital notes or telephone interview) make direct comparisons quite difficult. This is also shown by the difference in outcome over time with the same operation and the same team selecting patients, as demonstrated in Table 1.

Table 1 Published outcome data for neurosurgery

Overall, the impression is that capsulotomy and stereotactic limbic leucotomy have superior outcomes in the treatment of obsessive compulsive disorder, capsulotomy has a very significant impact in anxiety disorders, and stereotactic subcaudate tractotomy is the operation of choice in affective disorders and the only technique that has been used in bipolar affective disorder. Personality and social functioning The use of the older free-hand' techniques resulted in a considerable proportion of patients demonstrating troublesome postoperative changes. Since symptomatic improvement with poor social functioning would be considered to represent a poor outcome, social functioning has been included in the assessment of the effects of surgery. The only formal assessments of patients with obsessivecompulsive disorder(22) concluded that postoperative personality changes involved a return to mean subtest scores towards values seen in the general population. Early subjective reports for stereotactic subcaudate tractotomy suggest the presence of personality changes in 2 to 8 per cent of patients.(13) Neuropsychology All the studies show that there are no significant postoperative changes in global IQ measures. Detailed psychometric investigations report both improvements(27) and deficits(19,28,29) in particular tests. Altogether the studies demonstrate decreased performance in the early postoperative period followed by a recovery, but where the final general level of function is affected mostly by the disease state. Specific decrements in function depend upon the particular operative procedure (see Malizia(6) for a detailed review). Electroencephalogram and psychophysiology

Early postoperative increases in frontal slow-wave activity have been demonstrated in stereotactic subcaudate tractotomy and stereotactic anterior capsulotomy.(30,31) This may relate to a good clinical outcome, indicating that the lesions have a profound immediate effect on frontothalamic loops and that the extent of this effect has a predictive value on the final outcome, despite the fact that clinical improvement does not usually manifest itself for a number of months after surgery. Interestingly, in stereotactic limbic leucotomy the lesion site in the orbitofrontal cortex is selected by observing changes in autonomic response upon cortical stimulation with a probe at the time of surgery. The area of orbitofrontal cortex where these changes can be localized is precise, as moving the probe by 4 mm abolishes the responses. Furthermore, similar changes can be induced by stimulating particular areas of the anterior cingulate(32) and of the amygdala, the insula, and the hypothalamus, thus implying a widely distributed network of cortical modulation of autonomic function. Imaging
Structural

Patients with obsessivecompulsive disorder and anxiety disorders who had a stereotactic gamma anterior capsulotomy were assessed by magnetic resonance imaging.(33) Patients in whom the left-sided lesion was absent or minimal did not experience a good clinical outcome. In addition, a strong correlation was found between the adequacy of the lesion and the postoperative level of social functioning. In stereotactic subcaudate tractotomy (with an yttrium-rod insertion) prominent oedema, which caused a change in signal of between 23 and 48 per cent of the total brain volume, was observed 2 weeks postoperatively using magnetic resonance imaging.(34) However, the oedema does subside, and by 1 year only the lesion is visible with an estimated volume of between 200 and 1300 mm3. Loss of tissue and gliosis were reflected in significant ventricular enlargement, which had been measurable 3 months after surgery and had not progressed further at 6 months or 1 year. Structural magnetic resonance imaging is likely to remain a useful tool in verifying the extent and position of lesions.
Functional

The effects of stereotactic anterior capsulotomy have been examined using the 133Xe inhalation technique(35) and [11C]glucose positron emission tomography,(36) which have demonstrated medial frontal, orbitofrontal, and caudate nuclei decreases in brain metabolism postoperatively. These brain areas have all been demonstrated to have abnormal metabolism in obsessivecompulsive disorder, thus indicating that the capsulotomy affects relevant anatomical targets. Hexamethylpropylene amine oxide (HMPAO) single-photon emission tomography has been used to examine cerebral blood flow changes following stereotactic subcaudate tractotomy(6,37) in

patients with treatment-resistant affective disorders. Large decreases were observed in the orbitofrontal, frontal, anterior cingulate, and ventral striatal HMPAO signal 2 weeks after surgery. However, none of these were predictive of outcome. At 6 months there were significant decreases in HMPAO activity in low frontal (orbitofrontal) and low anterior cingulate cortex, while there were significant increases in the parietal lobe. These changes were mostly accounted for by outcome, whereby ventral anterior cingulate and frontal decreases were associated with a good outcome and parietal increases with a poor outcome. These findings are congruent with the notion that stereotactic subcaudate tractotomy affects brain anatomical circuits of human mood. Neurochemical While the mechanism by which lesions produce their clinical effects remains unclear, it is likely that brain neurochemistry and neuroendocrine parameters should change as a result of surgery. Changes in these indices have been investigated in relation to stereotactic subcaudate tractotomy and are comprehensively reviewed elsewhere.(6) The limitations of these studies are either due to the lack of healthy controls (applicable to studies of central cerebrospinal fluid and brain tissue) or to the common problem of relating peripheral chemical indices with changes in brain function. The most notable findings from these studies are a decrease in 5-HT1A binding in the superficial layer of the cortex of depressed patients when compared with surgical controls,(38) and an association between the extent of decreased urinary noradrenaline (norepinephrine)output 2 weeks postoperatively and outcome as measured at 6 months.(39)
OXFORD 8

Obsessivecompulsive disorders Expanded public attention to the issues of child molestation appears to have led to an increased incidence of individuals with obsessive compulsive disorders seeking evaluation driven by their fear that they may molest a child. These individuals deny sexual interest in children, and fail to show evidence of sexual interest in children in psychophysiological assessments. Therefore these patients should be treated for their obsessivecompulsive symptoms. Organic mental syndromes Organic brain disease resulting from strokes or brain injuries can profoundly effect the sexual behaviour of the patient. Since one of the primary functions of the cortex is to inhibit impulsivity, cortical injuries frequently lead to impulsive behaviour. Examination of paraphilacs for organic disease varies from treatment centre to treatment centre. When organic disease is suspected, outlying centres refer cases to larger medical facilities for more extensive evaluation. Consequently there have been no random studies of the prevalence of organic disease in paraphiliacs. However, specific medical centres report a relatively high occurrence of organic disease in paraphiliacs. Abnormal hormonal levels are found in 74 per cent of paraphiliacs, soft neurological signs in 27 per cent, chromosomal abnormalities in 24 per cent, seizure disorders in 9 per cent, dyslexia in 9 per cent, abnormal electroencephalograms in 4 per cent, major psychiatric disorders in 4 per cent, and mental retardation in 4 per cent.(15)

Factors suggesting the need for more extensive organic evaluations include the following.

The paraphilic individual reports altered states of consciousness or seizure-like symptoms prior to committing paraphilic acts. The paraphilic individual uses excess aggression during the commission of crime. The paraphilic act is atypical of this category of paraphilia. The paraphilic individual has abnormal body habits. The paraphilic individual's history suggests attention-deficit disorder, dyslexia, or mental retardation. The paraphilic individual displays sadistic or aggressive behaviour concomitant with problems of sexual identity or transvestic fetishism.

Aetiology and epidemiology

Aetiology Theoretical perspectives regarding the aetiology of paraphilias have evolved and shifted as Western society's understanding of mental disorders has increased. In the nineteenth century, paraphilic behaviour was considered sinful, as an expression of evil by the ungodly. As psychiatrists began to elucidate the organic factors causing mental illnesses, the idea of paraphilic behaviour strictly as the expression of sin gave way to the belief that organic disease was responsible for inappropriate sexual behaviour, such as that evolving from central nervous system complications of syphilis. By the early twentieth century, Freudian theory had identified paraphilic interests as inappropriately expressed unconscious conflicts. Dynamic theories have recently been combined with the cognitive-behavioural learning model, creating a more strongly unified theory to explain factors involved in paraphilic behaviour.
Combined dynamic and cognitive-behavioural model

According to psychoanalytic theory, initial paraphilic acts are direct consequences of misplaced sexual and aggressive drives. These drives would normally be channelled into appropriate gender-specific behaviour. However, in the paraphiliac, these sexual and aggressive drives are distorted by anxiety generated by fear of castration and separation from the mother. Inappropriate resolution of the Oedipal complex within the phallic stage of psychosexual development results in identification with the opposite gender parent. Identification with the opposite gender parent causes the child to make inappropriate object choices for libidinal cathexis. As a result, the child goes on to express a paraphilic interest that reflects his initial inappropriate identification with the opposite gender parent. The cognitive-behavioural learning model views the earliest choice or use of inappropriate objects or behaviour for sexual gratification as a somewhat random event, idiosyncratic to the individual, based upon the child's unique early experiences. Both theoretical orientations see the expression of such early interest in paraphilic objects or behaviour as resulting from the extent to which a child anticipates the consequence of acting on paraphilic interests. If the child is well socialized and has learned appropriate means of expressing sexual behaviour, or if anxiety or the fear of negative consequences resulting from expression of a paraphilic interest is dominant, the

initial paraphilic interest is not expressed by the youth. In contrast, when the expression of the earliest interest is unassociated with anxiety or anticipated negative consequences, the child expresses his first paraphilic behaviour. Often, the earliest experiences are found intriguing because of their novelty and uniqueness alone. Both the psychoanalytic and cognitive-behavioural models agree that, as time passes, the child's use of these paraphilic behaviours, fantasies, or images becomes paired and associated with the pleasurable experience of masturbation and orgasm. The association of paraphilic stimuli with sexual pleasure profoundly accelerates the child's use of paraphilic images, along with ensuring their repeated pairing and reinforcement by the power of genital pleasure and orgasm. Within the majority of cultures, sexual behaviour is considered to be a private activity that individuals should not discuss and that they should feel guilty for expressing. Aberrant sexual interests, which qualify as any sexual interests other than those that are traditionally accepted within the society, are especially proscribed. Because traditional societal rules and mores play such a large role in identifying what constitutes appropriate versus inappropriate sexual interest, the child learns that his interest in paraphilic stimuli and/or behaviour must be secretive, internalized, and concealed from others. The child with paraphilic interest surreptitiously begins to repetitively pair and associate paraphilic images with orgasm. When a child's paraphilic acts are discovered by others, and he or she is confronted by parents or authorities, the child is initially successful at denying or concealing his or her ongoing interest. The child becomes more clandestine in the use of paraphilic stimuli, with further pairings of each paraphilic stimulus with orgasm, along with added reinforcement from the excitement and anxiety of committing a forbidden act. By the time the young adult's paraphilic interests are apparent to others, there have been hundreds to thousands of associations between paraphilic stimuli and sexual pleasure. The young adult feels trapped by his or her paraphilic interest. Confrontation by others regarding the inappropriateness of the paraphilic behaviour is appreciated but, at the same time, the use of paraphilic stimuli may have become so ingrained that attempts by the individual to achieve sexual pleasure or orgasm without them are unsuccessful. The individual feels trapped by his own chronic use of paraphilic stimuli and unable to escape into satisfactory sexual experiences without the use of the paraphilic images, fantasies, or behaviour.(16)
The abused-abuser hypothesis

