essay--psychological intervention
199
In the next two chapters we discuss attention- deficit/ hyperactivity disorder (ADHD) or hyperkinesis, and Tourette syndrome. These are seen as neuropsychiat- ric disorders because of their frequent association with central nervous system manifestations. Neurological signs and symptoms are part of the definition of Tour- ette syndrome; it is also frequently associated with hy- perkinesis and learning difficulties. The situation with the attention-deficit disorders (ADDs) varies depending on the definitions used, but they are frequently associ- ated with learning difficulties. There is other evidence of brain involvement, such as epilepsy, clumsiness, lan- guage delay, and a variety of syndromes (e.g., fragile X syndrome, fetal alcohol syndrome). In this chapter we review the clinical presentation of ADHD, then sum- marize the literature on treatment. We refer only to the more robust studies from the vast literature on ADDs.
DefiNitioN
ADHD is characterized by reduced levels of concen- tration or attention, impulsivity, and overactivity or restlessness. There is no clear demarcation between extremes of normality and truly abnormal degrees of these behaviors (National Institute for Health and Clinical Excellence [NICE], 2008; J. Williams, 2008). ADHD has been variously named “minimal brain dys-
function,” “hyperkinesis,” “attention deficit disorder” (ADD), “attention deficit disorder with hyperactivity” (ADD-H), and “attention- deficit/hyperactivity disor- der” (ADHD) at different times and in different loca- tions. Opinions differ about the interchangeability of these names. DSM-5, used in the United States and Australia, refers to the diagnosis of “attention-deficit/ hyperactivity disorder.” In Europe, the ICD-10 (World Health Organization, 1993) classification system refers to “hyperkinetic disorder,” which, in practical terms, is a narrower term describing a subset of individuals with ADHD. Inattention is a required diagnostic criterion for hyperkinetic disorder but not for ADHD. Pervasiveness of the problems and absence of significant anxiety are also criteria for hyperkinetic disorder. Children so di- agnosed tend to have other neurodevelopmental delays (E. Taylor, Sandberg, Thorley, & Giles, 1991). J. Wil- liams (2008, p. 706) has proposed a relatively new con- ceptual framework for understanding ADHD, which he refers to as the “extended temporal difference model.” He explains that “ADHD does not have a single, simple, core problem, but rather . . . ADHD is a collection of disparate neurodevelopmental problems, which mainly cohere through the effectiveness of stimulants and the practical need for a very small number of diagnoses.” This theory is in contrast to that of Tripp and Wickens (2008), who, like Antrop et al. (2006), note the atypi- cal response of children with ADHD to positive rein-
c h a p t e r 6
AtteNtioN‑Deficit/hYPerActivitY DisorDer
C o p y r i g h t 2 0 1 5 . T h e G u i l f o r d P r e s s .
A l l r i g h t s r e s e r v e d . M a y n o t b e r e p r o d u c e d i n a n y f o r m w i t h o u t p e r m i s s i o n f r o m t h e p u b l i s h e r , e x c e p t f a i r u s e s p e r m i t t e d u n d e r U . S . o r a p p l i c a b l e c o p y r i g h t l a w .
EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY AN: 826932 ; Peter Fonagy, David Cottrell, Jeannette Phillips, Dickon Bevington, Danya Glaser, Elizabeth Allison.; What Works for Whom?, Second Edition : A Critical Review of Treatments for Children and Adolescents Account: s3094162.main.ehost
200 W h at Wo r k s f o r W h o m ?
forcement; these authors have proposed a “dopamine transfer deficit” as an explanation. It remains to be seen whether either of these models can fully explain the features of ADHD and characteristic responses to interventions.
DSM-5 describes three types of ADHD:
• Predominantly inattentive (six or more symptoms of inattention present up to age 16, or five or more symp- toms for adolescents ages 17 and older and adults. Symptoms of inattention must have been present for at least 6 months and are inappropriate for the devel- opmental level).
• Predominantly hyperactive– impulsive (six or more symptoms of hyperactivity– impulsivity present up to age 16, or five or more symptoms for adolescents ages 17 and older and adults. Symptoms of hyperactivity– impulsivity must have been present for at least 6 months to a level that is disruptive and inappropriate for the developmental level).
• Combined (both sets of symptoms).
Several symptoms must have been present before age 12 and some must have been present in two or more settings (in the DSM-IV description, some symptoms must have been present before age 7 and have persisted longer than 6 months). ICD-10 requires an age of onset before age 6.
Perceptions of hyperactivity may vary across cul- tures (E. M. Mann et al., 1992). However, Prendergast et al. (1988) found that if a diagnosis is based on the re- spective criteria of the DSM and ICD classification sys- tems, there is agreement on “caseness.” In other words, clinicians on either side of the Atlantic could be trained to use the other side’s criteria.
Since the vast majority of the relevant literature on attention deficit problems comes from the United States rather than Europe, in this chapter we use the term ADHD, as defined in DSM-5, to describe ADDs. It should be noted that inclusion criteria for studies will refer to DSM-IV (or earlier) criteria as the research re- viewed was conducted before DSM-5 was published.
PrevAleNce
Suspected hyperkinesis or ADHD is a frequent rea- son for referral to child and adolescent mental health services in the United States. Similarly, in the United Kingdom, the numbers of cases diagnosed and treated
have been increasing over the years. For example, in 2006, the total annual cost of prescribed stimulants and other drugs for ADHD in England was approximately £29 million, an increase of 20% on the previous year’s cost (National Health Service (NHS) Health and So- cial Care Information Centre, 2006; NHS Information Centre, 2007).
Prevalence varies according to the diagnostic sys- tem and criteria used. The American Psychiatric As- sociation (1994) estimated the prevalence of ADHD in school- age children at between 3 and 5%. E. Taylor et al. (1991) found that 1.7% of 7-year-old boys in a Lon- don population survey fulfilled criteria for the more restricted diagnosis of hyperkinetic disorder. The au- thors commented that this probably overestimated the true prevalence of hyperkinetic disorder in the general population, because it is more common in boys and in urban areas, and 7 years is the peak age for recognition of the disorder. They suggested that a hyperkinetic dis- order prevalence of at least 0.5% in the general popula- tion may be more accurate. Studies of the prevalence of the less restricted diagnosis of ADD-H or ADHD indi- cate a prevalence in children in the range of 4.2–12% (August, Ostrander, & Bloomquist, 1992; Bergeron, Valla, & Breton, 1992; NICE, 2008; Polanczyk et al., 2007). In the United Kingdom, Ford, Goodman, and Meltzer (2003) reported a prevalence of 2.23% for “any ADHD.” In adolescents, the prevalence of ADHD is in the range of 1.5–5.0% (Bergeron et al., 1992; Lewinsohn, Hops, Roberts, Seeley, & Andrews, 1993).
ADHD is thought to occur much more frequently in males. The American Psychiatric Association (1987) reports a male:female ratio in clinic- referred samples ranging from 9:1 to 6:1. This contrasts with a ratio of 3:1 from population- based studies (Szatmari, Offord, & Boyle, 1989).
cliNicAl PreseNtAtioN
The authors recommend reading the American Acad- emy of Pediatrics clinical practice guideline on the diagnosis and evaluation of ADHD (Subcommittee on Attention- Deficit/Hyperactivity Disorder et al., 2011) and the NICE (2008) clinical guidelines on ADHD.
Children with ADHD and, to a greater extent, hy- perkinesis show impaired functioning in most areas of their lives, including home and school settings. Their impulsivity and inattention lead to frequent criticism from relatives, peers, and teachers, and their hyperac-
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 201
tive and often aggressive behavior leads to peer rejec- tion.
The peak age of referral for ADHD or hyperkinesis is between 7 and 9 years, but peak age of onset appears to be in the preschool period at around 3 years. The onset of the more restricted diagnosis of hyperkinetic disorder is probably earlier, at age 2 or 3 years, because of the associated neurodevelopmental problems (Thor- ley, 1984). Children with attention problems without hyperactivity have later onset at around 6 years (S. M. Green, Loeber, & Lahey, 1991).
A meta- analysis of 18 studies, describing differences based on relevant variables, highlighted gender differ- ences in presentation (Gaub & Carlson, 1997). Clinic- referred girls with ADHD have lower levels of hyperac- tivity and fewer externalizing behavior problems than boys, but greater intellectual problems. The sexes did not differ in terms of impairment due to inattention, internalizing behavior, peer aggression, and being dis- liked. Among nonreferred children, however, boys with ADHD were more likely to be impaired by inattention, internalizing behavior problems, and peer relationship difficulties. These findings suggest that many of the gender differences may be attributable to referral bias; that is, there may be different thresholds for the referral of boys and girls. Boys are more likely to be referred for treatment than girls because of the increased rate of disruptive externalizing behavior problems in boys. This could mean that although boys with ADHD are generally more severely impaired than girls, studies of clinic samples fail to detect these differences, because only the most severely affected girls are referred. As mentioned earlier, the ratio of boys:girls attending treat- ment clinics with ADHD ranges from 6:1 to 9:1, despite the lower ratio of ADHD-affected boys:ADHD-affect- ed girls in the community of 3:1 (R. T. Brown, Madan- Swain, & Baldwin, 1991). Further research is needed to determine whether boys and girls with ADHD are receiving appropriate services.
The behaviors causing impairment in ADHD are dif- ficult to measure. Although there is an understanding of the behaviors that must be present for the diagnosis to be made, recognizing the point at which the behaviors in question become abnormal can be problematic, espe- cially in very young children. ADHD often has its onset in the second or third year of life, but since children this age are normally active and have a short attention span, diagnosis in this age group is more problematic than in older children. The varying presentation of ADHD also adds to the difficulty of accurate diagnosis.
