The Surgeon General's Report on Mental Health
As its name implies, attention-deficit/hyperactivity disorder (ADHD) is characterized by two distinct sets of symptoms: inattention and hyperactivity-impulsivity (see Table 3-3). Although these problems usually occur together, one may be present without the other to qualify for a diagnosis (DSM-IV). Inattention or attention deficit may not become apparent until a child enters the challenging environment of elementary school. Such children then have difficulty paying attention to details and are easily distracted by other events that are occurring at the same time; they find it difficult and unpleasant to finish their schoolwork; they put off anything that requires a sustained mental effort; they are prone to make careless mistakes, and are disorganized, losing their school books and assignments; they appear not to listen when spoken to and often fail to follow through on tasks (DSM-IV; Waslick & Greenhill, 1997).
The symptoms of hyperactivity may be apparent in very young preschoolers and are nearly always present before the age of 7 (Halperin et al., 1993; Waslick & Greenhill, 1997). Such symptoms include fidgeting, squirming around when seated, and having to get up frequently to walk or run around. Hyperactive children have difficulty playing quietly, and they may talk excessively. They often behave in an inappropriate and uninhibited way, blurting out answers in class before the teacher’s question has been completed, not waiting their turn, and interrupting often or intruding on others’ conversations or games (Waslick & Greenhill, 1997).
Many of these symptoms occur from time to time in normal children. However, in children with ADHD they occur very frequently and in several settings, at home and at school, or when visiting with friends, and they interfere with the child’s functioning. Children suffering from ADHD may perform poorly at school; they may be unpopular with their peers, if other children perceive them as being unusual or a nuisance; and their behavior can present significant challenges for parents, leading some to be overly harsh (DSM-IV).
Inattention tends to persist through childhood and adolescence into adulthood, while the symptoms of motor hyperactivity and impulsivity tend to diminish with age. Many children with ADHD develop learning difficulties that may not improve with treatment (Mannuzza et al., 1993). Hyperactive behavior is often associated with the development of other disruptive disorders, particularly conduct and oppositional-defiant disorder (see Disruptive Disorders). The reason for the relationship is not known. Some believe that the impulsivity and heedlessness associated with ADHD interfere with social learning or with close social bonds with parents in a way that predisposes to the development of behavior disorders (Barkley, 1998).
Even though a great many children with this disorder ultimately adjust (Mannuzza et al., 1998), some—especially those with an associated conduct or oppositional-defiant disorder—are more likely to drop out of school and fare more poorly in their later careers than children without ADHD. As they grow older, some teens who have had severe ADHD since middle childhood experience periods of anxiety or depression. This seems to be especially common in children whose predominant symptom is inattention (Morgan et al., 1996). Excellent reviews of ADHD can be found in DSM-IV and other sources5.
ADHD, which is the most commonly diagnosed behavioral disorder of childhood, occurs in 3 to 5 percent of school-age children in a 6-month period (Anderson et al., 1987; Bird et al., 1988; Esser et al., 1990; Pelham et al., 1992; Shaffer et al., 1996c; Wolraich et al., 1996). Pediatricians report that approximately 4 percent of their patients have ADHD (Wolraich et al., 1990), but in practice the diagnosis is often made in children who meet some, but not all, of the criteria recommended in DSM-IV (Wolraich et al., 1990) (see also Treatment later in this section). Boys are four times more likely to have the illness than girls are (Ross & Ross, 1982). The disorder is found in all cultures, although prevalences differ; differences are thought to stem more from differences in diagnostic criteria than from differences in presentation (DSM-IV).
The exact etiology of ADHD is unknown, although neurotransmitter deficits, genetics, and perinatal complications have been implicated. In the early post-World War II years, a number of pediatricians, neurologists, and child psychiatrists noted that brain-damaged children were often hyperactive (Strauss & Lehtinen, 1947; Eisenberg, 1957; Laufer & Denhoff, 1957). These observations led to the diagnostic concept of“minimal brain damage” (Wender, 1971), which was thought to be characterized by hyperactivity, inattention, learning difficulties, and a wide variety of behavior problems. However, large epidemiological studies (Rutter & Quinton, 1977) of grossly brain-damaged children with cerebral palsy, epilepsy, and so forth, did not find an excess of hyperactivity, and more recent imaging studies have found no evidence of gross brain damage in children with ADHD (Swanson et al., 1998). The past view that ADHD is a form of minimal brain damage has therefore been abandoned by experts. Many brain-damaged children are, if anything, significantly underactive.