An intriguing hypothesis for the aetiology of child molestation is that being molested by an adult when a child or adolescent may subsequently lead the victim to become a victimizer; this is called the abused-abuser hypothesis for child sexual abuse. A number of theories have been proposed regarding how such victimization could cause the victimization of others.(17,18) If abused as a youth, the child may recall these experiences with great clarity and associate them with masturbation and orgasm, thereby developing an interest in childadult sex. An emotionally deprived child abused by an adult who views the child in a positive light and showers him with gifts could himself see the experience as personally reinforcing and therefore continue similar behaviour into adulthood by molesting children himself. A further explanation is identification

with the aggressor, in which a fearful victim learns to cope with victimization by becoming a child molester and thus no longer fears being abused. The abused-abuser hypothesis probably emanated from early studies in which child molesters, when questioned about their prior sexual experiences with adults, reported a high incidence of having been abused as children. However, subsequent studies of the early sexual experiences of incarcerated child molesters, compared with the early sexual experiences of incarcerated nonchild-molesting sex offenders and incarcerated non-sex offenders, revealed that the prevalence of abuse in childhood within all three groups was not significantly different. This suggests that other factors must be responsible for determining the eventual consequences to and for the victims of child molestation. The abused-abuser hypothesis appears to be more relevant when examining juvenile sex offenders. When juveniles molest younger children, those younger child victims are prone to carry out similar sexual behaviour with peers or even younger children. With juvenile victims, the abused-abuser hypothesis appears to have greater relevance to the aetiology of child molestation.
Ethnological model

Ethnology has provided another explanation for an individual's sexual interest in children.(19) The purpose of intercourse, from an ethnological point of view, is to get one's genes in the gene pool'. Therefore the average male tends to be attracted to individuals more feminine and somewhat younger than him to maximize the likelihood of sexual behaviour perpetuating his genes. Although sexual experiences with individuals older or more masculine than him may be sexually pleasurable, from an ethnological perspective such activity will not lead to procreation and the maintenance of the species. Therefore the average adult male will be more sexually attracted to females somewhat younger than him. However, the ethnological process is imperfect, with some individuals being attracted to adolescent females (ephebophilia) or younger children (paedophilia). According to ethnological theory, the majority of males will be attracted to adult females somewhat younger than they are, but there will always be some individuals who are attracted to adolescent or much younger females and males. Epidemiology The paraphilias are predominantly disorders of males, but do occasionally occur in females. The estimate is that the male-to-female ratio is approximately 30 to 1. The higher prevalence of paraphilias in males probably results from the higher testosterone levels in males compared with females, leading to greater sexual drive at an earlier age. Estimating the incidence and prevalence of paraphilic behaviour is problematic for various reasons. First, it is difficult to obtain a representative sample of paraphiliacs upon which to base such estimates. Second, it is difficult to determine if information such as frequency of paraphilic behaviour would be more accurate if gathered from samples obtained from clinical populations

or from the general population. Both groups present challenges in attempts to elucidate epidemiological characteristics of the paraphilias. For example, in clinical populations it is common to see paraphiliacs who engage in a variety of paraphilic behaviours, even if they restrict those behaviours to one type of paraphilic diagnostic classification. To compound this problem, most paraphiliacs seen within the clinical environment do not fully divulge the extent of their paraphilic activities because they are afraid of criminal recourse. In the general population the accurate reporting of paraphilic interests and behaviours is influenced by the fact that paraphilias are socially undesirable and therefore are under-reported. Furthermore, most paraphiliacs fear arrest and prosecution by the criminal justice system; therefore, whether they are being assessed in a clinical environment or are being questioned as part of a survey sample, paraphiliacs are extremely adept at hiding the nature of their paraphilic interests.(10) Despite the problems indicated above, several studies have described and clarified some of the epidemiological characteristics of the paraphilias. A study of 561 subjects voluntarily seeking assessment and/or treatment of their sexual interests within a psychiatric setting found that the average number of paraphilic crimes and victims was substantial.(20) Participants had been involved in 291 737 paraphilic incidents with 195 408 victims. Additionally, each of the 560 participants were diagnosed following assessment and evaluation. Categories of paraphilic diagnoses are as follows: 19 per cent were diagnosed as paedophiles interested in extrafamilial girls; 13 per cent were diagnosed as paedophiles interested in extrafamilial boys; 13 per cent were diagnosed as incestuous paedophiles with an interest in girls; 4 per cent were diagnosed as incestuous paedophiles with an interest in boys; 11 per cent were diagnosed as rapists of adult females; 12 per cent were diagnosed as exhibitionists; 5 per cent were diagnosed as voyeurs; 5 per cent were diagnosed as frotteurs; 3 per cent were diagnosed as transsexuals; 3 per cent were diagnosed as transvestites; 2 per cent were diagnosed as sadists; 2 per cent were diagnosed as egodystonic homosexuals; the few remaining participants had carried out minor paraphilic behaviours. Utilizing data gathered from samples selected from general populations, three studies have provided information on the incidence and frequency of paraphilic behaviour. One study was conducted to assess the frequency of men's erotic fantasies during masturbation and intercourse.(21) Of the 94 males in this study who reported sexual fantasies, 62 per cent fantasized about sexual encounters with young girls, 33 per cent fantasized about raping adult females, 12 per cent fantasized about sadomasochistic encounters, 5 per cent fantasized about zoophilic encounters, and 3 per cent fantasized about sexual encounters with young boys. The other two studies were conducted with college students and were undertaken to investigate paraphilic sexual interests. One of these studies involved 193 male participants who were questioned regarding sexual interest in children.(22) Of these 193 participants, 21 per cent reported sexual attraction to children, 9 per cent fantasized about sexual encounters with a child, 5 per cent reported masturbation fantasies of sex with children, and 7 per cent responded that there was a likelihood of their becoming sexually involved with children if they could find a way to avoid criminal consequences for such activity. The second study involved 60 participants who were questioned about paraphilic interests, especially child molestation.(23) Of the 60 participants, a total of 65 per cent of this sample indicated involvement in a paraphilia; 3 per cent reported sexual encounters with girls under 12, 42 per cent reported voyeurism, 8 per cent indicated participation in telephone scatalogia, 35 per cent in frotteurism, 2 per cent in exhibitionism, and

5 per cent in coercive sexual encounters. These studies are limited because the samples are homogeneous and the participants were not questioned about every form of paraphilic interest. However, the three studies strongly indicate that males often fantasize about paraphilic behaviour and that these same males report engaging in paraphilic activities.

Treatment
Cognitive-behavioural treatment For the last 10 to 15 years, treatment for paraphiliacs has focused on two primary areas: cognitive-behavioural treatment and pharmacological treatment. Cognitive-behavioural treatment has generally focused on the following:

techniques to reduce or block inappropriate sexual arousal and/or to increase or maintain nondeviant appropriate sexual arousal(6,24,25,26,27,28,29,30,31,32,33,34 and 35) improving pro-social behaviour, including assertiveness training, anger-management training, social skills training, intimacy skills training, etc.(36,37,38 and 39) cognitive therapy to address distorted thinking patterns that the paraphiliac has used to justify his inappropriate sexual behaviour, as well as the establishment of empathy for the individuals he has victimized(36,40,41,42 and 43) relapse prevention, a long-term maintenance therapy which helps offenders to identify and manage more effectively situations that place them at risk of reoffending, to establish or improve social support systems, to develop a methodology for evaluating the effectiveness of their treatment, and to maintain a balanced lifestyle.(44,45,46,47,48,49,50,51,52 and 53)

Pharmacological treatment Pharmacological treatment for the paraphilias has been increasingly important in the last 20 years. Three categories of medication have proved effective for treating paraphiliacs: antiandrogens, hormonal agents, and selective serotonin reuptake inhibitors (SSRIs). Antiandrogens Ciproterone acetate has proved quite effective because of its antiandrogenic, antigonadotrophic, and progestational effects.(54) This agent blocks intracellular testosterone uptake as well as the intracellular metabolism of antiandrogens. As a consequence, it drastically reduces circulating testosterone and therefore reduces the paraphiliac's sexual drive. The oral dosage is usually 50 to 200 mg/ day but it is also available in an intramuscular form, requiring a dosage of 200 to 400 mg once every 1 or 2 weeks. Side-effects include liver damage, gynaecomastia (usually temporary and reversible), and reduction of sexual drive, fantasies, erections, frequency of masturbation, and sexual intercourse. This drug was first used in 1971, and a number of studies have shown it to be highly effective at reducing recidivism.(55) Ciproterone acetate is available throughout Europe and Canada, but is not available in the United States. Hormonal agents