The etiology for the primary problems of inattention and hyperactivity is heterogeneous (Curran & Taylor, 2000). In many cases, multiple secondary factors, such as parental disharmony, bullying, and additional learn- ing difficulties, produce the behavior problems that lead to referral.
At assessment, the parents and child should be in- terviewed. Interviews should include an inquiry about possible comorbidity and environmental causes of the problems. It is also important to ask about the home situation (e.g., parental conflict, financial hardship, parental mental disorder), and about the child’s educa- tional progress. This information can be obtained with use of standardized rating scales and an open report from the school; classroom observations are also in- formative. A variety of questionnaires are available: the best validated are the Child Behavior Checklist (Achenbach, 1991; Achenbach & Edelbrock, 1983), which includes a teacher report form, and the Conners Parent and Teacher Rating Scales (Goyette, Conners, & Ulrich, 1978). Children should be medically examined and have a comprehensive assessment of their learning abilities, including reading and writing. A helpful study of the performance of 621 children in neuropsychologi- cal testing found only 61.9% reliability from such tests (Pineda, Puerta, Aguirre, Garcia- Barrera, & Kam- phaus, 2007); these should therefore be used with cau- tion as an aid to diagnosis, although they can be helpful in identifying cognitive deficits. The clinical practice guideline (Subcommittee on Attention- Deficit/Hyper- activity Disorder et al., 2011) states that the evaluation of children and adolescents presenting with possible ADHD should include assessment for other emotional/ behavioral, developmental, and physical (e.g., tics or sleep apnea) conditions that may be comorbid with ADHD.
There is no evidence that screening for ADHD is beneficial. An article by Sayal et al. (2010) described a 5-year follow- up of a school- based randomized con- trolled trial (RCT) exploring the benefits of screen- ing (followed by an educational program for teachers of children identified as having significant features of ADHD) in children ages 4–5 years. The authors screened 1,662 children from 332 participating schools. There was no evidence of benefit in terms of long-term symptomatology; in fact, the authors suggested that the children who were identified to their teachers as having ADHD problems fared worse than children for whom nothing was done. The authors suggest that if school personnel decide to screen for ADHD, they
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
202 W h at Wo r k s f o r W h o m ?
should make further detailed assessments for learning difficulties or other comorbid psychiatric problems in any child identified by screening, because these other problems may be missed, and the child could easily be wrongly labeled by screening in this way.
comorbiDitY
Biederman, Newcorn, and Sprich (1991) reviewed the literature and reported that comorbidity with ADHD is high: 30–50% of children with ADHD have comor- bid conduct or oppositional disorders, 15–75% have a mood disorder, and 25% have an anxiety disorder. The type of comorbid problem seems to vary between two subtypes of ADHD: Those with hyperactivity are more likely to have a comorbid conduct disorder, while those without hyperactivity are more at risk of internalizing symptoms, such as anxiety and depression (Barkley, DuPaul, & McMurray, 1990a; D. P. Cantwell & Baker, 1992).
It has been reported that 50% of children with ADDs have speech or language impairments (Love & Thomp- son, 1988). Similar numbers go on to have reading and writing difficulties. Visuomotor problems and clumsi- ness are also common (Losse et al., 1991; E. Taylor et al., 1991). Hellgren, Gillberg, Gillberg, and Enerskog (1993) found that visuomotor problems persist into adolescence, placing the young person at higher risk of accidents than those in a control group. E. Taylor et al. (1991) similarly reported a higher rate of accidents in pervasively hyperactive youngsters. Richters et al. (1995) reported that 10–25% of children with ADHD have comorbid learning disorders. Problems with so- cial relationships are often associated with ADHD. T. Clark, Feehan, Tinline, and Vostanis (1999) reported a retrospective case study in which 65–80% of 49 sub- jects with ADHD reported significant difficulties in social interaction and communication.
GeNetic fActors
Several multicenter studies have now confirmed the significant impact of genetic factors, with estimates of heritability ranging between 60 and 90% (Asherson et al., 2007). Studies to clarify the exact genotypic abnor- malities are still under way. Asherson et al. reported the significant contribution of a gene affecting dopamine transport, DAT1, although the contribution is likely to
be fairly small. Lowe, Barry, Gill, and Hawi (2006) presented a list of genes that are possibly involved (multiple susceptibility genes), including several dopa- minergic and serotonergic genes, and the gene encod- ing synaptosomal- associated protein-25 (SNAP-25).
Polanczyk et al. (2007) reported a study demonstrat- ing the involvement of the alpha2A adrenergic recep- tor. In this study, methylphenidate was used as a treat- ment in children with ADHD symptoms who had the ADRA2A-1291 C → G polymorphism (the “G allele”) in the alpha2A receptor gene. There was a significant interaction effect between the presence of the G allele and treatment with methylphenidate over time, leading the authors to conclude that the (nor)adrenergic ner- vous system must be involved in modulating the action of methylphenidate. Studies are still under way to iden- tify the genes involved in the etiology and treatment response.
Studies are beginning to emerge that describe abnor- malities on brain scans of subjects with ADHD. Castel- lanos et al. (2002) reported smaller brain volumes in all areas apart from the caudate nucleus and suggested that these abnormalities were persistent and nonresponsive to treatment. Overmeyer et al. (2001) reported grey- matter deficits in the right frontal and posterior cin- gulate gyri and the basal ganglia on both sides. There were also some white- matter deficits.
NAtUrAl historY
Up to 80% of school- age children diagnosed with ADHD or hyperkinesis will have the same diagno- sis 5 years later or in adolescence (see Dreyer, 2006, for a helpful review of the continuity of ADHD from the preschool years and the reliability of the diagno- sis based on DSM-IV criteria). The disorder will per- sist into adulthood in up to 65% of cases (G. Weiss & Hechtman, 1993). At least one-third will be diag- nosed with a conduct disorder and/or have increased rates of substance abuse in adolescence, with problems persisting into adulthood in the form of trouble with the police and personality disorder. Mannuzza, Klein, and Moulton (2002) reported a 10.0- to 12.6-year follow- up study of 250 boys with ADHD and found that boys with pervasive problems (apparent in all set- tings) and those with problems identified by teachers only had worse outcomes in terms of antisocial behav- iors (29% had antisocial disorder) than those who had problems reported only by parents (no cases had anti-
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 203
social disorder). The factors most highly predictive of poor outcome include a family history of ADHD or hyperkinesis, psychosocial adversity, and comorbidity with conduct, mood, and anxiety disorders (Bieder- man et al., 1996). Biederman, Wilens, Mick, Spencer, and Faraone (1999) compared the rate of substance use disorder in adolescence in subjects with ADHD who had received medication (mean duration of treatment 1.7–7.1 years) and in those who had not. None of those treated with medication were receiving treatment at follow- up. The authors found that the untreated sub- jects were at significantly increased risk for substance misuse at follow- up compared with the controls. Medi- cation with stimulants reduced the risk of substance misuse, even in the subjects with ADHD and a comor- bid conduct disorder.
In an important meta- analysis of the impact of ADHD on academic achievement, T. W. Frazier, Youngstrom, Glutting, and Watkins (2007) reported a moderate to large discrepancy between individuals with ADHD and controls. Standardized reading mea- sures were the most impaired.
treAtmeNt
Relative to many other areas of child mental health difficulties, a large number of RCTs have examined the ADDs; these have been predominantly of physical treatments but also of nonphysical interventions. How- ever, the majority have been short-term and have as- sessed the response to a single treatment intervention, making it difficult to compare effectiveness in a given population. The Multimodal Treatment of Attention Deficit Hyperactivity Disorder (MTA) study, which we discuss first, was an attempt to improve on the length of follow- up and to enable between- group comparisons on a larger scale than had been attempted previously. It also included comparison with psychological interven- tions, although there are some concerns about the study design, which we discuss later in the chapter.
the mtA study
The MTA study, an 8-year, multimodal treatment study involving six sites, was set up by the National Institute of Mental Health (NIMH) to address unanswered ques- tions about ADHD and the possible benefits of multi- modal treatments for the disorder, compared with med- ication alone. Findings at 14 months (at the end of the
treatment phase of the study), 24 months, 36 months, and 96 months have been reported. Due to confusion over the results, there have been many discussions of the findings. The first official report (MTA Cooperative Group, 1999a) describes a sample of 539 children with carefully diagnosed DSM-IV combined- type ADHD. The children were randomly allocated to 14 months of treatment in one of four possible groups: (1) medica- tion; (2) intensive behavioral management involving the parents, child, and school; (3) these two treatments combined; or (4) standard care provided by a commu- nity team.
The parent training component of the behavioral in- tervention (Wells et al., 2000) was based on programs developed by Barkley (1987) and Forehand and Mac- Mahon (1981). There were 27 group sessions, with six families per group, plus eight individual sessions per family. The therapist also provided consultation to the school. The child- focused behavioral work was a summer treatment program developed by Pelham and Hoza (1996) as a summer camp and provided in a group format (Pelham et al., 2000). The same therapist who provided the parent training and school consultation supervised the child- focused work. The school- based intervention consisted of a part-time, behaviorally trained classroom assistant working directly with the child in the classroom, using methods suggested by J. M. Swanson (1992), and fortnightly consultation be- tween the therapist and the teacher targeting classroom behavioral management strategies, as described by Pel- ham and Waschbusch (1999).
