Other Mental Disorders in Children and Adolescents
The combined prevalence of the group of disorders known as anxiety disorders is higher than that of virtually all other mental disorders of childhood and adolescence (Costello et al., 1996). The 1-year prevalence in children ages 9 to 17 is 13 percent (Table 3-1). This section furnishes brief overviews of several anxiety disorders: separation anxiety disorder, generalized anxiety disorder, social phobia, and obsessive-compulsive disorder. Treatments for all but the latter are grouped together below.
Separation Anxiety Disorder
Separation anxiety is often associated with symptoms of depression, such as sadness, withdrawal, apathy, or difficulty in concentrating, and such children often fear that they or a family member might die. Young children experience nightmares or fears at bedtime.
About 4 percent of children and young adolescents suffer from separation anxiety disorder (DSM-IV). Among those who seek treatment, separation anxiety disorder is equally distributed between boys and girls. In survey samples, the disorder is more common in girls (DSM-IV). The disorder may be overdiagnosed in children and teenagers who live in dangerous neighborhoods and have reasonable fears of leaving home.
The remission rate with separation anxiety disorder is high. However, there are periods where the illness is more severe and other times when it remits. Sometimes the condition lasts many years or is a precursor to panic disorder with agoraphobia. Older individuals with separation anxiety disorder may have difficulty moving or getting married and may, in turn, worry about separation from their own children and partner.
The cause of separation anxiety disorder is not known, although some risk factors have been identified. Affected children tend to come from families that are very close-knit. The disorder might develop after a stress such as death or illness in the family or a move. Trauma, especially physical or sexual assault, might bring on the disorder (Goenjian et al., 1995). The disorder sometimes runs in families, but the precise role of genetic and environmental factors has not been established. The etiology of anxiety disorders is more thoroughly discussed in Chapter 4.
Generalized Anxiety Disorder
About half of all adults seeking treatment for this disorder report that it began in childhood or adolescence, but the proportion of children with this disorder who retain the problem into adulthood is unknown. The remission rate is not thought to be as high as that of separation anxiety disorder.
Young children do not articulate their fears, but may cry, have tantrums, freeze, cling, appear extremely timid in strange social settings, shrink from contact with others, stay on the side during social events, and try to stay close to familiar adults. They may fall behind in school, avoid school completely, or avoid social activities among children their age. The avoidance of the fearful situations or worry preceding the feared event may last for weeks and interferes with the individual’s daily routine, social life, job, or school. They may find it impossible to speak in social situations or in the presence of unfamiliar people (for review of social phobia, see DSM-IV; Black et al., 1997).
Social phobia is common, the lifetime prevalence ranging from 3 to 13 percent, depending on how great the fear is and on how many different situations induce the anxiety (DSM-IV; Black et al., 1997). In survey studies, the majority of those with the disorder were found to be female (DSM-IV). Often the illness is lifelong, although it may become less severe or completely remit. Life events may reassure the individual or exacerbate the anxiety and disorder.
Treatment of Anxiety
(Ollendick & King, 1998).
In addition, psychodynamic treatment to address underlying fears and worries can be helpful, and behavior therapy may reduce the child’s fear of separation or of going to school; however, the experimental support for these approaches is limited.
Preliminary research suggests that selective serotonin reuptake inhibitors may provide effective treatment of separation anxiety disorder and other anxiety disorders of childhood and adolescence. Two large-scale randomized controlled trials are currently being undertaken (Greenhill, 1998a, 1998b). Neither tricyclic antidepressants nor benzodiazepines have been shown to be more effective than placebo in children (Klein et al., 1992; Bernstein et al., 1998).
There is a strong familial component to OCD, and there is evidence from twin studies of both genetic susceptibility and environmental influences. If one twin has OCD, the other twin is more likely to have OCD if the children are identical twins rather than fraternal twin pairs. OCD is increased among first-degree relatives of children with OCD, particularly among fathers (Lenane et al., 1990). It does not appear that the child is simply imitating the relative’s behavior, because children who develop OCD tend to have symptoms different from those of relatives with the disease (Leonard et al., 1997). Many adults with either childhood- or adolescent-onset of OCD show evidence of abnormalities in a neural network known as the orbitofrontalstriatal area (Rauch & Savage, 1997; Grachev et al., 1998).