Medroxyprogesterone acetate is the primary hormonal agent that has been used in the United States, since initially reported by Heller et al.(56) Its effect results from the acceleration of testosterone-A-reductase in the liver which accelerates testosterone metabolism and thereby reduces testosterone levels. Medroxyprogesterone acetate also reduces plasma testosterone through the pituitary axis. It is not an antiandrogen. Significant side-effects have included liver damage, fatigue, weight gain, hot and cold flushes, headaches, gallbladder disease, diabetes, and thrombophlebitis. Historically, the dose was 300 to 400 mg of the injectable form of medroxyprogesterone acetate, but in recent years lower doses have been found to be equally effective without causing so many side-effects that the medication is discontinued by the paraphiliac. An alternative to injectable medroxyprogesterone acetate can be administered orally. Generally doses of less than 200 mg daily by mouth are effective at helping the paraphiliac gain control over his behaviour. More recently studies have been reported using luteinizing hormone-releasing hormone (LHRH) agonists, which initially accelerate the production of testosterone through the hypothalamicpituitary axis but then exhaust the axis and result in a dramatic reduction in testosterone to castrated levels. Since these drugs initially cause an acceleration of testosterone production, the non-steroidal antiandrogen flutamide is usually concomitantly administered at a dose of 250 mg three times daily for the first month of LHRH agonist use. After that, flutamide can be discontinued. The LHRH agonists have the advantage of not being true steroids, but polypeptides, and therefore do not cause many of the steroidal effects while still resulting in a dramatic reduction of testosterone and increased control over paraphilic urges. Since it was reported that SSRIs were effective in managing the treatment of exhibitionism, a number of authors have report their effectiveness in the treatment not only of other paraphiliacs, but also of those with hypersexuality.(57,58) The exact mechanism of action of the SSRIs is not completely understood, but it is suspected that their effectiveness results from a reduction of sexual drive and of the obsessive ruminations that accompany paraphiliacs' behaviour. The most extensively investigated SSRI has been sertraline, with the mean dose being 130 mg daily. Fluvoxamine, fluoxetine, and paroxetine have all been found to be effective in treating both the paraphilias and males with non-paraphilic hypersexuality. The main limitation of ciproterone acetate, medroxyprogesterone acetate, the LHRH agonists, and SSRIs is that they are only effective during clinical administration; to date, there is insufficient evidence to suggest persistent effectiveness following discontinuation of the medication. The SSRIs show great promise because of their greater ease of administration, lower cost, and lower side-effect profiles. These medications are traditionally prescribed as an adjunct to cognitive-behavioural treatment.
OXFORD 25

Obsessivecompulsive (anankastic) personality disorder The median prevalence rate of obsessivecompulsive PD, obtained from eight studies, was found to be 1.7 per cent. The rate of compulsive PD was especially high in a study in which the Personality Diagnostic Questionnaire was used (6.4 per cent).(3) However, lower rates were

reported with structured interviews. A community study, carried out at the Epidemiologic Catchment Area Program Baltimore site, found a prevalence of 1.7 per cent.(12) Males had a rate about five times higher than females. The disorder was also more frequent among white, highly educated, married, and employed subjects, and it was associated with anxiety disorders. However, the study derived the diagnosis from an interview originally not intended to diagnose PDs. This could mean that adaptive obsessivecompulsive traits, rather than a true' PD, were identified. Avoidant (anxious) personality disorder A total of seven studies have investigated the prevalence of avoidant PD in community samples, with a median prevalence rate of 0.7 per cent. Dependent personality disorder In eight studies in which the frequency of dependent PD was assessed, the median prevalence rate was 0.7 per cent. Passive-aggressive personality disorder The median prevalence rate of passive-aggressive PD, obtained from seven studies, was found to be quite high (1.7 per cent); interestingly, this type of PD has not been included either in DSMIV or in ICD-10.

Epidemiological studies of personality disorders carried out in psychiatric settings

Table 3 lists the median prevalence rates for any PDs found in 32 studies carried out in inpatient and outpatient psychiatric samples and published between 1981 and 1998. Only those prospective studies that surveyed homogeneous clinical samples (either inpatients or outpatients) of more than 100 subjects have been considered for this analysis. The second column shows the number of studies on which the median prevalence rate is based.

Table 3 Median prevalence rates of PDs among psychiatric patients in prospective studies including more than 100 subjects

In these studies subjects have been directly evaluated for the purpose of obtaining PD rates, by means of a standardized assessment instrument specific for PDs. Several other studies, which have evaluated only the prevalence of specified PDs in clinical samples, are not shown here.

In general, the prevalence of PDs among psychiatric outpatients and inpatients is quite high, with a majority of studies (n = 17) showing a PD prevalence rate higher than 50 per cent of the sample. However, it is difficult to draw more definite conclusions from these studies, because of substantial differences in sampling, diagnostic criteria, assessment methods, availability of mental health services, prevalence of Axis disorders, and sociocultural factors. There are, however, some consistencies across studies that deserve consideration. The most prevalent PD seems to be borderline, both in inpatient and in outpatient settings. The next most common PDs are schizotypal and histrionic. These three disorders are also characterized by the lowest social functioning. They are especially common in inpatient settings, as their symptomatology often results in the patient being admitted to hospital due to their suicidal behaviour, substance abuse, and cognitiveperceptual abnormalities. In outpatient settings, dependent and passive-aggressive PD are also common. Especially in inpatient settings, many people who meet the criteria for one PD also meet the criteria for other PDs.(45,46) The highest comorbidity rate appears to occur with borderline PD, with the frequent coexistence of borderline and histrionic PDs, followed by antisocial, schizotypal, and dependent PDs. With regard to comorbidity between PDs and Axis I disorders, the most common and beststudied patterns are between substance abuse and PDs, affective disorders and PDs, and anxiety disorders and PDs (particularly borderline, antisocial, avoidant, and dependent PDs). Other clinically significant associations have been found between bulimia nervosa and borderline PD, as well as between anorexia and avoidant PD.(47) High rates of PD (especially borderline and antisocial PDs) have also been detected in patients with selected medical conditions, such as HIV-positive patients.(48) Some studies have assessed the treated prevalence of PD using administrative data (e.g. discharge figures, psychiatric case register data, etc.). In the United States, using data from the 1993 National Hospital Discharge Survey, Olfson and Mechanic(49) found that almost 12 per cent of patients discharged from public general hospitals had a diagnosis of PD, compared with 11 per cent of patients from non-profit hospitals and 5 per cent of patients from proprietary general hospitals. In England and Wales, 7.6 per cent of all admissions and 8.5 per cent of first admissions over a 1-year period were diagnosed as having PDs.(50) Some investigations, which compared the hospital admission rates for PD over time, allow us to assess the impact of diagnostic changes. In Denmark, sex- and age-standardized rates of firstadmitted borderline patients significantly increased during the 16-year interval between 1970 and 1985, and this might be explained in terms of a change in diagnostic habits.(51) In the United States, comparing the diagnoses given to inpatients in a large university-affiliated mental hospital in the last 5 years of the DSM-II era (n = 5143) with those given in the first 5 years of the DSM-III era (n = 5771), a marked increase (from 19 per cent to 49 per cent) was found in the diagnosis of PD, together with a decrease in the diagnosis of schizophrenia and a corresponding increase in the diagnosis of affective disorders.(52)

The epidemiological findings in treated samples are especially important if we bear in mind that the presence of a PD among those suffering from other mental disorders can be a major predictor of the natural history and treatment outcome. Therefore an important clinical implication of these findings is that patients in treatment because of severe Axis I disorders must be carefully assessed with an assessment instrument specific for PDs, because of the high likelihood of diagnosing a PD and the subsequent need to adjust their treatment accordingly.

Epidemiological studies of personality disorders carried out in other settings

A few epidemiological studies on PDs have been carried out among patients attending primary healthcare settings; in these studies between 5 and 8 per cent of patients have been identified as having a primary diagnosis of PD.(45) When the assessment is made independently of the primary diagnosis, however, the average prevalence rate can rise several-fold because of state effects. There are also indications that people showing certain PDs are high users of medical services. In other institutional settings, such as prisons, several studies have found very high rates of PDs. In the United Kingdom, two large-scale studies have recently been completed; in the first, carried out among 750 prisoners representing a 9 per cent cross-sectional sample of the entire male unconvicted population, a PD was diagnosed in 11 per cent of the sample.(53) In the second study, a representative sample of the entire prison population of England and Wales was evaluated; a subsample was assessed with the SCID-II administered by a clinician.(54) The prevalence rates for any PD were 78 per cent for male remand prisoners, 64 per cent for male sentenced prisoners, and 50 per cent for female prisoners. High rates of borderline and antisocial PDs have also been found in a sample (n = 805) of women felons entering prison in a North American State.(55)

Conclusions
Up to 20 years ago, the epidemiology of PDs had not received the same amount of attention as that of many other psychiatric disorders. Since then the situation has changed, and we now have data on the prevalence of PD in the community and in psychiatric facilities. Community data come primarily from eight studies, with a total sample of 4518 subjects from three countries (Germany, the United Kingdom, and the United States). There are excellent national and crossnational epidemiological data on antisocial personality disorder based on the same diagnostic methods. There are almost no data on other PDs from countries other than the United States, the United Kingdom, and Germany. One important methodological problem is that some PDs have a very low prevalence rate. Consequently, epidemiological surveys carried out among the general population may require very large samples in order to identify a sufficient number of cases to study demographic correlates and the association of PD with other psychiatric disorders. Future studies should try to address this problem and provide us with more definite epidemiological data. These data will also be invaluable in showing the validity of current classifications and in better delineating the boundaries between different PDs.
OXFORD 28

Obsessivecompulsive disorder The characteristic features of obsessi vecompulsi ve dis-o rder (ocd) are obsessional thinking and compulsive Behavio r. Obsessive thinking includes recur rent persis-tent thoughts, impulses and images that cause ma rked Anxiety or dist ress. Compulsive behavior include repet-iti ve behavio r, rituals or mental acts done to p revent or R educe anxiety. O ther featu r es include indecisiv eness And inability to take action. Many patients with ocd Experience significant deg r ees of anxiety, depr ession And depersonalization (see figur e 5.10 ). Ocd is Uncommon in the general population, but minor Obsessional symptoms are fairly common. The 1-month prevalence rates are estimated to be about 1% For men and 1.5% for women ATLAS 1 ATLAS 6/79 Class 4 Obsessions (AUTISTIC SPECTRUM) Children with ASD usually have all-absorbing narrow interests which persist Over time to the exclusion of all other topics. When young, these interests may Centre around Thomas the Tank Engine and dinosaurs, but as they grow older These earlier interests are superseded by subjects which require an almost Encyclopaedic detailed knowledge about one or a few narrow areas of interest Such as timetables. They do not usually involve piecing information together in a Novel and creative way, and do not generally demand knowledge, experience or An attraction for human interactions involving empathy and imagination soap Operas and dramas high in emotional content are of little interest to children