Medication management in the MTA study began with a double- blind titration of methylphenidate to identify the optimal dose. For children who did not respond adequately to any tested dose of methylphe- nidate, alternative medications (dextroamphetamine [dexamfetamine; see “Methylphenidate and Dexamfet- amine” section later in the chapter], pemoline, imipra- mine and, if necessary, others) were titrated. The care- fully titrated medication treatment, when compared with standard care or behavioral management, was optimal for parents’ and teachers’ ratings of inatten- tion, and teachers’ ratings of hyperactivity– impulsivity. There were no other differences between medication and behavioral management. The combined interven- tion was also superior to behavioral intervention alone on the previously described parameters. Moreover, it was significantly better than behavioral intervention alone in three particular areas: parent- rated opposi- tional/aggressive behaviors, internalizing symptoms,
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
204 W h at Wo r k s f o r W h o m ?
and reading achievement scores. The authors conclude that “Combined behavioral intervention and stimulant medication— multimodal treatment, the current cri- terion standard for ADHD interventions— yielded no significantly greater benefits than medication manage- ment for core ADHD symptoms” (MTA Cooperative Group, 1999a, p. 1078). However, the combined treat- ment was superior to the standard community treat- ment in respect of internalizing symptoms, opposi- tional behaviors, peer relationships, and parent– child interactions. The authors concluded that there is evi- dence that long-term (14 months) stimulant medication is beneficial and safe, and that “carefully monitored” drug treatment may render intensive behavioral inter- ventions unnecessary.
A second article (MTA Cooperative Group, 1999b) reported on the moderators of treatment response. In the presence of comorbid anxiety (present in 34% of the sample) there was a trend toward a better response to the combined treatment than to medication alone, and behavioral intervention alone was better than the standard community care. The presence of comorbid anxiety reduced the relative advantage of medication over the other treatments, but it did not reduce the rate of response to medication. The presence of comorbid oppositional or conduct problems had no impact on the response rate. The authors conclude that improve- ments in ADHD symptoms resulting from medication can lead to a reduction in anxiety symptoms; therefore, some of the anxiety present in ADHD-affected chil- dren may be generated by their distress about problems resulting from their ADHD.
In the first commentary on the responses to the find- ings, P. S. Jensen (1999) suggested that the findings do not mean that behavioral treatment is ineffective. Pel- ham (1999) made a similar argument. First, he noted that the behavioral treatment had a large effect size (ES) improvement (0.9–1.3) from baseline to end point across all measures. Second, he pointed out that the medical management in the MTA study did not differ from withdrawn behavioral treatment (which faded out several months before the 14-month review) for most of the measures. Third, there was no difference between the behavioral treatment and standard community treatment for all of the measures. On this basis, he con- cluded that behavioral treatment is nearly as effective as ongoing medical management.
Hinshaw et al. (2000) analyzed in more detail the changes in parenting that led to an improvement in chil- dren’s social behaviors at school, and found a signifi-
cant correlation between the reduction in negative/inef- fective parental discipline and reduction in children’s school- based disruptive behaviors. Hoza et al. (2000, p. 569) also reported that, based on the MTA study sample, “Low self- esteem in mothers, low parenting efficacy in fathers, and fathers’ attributions of noncom- pliance to their ADHD child’s insufficient effort and bad mood” were all associated with a poorer outcome of treatment for the child. Pelham (1999) pointed out the relevance of this finding for parents, who often pre- fer the combined or behavioral approaches. Another important finding, supported by previous studies, was that the combined approach required a 20% lower dose of medication than the dose required when the medica- tion was used alone. The fact that the beneficial effects of the behavioral treatment appeared to persist several months after treatment ceased supports the notion that combining treatments may allow earlier discontinua- tion of medication.
At 24 months (i.e., 10 months after treatment cessa- tion), the groups of children who had received carefully monitored medication management (either medication alone or combined medication and behavioral treat- ment) showed significantly greater improvements in ADHD and oppositional defiant disorder (ODD) symp- toms than the groups that had received behavioral in- terventions only or the standard community treatment. The combined treatment group was still showing sig- nificantly greater reductions in internalizing symptoms, and better teacher- rated social skills, parent– child re- lations, and reading achievement, compared with the community treatment and behavioral treatment groups. The medication- only group did not fare significantly better than these two groups in these domains. Howev- er, the ESs were reduced by approximately 50% at the 14-month follow- up. P. S. Jensen et al. (2007) reported that an analysis of treatments the groups were receiving at the time of the 24-month follow- up revealed changes in status, such that children originally assigned to the behavioral group may have commenced medication, and those originally assigned to the medication and combined groups may have stopped medication. The authors suggest that this finding could explain the re- duction in ES and between- group differences.
P. S. Jensen et al. (2007) also reported on the 36- month follow- up. At that time, there were no differ- ences between the treatment groups in relation to the five outcome measures of parent- and teacher- rated ADHD and ODD symptoms, reading achievement, so- cial skills, and functional impairment. However, there
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 205
were large improvements in all the groups, with the fol- lowing ranges of ESs for improvement from baseline to 36 months: 1.6–1.7 for ADHD, 0.7 for ODD, 0.9–1.0 for impairment, 0.8–0.9 for social skills, and 0.1–0.2 for reading. Unfortunately, in the absence of a no- treatment control group, it is difficult to know whether this represents a natural tendency to improve; the lack of such a control group has been the main reason for criticism of the otherwise well- designed MTA study.
Molina et al. (2007) reported on rates of delinquency and emerging substance use 36 months after the end of treatment in the MTA study. At that point, the children were between 11 and 13 years of age and were there- fore still at the age where experimentation, rather than substance dependence, was more likely. Delinquency rates in the MTA sample were significantly higher than those of local control children (27.1 vs. 7.4%) at 36- month follow- up. There was also a significant differ- ence in substance use rates in the MTA study subjects compared with controls, with 11.7% of children in the MTA study reporting lifetime use of any substance at 24 months, compared with 5.6% of controls unaffect- ed by ADHD. By 36 months, 17.4% of children in the MTA study reported lifetime use, compared with 7.8% of controls. Twenty-five percent of the original sample evidenced moderate or severe delinquency at 36-month follow- up. There had been a significant decrease in the delinquency rate during the initial 24 months of the study (i.e., 14 months of treatment and 10 months of follow- up). The increase had therefore occurred in the last 12 months of follow- up. It was determined that pre- scription medication was not a risk factor for substance misuse. The children who had received the intensive behavior therapy reported less severe substance misuse at 24 months, but not at 36 months. There were no other differences between treatment groups.
Summary of MTA Study Findings
Table 6.1 summarizes findings from the MTA study, comparing the four treatment groups: (1) intensive med- ication, which required close monitoring of medication and titration of optimal dose; (2) behavioral approach, which combined behavioral work with children, parent training, school consultation, and behavioral input in the classroom; (3) combination of behavioral interven- tion and medication; and (4) usual community treat- ment. Table 6.1 shows the aggregated findings reported at 14 months (end of treatment) and at 24 months (10- month follow- up).
tABle 6.1. summary of findings from the mtA study
Parameter Finding
Parent-rated
Hyperactivity No difference between groups
Impulsivity No difference between groups
Inattention Intensive medication best, but no benefit of combined treatment
Aggression and oppositional behaviors
Combined treatment significantly better than behavioral only
Parent–child interactions
Combined treatment significantly better than behavioral only
Teacher-rated
Hyperactivity Combined treatment and intensive medication best
Inattention Combined treatment and intensive medication best
Impulsivity Combined treatment and intensive medication best
Peer relationships Combined treatment best
Child/adolescent and parent report
Delinquency and substance use
No differences between treatments in delinquency rates; substance misuse reduced with intensive medication
Researcher observations
Internalizing disorders Combined treatment best
Reading ability Combined treatment best
Comorbid anxiety effects
Combined treatment best; behavioral treatment alone better than routine care; intensive medication often as effective as in nonanxious children, but effect size lower
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
206 W h at Wo r k s f o r W h o m ?
By the 36-month follow- up, the differences between treatment groups had disappeared. It is hypothesized that this was due to significant movement between types of treatment as patients returned to their usual community care; for example, those who had previous- ly received medication went on to receive behavioral treatment, and vice versa.
At 6–8 years (96-month follow- up; Molina et al., 2009) the type or intensity of the original 14-month treatment no longer had a significant effect on outcome measures analyzed. The authors found that prognosis was determined by symptom severity, with children with the DSM-IV ADHD combined type (inattention and hyperactivity) faring worse. This suggests that any between- group differences were lost when the intensive interventions ended. Most children showed some im- provement, but this was not to the degree required for normalization. It is not possible to say whether long- term intensive medication would have provided an ad- vantage. In fact, 62% of the children taking medication at 14 months post-MTA treatment had stopped medi- cation at the 6- to 8-year follow- up point. The authors comment on the need to look at motivational issues for adolescents in a bid to improve adherence to medica- tion. They also suggest the need for periodic psychoso- cial interventions.
Physical treatments
Psychostimulants
Methylphenidate and Dexamfetamine
Stimulants are the drugs most frequently prescribed for children with ADHD. They have been shown to be effective in the control of hyperactive and aggressive behavior. Two stimulants are available, methylpheni- date and dexamfetamine (also known as dexamphet- amine or dextroamphetamine). They are both fast- or immediate- acting drugs; after oral ingestion, they begin to work within the first hour. Effects last no more than 4 hours. Several authors (reviewed by Wilens & Biederman, 1992) have reported that the serum concen- tration and onset of action of methylphenidate are simi- lar, with a half-life for both concentration and action of approximately 3 hours. In contrast, the blood level of dexamfetamine peaks at 4 hours and consequently has a longer half-life of 6–8 hours.