In the late 1970s, it was postulated that the core problem in hyperkinetic children was one of inattention (Douglas & Peters, 1979). This view led, in 1980, to the adoption, in the official DSM-III (American Psychiatric Association, 1980) nomenclature, of the new diagnostic label attention-deficit disorder.
Because the symptoms of ADHD respond well to treatment with stimulants, and because stimulants increase the availability of the neurotransmitter dopamine, the“dopamine hypothesis” has gained a wide following. The dopamine hypothesis posits that ADHD is due to inadequate availability of dopamine in the central nervous system. The neurotransmitter dopamine plays a key role in initiating purposive movement, increasing motivation and alertness, reducing appetite, and inducing insomnia, effects that are often seen when a child responds well to methylphenidate. The dopamine hypothesis has thus driven much of the recent research into the causes of ADHD.
The fact that ADHD runs in families suggests that inheritance is an important risk factor. Between 10 and 35 percent of children with ADHD have a first-degree relative with past or present ADHD. Approximately one-half of parents who had ADHD have a child with the disorder (Biederman et al., 1995). Over the past decade, a large number of twin studies have shown that, when ADHD is present in one twin, it is significantly more likely also to be present in an identical twin than in a fraternal twin (Goodman & Stevenson, 1989). These findings have led geneticists to estimate that genes are important in a high proportion of children with ADHD.
Research to pinpoint abnormal genes is honing in on two genes: a dopamine-receptor (DRD) gene on chromosome 11 and the dopamine-transporter gene (DAT1) on chromosome 5 (Cook et al., 1995; Smalley et al., 1998). Several studies have found evidence that children with ADHD have genetic variations in one of the dopamine-receptor genes (DRD4), although the largest of these studies suggests that the presence of such a variation is associated with only a modest increase in the risk of developing ADHD (Smalley et al., 1998). Several other studies have found evidence for abnormalities of the dopamine-transporter gene (DAT1) in children with very severe forms of ADHD (Cook et al., 1995; Gill et al., 1997; Waldman et al., 1998).
Yet for most children with ADHD, the overall effects of these gene abnormalities appear small, suggesting that nongenetic factors also are important. Although none of the many imaging studies have found evidence of gross brain damage, some investigators have suggested that exposure to toxins, such as lead, or episodes of oxygen deprivation for the fetus, as may occur during some complications of pregnancy, may adversely affect dopamine-rich areas of the brain. These theories support observations that hyperactivity and inattention are more common in children whose mothers smoked during pregnancy (Nichols & Chen, 1981), in children who have been exposed to high quantities of lead (Needleman et al., 1990), and in children who had a lack of oxygen in the neonatal period (Whittaker et al., 1997).
Some investigators have noted that the parents of hyperactive children are often overintrusive and overcontrolling (Carlson et al., 1995). It has therefore been suggested that such parental behavior is another possible risk factor for ADHD. However, others have noted that, when children are treated with methylphenidate, there is a reduction in parental negativity and intrusiveness. This suggests that the observed overintrusive and overcontrolling behavior of the parent is a response to the child’s behavior rather than the cause (Barkley et al., 1985).
The American Academy of Child and Adolescent Psychiatry (AACAP) published “practice parameters” (i.e., guidelines for clinical practice) on the diagnosis and treatment of ADHD. The AACAP parameters include an extensive literature review, detailed descriptions of the clinical presentation of the disorder, and recommendations for treatment. The practice parameters state that“the cornerstones of treatment are support and education of parents, appropriate school placement, and pharmacology” (AACAP, 1991). These practice parameters evolved out of research relating to two major types of treatment: pharmacological treatment and psychosocial treatment, particularly behavioral modification, as well as multimodal treatment, the combination of psychosocial and pharmacological treatments.
Pharmacological treatment with psychostimulants is the most widely studied treatment for ADHD. Stimulant treatment has been used for childhood behavioral disorders since the 1930s (Bradley, 1937). Psychostimulants are highly effective for 75 to 90 percent of children with ADHD. At least four separate psychostimulant medications consistently reduce the core features of ADHD in literally hundreds of randomized controlled trials: methylphenidate, dextroamphetamine, pemoline, and a mixture of amphetamine salts (Spencer et al., 1995; Greenhill, 1998a, 1998b; Greenhill et al., 1998).