With ASD. School interests are unlikely to include arts subjects which involve Subjective responses maths, science and IT are subjects generally preferred. A child with ASD will often talk about his interests in an obsessive way and Try to impose his topic of conversation. He is often only interested in others Insofar as they share his interests, and cannot understand that others do not Necessarily share a passion for these same hobbies. As he cannot read non-verbal Cues, he is frequently unaware when others have lost interest. Repetitive play and obsessive interests do undoubtedly help to reduce stress In children with ASD but they may be a substitute for real conversations and Imaginative play which the child finds difficult too much time spent on the Preferred subject will reduce the time available for practising other sorts of Interaction. Strategies Allocate specific times or places when the childs favoured topic of Conversation can be discussed. Deflect discussion at other times, gently but Firmly returning the child to the activity in hand. Give the child talk tickets as a reward for successfully completing a teacher-directed task which he can exchange for time to talk on his chosen topic. If possible, use the favourite topic to practise conversational skills, perhaps Including one or two other children. Try to extend the favourite topic and relate it to other similar topics even to The current teaching topic where possible. BURN Obsessive compulsive disorder Most of us have experienced obsessional behaviour as children avoiding the cracks in the pavement to avoid catastrophic

consequences is the commonest. Sportsmen and actors are notorious for such rituals the tennis player who has to bounce the ball three times before serving, the leading lady who cannot play without something green in her costume. These superstitious behaviours have much in common with obsessive compulsive disorder (OCD). In this disorder the patient has to repeat activities or thoughts (classically hand washing or checking and counting rituals) a set number of times or in a set order to ward off anxiety or feared consequences. In the obsessional form (where there are often no external rituals) the problem is repetitive thoughts, often about awful outcomes (contamination with dirt or germs, or a fear of shouting out something blasphemous or offensive). The hallmark of OCD is that the thoughts or actions are repeated , resisted, and distressing . It isnt a harmless superstition or quirk but can dominate and ruin lives. Compulsive cleaners, for instance, end up exhausted because they are never finished cleaning over and over again. Obsessional ruminators cant hold down a job because they are distracted with repeating their thoughts or counting and may wear out their partners as they seek constant reassurance about their worries. OCD tends to be associated with specific personality traits neat, tidy, conscientious. Most of us recognize obsessional features in ourselves and yet the full disorder seems so bizarre. Indeed, 23 What is psychiatry?

sufferers are often slow to seek help because they consider it so strange and incomprehensible they are embarrassed by it. It has been subject to psychological over-interpretation (Chapter 4) and only recently have effective treatments been developed (behaviour therapy and antidepressants in milder cases).

Obsessive-Compulsive Disorder (CHILD AND ADOLESCENE) Obsessive-compulsive disorder (OCD) is characterized by recurrent, time-consuming obsessions or compulsive behaviors that cause distress and/or impairment. The obsessions may be repetitive intrusive images, thoughts, or impulses. Often the compulsive behaviors, such as hand-washing or cleaning rituals, are an attempt to displace the obsessive thoughts. There is a strong familial component to OCD, and there is evidence from twin studies of both genetic susceptibility and environmental influences. Tic disorders and OCD may have similar genetic origins, as they tend to co-occur in families. There is evidence that about 10% of the individuals with OCD may have the symptoms precipitated by pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS). Although evidence remains equivocal, antineuronal antibodies P.123 formed against the group A beta-hemolytic streptococcal cell wall antigens may cross-react with caudate neural tissue. This should be considered for children whose symptoms correlate with recurrent strep infections. About half of all adults seeking treatment for OCD report that it began in childhood or adolescence. Epidemiology As a group, anxiety disorders affect up to 20% of youth up to age 18 years. Clinically, anxiety disorders are diagnosed equally in men and women, but in epidemiologic samples they are more frequently found in women. Generalized anxiety disorder is thought to affect 3 to 6% of youth. Specific phobias affect about 3% of children. Girls tend to suffer from phobias more commonly than boys, but for both genders the disorder wanes with age. Social phobia has been estimated to affect 1% of children and adolescents at any point in time. Lifetime prevalence may be as high as 13%. Etiology and Risk Factors Many investigators postulate that children are born with biologically or constitutionally predetermined temperaments, some of which are a liability for the development of anxiety disorders. Familial factors, both genetic and environmental, contribute to anxiety disorders. General anxiety disorder and major depression seem to have the same genetic risk factors. On the basis of new neuroimaging studies, it is hypothesized than anticipatory anxiety is associated with the cingulated portion of the limbic system, phobic avoidance is associated with the prefrontal cortex, and panic is associated with the brainstem. Serotonin receptor site dysregulation is also posited. The biological vulnerabilities are then variably affected by environmental factors to form clinically significant anxiety symptoms. The environmental factors may be diverse, including neurobiological insults, exposure to trauma, emotionally unavailable parents who are not attuned to the child's needs, or, for the most vulnerable youth, simple uncertainties such as

peer teasing or parental discord. Infants who are temperamentally inhibited have higher rates of anxiety disorders in later life. Prospective studies have shown an increased risk of multiple anxiety disorders in middle childhood for children who were classified as behaviorally inhibited as preschoolers. P.124

Clinical Vignette You are called as a consultant to a school to assess Jenn, a 13-year-old eighth-grade girl who has refused to come to school. You find out that Jenn has a long history of multiple absences in school, but this year she came only a few days at the beginning and now is not coming at all. Her mother reports that she has attempted to drag her daughter to the car, but Jenn screams and scratches. When she has managed to drive her to school, Jenn will not get out of the car, and even the school social worker and principal cannot persuade her. When you get more history you find that Jenn suffered from separation issues in preschool and kindergarten. The school nurse knows her well, as she visits frequently with complaints of stomachaches and headaches. She refuses to speak in class. She avoids the lunch room and is quiet and nonparticipative in class. Despite this, she had been getting straight A's in school until recently, as not coming to school has negatively impacted her grades. Homework sent to her home is done completely and neatly. Jenn has one friend with whom she spends time. You suspect social anxiety with school phobia. What else should you do? First, talk with the primary care doctor. Chances are the girl has been a frequent visitor there with a variety of somatic complaints. Does she have any medical problems? Try to ascertain if she has been traumatized (bullied, etc.) in school as a reason for her refusal. Get a family history and developmental history from the parents. Discuss the course of the difficulties with the school. Interview Jenn and ask about depressive, anxiety, and psychotic symptoms. If she is suffering from school phobia or social anxiety, consider medication (SSRI), psychotherapy, and an intensive, slow but progressive reintegration into school (often starting with tutoring after school). This is one of the few times that benzodiazepines are indicated in child and adolescent psychiatry. Rapid and effective treatment of the anxiety to help the child be able to get back to school will substantially improve prognosis. Assessment When assessing the child or adolescent for whom you suspect an anxiety disorder, it is important to consider other psychiatric disorders, as well as potential for comorbidities (Table 15.1). Table 15.1. Assessment Essentials for Anxiety Disorders 1. Rule out physical causes such as hyperthyroidism, side effects to medications (allergy/asthma medications, etc.), substance abuse, or other medical conditions. 2. Get data from multiple sources. Children are often reluctant to talk about their worries. Be sure to get a family genetic history of anxiety disorders, as well as depression and other mood disorders and tics. 3. Younger children may better communicate their anxieties through drawings or play techniques. 4. Determine the trigger(s) for the anxiety. Does the anxiety only occur in a specific situation? Does it occur out of the blue? Does it occur in anticipation of something? Is it pervasive? 5. Understand the environmental and family factors that may affect the youth's anxiety. How do the parents react? Are there family conflicts or other stresses contributing to the

anxiety? 6. Screen for comorbid psychiatric disorders: mood disorders, psychosis, eating disorders, tic disorders, and disruptive behavior disorders. 7. Consider the use of symptom rating scales to better categorize, understand, and monitor the child's anxieties. Yale-Brown Obsessive Compulsive Scale (Y-BOCS), the Screen for Child Anxiety Related Emotional Disorders (SCARED), the Social Phobia and Anxiety Inventory for Children (SPAI-C), and the Revised Children's Manifest Anxiety Scale (RCMAS) are suggestions. P.125

Children may suffer from both internalizing and externalizing disorders. Co-occurrence of the inattentive type of ADHD and anxiety disorder is not infrequent. If using a stimulant, start low and go slow to minimize the risk of increasing anxiety. Tip Inquire about caffeine intake and counsel to minimize it. Caffeine is a known cause of anxiety. Treatment Treatment is multimodal and requires a thoughtful, stepwise intensive treatment of the child and adolescent. Table 15.2 gives the essentials of treatment, elaborating on the types of P.126 psychosocial and medication treatment options available. For mild to moderate anxiety, evidence-based psychotherapies and psychoeducation should be used first, with adjunctive medication if necessary. There is evidence that some milder forms of anxiety may have a more prolonged course if medications are started initially. However, for disabling anxiety, consider concomitant psychotherapy and medication. Table 15.2. Essentials of Multimodal Treatment of Anxiety 1. Psychoeducation of parent and child about the nature of anxiety, how it can affect family relationships, how family members can inadvertently perpetuate the symptoms through their own anxiety, and how the family can support the child in overcoming his or her anxiety. 2. Cognitive-behavioral therapy should comprise first-line treatment. There are evidencebased treatments for OCD (exposure and response prevention), phobias, and other anxiety disorders. 3. School intervention when the anxiety is seriously impairing school functioning. 4. Medications 1st line: SSRIsremember that SSRIs can induce anxiety or even panic symptoms in vulnerable individuals so start low and go slow. Sometimes, benzodiazepines are started concurrently with an SSRI and later tapered once the SSRI confers therapeutic benefits. The SSRIs that are FDA approved for OCD in children include fluoxetine, sertraline, and fluvoxamine (chlomipramine, a TCA, is also approved). 2nd line: benzodiazepines such as alprazolam, lorazepam, and clonazepam can be useful in the short-term treatment of anxiety, e.g., to reintegrate the child into school. Remember to taper slowly to avoid rebound anxiety.

3rd line: alpha-2a-agonistsguanfacine and clonidine may be useful for symptoms of hyperautonomic arousal such as palpitations and tachypnea. Others: tricyclic antidepressants (TCAs)requires EKG and blood level monitoring, but may be effective. Buspironea few case reports of effectiveness in mild anxiety. Anticonvulsant agentscase reports for the use of gabapentin, topiramate, and oxcarbazepine. Consider using when other agents have been ineffective. Antipsychotic agentsmay be useful when all other medications have not been successful or in children with borderline reality testing and high levels of agitation. P.127

Anxiety disorders affect a large portion of children and adolescents, causing them tremendous suffering and interfering with optimal development in many domains (social, academic, and life skills). Currently, cognitive behavioral treatments are the best supported interventions and should comprise the first line of treatment. Pharmacotherapy can augment psychosocial treatments individualized to each youth's circumstances and response to psychotherapeutic interventions. Early and effective treatments may improve long-term prognosis.
ObsessiveCompulsive Disorder (CLINICAL GUIDE) DIAGNOSIS Obsessivecompulsive disorder (OCD) is an intriguing and often debilitating syndrome characterized by the presence of two distinct phenomena: obsessions and compulsions. Obsessions are intrusive, recurrent, un-wanted ideas, thoughts, or impulses that are diffi cult to dismiss, despite their disturbing nature. Compulsions are repetitive behaviors, either observable or mental, that are intended to reduce the anxiety engendered by obsessions. Both obsessions and compulsions have been described in a wide variety of mental and neu-rological disorders. However, obsessions and compul-sions that clearly interfere with the functioning and/or cause signifi cant distress are the hallmark of OCD (see DSM-IV-TR diagnostic criteria, page 317).