Recent research suggests that some children with OCD develop the condition after experiencing one type of streptococcal infection (Swedo et al., 1995). This condition is referred to by the acronym PANDAS, which stands for Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections. Its hallmark is a sudden and abrupt exacerbation of OCD symptoms after a strep infection. This form of OCD occurs when the immune system generates antibodies to the streptococcal bacteria, and the antibodies cross-react with the basal ganglia13 of a susceptible child, provoking OCD (Garvey et al., 1998). In other words, the cause of this form of OCD appears to be antibodies directed against the infection mistakenly attacking a region of the brain and setting off an inflammatory reaction.
The selective serotonin reuptake inhibitors appear effective in ameliorating the symptoms of OCD in children, although more clinical trials have been done with adults than with children. Several randomized, controlled trials revealed SSRIs to be effective in treating children and adolescents with OCD (Flament et al., 1985; DeVeaugh-Geiss et al., 1992; Riddle et al., 1992, 1998). The appropriate duration of treatment is still being studied. Side effects are not inconsequential: dry mouth, somnolence, dizziness, fatigue, tremors, and constipation occur at fairly high rates. Cognitive- behavioral treatments also have been used to treat OCD (March et al., 1997), but the evidence is not yet conclusive.
Autism, the most common of the pervasive developmental disorders (with a prevalence of 10 to 12 children per 10,000 [Bryson & Smith, 1998]), is characterized by severely compromised ability to engage in, and by a lack of interest in, social interactions. It has roots in both structural brain abnormalities and genetic predispositions, according to family studies and studies of brain anatomy. The search for genes that predispose to autism is considered an extremely high research priority for the National Institute of Mental Health (NIMH, 1998). Although the reported association between autism and obstetrical hazard may be due to genetic factors (Bailey et al., 1995), there is evidence that several different causes of toxic or infectious damage to the central nervous system during early development also may contribute to autism. Autism has been reported in children with fetal alcohol syndrome (Aronson et al., 1997), in children who were infected with rubella during pregnancy (Chess et al., 1978), and in children whose mothers took a variety of medications that are known to damage the fetus (Williams & Hersh, 1997).
Cognitive deficits in social perception likely result from abnormalities in neural circuitry. Children with autism have been studied with several imaging techniques, but no strongly consistent findings have emerged, although abnormalities in the cerebellum and limbic system (Rapin & Katzman, 1998) and larger brains (Piven, 1997) have been reported. In one small study (Zilbovicius et al., 1995), evidence of delayed maturation of the frontal cortex was found. The evidence for genetic influences include a much greater concordance in identical than in fraternal twins (Cook, 1998).
Thirty years of research demonstrated the efficacy of applied behavioral methods in reducing inappropriate behavior and in increasing communication, learning, and appropriate social behavior. A well-designed study of a psychosocial intervention was carried out by Lovaas and colleagues (Lovaas, 1987; McEachin et al., 1993). Nineteen children with autism were treated intensively with behavior therapy for 2 years and compared with two control groups. Followup of the experimental group in first grade, in late childhood, and in adolescence found that nearly half the experimental group but almost none of the children in the matched control group were able to participate in regular schooling. Up to this point, a number of other research groups have provided at least a partial replication of the Lovaas model (see Rogers, 1998).
Several uncontrolled studies of comprehensive center-based programs have been conducted, focusing on language development and other developmental skills. A comprehensive model, Treatment and Education of Autistic and Related Communication Handicapped Children (TEACCH), demonstrated short-term gains for preschoolers with autism who received daily TEACCH home-teaching sessions, compared with a matched control group (Ozonoff & Cathcart, 1998). A review of other comprehensive, center-based programs has been conducted, focusing on elements considered critical to school-based programs, including minimum hours of service and necessary curricular components (Dawson & Osterling, 1997).