Short‑ Acting Stimulants. Methylphenidate remains the most frequently used psychostimulant (J. M. Swanson,
Lerner, & Williams, 1995), despite little evidence for its superiority over other stimulant preparations. It has a short half-life, which often necessitates several doses throughout the day. Up to 40% of children with ADHD do not respond to methylphenidate, although they may respond to other stimulants; however, there is no way of predicting which child will respond to which stimu- lant preparation. In a summary of practice parameters for the use of stimulant medication, Greenhill et al. (2001a) reported that, at the time of writing, 152 RCTs indicated that 65–75% of patients improve with stimu- lants, compared with 5–30% of patients treated with placebo.
Schachter, Pham, King, Langford, and Moher (2001) reported a meta- analysis of studies (not including the MTA) of short- acting methylphenidate. They included 62 randomized trials involving 2,897 children and ado- lescents up to the age of 18 with DSM-III-diagnosed ADDs, with diagnosis made in a systematic and re- producible way. The average length of the trials was 3 weeks, with some lasting up to a maximum of 28 weeks. The overall ES, as reported by teachers, was 0.78 (95% confidence interval [CI] [0.64, 0.91]). The ES reported by parents was 0.54 (CI [0.40, 0.67]). The adverse- event profile (expressed as number needed to harm [NNH]) showed that four children needed to be treated to get one with drug- induced appetite suppres- sion; 22, for the side effect of headache; 26, for insom- nia; and 367, for drowsiness.
A more recent meta- analysis by Faraone and Buite- laar (2010) of RCTs of stimulant therapy for ADHD, published since 1979, identified 23 trials meeting cri- teria of the meta- analysis. The trials studied 11 drugs and used 19 outcome measures related to core ADHD symptoms; the authors therefore had to use some com- plex statistical analyses for the meta- analysis. They found that the ESs for amphetamine products were sig- nificantly larger than those for methylphenidate. Trans- lating this into the number needed to treat (NNT) in order for there to be a significant benefit compared with placebo, two patients needed to be treated with amphet- amine, compared to 2.6 patients with methylphenidate.
A helpful review by Pietrzak, Mollica, Maruff, and Snyder (2006) summarizes the findings of 40 studies of the cognitive effects of immediate- release stimulant medication. The authors report that 63.5% of studies showed some improvement in cognitive functioning.
Long‑Term Stimulant Treatment. Hechtman and Green- field (2003) reported a review of RCTs of stimulant use
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 207
for up to 2 years. Review of two long-term multimodal studies indicated that well- titrated stimulant medica- tion continued to be beneficial for at least 14 months. Even with treatment, subjects with ADHD had less favorable outcomes than nonaffected controls, but the authors note that the treated ADHD group had fewer car accidents and better social skills and self- esteem compared to untreated affected subjects.
Long‑ Acting Methylphenidate. A long- acting formula- tion of methylphenidate (Concerta), designed to provide cover for 12 hours, has been compared with immediate- release methylphenidate in 68 children between ages 6 and 12 years, in structured and unstructured settings, over a trial period of 3 years (Pelham et al., 2001). The study found no significant differences in responses to the two drug preparations, apart from significantly bet- ter parent- reported behavior in the evening for the chil- dren taking the long- acting preparation, compared with those taking the immediate- release preparation. There were no significant side effects from either preparation.
Equasym XL, a controlled- release preparation of methylphenidate similar to Concerta, has been found to be equivalent to immediate- release preparations in onset of action. The peak plasma concentrations are at 1.5 and 4.5 hours (V. R. Anderson & Keating, 2007). Clinical efficacy and high levels of tolerability have been demonstrated in two well- designed placebo- controlled studies (Findling et al., 2006a; Greenhill, Findling, Swanson, & ADHD Study Group, 2002). In the study by Findling et al. (2006a), 318 children between ages 6 and 12 years with a DSM-III-R diagnosis of ADHD were randomized to receive placebo, immediate- release methylphenidate (Ritalin), or controlled- release meth- ylphenidate (Equasym XL). There were equally signifi- cant clinical improvements in the two treatment arms. Overall, the number of children with adverse reactions reported was highest in the placebo group. Adverse reactions leading to withdrawal occurred in 3% of the group receiving the immediate- release preparation and 2.2% of the controlled- release group. The most fre- quent side effect for both preparations was headache. One child treated with the controlled- release prepara- tion developed neutropenia, but this was not thought to be related to the treatment.
Banaschewski et al. (2006) carried out a meta- analysis of studies (published and unpublished) on the use of long- acting medications in ADHD and hy- perkinetic disorder. They discuss the four long- acting methylphenidate hydrochloride preparations available,
namely, Ritalin LA, Equasym XL, Concerta XL, and Medikinet retard, all of which provide a combination of immediate- release and delayed- release methylphe- nidate. The authors report that, in summary, Equasym XL was more effective than Concerta XL early in the day, while Concerta XL was more effective than Equasym XL later in the day. The authors emphasize that the drugs should be used to meet patients’ needs according to their preferences and that no stimulant drug is superior to another overall. Mean ESs were very similar, between 0.8 and 1.0. The ES for atomoxetine and modafinil (both discussed later in this chapter) was lower, at 0.6. However, the atomoxetine study results may have been adversely affected by the fact that the “clinician rating” used in most studies was parent inter- views, and in one study (M. Weiss et al., 2005) teacher reports, whereas the stimulant studies used direct clini- cian observations.
Further studies are required, particularly using iden- tical methodology, so that more direct comparisons can be made. Better designed quality- of-life studies are also needed but, based on the findings outlined earlier, Concerta may be beneficial when children need to be medicated in the evenings and compliance is poor.
Table 6.2 summarizes information regarding the profiles of the slow- release stimulants in comparison with immediate- release methylphenidate.
Long‑ Acting Dexamfetamine. Lisdexamfetamine, an amphetamine prodrug, is a compound of dexamfet- amine and the amino acid lysine that is inactive until digested and absorbed into the bloodstream. It has been licensed for the treatment of ADHD in 6- to 12-year- olds in the United States since 2007, in Canada in the same age group since 2009, and in the United Kingdom in children 6 years and older since early 2013. Lisdex- amfetamine is converted to dexamfetamine by hydro- lytic enzymes located on erythrocytes. This leads to an onset of action 1–3 hours after taking the drug. The du- ration of action has been reported to be up to 13 hours. Lisdexamfetamine was developed due to its lower po- tential for abuse compared with dexamfetamine (see Blick & Keating, 2007, for a helpful review). It is given once a day at doses of 30–70 mg.
Elbe, MacBride, and Reddy (2010) have reported a literature review of five RCTs demonstrating the effi- cacy of lisdexamfetamine. Two of the larger trials were reported by Biederman and colleagues (2007a; Bieder- man, Krishnan, Zhang, McGough, & Findling, 2007b), who describe a study of the use of lisdexamfetamine
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
208 W h at Wo r k s f o r W h o m ?
dimesylate in 290 school- age children. The efficacy and side effects in this study were comparable to those in a trial of 584 children (Biederman, Lopez, Boellner, & Chandler, 2002) using extended- release mixed am- phetamine salts (Adderall XR; see below), and also to those of other controlled- release stimulants. The drug performed well over 8 hours at doses of 30–70 mg. A further RCT by Findling et al. (2011) investigated the treatment of 265 adolescents ages 13–17 years and, similarly, demonstrated significant improvements with doses of between 30 and 70 mg per day, without sig- nificant adverse effects. Unfortunately, the study lasted only 4 weeks.
As with all stimulants, monitoring of the patients’ height and weight is essential. This has recently been emphasized in a study by Faraone, Spencer, Kollins, and Glatt (2010), in which 281 children ages 6–13 years receiving lisdexamfetamine dimesylate treatment were followed up over a 15-month period. There was no control group. There was a significant reduction in the children’s weight, body mass index (BMI), and height relative to age- appropriate standards, although the au- thors commented that overall, the children entering the study were taller and heavier than the average for the population.
Adderall
Adderall, a long- acting amphetamine compound, is a 75:25 ratio of dextro- and levoamphetamine. It has been shown to have a longer half-life than methylphenidate (J. Swanson et al., 1998), which may be advantageous in reducing the frequency with which the drug needs to
be taken during the day. It is available as immediate- release and extended- release preparations.
Ahmann et al. (2001) reported an RCT of the use of Adderall in 154 children and adolescents ages 5–18 years. The subjects all fulfilled criteria for DSM-IV di- agnosis of ADHD (hyperactive– impulsive, inattentive, or combined) using standardized measures such as the Conners Parent and Teacher Rating Scales (Conners, 1989), the ADD-H Comprehensive Teachers’ Rating Scale (Ullman, Sleator, & Spraugue, 1988), and the Child Behavior Checklist (Achenbach, 1991). In this study, Adderall was found to be very effective, with side effects similar to those of other stimulants. Bieder- man et al. (2002) have similarly reported the benefits of Adderall XR (the extended- release formulation) in an RCT of 509 6- to 12-year-olds with DSM-IV diag- nosed ADHD. This study used standardized assess- ment scales such as the Conners Rating Scales.
Faraone and Biederman (2002) reported a meta- analysis of six studies of Adderall and confirmed its efficacy, compared with placebo, in reducing ADHD symptoms and aggression. In a second report, Faraone, Biederman, and Roe (2002) reported a meta- analysis comparing Adderall with methylphenidate. They found that, on the basis of results from four small studies, cov- ering a total of 186 subjects, Adderall is significantly superior to methylphenidate in terms of reduction of ADHD symptoms and aggression.