OCDs clinical presentation is characterized by phe-nomenological subtypes based on the content of the obsessions and corresponding compulsions. The list of subtypes in the Yale-Brown ObsessiveCompulsive Scale (Y-BOCS) (Table 31-1) was generated on the basis of clinical interviews with OCD patients in the 1980s. The basic types of obsessions and compulsions seem to be consistent across cultures. The most com-mon obsession is the fear of contamination, followed by pathological doubt, a need for symmetry, and ag-gressive obsessions. The most common compulsion is checking, which is followed by washing, symmetry, the need to ask or confess, and counting. Children with OCD present most commonly with washing compul-sions, which are followed by repeating rituals. Most individuals with OCD have multiple obses-sions and compulsions over time, with a particular fear or concern dominating the clinical picture at any one time. The presence of obsessions without compul-sions, or compulsions without obsessions, is unusual. In the DSM-IV OCD fi eld trial of 431 individuals, only 2% had predominantly obsessions and 2% had predominantly compulsions; the remaining 96% en-dorsed both obsessions and compulsions Individuals who appear to have obsessions without compulsions frequently have unrecognized reassurance rituals or mental compulsions, such as repetitive, ritualized praying, in addition to their obsessions. Pure com-pulsions are also unusual in adults, although they do occur in children, especially in the young (e.g., 6 to

8 years of age). Most people have both mental and behavioral compulsions; in the DSM-IV fi eld trial, 79.5% reported having both mental and behavioral compulsions, 20.3% had behavioral compulsions only, and 0.2% had only mental compulsions. Contamination obsessions are the most frequently encountered obsessions in OCD. Such obsessions are usually characterized by a fear of dirt or germs. For ex-ample, a 38-year-old computer programmer was exces-sively preoccupied with the thought that her apartment would become dirty. She had never allowed a visitor into her apartment or worn a coat during the winter, be-cause she feared that she would be unable to protect her apartment from dirt brought inside by either a visitor or a coat. Excessive washing is the compulsion most com-monly associated with contamination obsessions. This behavior usually occurs after contact with the feared object; however, proximity to the feared stimulus is of-ten suffi cient to engender severe anxiety and washing compulsions, even though the contaminated object has not been touched. Most individuals with washing com-pulsions perform these rituals in response to a fear of contamination, but these behaviors occasionally occur in response to a drive for perfection or a need for sym-metry. Some individuals, for example, repeatedly wash themselves in the shower until they feel right or must wash their right arm and then their left arm the same number of times. Need for symmetry is a term that describes a drive

to order or arrange things perfectly or to perform cer-tain behaviors symmetrically or in a balanced way. Clinical Guide to the Diagnosis and Treatment of Mental Disorders. M. B. First and A. Tasman 2 0 0 6 Jo h n W i l ey & So n s , L t d . I SBN 0- 470 - 01915-8 Chapter 31 Anxiety Disorders: ObsessiveCompulsive Disorder 317 TABLE 31-1 Yale-Brown ObsessiveCompulsive Scale Symptom Checklist Aggressive obsessions Fear might harm others Fear might harm self Violent or horrifi c images Fear of blurting out obsessions or insults Fear of doing something embarrassing Fear of acting on other impulses (e.g., robbing a bank, stealing groceries, overeating) Fear of being responsible for things going wrong (e.g., others will lose their job because of the patient) Fear something terrible might happen (e.g., fi re, burglary) Other Contamination obsessions Concerns or disgust with bodily waste (e.g., urine, feces, saliva) Concern with dirt or germs Excessive concern with environmental contaminants (e.g., asbestos, radiation, toxic wastes)

Excessive concern with household items (e.g., cleansers, solvents, pets) Concerned will become ill Concerned will become ill (aggressive) Other Sexual obsessions Forbidden or perverse sexual thoughts, images, or impulses Content involves children Content involves animals Content involves incest Content involves homosexuality Sexual behavior toward others (aggressive) Other Hoarding or collecting obsessions Religious obsessions Obsession with need for symmetry or exactness Miscellaneous obsessions Need to know or remember Fear of saying certain things Fear of not saying things just right Intrusive (neutral) images Intrusive nonsense sounds, words, or music Other Somatic obsessioncompulsion

Cleaning or washing compulsions Excessive or ritualized hand washing Excessive or ritualized showering, bathing, brushing the teeth, or grooming Involves cleaning of household items or inanimate objects Other measures to prevent contact with contaminants Counting compulsions Checking compulsions Checking that did not or will not harm others Checking that did not or will not harm self Checking that nothing terrible did or will happen Checking for contaminants Other Repeating rituals Ordering or arranging compulsions Miscellaneous compulsions Mental rituals (other than checking or counting) Need to tell, ask, or confess Need to touch Measures to prevent Harm to self Harm to others Ter r ible conse quences Other 300.3 O BSESSIVECOMPULSIVE DISORDER

A. Either obsessions or compulsions: Obsessions as defi ned by (1), (2), (3), and (4) : (1) recurrent and persistent thoughts, impulses, or im-ages that are experienced, at some time during the disturbance, as intrusive and inappropriate and that cause marked anxiety or distress (2) the thoughts, impulses, or images are not simply excessive worries about real-life problems (3) the person attempts to ignore or suppress such thoughts, impulses, or images, or to neutralize them with some other thought or action (4) the person recognizes that the obsessional thoughts, impulses, or images are a product of his or her own mind (not imposed from without as in thought insertion) Compulsions as defi ned by (1) and (2) : (1) repetitive behaviors (e.g., hand washing, order-ing, checking) or mental acts (e.g., praying, counting, repeating words silently) that the person feels driven to perform in response to an obsession, or according to rules that must be applied rigidly (2) the behaviors or mental acts are aimed at prevent-ing or reducing distress or preventing some dreaded event or situation; however, these behaviors or mental acts either are not connected in a realistic way with what they are designed to neutralize or prevent or are clearly excessive B. At some point during the course of the disorder, the

person has recognized that the obsessions or compul-sions are excessive or unreasonable. Note: This does not apply to children. C. The obsessions or compulsions cause marked distress, are time consuming (take more than 1 hour a day), or signifi cantly interfere with the persons normal rou-tine, occupational (or academic) functioning, or usual social activities or relationships. D. If another Axis I disorder is present, the content of the obsessions or compulsions is not restricted to it (e.g., preoccupation with food in the presence of an eat-ing disorder; hair pulling in the presence of trichotil-lomania; concern with appearance in the presence of body dysmorphic disorder (BDD); preoccupation with drugs in the presence of a substance use disorder; pre-occupation with having a serious illness in the pres-ence of hypochondriasis; preoccupation with sexual urges or fantasies in the presence of a paraphilia; or guilty ruminations in the presence of major depressive disorder). E. The disturbance is not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medica-tion) or a general medical condition. Specify if: With poor insight: if, for most of the time during the cur-rent episode, the person does not recognize that the ob-sessions and compulsions are excessive or unreasonable Reprinted with permission from the Diagnostic and Statistical Manual of Mental Disorders, 4th ed., Text Rev. Copyright 2000 American Psychiatric Association. DSM-IV-TR Diagnostic Criteria

318 Clinical Guide to the Diagnosis and Treatment of Mental Disorders Individuals describe an urge to repeat motor acts until they achieve a just right feeling that the act has been completed perfectly. Individuals with a prominent need for symmetry may have little anxiety but rather describe feeling unsettled or uneasy if they cannot repeat actions or order things to their satisfaction. Individuals with a need for symmetry frequently present with obsessional slowness, taking hours to perform acts such as grooming or brushing their teeth. A 23-year-old cook spent 2 hours a day brushing his teeth in a symmetrical fashion and as a result developed gingival erosion. He reported being exquisitely aware of exactly how the toothbrush touched each surface of each tooth and of how he placed the toothbrush and cup down after fi nishing. He was unable to describe any obsession or fear about not performing this task adequately but rather felt unable to stop until he had brushed completely, despite warnings from his dentist about the harm he was causing. Individuals with somatic obsessions are worried about the possibility that they have or will contract an illness or disease. In the past, the most common so-matic obsessions consisted of fears of cancer or vene-real diseases. However, a fear of developing AIDS has become increasingly common. Checking compulsions consisting of checking and rechecking the body part of concern, as well as reassurance seeking, are commonly

associated with this fear. For example, a 29-year-old fi refi ghter spent 3 hours a day examining his throat in the mirror and palpating his lymph nodes to determine whether he had throat cancer. People with sexual or aggressive obsessions are plagued by fears that they might harm others or commit a sexually unacceptable act such as molestation. Often, they are fearful not only that they will commit a dread-ful act in the future but also that they have already com-mitted the act. Individuals are usually horrifi ed by the content of their obsessions and are reluctant to divulge them. It is striking that the content of these obsessions tends to consist of ideas that individuals fi nd particu-larly abhorrent. A 32-year-old librarian who wanted to be a good mother had intrusive thoughts of stabbing her daughter. Individuals with these highly distressing obsessions frequently have checking and confession or reassurance rituals. They may report themselves to the police or repeatedly seek out priests to confess their imagined crimes. For example, a 29-year-old secretary constantly checked the local news to be certain that she had not murdered someone. An unsolved murder case caused her tremendous anxiety and led to extensive re-assurance rituals. Pathological doubt is a common feature of individu-als with OCD who have a variety of different obsessions and compulsions. Individuals with pathological doubt are plagued by the concern that, as a result of their carelessness, they will be responsible for a dire event.