The antipsychotic drug, haloperidol, has been shown to be superior to placebo in the treatment of autism (Perry et al., 1989; Locascio et al., 1991), although a significant number of children develop dyskinesias as a side effect (Campbell et al., 1997). Two of the SSRIs, clomipramine (Gordon et al., 1993) and fluoxetine (McDougle et al., 1996), have been tested, with positive results, except in young autistic children, in whom clomipramine was not found to be therapeutic, and who experienced untoward side effects (Sanchez et al., 1996). Of note, preliminary studies of some of the newer antipsychotic drugs suggest that they may have fewer side effects than conventional antipsychotics such as haloperidol, but controlled studies are needed before firm conclusions can be drawn about any possible advantages in safety and efficacy over traditional agents.
Disruptive disorders, such as oppositional defiant disorder and conduct disorder, are characterized by antisocial behavior and, as such, seem to be a collection of behaviors rather than a coherent pattern of mental dysfunction. These behaviors are also frequently found in children who suffer from attention-deficit/hyper-activity disorder, another disruptive disorder, which is discussed separately in this chapter. Children who develop the more serious conduct disorders often show signs of these disorders at an earlier age. Although it is common for a very young children to snatch something they want from another child, this kind of behavior may herald a more generally aggressive behavior and be the first sign of an emerging oppositional defiant or conduct disorder if it occurs by the ages of 4 or 5 and later. However, not every oppositional defiant child develops conduct disorder, and the difficult behaviors associated with these conditions often remit.
Oppositional defiant disorder (ODD) is diagnosed when a child displays a persistent or consistent pattern of defiance, disobedience, and hostility toward various authority figures including parents, teachers, and other adults. ODD is characterized by such problem behaviors as persistent fighting and arguing, being touchy or easily annoyed, and deliberately annoying or being spiteful or vindictive to other people. Children with ODD may repeatedly lose their temper, argue with adults, deliberately refuse to comply with requests or rules of adults, blame others for their own mistakes, and be repeatedly angry and resentful. Stubbornness and testing of limits are common. These behaviors cause significant difficulties with family and friends and at school or work (DSM-IV; Weiner, 1997). Oppositional defiant disorder is sometimes a precursor of conduct disorder (DSM-IV).
In different studies, estimates of the prevalence of ODD have ranged from 1 to 6 percent, depending on the population sample and the way the disorder was evaluated, but not depending on diagnostic criteria. Rates are lower when impairment criteria are more strict and when information is obtained from teachers and parents rather than from the children alone (Shaffer et al., 1996a). Before puberty, the condition is more common in boys, but after puberty the rates in both genders are equal.
In preschool boys, high reactivity, difficulty being soothed, and high motor activity may indicate risk for the disorder. Marital discord, disrupted child care with a succession of different caregivers, and inconsistent, unsupervised child-rearing may contribute to the condition.
Children or adolescents with conduct disorder behave aggressively by fighting, bullying, intimidating, physically assaulting, sexually coercing, and/or being cruel to people or animals. Vandalism with deliberate destruction of property, for example, setting fires or smashing windows, is common, as are theft; truancy; and early tobacco, alcohol, and substance use and abuse; and precocious sexual activity. Girls with a conduct disorder are prone to running away from home and may become involved in prostitution. The behavior interferes with performance at school or work, so that individuals with this disorder rarely perform at the level predicted by their IQ or age. Their relationships with peers and adults are often poor. They have higher injury rates and are prone to school expulsion and problems with the law. Sexually transmitted diseases are common. If they have been removed from home, they may have difficulty staying in an adoptive or foster family or group home, and this may further complicate their development. Rates of depression, suicidal thoughts, suicide attempts, and suicide itself are all higher in children diagnosed with a conduct disorder (Shaffer et al., 1996b).
The prevalence of conduct disorder in 9- to 17-year-olds in the community varies from 1 to 4 percent, depending on how the disorder is defined (Shaffer et al., 1996a). Children with an early onset of the disorder, i.e., onset before age 10, are predominantly male. The disorder appears to be more common in cities than in rural areas (DSM-IV). Those with early onset have a worse prognosis and are at higher risk for adult antisocial personality disorder (DSM-IV; Rutter & Giller, 1984; Hendren & Mullen, 1997). Between a quarter and a half of highly antisocial children become antisocial adults.