Methylphenidate Patches
Several studies have shown the benefit of methylpheni- date patches. The literature on methylphenidate patches
tABle 6.2. Profiles of slow‑release stimulants in Comparison to immediate‑release methylphenidate
Methylphenidate IR BID Concerta XL Equasym XL Medikinet retard Adderall XR
Formulation 100% IR 22% IR, 78% ER 30% IR, 70% ER 50% IR, 50% ER 50% IR, 50% ER
Duration of action 8 hours 12 hours 8 hours 7 hours 8–10 hours
Dosage 1.6 mg/kg/day; maximum 60 mg/ day
2 mg/kg/day; maximum 72 mg/ day
2 mg/kg/day; maximum 72 mg/ day
2 mg/kg/day; maximum 72 mg/ day
10–30 mg/day; maximum 30 mg/ day
Time course of action/serum concentration
Peaks at around 2 hours and 6 hours; trough at 5 hours
Peaks at 2 and 10 hours
Larger effect in first 4–5 hours and peak at 5 hours
Peak at 2 hours; larger effect in first 4–5 hours
Peak at 7 hours
Note. IR, immediate release; ER, extended release; BID, twice daily.
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 209
has been well reviewed by V. R. Anderson and Scott (2006), who give helpful advice on dosage, which is dependent on patch size. The patches are effective for 9 hours, with the smallest patch (12.5 cm) releasing a 10-mg dose and the largest (37.5 cm) releasing a 30-mg dose. They report that there have been at least two randomized controlled multicenter studies confirming the efficacy of the patches, which have been approved by the U.S. Food and Drug Administration for use in children ages 6–12 years. One of the studies to which they referred was a poster presentation. The other, a published study by McGough et al. (2006b), was fairly well designed, with 80 participants between ages 6 and 12 years showing significant improvement with patch treatment compared to placebo. Unfortunately, side effects were not assessed, although “pinkness” at the patch site was common.
Faraone and Giefer (2007) reported that although there was no evidence of serious side effects among 127 children ages 6–12 after 36 months of methylphenidate patch treatment, there was significant slowing of gains in height and weight, particularly in children who were taller or heavier than average for the population at the start of the study; the same was not true for smaller children with lower centiles for weight, height, and BMI at baseline. The reduction in growth was greatest in the first 12 months of the study; after that, there was attenuation of the losses, particularly for weight and BMI and to a lesser degree for height.
Treatment Protocols, Dosage, and Side Effects
There appears to be reasonable consensus among ex- perts on both sides of the Atlantic (Greenhill, 1992; NICE, 2008; Subcommittee on Attention- Deficit/ Hyperactivity Disorder et al., 2011; J. M. Swanson et al., 1993; E. Taylor, 1994) about the treatment proce- dure for stimulants. Treatment should commence with a suboptimal dose (5 mg methylphenidate or 2.5 mg dexamfetamine in school- age children) in order to re- duce the risk of side effects. There is great individual variation in responsiveness: Greenhill et al. (2001b) described the findings of the MTA study in relation to optimal dosages for children ages 7–10 years with DSM-IV-diagnosed ADHD (combined type) and found that 22% of subjects responded to 15 mg/day or less, 25% responded to 16–34 mg/day, and 30% responded to 35 mg/day or more. The dose should be increased gradually while monitoring for effectiveness and any adverse reactions. The recommended average daily
dose of methylphenidate is 10–40 mg per day, divided between two and six doses; the maximum dose is 60 mg. An adequate trial period for immediate- release or slow- release preparations is between 4 and 8 weeks.
Overall, 20–25% of patients who respond poorly to one treatment will respond well to another (Dulcan, 1990). Side effects are common and dose- related but rarely serious. Probably the most severe side effect is precipitation of a psychosis (A. S. Bloom, Russell, Weisskopf, & Blackerby, 1988), but this is rare. Pos- sible adverse reactions include appetite suppression, insomnia, irritability, mood changes, nausea and vom- iting, growth suppression, and an increase in heart rate and/or blood pressure. Interestingly, Barkley, McMur- ray, Edelbrock, and Robbins (1990b) reported a high (> 50%) rate of side effects with placebo; some of these, such as anxiety and sadness, decreased with stimulant treatment. Height and weight should be monitored reg- ularly. Studies have demonstrated a significant negative impact of methylphenidate on the weight and height of treated children relative to untreated peers (Charach, Figueroa, Chen, Ickowicz, & Schachar, 2006) and to children who had annual drug holidays (R. G. Klein, Landa, Mattes, & Klein, 1988). The seriousness of the effects on weight and height depend on the individual’s baseline weight and height. Pulse and blood pressure should be monitored after an increase in dose. Some children may have a behavioral rebound 5 hours after the last dose. In the case of a severe rebound, adding clonidine may be necessary. Alternatively, a small dose of stimulant given late in the afternoon may reduce the rebound effect.
Effect on the Primary Symptoms
Over 100 medication trials have confirmed the benefi- cial effects of stimulants for the treatment of inattention, hyperactivity, and impulsivity. In the United Kingdom, the first NICE (2000) review of the evidence for meth- ylphenidate treatment concluded that methylphenidate is recommended for children with a diagnosis of severe ADHD or hyperkinetic disorder. In the most recent NICE (2008) guidelines, stimulants are still the recom- mended first-line treatment for severe ADHD. There is no doubt that stimulant medication relieves the primary symptoms, at least in the short term. There is less cer- tainty, however, regarding the extent of the benefit. In a helpful review of the literature, J. M. Swanson et al. (1993) reported that there is fairly good consensus that the ES of stimulant treatment on the primary symptoms
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
210 W h at Wo r k s f o r W h o m ?
of ADHD or hyperkinetic disorder is 0.8, and that 70% of treated children respond. In contrast, the ES on aca- demic performance is approximately 0.4. The placebo response rate appears to be approximately 30%.
Some studies have looked at laboratory measures of the symptoms of ADHD, such as impulsivity. Two types of impulsivity have been described in the litera- ture. “Behavioral impulsivity” refers to a disregard for the consequences of an action and a lack of tolerance for any delay in gratification. “Cognitive impulsivity” refers to a tendency to give answers without adequate consideration to their accuracy. The Matching Familiar Figures Test (Kagan, Rosman, Day, Albert, & Phillips, 1964) was developed to assess cognitive impulsivity. Unfortunately, it is unclear whether inaccuracy alone, or a fast and inaccurate response, is characteristic of children with ADHD. Other tests have been developed, such as the Continuous Performance Test (Rosvold, Mirsky, Sarason, Bransome, & Beck, 1956) and the Test of Everyday Attention for Children (T. Manly, Robertson, Anderson, & Nimmo-Smith, 1998), but these are thought not to be specific to impulsivity. Using the Matching Familiar Figures Test, several au- thors have reported a decrease in errors when the child is treated with a stimulant (Losier, McGrath, & Klein, 1996; Tannock, Schachar, & Logan, 1995). Similar ap- proaches have used other test paradigms, such as M. A. Malone and Swanson’s study (1993) using a word- matching test, and a study by Bedard, Ickowicz, and Tannock (2002) using the Stroop Color and Word Test (C. J. Golden, 1978). As a result of these studies, re- searchers concluded that control children made fewer impulsive errors than children with ADHD who are receiving placebo. Children with ADHD who were on medication performed nearly as well as the nonaffected control group, and significantly better than untreated children with ADHD.
More recently, studies have examined the response to medication by monitoring brain performance using im- aging techniques. Akay et al. (2006) reported a single- photon emission computed tomography (SPECT) study of children with DSM-IV-diagnosed ADHD. The SPECT technique detects variations in cerebral blood flow. Akay et al. (2006) summarized the findings to date on reduced blood flow in the prefrontal and tem- poral cortices and the cerebral hemispheres. In their small study of nine patients, they found that there was a visually increased perfusion of the frontal and right lateral temporal cortices during 2 months of stimulant treatment and up to 2 months after treatment discon-
tinuation. However, this apparent increase in perfusion was not confirmed by semiquantitative analyses, and the authors hypothesized that variability in D2 dopa- mine receptor availability and the small sample size affected the analysis.
The influence of genetics on response to treatment with stimulants has been explored by Cheon, Kim, and Cho (2007). In this study, 83 Korean children with a DSM-IV diagnosis of ADHD were treated with meth- ylphenidate for 8 weeks. Between 71.1 and 80.0% of those with a good response to treatment had the 4/4 genotype at DRD4 (the gene that encodes dopamine receptor D4); in contrast, between 31.6 and 37.7% of poor responders had the 4-repeat allele. Children who were homozygous for the 4-repeat allele had most severe ADHD symptoms at the start of the trial and showed the greatest response to treatment. D. L. Gilbert et al. (2006) looked at short- interval cortical inhibition (SICI) in the motor cortex, measured by transcranial magnetic stimulation; SICI is reduced in children with ADHD compared with controls. These authors inves- tigated whether a genetic variation in the dopamine transporter gene (DAT1), a site of action of methylphe- nidate, would influence the effect of methylphenidate and atomoxetine on SICI. Both medications increased SICI in individuals heterozygous for the gene variant but not in homozygotes.