They may worry, for example, that they will start a fi re because they neglected to turn off the stove before leav-ing the house. Although many individuals report being fairly certain that they performed the act in question (e.g., locking the door, unplugging the hairdryer, pay-ing the correct amount on a bill), they cannot dismiss the nagging doubt What if? Excessive doubt and as-sociated feelings of excessive responsibility frequently lead to checking rituals. For example, individuals may spend several hours checking their home before they leave. As with contamination obsessions, pathological doubt can lead to marked avoidance behavior. Some in-dividuals become housebound to avoid the responsibil-ity of potentially leaving the house unlocked. There has been considerable interest in the role of insight, or awareness, in OCD. An ability to recognize the senselessness of the obsessions and the ability to resist obsessional ideas have been considered as the fundamental components of OCD. However, research fi ndings during the past decade have demonstrated a continuum of insight in this disorder, which ranges from excellent (i.e., complete awareness of the senselessness of the content of the obsessions), through poor insight, to delusional thinking (i.e., the obsessions are held with delusional conviction). Combining data from a number of studies, 2025% of individuals with OCD at some point during their illness are fairly convinced that their obsessions are realistic and that consequences other

than anxiety would occur if they did not perform their compulsions. Nonetheless, most people with OCD are aware that other people think their symptoms are un-realistic and that the obsessions are caused by a mental disorder. To refl ect the fact that many individuals lack insight, DSM-IV-TR includes a specifi er With Poor Insight that applies to an individual who, for most of the time in the current episode, does not recognize that the obsessions or compulsions are excessive or un-reasonable. DSM-IV-TR also acknowledges that the beliefs that underlie OCD obsessions can be delusional and notes that, in such cases, an additional diagnosis of delusional disorder or psychotic disorder not otherwise specifi ed may be appropriate. Women appear to develop OCD slightly more fre-quently than do men. A predominance of males has been observed in child and adolescent OCD populations. OCD frequently occurs in association with other Axis I disorders. In a study of 100 individuals with primary OCD, 67 had a lifetime history of major de-pressive disorder and 31 had symptoms that met criteria for current major depressive disorder. Although it may be diffi cult to distinguish a primary from a secondary Chapter 31 Anxiety Disorders: ObsessiveCompulsive Disorder 319 diagnosis, some individuals with OCD view their de-pressive symptoms as occurring secondary to the de-moralization and hopelessness accompanying their OCD and report that they would not be depressed if

they did not have OCD. However, others view their ma-jor depressive symptoms as occurring independently of their OCD symptoms, which may be less severe when they cycle into an episode of major depression, because they feel too apathetic to be as concerned with their obsessions and too fatigued to perform compulsions. Conversely, OCD symptoms may intensify during de-pressive episodes. Although fi ndings have varied, the generally ac-cepted frequency of tic disorders in individuals with OCD is far higher than in the general population, with a rate of approximately 510% for Tourettes Disorder and 20% for any tic disorder. Conversely, individuals with Tourettes disorder have a high rate of comorbid OCD, with 3040% reporting obsessivecompulsive symptoms. The likelihood of childhood onset of OCD is greater in this group, and the presence of tics is as-sociated with more severe OCD symptoms in children. There is an increased rate of both OCD and tic disor-ders in the fi rst-degree relatives of OCD probands with a family lifetime history of tics and an increased fre-quency of tic disorders in the fi rst-degree relatives of OCD probands compared to controls. Studies of individuals with schizophrenia or schizoaf-fective disorder have found rates of OCD ranging from 8% to 46%. This strikingly large range is most likely due to the OCD criteria used (i.e., subclinical OCD symptoms versus OCD symptoms severe enough to cause signifi cant impairment or distress). Regardless, it

is clear that a signifi cant number of people with schizo-phrenia have OCD symptoms that require assessment, and may benefi t from treatment. The relationship between OCD and personality dis-orders, particularly obsessivecompulsive personality disorder (OCPD), has received considerable attention. Early observations noted the presence of OCPD traits in individuals with OCD. Systematic studies, however, have yielded inconsistent fi ndings. Until the mid-1980s, OCD was considered extremely rare. This perception was based on studies from the 1950s and 1960s that examined the frequency of men-tal disorders in inpatient and outpatient settings. The results of a large epidemiological study, the national ECA survey, conducted in the United States in 1984, painted a different picture of OCDs prevalence. This study found that OCD was the fourth most common mental disorder (after the phobias, substance use disor-ders, and major depressive disorder), with a prevalence of 1.6% over 6 months and a lifetime prevalence of 2.5%. Although the ECA survey has been criticized as overestimating OCDs prevalence, a subsequent study in the United States and several epidemiological stud-ies in other countries have supported its fi ndings. Course Age at onset usually refers to the age when OCD symp-toms (obsessions and compulsions) reach a severity level, wherein they lead to impaired functioning or

signifi cant distress or are time consuming (i.e., meet DSM-IV-TR criteria for the disorder). Reported age at onset is usually during late adolescence. People with OCD, however, usually describe the onset of minor symptoms in childhood, well before the onset of symp-toms meeting the full criteria for the disorder. In several studies, earlier age at onset has been as-sociated with an increased rate of OCD in fi rstdegree relatives. These data suggest that there is a familial type of OCD characterized by early onset. Age at onset of OCD may also be a predictor of course. The vast majority of individuals report a gradual worsening of obsessions and compulsions prior to the onset of full-criteria OCD, which is followed by a chronic course. However, a subtype of OCD that begins before puberty and is characterized by an episodic course with in-tense exacerbations has been described. Exacerbations of OCD symptoms in this subtype have been linked with Group A beta-hemolytic streptococcal infections, which has led to the subtype designation of pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS). In a study of 50 children with PANDAS, the average age of onset was 7.4 years. Whether the course of illness in indi-viduals with PANDAS continues to be episodic into adulthood, or, as is the case with postpubertal onset, tends to be chronic, is not known. The course of OCD is usually waxing and waning that is, once an individual acquires OCD, obsessions

or compulsions, or both, are present continuously, with varying degrees of intensity over time. Relatively few individuals have either a progressively deteriorating course or a truly episodic course. Differential Diagnosis OCD is sometimes diffi cult to distinguish from cer-tain other disorders. Obsessions and compulsions may appear in the context of other syndromes, which can raise the question whether the obsessions and compul-sions are a symptom of another disorder or whether both OCD and another disorder are present. A general guideline is that if the content of the obsessions is not 320 Clinical Guide to the Diagnosis and Treatment of Mental Disorders limited to the focus of concern of another disorder (e.g., an appearance concern, as in body dysmorphic disor-der [BDD], or food concerns, as in an eating disorder) and if the obsessions or compulsions are preoccupying as well as distressing or impairing, OCD should gener-ally be diagnosed. Diagnostic dilemmas may also arise when it is unclear whether certain thoughts are obses-sions or whether, instead, they are ordinary worries, ruminations, overvalued ideas, or delusions. In a simi-lar vein, questions may develop about whether certain behaviors constitute true compulsions or whether they should instead be conceptualized as impulses, tics, or addictive behaviors. Both OCD and the other anxiety disorders are char-acterized by the use of avoidance to manage anxiety.

However, OCD is distinguished from these disorders by the presence of compulsions. For individuals with preoccupying fears or worries but no rituals, several other features may be useful in establishing the diagno-sis of OCD. In social phobia and specifi c phobia, fears are circumscribed and related to specifi c triggers (in specifi c phobia) or social situations (in social phobia). As many as 60% of people with OCD experience full-blown panic symptoms. However, unlike panic disor-der, in which panic attacks occur spontaneously, panic symptoms occur in OCD only during exposure to spe-cifi c feared triggers such as contaminated objects. The worries that are present in generalized anxiety disor-der (GAD) are more egosyntonic and involve an exag-geration of ordinary concerns, whereas the obsessional thinking of OCD is more intrusive, is limited to a spe-cifi c set of concerns (e.g., contamination, blasphemy), and usually has an irrational, senseless, or unreason-able quality. One question is how to differentiate OCD from psy-chotic disorders such as schizophrenia and delusional disorder. Another question is how to distinguish OCD with insight from OCD without insight (delusional OCD). One distinguishing feature between OCD and the psychotic disorders is that the latter are not character-ized by prominent ritualistic behaviors. If compulsions are present in an individual with prominent psychotic symptoms, the possibility of a comorbid OCD diagnosis should be considered. Furthermore, although schizo-phrenia may be characterized by obsessional thinking, other characteristic features of the disorder, such as prominent hallucinations or thought disorder, are also

present. With regard to delusional disorder, paranoid and grandiose concerns are generally not considered to fall under the OCD rubric. However, some other types of delusional disorder, such as the somatic and jealous types, seem to bear a close resemblance to OCD and are not always easily distinguished from it. The second issue noted abovehow to distinguish OCD with insight from OCD without insightis com-plex. As previously discussed, insight in OCD is increasingly being recognized as spanning a spectrum from good to poor to absent. Both clinical observations and research fi ndings indicate that some individuals hold their obsessional concerns with delusional inten-sity, and believe that their concerns are reasonable. In DSM-IV-TR, delusional OCD may be double coded as both OCD and delusional disorder or as both OCD and psychotic disorder not otherwise specifi ed; in other words, individuals with delusional OCD would receive both diagnoses. This double coding refl ects the fact that it is unclear whether OCD with insight and OCD with-out insight constitute the same or different disorders. Further research using validated scales to assess insight in OCD is needed to shed light on this question. Differential diagnosis questions have been raised with regard to kleptomania, trichotillomania, patho-logical gambling, and other disorders involving im-pulsive behaviors. Several features have been said to distinguish these disorders from OCD. For example,

compulsionsunlike behaviors of the impulse con-trol disordersgenerally have no gratifying element, although they do diminish anxiety. In addition, the af-fective state that drives the behaviors associated with these disorders may differ. In OCD, fear is frequently the underlying drive that leads to compulsions, which, in turn, decrease anxiety. In the impulse control dis-orders, individuals frequently describe heightened ten-sion, but not fear, preceding an impulsive behavior. Complex motor tics of Tourettes disorder may be diffi cult to distinguish from OCD compulsions. Both tics and compulsions are preceded by an intrusive urge and are followed by feelings of relief. However, OCD compulsions are usually preceded by both anxiety and obsessional concerns, whereas, in Tourettes disorder, the urge to perform a tic is not preceded by an obses-sional fear. This distinction breaks down to some extent when considering the just right perceptions of some individuals with OCD. The just right perception re-fers to the need to perform a certain motor action, such as touching, tapping, checking, ordering, arranging, or counting, until it feels right. Determining when an action has been performed enough or perfectly may depend on tactile, visual, or auditory perceptions. In a study of in-dividuals with Tourettes disorder and OCD symptoms, most individuals could distinguish between the mental urge to do something repeatedly until it felt right and a physical urge to perform a motor tic. However, it is sometimes diffi cult for mental health professionals to

distinguish between complex tics and compulsions, es-pecially when an individual has both disorders. Chapter 31 Anxiety Disorders: ObsessiveCompulsive Disorder 321 Fears of illness that occur in OCD, referred to as so-matic obsessions, may be diffi cult to distinguish from hypochondriasis. Usually, however, individuals with somatic obsessions have other current or past classic OCD obsessions unrelated to illness concerns. Individ-uals with OCD also often engage in classic OCD rituals, such as checking or reassurance seeking, in an attempt to diminish their illness concerns. Unlike individuals with OCD, individuals with hypochondriasis experi-ence somatic and visceral sensations. BDD, a preoccu-pation with an imagined or slight defect in appearance (e.g., thinning hair, facial scarring, or a large nose), has many similarities to OCD. Individuals with BDD expe-rience obsessional thinking about the supposed defect and usually engage in associated repetitive ritualistic behaviors, such as mirror checking and reassurance seeking. Preliminary evidence suggests that BDD also appears similar to OCD in terms of age of onset, course of illness, and other variables. Nonetheless, emerging data suggest that there are some important differences between the two disorders and they are currently classi-fi ed separately in DSM-IV-TR. Insight, for example, is more frequently impaired in BDD than in OCD. If the content of a individuals obsessions involves a concern about a supposed defect in appearance, BDD, rather than OCD, is the diagnosis that should be given.