The etiology of conduct disorder is not fully known. Studies of twins and adopted children suggest that conduct disorder has both biological (including genetic) and psychosocial components (Hendren & Mullen, 1997). Social risk factors for conduct disorder include early maternal rejection, separation from parents with no adequate alternative caregiver available, early institutionalization, family neglect, abuse or violence, parents’ psychiatric illness, parental marital discord, large family size, crowding, and poverty (Loeber & Stouthamer-Loeber, 1986). These factors are thought to lead to a lack of attachment to the parents or to the family unit and eventually to lack of regard for the rules and rewards of society (Sampson & Laub, 1993). Physical risk factors for conduct disorder include neurological damage caused by birth complications or low birthweight, attention-deficit/hyperactivity disorder, fearlessness and stimulation-seeking behavior, learning impairments, autonomic underarousal, and insensitivity to physical pain and punishment. A child with both social deprivation and any of these neurological conditions is most susceptible to conduct disorder (Raine et al., 1998).
Since many of the risk factors for conduct disorder emerge in the first years of life, intervention must begin very early. Recently, screening instruments have been developed to enable earlier identification of risk factors and signs of conduct disorder in young children (Feil et al., 1995). Studies have shown a correlation between the behavior and attributes of 3-year-olds and the aggressive behavior of these children at ages 11 to 13 (Raine et al., 1998). Measurements of aggressive behaviors have been shown to be stable over time (Sampson & Laub, 1993). Training parents of high-risk children how to deal with the children’s demands may help. Parents may need to be taught to reinforce appropriate behaviors and not harshly punish transgressing ones, and encouraged to find ways to increase the strength of the emotional ties between parent and child. Working with high-risk children on social interaction and providing academic help to reduce rates of school failure can help prevent some of the negative educational consequences of conduct disorder (Johnson & Breckenridge, 1982).
By applying criteria established by the American Psychological Association Task Force (see earlier) to the 82 studies, two treatments met criteria for well- established treatment and 10 for probably efficacious treatment. Two well-established treatments, both directed at training parents, succeeded in reducing problem behaviors. The two treatments were a parent training program based on the manual Living With Children (Bernal et al., 1980) and a videotape modeling parent training (Spaccarelli et al., 1992). The first teaches parents to reward desirable behaviors and ignore or punish deviant behaviors, based on principles of operant conditioning. The second provides a series of videotapes covering parent-training lessons, after which a therapist leads a group discussion of the videotape lessons. The identification of 12 treatments as well-established or probably efficacious is very encouraging because of the potential to intervene effectively with youth at high risk of poor outcomes. A new and promising approach for the treatment of conduct disorder is multisystemic therapy, an intensive home- and family-focused treatment that is described under Home-Based Services.
Despite strong enthusiasm for improving care for conduct-disordered youth, there are important groups of children, specifically girls and ethnic minority populations, who were not sufficiently represented in these studies to ensure that the identified treatments work for them. Other issues raised by Brestan and Eyberg (1998) are cost-effectiveness, the sufficiency of a given intervention, effectiveness over time, and the prevention of relapse.
No drugs have been demonstrated to be consistently effective in treating conduct disorder, although four drugs have been tested. Lithium and methylphenidate have been found (one double-blind placebo trial each) to reduce aggressiveness effectively in children with conduct disorder (Campbell et al., 1995; Klein et al., 1997b), but in two subsequent studies with the same design, the positive findings for lithium could not be reproduced (Rifkin et al., 1989; Klein, 1991). In one of the latter studies, methylphenidate was superior to lithium and placebo. A third drug, carbamazepine, was found in a pilot study to be effective, but multiple side effects were also reported (Kafantaris et al., 1992). The fourth drug, clonidine, was explored in an open trial, in which 15 of 17 patients showed a significant decrease in aggressive behavior, but there were also significant side effects that would require monitoring of cardiovascular and blood pressure parameters (Kemph et al., 1993).
Since the early 1990s there has been a“sharp resurgence” in the misuse of alcohol and other drugs by adolescents (Johnston et al., 1996). A recent review, focusing particularly on substance abuse and dependence, synthesizes research findings of the past decade (Weinberg et al., 1998). The authors review epidemiology, course, etiology, treatment, and prevention and discuss comorbidity with other mental disorders in adolescents. All of these issues are important to public health, but none is more relevant to this report than the co-occurrence of alcohol and other substance use disorders with other mental disorders in adolescents.