Influence of Age on Response to Stimulant Treatment
Adolescents. Studies relevant to this age group have been reviewed by P. Hazell (2007), who reported that a meta- analysis of eight RCTs between 1988 and 1991, which included 199 participants ages 12–18 years with DSM-IV-diagnosed ADHD, found an ES of 0.94. He points out that since 1996, several new slow- release preparations have been shown to be equally effective in adolescents and also notes that use of stimulant medi- cation does not predispose adolescents to substance misuse. In a meta- analysis of six follow- up studies (two with follow- up in adolescence and four in young adult- hood), Wilens, Faraone, Biederman, and Gunawardene (2003) reported that stimulant treatment of ADHD led to a reduced rate of substance misuse in adolescence. In another meta- analysis of seven studies (five prospec- tive) with 766 subjects, 98% of whom had been treated with stimulants, Faraone and Wilens (2003) suggested that not only was there no increased risk of substance misuse, but the treatment also reduced the risk of sub-
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 211
stance misuse. However, S. M. Gordon, Tulak, and Troncale (2004) found that approximately 20% of pa- tients will misuse or sell their medication; these authors recommend switching such patients to a nonstimulant treatment. In a small controlled study of the subjective response to methylphenidate in adolescents, C. A. Mar- tin, Guenthner, Bingcang, Rayens, and Kelly (2007) concluded that abuse potential is a risk because a sig- nificant number of adolescents in this study reported subjectively feeling better when taking stimulants.
D. J. Cox et al. (2006) reported a study of 35 adoles- cents in which they compared slow- release methylphe- nidate, dexamfetamine, and placebo. They found that extended- release methylphenidate, at a dose of 72 mg, led to significant improvements in driving skills and safety.
Preschool Children. In a literature review of all stud- ies of treatments for this age group, McGoey, Eckert, and DuPaul (2002) concluded that stimulant treatment, parent training, and classroom behavior management are all effective. However, at that time, there were only 14 nonpharmacological studies in the literature. Con- nor (2002) reported a review of stimulant medication studies in preschoolers and suggested that there is rea- sonable evidence to support using stimulant medication in this age group if behavioral approaches, including parent training, have failed. Greenhill et al. (2006b) reported the results of the NIMH multisite study of preschoolers with a DSM-IV consensus diagnosis of ADHD. This study of children ages 3.0–5.5 years with ADHD examined the response to three daily doses of immediate- release methylphenidate. A total of 303 preschoolers were enrolled. There was a significant improvement with stimulant medication compared to placebo. The daily maintenance dose of methylpheni- date ranged from 7.5 to 30 mg. However, the ESs were smaller than those reported for older children. Parents reported an ES of 0.35 for preschoolers compared with 0.52 in school- age children, and teachers reported an ES of 0.43 in preschoolers compared with 0.75 in school- age children. The authors reported higher rates of emotional lability in the preschoolers.
An additional aspect of this study, reported by Mc- Gough et al. (2006a), was an attempt to identify a genetic contribution to the response. Possible associa- tions were found for the DRD4 promoter and SNAP- 25. The SNAP-25 variant was associated with tics and buccal– lingual movements, in addition to irritability. The DRD4 variant was associated with picking. Fur-
ther studies are clearly required to confirm these ge- netic influences on the response to stimulant treatment in preschoolers.
Ghuman et al. (2007) showed that, when comorbid disorders are present, the response to stimulants in pre- schoolers is reduced, as is the case in other age groups. In this study, the children in a high- comorbidity sub- group had an ES of –0.37 with a 7.5-mg dose of meth- ylphenidate three times a day (relative to placebo), compared with high ESs of 0.89 in children with no comorbidity, 1.00 in those with low comorbidity, and 0.56 in children with moderate levels of comorbidity.
Abikoff et al. (2007) explored the outcomes for the same sample in more detail. Unfortunately, there was a dropout rate of 45% in the placebo group and 15% in the methylphenidate group, due to deterioration in be- havior. This may have adversely affected the power of the study; hence, the treatment effect reported may be an underestimate. The authors suggest that the 4-week treatment phase may have been too short and the dose of medication too low. They also recommend adding a parenting program in future studies.
In summary, regardless of age, 75% of children with ADHD show normalization of inattention, hyperactiv- ity, and impulsivity when treated with stimulants. It is not possible to predict reliably which children will show a good response. In addition, 70% of children with ADHD and comorbid aggression show significant improvement in aggressive behaviors. However, proso- cial behaviors do not improve.
Stimulant Use in the Presence of Comorbid Disorders
Ter- Stepanian, Grizenko, Zappitelli, and Joober (2010) found that most children with ADHD have comorbidi- ties. In this study of 267 children ages 6–12 years, in which boys were overrepresented (77.9% of the sam- ple), 40.8% had ODD, 27.7% had conduct disorder, 47.2% had an anxiety disorder, and 7.9% had a depres- sive disorder.
Internalizing Disorders. T. Spencer et al. (1996) re- viewed the findings from 11 studies that examined the response of children with ADHD and comorbid inter- nalizing symptoms to stimulant medication. Most of these studies reported that the presence of a comorbid internalizing disorder, such as anxiety or depression, reduced the response to stimulant medication. In a well- designed trial with a sample of 40 children with
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
212 W h at Wo r k s f o r W h o m ?
ADHD and varying degrees of internalizing symp- toms, DuPaul, Barkley, and McMurray (1994) found that significantly fewer children in a high- internalizing group exhibited a positive response to methylphenidate compared to subjects with fewer internalizing symp- toms. Fifty percent of the children with significant co- morbid internalizing symptoms either failed to respond or had an adverse reaction to methylphenidate; 25% of the children with a comorbid internalizing disorder showed deterioration in behavior (based on teacher reports of classroom behavior) with stimulant use, compared with 9% in the groups with fewer internal- izing disorder symptoms. The study by Ter- Stepanian et al. (2010) confirmed the poor response of children to methylphenidate when comorbid anxiety was pres- ent. In contrast, there was a positive response in the presence of comorbid conduct disorder, although this latter result must be interpreted with caution because the duration of the study was only 2 weeks, comprising 1 week of placebo followed by crossover to 1 week of methylphenidate.
Analysis of the data from the MTA study by J. S. March et al. (2000) revealed no adverse response to stimulant medication in children with comorbid inter- nalizing disorders, but it did highlight the need for ad- ditional psychosocial interventions for this group.
More recently, Goez, Back- Bennet, and Zelnik (2007) found that some children with DSM-IV-diag- nosed ADHD and anxiety improved with stimulant medication, and a larger proportion (58.62%) worsened (a bimodal distribution). They found similar results in children with comorbid ODD. They suggested that in these children the inattention is secondary to the other problems; hence, this represents a separate group re- quiring different interventions.
In summary, while earlier studies suggested that only 50% of children with ADHD and comorbid depression and/or anxiety significantly benefited from stimulant treatment, and that there was an increased risk of ad- verse reactions and deterioration in mental state, more recent findings suggest that, at least for comorbid anxi- ety, this may not be the case. The findings from the MTA study also suggest that when ADHD is comor- bid with internalizing disorders, additional behavioral interventions may be appropriate and more beneficial than when ADHD is present on its own.
Manic Symptoms. There is considerable debate about how to diagnose children with ADHD who present with manic symptoms such as irritability and lability
of mood but do not show all the symptoms of a bipolar disorder. Galanter et al. (2003) reported on the sample used for the MTA study (579 children ages 7.0–9.9 years). Children in the sample with symptoms of mania showed a good response to stimulant treatment. Galant- er et al. and Sarampote, Efron, Pearl, Robb, and Stein (2002) have suggested that the high reported rates of manic symptoms in individuals with ADHD may be due to stimulant rebound. There was no evidence to support this suggestion in the study by Galanter et al. (2003), and they concluded that, on the basis of their short-term study (4 weeks), there was no contraindica- tion to a trial of methylphenidate in the presence of pos- sible bipolar symptoms. They recommend a carefully monitored trial of stimulant medication.
Conduct Disorder and Aggression. T. Spencer et al. (1996) reported on 17 controlled studies of stimulant treatment in ADHD. They concluded from their review of these studies that stimulants produce significant improvement in ADHD symptoms in children with comorbid aggression. They also found that stimulants reduce both verbal and physical aggression in these individuals. A meta- analysis of 28 studies by Connor, Glatt, Lopez, Jackson, and Melloni (2002) similarly confirmed the benefits of stimulants in reducing ag- gression in children or adolescents below age 18 years.
In a well- designed RCT of 85 children and ado- lescents ages 6–16 years, Sinzig et al. (2007) showed that long- acting methylphenidate was effective in the treatment of ADHD and comorbid DSM-IV-diagnosed oppositionality and aggression. Symptomatic improve- ment was more favorable in school than in the home.
In summary, 70% of children with ADHD and co- morbid aggression show a significant reduction of ag- gressive behaviors when treated with stimulants. How- ever, prosocial behaviors do not improve. For further discussion of this topic, see Chapter 4.
There has been one pilot study on supplementation of stimulant medication with risperidone for the treat- ment of aggression in children (Armenteros, Lewis, & Davalos, 2007). However, this study had several limita- tions, particularly its short duration (4 weeks), which may have led to an underestimation of side effects, and the small sample size (25 children ages 7–12 years). A well- designed controlled study is needed before defi- nite recommendations can be made.
Learning Disabilities. ADHD and hyperkinesis are ap- proximately four times as prevalent in children with
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 213
learning disabilities as in non- learning- disabled chil- dren (Biederman et al., 1991). There is no consistent view in the literature regarding the efficacy of medi- cation in children with ADHD and comorbid learning disabilities, because most studies have excluded such children from their sample. The few that have includ- ed learning- disabled subjects have tended to contain only those with mild deficits. Three studies (Aman, Kern, McGhee, & Arnold, 1993; Handen, McAuliffe, Janosky, Feldman, & Breaux, 1995; Payton, Burkhart, Hersen, & Helsel, 1989) have reported a response rate averaging 65%, which, although useful, is lower than the rate of 70–80% reported in the literature for school- age children without learning disabilities. Aman, Bui- can, and Arnold (2003) pointed out that the response of children with DSM-III-diagnosed ADD and addi- tional learning disabilities is more variable than that of children with normal IQ. Handen, Feldman, Lurier, and Murray (1999) also reported a small (11 subjects) RCT that confirmed preschool children with learning disabilities respond to methylphenidate at a rate simi- lar to that of school- age children, although there was a suggestion of a higher rate of side effects in the younger age group.