Obsessivecompulsive personality disorder is a life-long maladaptive personality style characterized by perfectionism, excessive attention to detail, indecisive-ness, rigidity, excessive devotion to work, restricted af-fect, lack of generosity, and hoarding. OCD and OCPD have historically been considered variants of the same disorder on a continuum of severity, with OCD viewed as the more severe manifestation of illness. Contrary to this notion, studies using structured interviews to establish diagnosis have found that not all individu-als with OCD also have OCPD. One reason for the perception that these disorders are linked lies in the frequency of several OCPD traits in individuals with OCD. In one study, the majority of 114 individuals with OCD had perfectionism and indecisiveness (82 and 70, respectively). In contrast, other OCPD traits, such as restricted affect, excessive devotion to work, and rigid-ity, were seen infrequently. Although perfectionism and indecisiveness are rela-tively common traits in individuals with OCD, the dis-tinction between OCD and OCPD is important, and sev-eral guidelines may be useful in distinguishing them. Unlike OCPD, OCD is characterized by distressing, timeconsuming egodystonic obsessions and repetitive rituals aimed at diminishing the distress engendered by obsessional thinking. One of the hallmarks that has been traditionally used to distinguish OCD from OCPD is that, in contrast, OCPD features are considered ego-syntonic. In addition, as previously noted, the traits of restricted affect, excessive devotion to work, and rigid-ity are generally characteristic of OCPD but not OCD. Although useful, these guidelines are not absolute, and

some individuals defy easy categorization. Some indi-viduals, for example, spend hours each day engaged in egosyntonic behaviors such as excessive cleaning; such individuals may seek treatment not because they are disturbed by their behaviors but because the behaviors cause problems in functioning or family friction. It is unclear whether some of these individuals should be diagnosed with OCPD or subthreshold OCD. TREATMENT Both pharmacologic and behavioral therapies have proved effective for OCD. The majority of control-led treatment trials have been performed with adults aged 18 to 65 years. However, these therapies have been shown to be effective for individuals of all ages. In general, children and the elderly tolerate most of these medications well. For children, lower doses are indicated because of lower body mass. For instance, the recommended dose for clomipramine in children is up to 150 mg/day (3 mg/kg/day) versus 250 mg/day in adults. Use of lower doses should also be considered in the elderly because their decreased ability to metabo-lize medications can increase the risk of side effects and toxicity. Behavioral therapy has also been used successfully in all age groups, although when treating children with this modality it is usually advisable to use a parent as a cotherapist. A fl owchart that outlines treatment options for OCD is shown in Figure 31-1.

In general, the goals of treatment are to reduce the frequency and intensity of symptoms as much as pos-sible and to minimize the amount of interference the symptoms cause. It should be noted that few individuals experience a cure or complete remission of symptoms. Instead, OCD should be viewed as a chronic illness with a waxing and waning course. Symptoms are of-ten worse during times of psychosocial stress. Even when on medication, individuals with OCD are often upset when they experience even a mild symptom ex-acerbation, anticipating that their symptoms will revert to their worst, which is rarely the case. Anticipating with the individual that stress may make the symptoms worse can often be helpful in long-term treatment. Somatic Treatments The most extensively studied agents for OCD are medi-cations that affect the serotonin system. The principal 322 Clinical Guide to the Diagnosis and Treatment of Mental Disorders pharmacologic agents used to treat OCD are the SRIs, which include clomipramine, fl uoxetine, fl uvoxamine, sertraline, paroxetine, citalopram, and escitalopram. The tricyclic antidepressant clomipramine is among the most extensively studied pharmacological agents in OCD. This drug is unique among the antiobsessional agents in that in addition to its potency as an SRI, it has signifi cant affi nity for noradrenergic, dopaminergic,

muscarinic, and histaminic receptors. The most com-mon side effects were those typical of the tricyclic anti-depressants, including dry mouth, dizziness, tremor, fatigue, somnolence, constipation, nausea, increased sweating, headache, mental cloudiness, and sexual dys-function. Previous data have indicated that at doses of 300 mg/day or more, the risk of seizures is 2.1%, but at doses of 250 mg/day or less, the risk of seizures is low (0.48%) and comparable to that of other tricyclic antidepressants. It is therefore recommended that doses of 250 mg/day or less be used. Recent studies of IV clomipramine have been partic-ularly promising because it seems to have a quicker on-set of action and fewer side effects than the oral form, and it may be effective even in individuals who do not respond to oral clomipramine. Oral clomipramine, like other SRIs, usually takes a minimum of 4 to 6 weeks to produce a clinically signifi cant clinical response, but Figure 31-1 Flowchart of treatment options for OCD. Treatment of OCD Also consider Pharmacotherapy 1st-line SRIs: sertraline200 mg fluvoxamine 300 mg 2nd-line SRIs: clomipramine 250 mg fluoxetine 80 mg paroxetine 60 mg

Augmentors if partial response: buspirone, lithium, clonazepam, trazodone, alprazolam, liothyronine Try a 2nd SRI/Behavior therapy Still unresponsive r edi s noc osl A For personality disorder 1. Psychotherapy 2. Day program 3. Halfway houses For severely disabled patients Consider neurosurgery: 1. After a minimum of two adequate medication trials with augmentation 2. After behavioral treatment 3. Without severe personality disorder Behavioral therapy Exposure with response prevention Reasons of poor response: 1. Poor compliance 2. Comorbid depression 3. Use/abuse of CNS depressants 4. Delusions

Chapter 31 Anxiety Disorders: ObsessiveCompulsive Disorder 323 in at least one study using IV pulse dosing, individu-als showed a response within 4.5 days. The reasons for this unique response are not fully understood, but it is postulated that the IV preparation avoids fi rst-pass hepatoenteric metabolism, leading to increased bioa-vailability of the parent compound clomipramine. This in turn may play a role in rapidly desensitizing sero-tonergic receptors or initiating changes in postsynaptic serotonergic neurons. Although studies of IV clomi-pramine for obsessional states date as far back as 1973, this preparation is still not FDA-approved for clinical use in the United States. Cardiac monitoring is recom-mended during the use of IV clomipramine. Fluoxetine (as well as fl uvoxamine, sertraline, par-oxetine, citalopram and escitalopram) is often referred to as a selective serotonin reuptake inhibitor (SSRI) because it has a far more potent effect on serotonergic than on noradrenergic or other neurotransmitter sys-tems. Despite their different chemical structures, all of the SSRIs appear to have similar effi cacy in treating OCD. Fluoxetine and the other SSRIs have fewer side effects than clomipramine, refl ecting its more selective mechanism of action. The most common side effects are headache, nausea, insomnia, anorexia, dry mouth, somnolence, nervousness, tremor, and diarrhea. Side effects occur more frequently at higher doses. Most studies of other medications for OCD have consisted of only case reports or small samples. One small trial

suggested that venlafaxine, a medication which, like clomipramine, inhibits the reuptake of both serotonin and norepinephrine, may hold some promise. The effi cacy of each SSRIclomipramine, fl uox-etine, fl uvoxamine, sertraline, paroxetine, and citalopramis supported by existing data. During the past 10 years, at least seven head-to-head SRI comparison studies have been done. All of the studies found that the agents studied were equally effi cacious, although they may have been underpowered to detect differences among medications. However, several meta-analyses of OCD trials, which compared SRIs across large pla-cebo-controlled multicenter trials, lend some support to the notion that clomipramine might be more effec-tive than the more selective agents. However, like most meta-analyses, these studies are fl awed by factors that include variations in the study protocol, sample size, and the number of treatment-resistant and treatment-nave subjects. The meta-analyses do support a trial of clomipramine in all individuals who do not respond to SRIs, even though clomipramine tends to cause more side effects. It is worth noting that the SSRIs, via their effect on the liver cytochrome system, can inhibit the metabolism of certain other drugs. Fluoxetine can elevate blood levels of a variety of coadministered drugs, including tricyclic antidepressants (such as clomipramine), car-bamazepine, phenytoin, and trazodone. However, the

other SSRIs (with the exception of citalopram) can theoretically cause similar elevations, although fewer reports on such interactions are currently available. Some clinicians have taken advantage of these interac-tions by carefully combining fl uvoxamine with clomi-pramine in order to block clomipramines metabolism to desmet hylclom ipr a m i ne ; t h is i n t u r n favor s seroton i n reuptake inhibition provided by the parent compound rather than the norepinephrine reuptake inhibition pro-vided by the metabolite. However, caution should be exercised with this approach since the elevation in clo-mipramine levels, and perhaps other compounds, can be nonlinear and quickly lead to dangerous toxicity. At the very least, clomipramine levels should be carefully monitored. All of the SSRIs are generally well tolerated, with a relatively low percentage of individuals experiencing notable side effects or discontinuing them because of side effects. In addition, these compounds are unlikely to be lethal in overdose, except for clomipramine, which can lead to cardiac arrhythmias and death. All these agents can cause sexual side effects, ranging from anorgasmia to diffi cultly with ejaculatory function. However, such symptoms are not readily volunteered by the individual; thus it is important to ask. Should such symptoms be experienced, conservative measures may include dosage reduction, transient drug holidays