According to the National Comorbidity Study, 41 to 65 percent of individuals with a lifetime substance abuse disorder also have a lifetime history of at least one mental disorder, and about 51 percent of those with one or more lifetime mental disorders also have a lifetime history of at least one substance use disorder (Kessler et al., 1996). The rates are highest in the 15- to 24-year-old age group (Kessler et al., 1994). The cross-sectional data on association do not permit any conclusion about causality or clinical prediction (Kessler et al., 1996), but an appealing theory suggests that a subgroup of the population abuses drugs in an effort to self-medicate for the co-occurring mental disorder. Little is actually known about the role of mental disorders in increasing the risk of children and adolescents for misuse of alcohol and other drugs. Stress appears to play a role in both the process of addiction and the development of many of the comorbid conditions.
The review by Weinberg and colleagues (1998) provides more detail on epidemiology and assessment of alcohol and other drug use in adolescents and describes several effective treatment approaches for these problems. A meta-analysis and literature review (Stanton & Shadish, 1997) concluded that family-oriented therapies were superior to other treatment approaches and enhanced the effectiveness of other treatments. Multisystemic family therapy, discussed elsewhere in this chapter, is effective in reducing alcohol and other substance use and other severe behavioral problems among adolescents (Pickrel & Henggeler, 1996).
Eating disorders are serious, sometimes life- threatening, conditions that tend to be chronic (Herzog et al., 1999). They usually arise in adolescence and disproportionately affect females. About 3 percent of young women have one of the three main eating disorders: anorexia nervosa, bulimia nervosa, or binge-eating disorder (Becker et al., 1999). Binge-eating disorder is a newly recognized condition featuring episodic uncontrolled consumption, without compensatory activities, such as vomiting or laxative abuse, to avert weight gain (Devlin, 1996). Bulimia, in contrast, is marked by both binge eating and by compensatory activities. Anorexia nervosa is characterized by low body weight (< 85 percent of expected weight), intense fear of weight gain, and an inaccurate perception of body weight or shape (DSM-IV). Its mean age of onset is 17 years (DSM-IV).
The causes of eating disorders are not known with precision but are thought to be a combination of genetic, neurochemical, psychodevelopmental, and sociocultural factors (Becker et al., 1999; Kaye et al., 1999). Comorbid mental disorders are exceedingly common, but interrelationships are poorly understood. Comorbid disorders include affective disorders (especially depression), anxiety disorders, substance abuse, and personality disorders (Herzog et al., 1996). Anorexia nervosa has the most severe consequence, with a mortality rate of 0.56 percent per year (or 5.6 percent per decade) (Sullivan, 1995), a rate higher than that of almost all other mental disorders (Herzog et al., 1996). Mortality is from starvation, suicide, or electrolyte imbalance (DSM-IV). The mortality rate from anorexia nervosa is 12 times higher than that for other young women in the population (Sullivan, 1995).
Treatment of eating disorders entails psychotherapy and pharmacotherapy, either alone or in combination. Treatment of comorbid mental disorders also is important, as is treatment of medical complications. There are some controlled studies of the efficacy of specific treatments for adults with bulimia and binge-eating disorder (Devlin, 1996), but fewer for anorexia nervosa (Kaye et al., 1999). Controlled studies in adolescents are rare for any eating disorder (Steiner and Lock, 1998). Pharmacological studies in young adult women found conflicting evidence of benefit from antidepressants for anorexia and some reduction in the frequency of binge eating and purging with tricyclic antidepressants, monoamine oxidase inhibitors, and SSRIs (see Jimerson et al., 1993; Jacobi et al., 1997). Studies mostly of adult women find cognitive-behavioral therapy and interpersonal therapy to be effective for bulimia and binge-eating disorder (Fairburn et al., 1993; Devlin, 1996; Becker et al., 1999). Clearly, more research is warranted for the treatment of eating disorders, especially because a sizable proportion of those with eating disorders have limited response to treatment (Kaye et al., 1999).
10 Contingency management attempts to alter behavior by manipulating its consequences through the behavioral principles of shaping, positive reinforcement, and extinction.