An important subgroup is children with fragile X syndrome, most of whom have ADHD or hyperkinesis. Hagerman, Murphy, and Wittenberger (1988) reported a double- blind crossover study of a small sample (N = 15) of children with ADHD and fragile X syndrome who were treated with methylphenidate, dexamfet- amine, or placebo. The children showed significant behavioral improvement in response to stimulant medi- cation.
Stimulants are therefore often beneficial in the pres- ence of a comorbid learning disability; however, the more severe the learning disability, the poorer the re- sponse.
Specific Learning Disabilities. Many children with ADHD or hyperkinesis have a comorbid specific learning disability, such as a specific reading disabil- ity. Bental and Tirosh (2008) reported on a study of 25 boys ages 7.9–11.7 years with DSM-IV-diagnosed ADHD and reading disorder, who participated in a double- blind study of methylphenidate and placebo. The authors found a significant improvement in word accuracy, nonword accuracy, and rapid naming with the active drug. However, in another study, Grizenko, Bhat, Schwartz, Ter- Stepanian, and Joober (2006) found that children with a mathematics disability had a poorer
response (37%) to methylphenidate than those with a reading disability (67%); those with no disability had a 75% response to methylphenidate. The authors suggest that the explanation is that children with mathematics disabilities have higher executive dyscontrol than do children with pure forms of ADHD.
In summary, in the presence of ADHD and a spe- cific learning disability, stimulant therapy facilitates improvement in the learning disability.
Pervasive Developmental Disorders. Two studies (Bir- maher, Quintana, & Greenhill, 1988; Handen, Johnson, & Lubetsky, 2000) have suggested that moderate to severe social withdrawal can be an outcome of higher doses of stimulants (0.6 mg/kg) and recommend the lower dose of 0.3 mg/kg. However, a more recent report (Posey et al., 2006) of an open-label study of atomox- etine use in sixteen 6- to 14-year-olds with DSM-IV- diagnosed ADHD and autism, Asperger syndrome, or pervasive developmental disorder (PDD) not otherwise specified found that 14 of the 16 subjects benefited. A second study by Posey et al. (2007), describing a 4-week blinded crossover study of methylphenidate in 66 children with a mean age of 7.5 years, showed sig- nificant benefit without significant side effects. These more recent findings support the use of these treatments when ADHD is comorbid with autistic spectrum disor- ders. In Posey et al.’s second study, six of the original sample of 72 subjects were excluded because they could not tolerate methylphenidate. Therefore, it would be ad- visable initially to use a low test dose in this group of children and adolescents. Further studies are required in this area.
Seizure Disorders. Until recently, the use of stimulants was avoided in children with ADHD and comorbid seizures, because stimulants were thought to lower the seizure threshold. Some clinicians would prescribe dexamfetamine but not methylphenidate, because it was thought that dexamfetamine raised the seizure threshold. An extremely helpful review of the treatment of ADHD and comorbid epilepsy (Torres, Whitney, & Gonzalez- Heydrich, 2008) highlights the fact that the majority of studies to date have evaluated the risks and benefits of stimulant medication when the two condi- tions coexist. The authors emphasize the need for good communication between the clinicians treating the epi- lepsy and the ADHD, as well as the child’s caregivers and school. The studies report a 70% response rate to stimulants. They recommend the use of stimulants first,
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
214 W h at Wo r k s f o r W h o m ?
because these treatments have the most evidence for ef- ficacy, with atomoxetine next, and clonidine or guanfa- cine as third-line treatments.
Stimulant Use with Comorbid Tics and Tourette Syn‑ drome. The use of stimulant treatment in children with ADHD and comorbid tics or Tourette syndrome is dis- cussed in Chapter 7.
Adaptation to Psychosocial Environments
Social Relationships. Children with ADHD and hy- perkinesis usually have serious social difficulties, but there have been few studies of the effects of stimulants on social relationships. The studies that exist suggest a benefit in this area, showing that children who re- spond positively to stimulant medication receive in- creased warmth and decreased maternal criticism, with increased frequency of maternal contact (Schachar, Taylor, Wieselberg, Thorley, & Rutter, 1987; Whalen & Henker, 1991). In support of the suggestion that medi- cation alone improves social relationships, several au- thors have reported a decreased incidence of arguments and fighting in stimulant- treated boys with ADHD (Barkley, McMurray, Edelbrock, & Robbins, 1989). Children also become more popular with their peers, but often some social difficulties remain (Gadow, Nolan, Sverd, Sprafkin, & Paolicelli, 1990; Whalen et al., 1989). Approximately 70% of children and adoles- cents treated with medication show a good response. Nevertheless, J. M. Swanson et al. (1993) concluded that although stimulants may reduce negative behav- iors, they do not increase positive prosocial behaviors; other therapies are required if social relationships are to be fully normalized.
Academic Performance. Discussion of the benefits, if any, of stimulant medication in relation to learning in children with ADHD has been ongoing for many years. In a meta- analysis of 135 studies, Kavale (1982) found moderate positive effects favoring drug treat- ment. The results revealed a 15% increase in achieve- ment for those treated with methylphenidate. Rapport, Denny, DuPaul, and Gardner (1994) evaluated the ef- fects of methylphenidate on the classroom behavior and academic performance of 76 subjects ages 6–11 years. Standardized statistical assessments showed that medication significantly improved attention, academic efficiency, and teacher ratings of classroom behavior. Doses of 10 mg or above significantly increased per-
formance in all three areas compared to placebo or a 5-mg dose.
Overall, 76% of the sample showed either improved or normalized attention (72% normalized), with 24% showing no change; 53% showed improved or normal- ized academic efficiency (50% normalized), and 47% showed no change in academic efficiency. Academic accuracy stabilized at the dose of 10 mg and did not im- prove further at higher doses. In contrast, the numbers of children with normalized attention and classroom behavior increased as the dose of methylphenidate was increased toward 20 mg. Hence, for some children, academic performance may not improve in line with changes in behavior. Moreover, the children who failed to show improvements in academic accuracy were less likely to show improvements in attention.
Some authors (Alto & Frankenberger, 1995; C. L. Carlson, Pelham, Swanson, & Wagner, 1991) have suggested that mathematical skills are particularly improved by methylphenidate. However, Zentall and Ferkis (1993) failed to find evidence to support this finding. Another question is the effect of the dose on learning; J. M. Swanson, Cantwell, Lerner, McBurnett, and Hanna (1991) have suggested that high doses of stimulant may produce such an “overfocusing” effect that it impedes learning. These authors suggest that higher dosages may maximize achievement in some tasks and lower dosages may optimize other tasks, such as those demanding greater cognitive effort.
Longer-term follow- up studies have often failed to confirm the beneficial effects of stimulants on aca- demic performance. Weber, Frankenberger, and Heil- man (1992) studied 22 subjects with ADHD and used a group achievement test before and after 1–2 years of methylphenidate or control treatment to assess the re- sponse to treatment. They concluded that methylpheni- date did not improve achievement.
Alto and Frankenberger (1995) reported a study of 17 children ages 7–8 years with ADHD. Each child was matched with a control child by age, gender, and cognitive abilities. The authors found that even when subjects were matched on verbal cognitive scores, they showed significantly lower achievement than controls in the areas of word analysis and reading. The perfor- mance of subjects with ADHD in mathematics showed an increase in rate of learning approaching significance when compared with the control group. Listening skills were more likely to respond to medication than vocabu- lary. Increasing the dosage further increased listening but had the opposite effect on vocabulary. The authors
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 215
concluded that although the children with ADHD were behind their normal peers in most areas of learning at the start of the study, once placed on methylphenidate their learning rate was equal to that of the controls, so that they did not fall further behind, although they did remain behind in reading skills. The study lasted 1 year; it was not possible to assign the children to a pla- cebo treatment, because the parents and teachers were reluctant to consent.
In practice, medication alone is unlikely to be ad- equate for optimal academic achievement. Studies such as that by Pelham, Milich, and Walker (1986) reveal that when behavioral approaches incorporating a re- ward system were combined with methylphenidate, performance was maximized.
In summary, although attention and output during academic tasks improve by 70% with stimulant medi- cation, efficiency and accuracy show only approximate- ly 50% improvement. In the long term, subjects with ADHD receiving stimulant medication do not achieve as much as nonaffected control children, probably be- cause they are too far behind to catch up fully. Further studies are required to verify this as the cause of poorer long-term academic progress.
Summary
Stimulant treatment achieves normalization of inatten- tion, hyperactivity, and impulsivity in 75% of children with ADHD. It is not possible to predict with certainty which children will show a good response. Attention and output during academic tasks improve by 70% with stimulant medication, but efficiency and accuracy show less improvement. Prosocial behaviors do not improve.
The outcome of ADHD is still uncertain, but it ap- pears to be better with treatment than without it. Indi- viduals with ADHD who are treated with stimulants still do not achieve as much as nonaffected individuals in the long term, probably because they have already fallen too far behind their unaffected peers to catch up. Nevertheless, young adults treated for ADHD with stimulants in childhood and adolescence have fewer ac- cidents, improved social skills, and happier childhood memories compared with individuals whose ADHD was not treated.