for a special weekend or occasion, or switching to an-other SSRI since individuals may not have the same degree of dysfunction with a different agent. However, if the clinician feels that it is critical to continue with the same agent, various treatments have been reported in the literature. Usually taken within a few hours of sexual activity, no one agent has been shown to work consistently. Among those that have been tried are yohimbine, buspirone, cyproheptadine, ropinirole, bu-proprion, dextroamphetamine, methylphenidate, aman-tidine, and nefazodone, to name a few. If an individual has had only a partial response to an antiobsessional agent of adequate dose and dura-tion, the next question is whether to change the SSRI or add an augmenting agent. Current clinical practice suggests that if there is no response at all to an SSRI, it may be best to change to another SSRI. However, if there has been some response to treatment, an augmen-tation trial of at least 2 to 8 weeks may be warranted. No augmentation agent has been fi rmly established as effi cacious. Although many augmentation agents ap-peared promising in open trials, they failed to be ef-fective in more systematic trials although some of the 324 Clinical Guide to the Diagnosis and Treatment of Mental Disorders later studies did not report response to the SRI alone, leaving unanswered the question of whether some augmentation strategies may be effective in partial SSRI responders. Many questions about augmenta-tion remain unanswered, including the optimal dura-tion of augmentation, comparative effi cacy of different

agents, predictors of response, and mechanism of ac-tion. Nonetheless, these agents do help some individu-als signifi cantly, and thus their systematic use should be considered (see Table 31-2). In individuals with severe symptoms or comorbid psychosis or tic disorder, pimozide 13 mg/day, haldol 210 mg/day, and other neuroleptic agents (risperidone 28 mg/day and olanzapine 2.510 mg/day) have been used with some success. However, the use of a neu-roleptic agent should be considered carefully in light of the risk of extrapyramidal symptoms and side effects such as weight gain, lethargy, and tardive dyskinesia. Thus, when a neuroleptic drug is used, target symp-toms should be established before beginning treatment and the medication discontinued within several months if target symptoms do not improve. The use of lithium (300600 mg/day) and buspirone (up to 60 mg/day) as augmentation agents has also been explored. Both agents looked promising in open tri-als but failed to be effective in more systematic trials. Augmentation with fenfl uramine (up to 60 mg/day), clonazepam (up to 5 mg/day), clonidine (0.10.6 mg/ day), and trazodone (100200 mg/day), as well as the combination of clomipramine with any of the SSRIs, has had anecdotal success but has not been evaluated in methodologically rigorous studies. Some potential aug-menting agents and their dosage ranges are presented in Table 31-2.

Occasionally, even after receiving adequate phar-macotherapy (including augmentation), adequate behavioral therapy, and a combination of behavioral therapy and pharmacotherapy, individuals may still ex-perience intractable OCD symptoms. Such individuals may be candidates for neurosurgery. Although criteria for who should receive neurosurgery vary, it has been suggested that failure to respond to at least 5 years of systematic treatment is a reasonable criterion. The pro-cedures that have been most successful interrupt tracts involved in the serotonin system. The surgical proce-dures usedanterior capsulotomy, cingulotomy, and limbic leukotomyall aim to interrupt the connection between the cortex and the basal ganglia and related structures. Current stereotactic surgical techniques in-volve the creation of precise lesions, which are often only 10 to 20 mm, to specifi c tracts. These procedures have often been done with radio-frequency heated elec-trodes and more recently with gamma knife techniques. Postsurgical risks have been minimized, and in some cases cognitive function and personality traits improve along with symptoms of OCD. Psychosocial Treatments Behavioral therapy is effective for OCD both as a pri-mary treatment and as an augmentation agent. This form of therapy is based on the principle of exposure and response prevention. The individual is asked to en-dure, in a graduated manner, the anxiety that a specifi c obsessional fear provokes while refraining from com-pulsions that allay that anxiety. The principles behind

the effi cacy of behavioral treatment are explained to the individual in the following way. Although compulsions, either covert or overt, usually immediately relieve anxi-ety, this is only a short-term solution; the anxiety will ultimately return, requiring the performance of another compulsion. However, if the individual resists the anxi-ety and urge to ritualize, the anxiety will eventually decrease on its own (i.e., habituation will occur), and the need to perform the ritual will eventually disappear. Thus, behavioral therapy helps the individual habituate to the anxiety and extinguish the compulsions. Compulsions, especially overt behaviors like wash-ing rituals, are more successfully treated by behavioral Table 31-2 Potential Augmenting Agents for Treatment-Resistant ObsessiveCompulsive Disorder Augmenting Agent Suggested Dosage Range * Lithium 300600 mg/day Clonazepam 13 mg/day Tryptophan 210 g/day Trazodone 100 200 mg/day Buspirone 1560 mg/day Alprazolam 0.52 mg/day Methylphenidate 1030 mg/day Haloperidol 210 mg/day

Pimozide 210 mg/day Nifedipine 10 mg t.i.d. Liothyronine sodium 1025 mg/day Clonidine 0.10.6 mg/day Fenfl uramine Up to 60 mg/day * Add these to an ongoing trial of antidepressant medication. It should be noted that most of these dosages have not been tested with rigorous clinical trials but simply represent some of the reported doses tried in the current literature. Some would not recommend augmentation unless the initial treatment showed some response. Use with caution there have been some reports of elevated lithium levels with ongoing fl uoxetine treatment. Because the use of l-tryptophan has been implicated in an increased incidence of eosinophilia, the authors advise against the prescribing and use of this agent until the issue is resolved. Source : Jenike MA (1991) Management of patient with treatment-resistant obsessivecompulsive disorder. In Current Treatments of ObsessiveCompulsive Disorder, Pato MT and Zohar J (eds). Copyright, American Psychiatric Press, Washington DC, p. 146. Chapter 31 Anxiety Disorders: ObsessiveCompulsive Disorder 325 therapy than are obsessions alone or covert rituals like mental checking. This is because covert rituals are harder to physically resist than are rituals like hand

washing and checking a door. It has been reported that washing rituals are the most amenable to behavioral treatment, followed by checking rituals and then men-tal rituals. For rituals that do not constitute overt behaviors, techniques other than exposure and response preven-tion have been used in conjunction with exposure and response prevention. These approaches include imagi-nal fl ooding and thought stopping. In imaginal fl ood-ing, the anxiety provoked by the obsessions is evoked by continually repeating the thought, often with the help of a continuous-loop tape or the reading of a script composed by the individual and therapist, un-til the thought no longer provokes anxiety. In thought stopping, a compulsive mental ritual (e.g., continually repeating a short prayer in ones head) is stopped by simply shouting, making a loud noise, or snapping a rubber band on the wrist in an attempt to interrupt the thought. In the early stages of treatment, a behavioral assess-ment is performed. During this assessment, the con-tent, frequency, duration, amount of interference and distress, and attempts to resist or ignore the obsessions and compulsions are catalogued. An attempt is made to clarify the types of symptoms, any triggers that bring on the obsessions and compulsions, and the amount and type of avoidance used to deal with the symptoms. The individual, usually with the help of a therapist, then develops a hierarchy of situations according to the amount of anxiety they provoke. During treatment,

individuals gradually engage in the anxiety-provoking situations included in their hierarchy without perform-ing anxiety-reducing rituals. Despite its effi cacy, behavioral therapy has limita-tions. To begin with, about 1525% of individuals refuse to engage in behavioral treatment initially or drop out early in treatment because it is so anxiety provoking. Behavioral treatment fails in another 25% of individuals for a variety of other reasons, includ-ing concomitant depression; the use of central nervous system depressants, which may inhibit the ability to habituate to anxiety; lack of insight; poor compliance with homework, resulting in inadequate exposure; and poor compliance on the part of the therapist in enforc-ing the behavioral paradigm. Thus, overall, 50 70% of individuals are helped by this form of therapy. Behavior therapy can be used as the sole treatment of OCD, particularly with individuals whose contamina-tion fears or somatic obsessions make them resistant to taking medications. Behavioral treatment is also a powerful adjunct to pharmacotherapy. Some research appears to indicate that combined treatment may be more effective than pharmacotherapy or behavioral therapy alone, although these fi ndings are still prelimi-nary. Some studies have even suggested that adding pharmacotherapy to behavior therapy may be particu-larly helpful in reducing obsessions, while compulsions respond to behavior therapy. From a clinical perspec-tive, it may be useful to have individuals begin treat-ment with medication to reduce the intensity of their symptoms or comorbid depressive symptoms if present;

individuals may then be more amenable to experienc-ing the anxiety that will be evoked by the behavioral challenges they perform. The data on the discontinu-ation of behavioral therapy are encouraging. Overall, about 75% of individuals continue to do well at follow-up, but are symptom free. The use of psychotherapeutic techniques of either a psychoanalytic or a supportive nature has not been proved successful in treating the specifi c obsessions and compulsions that are a hallmark of OCD. How-ever, the more characterological aspects that are part of OCPD may be helped by a more psychoanalyti-cally oriented approach. The defense mechanisms of reaction formation, isolation, and undoing, as well as a pervasive sense of doubt and need to be in control, are hallmarks of the obsessivecompulsive character. In therapy the individual must be encouraged to take risks and learn to feel comfortable with, or at least less anxious about, making mistakes and to accept anxi-ety as a natural and normal part of human experience. Techniques for meeting such goals in treatment may include the therapists being relatively active in therapy to ensure that the individual focuses on the present rather than getting lost in perfectly recounting the past, as well as the therapists being willing to take risks and present herself or himself as less than perfect. COMPARISON OF DSM-IV-TR AND ICD-10 DIAGNOSTIC CRITERIA The ICD-10 Diagnostic Criteria for Research for

ObsessiveCompulsive Disorder differentiate between obsessions and compulsions on the basis of whether they are thoughts, ideas, or images (obsessions) or acts (compulsions). In contrast, DSM-IV-TR distinguishes between obsessions and compulsions on the basis of whether the thought, idea, or image causes anxiety or distress or prevents or reduces it. Thus, in DSM-IV-TR, there can be cognitive compulsions that would be considered obsessions in ICD-10. In addition, ICD-10 sets a minimum duration of at least 2 weeks, whereas DSM-IV-TR has no minimum duration. CHAPTER 32 Anxiety Disorder