These medications are metabolized, leave the body fairly quickly, and work for 1 to 4 hours. Administration is timed to meet the child’s school schedule, to help the child pay attention and meet his or her academic demands, and to mitigate side effects. These medications have their greatest effects on symptoms of hyperactivity, impulsivity, and inattention and the associated features of defiance, aggression, and oppositionality. They also improve classroom performance and behavior and promote increased interaction with teachers, parents, and peers. Small effects were found on learning and school achievement (see reviews by Barkley, 1990; Pelham, 1993; Swanson et al.,1993, 1995b; Greenhill et al., 1998; Cantwell, 1996a; Spencer et al., 1996.) However, psychostimulants do not appear to achieve long-term changes in outcomes such as peer relationships, social or academic skills, or school achievement (Pelham et al., 1998).
Children who do not respond to one stimulant may respond to another (Elia et al., 1991; Elia & Rapoport, 1991). Children should be reevaluated without the medication to see if stimulant treatment is still indicated. Many families choose to have their child take a“drug holiday” on weekends and vacations to reduce overall exposure, but the utility of this strategy has not been demonstrated (AACAP, 1991).
Stimulants are usually started at a low dose and adjusted weekly (AACAP, 1991). A recent study demonstrated that the practice of dosing methylphenidate on the basis of body weight fails to predict the optimal dose of medication (Rapport & Denney, 1997). One of the goals of the recently completed NIMH Multimodal Treatment Study of ADHD (described more fully below) was to develop medication strategies to guide“best dose,” dose changes, management of side effects, and integration with other treatments (Greenhill et al., 1996).
Common stimulant side effects include insomnia, decreased appetite, stomach aches, headaches, and jitteriness. Some children may develop tics, but a recent study suggests that they disappear with continued treatment (Gadow et al., 1995). Rebound activation (i.e., a sudden increase in attention deficit and hyperactivity) has been noted anecdotally after the child’s last dose of medication wears off (Johnston et al., 1988). Most of the side effects are mild, recede over time, and respond to dose changes. Children rarely experience cognitive impairment, which, if it does occur, can be resolved with reduction or cessation of the drug (Cantwell, 1996). A few cases of psychosis have been reported. Pemoline has been associated with hepatotoxicity, so monitoring of liver function is necessary. Two studies have shown no long-term effects of stimulants on later height or weight (Klein & Mannuzza, 1988; Vincent et al., 1990). Nonetheless, regular precautionary monitoring of weight and height for children on stimulants is recommended.
For children with ADHD who do not respond to stimulants (10 to 30 percent) or cannot tolerate the side effects, there are other useful medications. The antidepressant bupropion has been found to be superior to placebo, although the response is not as strong as that found with stimulants (Cantwell, 1998). Bupropion can also be used as an adjunct to augment stimulant treatment. Well-controlled trials have shown tricyclic antidepressants to be superior to placebo but less effective than stimulants (Elia et al., 1991; Elia & Rapoport, 1991). Reports of sudden death of a few children in the early 1990s on the tricyclic compound desipramine led to great caution with the use of tricyclics in children (Riddle et al., 1991).
Considerable controversy surrounds the use of central alpha-adrenergic blocking drugs, such as clonidine and guanfacine, to treat ADHD. There is some evidence that clonidine is effective for ADHD when it occurs with a tic disorder (Hunt, 1987; Hunt et al., 1990, 1995). Caution is warranted in view of the four cases of sudden death that have been reported in children taking methylphenidate and clonidine together and of a number of reports of nonfatal cardiac side effects in children taking clonidine alone or in combination (Swanson et al., 1995a).
Neuroleptics have been found to be occasionally effective (Green, 1995), yet the risk of movements disorders, such as tardive dyskinesia, makes their use problematic. Lithium, fenfluramine, or benzodiazapines have not been found to be effective treatments for ADHD (Cantwell, 1996a; Green, 1995), nor have SSRIs, such as fluoxetine (Goldman et al., 1998). Furthermore, more than 20 studies have shown that dietary manipulation (e.g., the Feingold diet) is not efficacious (Mattes & Gittelman, 1981), and controlled studies failed to demonstrate that sugar exacerbates the symptoms of children with ADHD (Milich & Pelham, 1986).
Important options for the management of ADHD are psychosocial treatments, particularly in the form of training in behavioral techniques for parents and teachers. Behavioral techniques, which are described more fully below, typically employ“time-out,” point systems and contingent attention (adults reinforcing appropriate behavior by paying attention to it). Psychosocial treatments are useful for the child who does not respond to medication at all or for whom the therapeutic benefits of the medication have worn off and for the child who responds only partially to medication or cannot tolerate medication. In addition, some families express a strong preference not to use medication. Even children who are receiving medication may continue to have residual ADHD symptoms or symptoms from other disorders, such as oppositional defiant disorder or depression, which make specialized child management skills necessary and helpful (see next section, Multimodal Treatments). Furthermore, children with ADHD can present a challenge that puts significant stress on the family. Skills training for parents can help reduce this stress on parents and siblings.
The main psychosocial treatments for ADHD are behavioral training for parent and teacher, as well as systematic programs of contingency management (this behavioral technique is described in more detail in the Treatment section later in this chapter). Of these options, systematic programs of intensive contingency management conducted in specialized classrooms or summer camps with the setting controlled by highly trained individuals is the most effective (Abramowitz et al., 1992; Carlson et al., 1992; Pelham & Hoza, 1996). The efficacy of behavioral training of teachers is well-established, while the evidence for parent training is less solid, according to the criteria, noted earlier, promulgated by the American Psychological Association Task Force (Pelham et al., 1998). There is, however, indirect support for the effectiveness of parent training in the literature, demonstrating the efficacy of parent training for children with oppositional defiant disorder who share many characteristics with children who have ADHD (see section on Disruptive Disorders).
A number of studies have compared parent training (Gittelman et al., 1980; Firestone et al., 1986; Horn et al., 1987, 1990, 1991; Pelham et al., 1988) or school-based behavioral modification (Gittelman et al., 1980; Pelham et al., 1988) with the use of stimulants. Most of the studies are of outpatient behavioral therapy programs in which parents meet in groups and are taught behavioral techniques such as time out, point systems, and contingent attention. Teachers are taught similar classroom strategies, as well as the use of a daily report card for parents that evaluates the child’s in-school behavior. The improvements in the symptoms of ADHD achieved with psychosocial treatments are not as large as those found with psychostimulants (Pelham et al., 1998). Behavioral interventions tend to improve targeted behaviors or skills but are not as helpful in reducing the core symptoms of inattention, hyperactivity, or impulsivity. Questions remain about the effectiveness of these treatments in other settings. To be fully effective, treatments for ADHD need to be conducted across settings (school, home, community) and by different people (e.g., parents, teachers, therapists)—a consistency and comprehensiveness that can be hard to achieve.
Cognitive-behavioral therapy (CBT), primarily training in problem solving and social skills, has not been shown to provide clinically important changes in behavior and academic performance of children with ADHD (Pelham et al., 1998). However, CBT might be helpful in treating symptoms of accompanying disorders such as oppositional defiant disorder, depression, or anxiety disorders (Abikoff, 1985; Hinshaw & Ehardt, 1991; Lochman, 1992).
Although there are no studies evaluating the efficacy of psychoeducation as a treatment modality for ADHD, providing information to parents, children, and teachers about ADHD and treatment options is considered critical in the development of a comprehensive treatment plan (AACAP, 1991). Educational accommodations for children with ADHD are federally mandated, and mental health providers are required to ensure that patients and families have access to adequate and appropriate educational resources. Organizations such as Children and Adults with Attention Deficit Disorder (CHADD) and the National Attention Deficit Disorder Association can be helpful sources of information and support for families.
Many researchers and families have long suspected that multimodal treatment—medication used together with multiple psychosocial interventions in multiple settings—should be more effective than medication alone. Multimodal treatment has thus been used in the absence of empirical support (Hechtman, 1993). To determine whether multimodal treatment is indeed effective, the recent NIMH Multimodal Treatment Study of ADHD (called the MTA Study) examined three experimental conditions: medication management alone, behavioral treatment alone, or a combination of medication and behavioral treatments. The study compared the effectiveness of these three treatment modes with each other and with standard care provided in the community (the control group). The behavioral treatment condition consisted of parent training, a school intervention, and a summer treatment program. The MTA Study was also designed to determine the relative benefits of these treatments over time (Richters et al., 1995). All subjects were treated for 14 months and then followed for an additional 22 months.
Results of the MTA Study comparing the 14-month outcomes of 579 children randomly assigned to one of the four treatment conditions were presented in the fall of 1998 (MTA Cooperative Group, 1998). At 14 months, medication and the combination treatment were generally more effective than the behavioral treatment alone or the control treatment. Notably, the combined treatment resulted in significant improvement over the control condition in six outcome areas—social skills, parent child relations, internalizing (e.g., anxiety) symptoms, reading achievement, oppositional and/or aggressive symptoms, and parent and/or consumer satisfaction—whereas the single forms of treatment (medication or behavior therapy) were each superior to the control condition in only one to two of these domains. The conclusions from this major study are that carefully managed and monitored stimulant medication, alone or combined with behavioral treatment, is effective for ADHD over a period of 14 months. Addition of behavioral treatment yields no additional benefits for core ADHD symptoms but appears to provide some additional benefits for non-ADHD-symptom outcomes.
Overprescription of Stimulants
Concerns have been raised that children, particularly active boys, are being overdiagnosed with ADHD and thus are receiving psychostimulants unnecessarily. However, recent reports found little evidence of overdiagnosis of ADHD or overprescription of stimulant medications (Goldman et al., 1998; Jensen et al., 1999). Indeed, fewer children (2 to 3 percent of school-aged children) are being treated for ADHD than suffer from it. Treatment rates are much lower for selected groups such as girls, minorities, and children receiving care though public service systems (Bussing et al., 1998a, 1998b). However, there have been major increases in the number of stimulant prescriptions since 1989 (Hoagwood et al., 1998), and methylphenidate is being manufactured at 2.5 times the rate of a decade ago (Goldman et al., 1998). Most researchers believe that much of the increased use of stimulants reflects better diagnosis and more effective treatment of a prevalent disorder. Medical and public awareness of the problem of ADHD has grown considerably, leading to longer treatment, fewer interruptions in treatment, and increased treatment of adults. Adolescents and younger girls with ADHD, who were underdiagnosed in the past, are being identified and treated.
Nonetheless, some of the increase in use may reflect inappropriate diagnosis and treatment. In one study, the rate of stimulant treatment was twice the rate of parent-reported ADHD, based on a standardized psychiatric interview (Angold & Costello, 1998). While many children who do meet the full criteria for ADHD are not being treated, the majority of children and adolescents who are receiving stimulants did not fully meet the criteria. These findings may reflect a failure of proper, comprehensive evaluation and diagnosis rather than a failure of the diagnostic criteria, which are clear and validated by research (Angold & Costello, 1998). A diagnosis of ADHD requires the presence of impairing ADHD symptoms in multiple settings for at least 6 months. Although fidgeting and not paying attention are normal, common childhood behaviors, DSM-IV criteria reserve a diagnosis of ADHD for children in whom such frequent behavior produces persistent and pervasive dysfunction. An adequate diagnostic evaluation requires histories to be taken from multiple sources (parents, child, teachers), a medical evaluation of general and neurological health, a full cognitive assessment including school history, use of parent and teacher rating scales, and all necessary adjunct evaluation (such as assessment of speech, language). These evaluations take time and require multiple clinical skills. Regrettably, there is a dearth of appropriately trained professionals.
Family practitioners are more likely than either pediatricians or psychiatrists to prescribe stimulants and less likely to use diagnostic services, provide mental health counseling, or provide followup care (Hoagwood et al., 1998). The American Academy of Pediatrics published a policy statement in 1996 on the use of medication for children with attentional disorders, concluding that use of medication should not be considered the complete treatment program for children with ADHD and should be prescribed only after a careful evaluation (American Academy of Pediatrics Committee on Children With Disabilities and Committee on Drugs, 1996).
Safety of Long-Term Stimulant Use
Even though the MTA Study found no safety issues over a 14-month period (Greenhill et al., 1998), concerns have been raised about the longer term safety of stimulant treatment. Since ADHD has an early onset and requires an extended course of treatment, research is needed to examine the long-term safety of treatment and to investigate whether other forms of treatment could be combined with psychostimulants to lower their dose as well as to reduce other problem behaviors found with ADHD. Such combined treatments could be targeted for symptoms of disorders that often accompany ADHD, such as conduct disorder, substance abuse, and learning disabilities, and could be targeted to improve overall functioning (Laufer, 1971; Gittelman et al., 1985).
Because stimulants are also drugs of abuse and because children with ADHD are at increased risk for a substance abuse disorder, concerns have also been raised about the potential for abuse of stimulants by children taking the medication or diversion of the drug to others. While stimulants clearly have abuse potential, the rate of lifetime nonmedical methylphenidate use has not significantly increased since methylphenidate was introduced as a treatment for ADHD, suggesting that abuse is not a major problem (Goldman et al., 1998). Case reports describing abuse by children prescribed stimulants for ADHD are rare (Hechtman, 1985).
5 Taylor, 1994; Cantwell, 1996; Waslick & Greenhill, 1997; Barkley, 1998; and NIH Consensus Statement 110, 1998.