When comorbid problems are present, stimulant medication is beneficial to varying levels depending on the comorbidity. In the case of conduct disorder, 70% of children with ADHD show significant improvement in both aggressive behaviors and ADHD behaviors
when treated with stimulants. In contrast, the consen- sus until recently was that when ADHD is comorbid with depression and/or anxiety, only 50% of children significantly benefit from stimulant treatment. In addi- tion, there was a presumed increased risk of adverse re- actions and deterioration in mental state. The findings from three more recent studies have concluded that, at least in the case of anxiety, this is not so.
Stimulants are also often beneficial in the presence of a comorbid generalized learning disability; however, the more severe the learning disability, the poorer the response. They also lead to improvements in the learn- ing problems of some specific learning disabilities. Children with PDD and ADHD have recently been found to benefit from stimulant therapy without show- ing an increase in stereotypies. Stimulants are safe to use in the presence of epilepsy as long as the underlying seizure disorder is appropriately treated.
Atomoxetine
Atomoxetine is a highly selective norepinephrine reuptake inhibitor. It has been shown to be effective in several studies. Kratochvil et al. (2006) reported a meta- analysis of long-term atomoxetine treatment in 272 children ages 6 and 7 years who met diagnostic criteria for a DSM-IV diagnosis of ADHD. Atomox- etine was found to be beneficial over 2 years of treat- ment with no significant risk of major side effects. In another meta- analysis of atomoxetine, Cheng, Chen, Ko, and Ng (2007) found that the NNT for treatment response was 3.4, and the NNT for relapse preven- tion was 10.3. Patients with more pronounced baseline ADHD symptoms had the greatest reduction in symp- toms. The groups showing the least response were male children with hyperactive– impulsive subtype ADHD and comorbid ODD. The most common side effects were appetite suppression (NNH = 8.81), abdominal pain (NNH = 22.48), vomiting (NNH = 29.96), and dyspepsia (NNH = 49.38). Side effects were worse in younger children and those with more severe baseline hyperactive– impulsive symptoms. Oppositional symp- toms reduced and quality of life improved.
Newcorn et al. (2006) reported an RCT of long-term treatment of ADHD with atomoxetine and found that a low dose of 0.5 mg/kg body weight was as effective as a higher dose of 1.2–1.8 mg/kg. This study involved 459 children, ages 6–16 years, with a DSM-IV diagnosis of ADHD. Treatment was initiated at the higher dose; 229 children responded positively and were subsequently
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
216 W h at Wo r k s f o r W h o m ?
randomized either to continue on the higher dose or move to the lower dose. The only significant difference in adverse reactions was a higher rate of affective labil- ity in the lower dose group.
G. A. Carlson et al. (2007) reported a small study of 25 children ages 6–12 years who met diagnostic criteria for a DSM-IV diagnosis of ADHD. The study investigated a combination of atomoxetine and meth- ylphenidate and showed that this was safe and effec- tive in those who did not respond to atomoxetine alone. However, it is difficult to draw conclusions given the small sample.
Given that there have been reports of suicidal ideation and depression as side effects of atomoxetine treatment, Bangs et al. (2008b) undertook a meta- analysis com- paring suicidal ideation in atomoxetine- treated patients and placebo- treated controls. The authors compared 12 placebo- controlled studies of atomoxetine, with a treat- ment duration of 6–18 weeks, and five trials of meth- ylphenidate, with a duration of 6–9 weeks. The mean daily dose of atomoxetine in the trials was 1.3–1.6 mg/kg. The mean daily dose of methylphenidate was 0.9–1.1 mg/kg. The age range of the children and ado- lescents in these studies was 6–17 years. Although sui- cidal ideation was uncommon in the studies, it was sig- nificantly more common in atomoxetine- treated than in methylphenidate- treated groups. However, there was no difference between atomoxetine and methylpheni- date in terms of the risk of inducing suicide.
Comorbid ODD
Dittman et al. (2011) reported an RCT of atomox- etine in 181 German children ages 6–17 years with ADHD and comorbid ODD. Both ADHD and ODD improved during 9 weeks of atomoxetine treatment. The authors reported that both ODD and ADHD be- haviors significantly improved. However, a high rate of side effects— sleep disorders, fatigue, nausea, and gastrointestinal complaints— was reported in the first 3 weeks of treatment, occurring in 60% of cases re- ceiving fast- titration atomoxetine treatment and 44% of cases receiving slow- titration treatment. Partici- pants in the fast- titration schedule received 0.5 mg/ kg atomoxetine for 7 days and a subsequent increase to the target dose of 1.2 mg/kg. Participants in the slow- titration schedule received 0.5 mg/kg for 7 days, followed by 0.8 mg/kg for 7 days and thence 1.2 mg/ kg. The findings suggest that a slow- titration schedule is preferable— possibly with an even slower increase
than the “slow titration rate” described in this study. The authors suggest that the effect of atomoxetine on ODD may be direct rather than secondary to the re- duction in ADHD symptoms.
Bangs et al. (2008a) reported a study of 226 children ages 6–12 years who met diagnostic criteria for a DSM- IV diagnosis of ADHD. The children were randomized to atomoxetine or placebo. There were significant im- provements in both oppositionality and ADHD symp- toms at 2 and 5 weeks postbaseline, but not at 8 weeks. Biederman et al. (2007c) reported similar findings from a meta- analysis of studies of comorbid ADHD and ODD.
Comorbid Depression
In an RCT, Bangs et al. (2007) studied the effects of atomoxetine in 142 adolescents ages 12–18 years with DSM-IV-diagnosed ADHD and moderate depression. They found atomoxetine to be as effective in reducing ADHD symptoms over the 9 weeks of the trial as it had been shown in trials in adolescents with ADHD and no comorbid depression. The atomoxetine had no impact on the depression. There was no precipitation of mania, worsening of depression, or significant increase in sui- cidal ideation, although the latter did increase a little.
Comorbid Anxiety
D. Geller et al. (2007) reported a well- designed RCT in 176 children and adolescents ages 8–17 years with ADHD and comorbid anxiety. The children met DSM- IV criteria for a diagnosis of ADHD and generalized anxiety disorder, separation anxiety disorder, and/or social phobia. The authors reported a significant reduc- tion in symptoms of both ADHD and anxiety.
Side Effects
D. Michelson et al. (2007) reported on a large, inter- esting, well- conducted study exploring the influence of genetics on the response to atomoxetine. The study investigated the effect of variation in the gene for cyto- chrome P450 2D6 (CYP2D6), the enzyme pathway that breaks down atomoxetine, in a sample of children and adolescents ages 6–18 years (589 receiving atomox- etine and 294 receiving placebo). The authors reported that poor metabolizers had highly significant superior responses to medication compared with better metabo- lizers. However, unsurprisingly, the poor metabolizers
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 217
had more severe side effects, with significantly (p < .001) increased heart rate and diastolic blood pressure, and smaller increases in weight (p < .05). The authors concluded that tolerability was good in both genetic groups and, in particular, there was no significant ad- verse effect on the cardiac QTc interval.
Wang et al. (2007) reported a multicenter study of a large sample of 330 children ages 6–16 years who met criteria for a DSM-IV diagnosis of ADHD. The chil- dren were randomly allocated to treatment with either atomoxetine or methylphenidate. The authors reported no differences in outcome using standardized outcome measures, but the atomoxetine group reported signifi- cantly more frequent side effects: anorexia (37.2 vs. 25.3%, p = .024), nausea (20.1 vs. 10.2%, p = .014), somnolence (26.2 vs. 3.6%, p = .0001), dizziness (15.2 vs. 7.2%, p = .024), and vomiting (11.6 vs. 3.6%, p < .001). Unfortunately, there was no control group; in ad- dition, atomoxetine treatment was initiated at a faster rate than is usually recommended (titrated to 1.2 mg/ kg on Day 5). The impact of this more rapid initiation of medication on side effects is uncertain, but it may be relevant.
Summary
Atomoxetine can be useful if stimulant treatment has been found to be ineffective or when stimulants are contraindicated due to substance misuse in the fam- ily, when there is comorbidity with tics and/or anxiety, patient preference, or when a 24-hour response is es- sential. Atomoxetine should commence at a dose of 0.5 mg/kg and, after a minimum of 7 days, be increased gradually to the recommended dose of 1.2 mg/kg. Side effects may be reduced by giving atomoxetine in the evening. Alternatively, it can be divided into morning and evening doses.
Common side effects of atomoxetine are nausea and appetite loss. Insomnia and constipation may occur. Earlier concerns about potential liver problems appear to have been related to a serious idiosyncratic (allergic- type) reaction, which can also occur with stimulants; none of the recent studies has identified further cases. Atomoxetine can precipitate seizures in someone al- ready predisposed to them, but there is no evidence that it is contraindicated in the presence of well- controlled seizure conditions.
Table 6.3 summarizes information for clinicians regarding the use of stimulants and atomoxetine in ADHD.
Tricyclic Antidepressants
The tricyclic antidepressants are the most frequently used alternatives to stimulants and atomoxetine in ADHD treatment. T. Spencer et al. (1996) reviewed the literature on their use in ADHD or hyperkinesis and reported that 24 of the 26 (92%) studies of tricy- clic antidepressant treatment of ADHD in latency- age children indicated significant behavioral improvements compared to placebo. Twelve studies evaluated the re- sponse to imipramine; nine, to desipramine; three, to amitriptyline; four, to nortriptyline; and one, to clomip- ramine. Other studies that have tended to use higher dosages report a better outcome than the earlier studies (Biederman, Gastfriend, & Jellinek, 1986; Gastfriend, Biederman, & Jellinek, 1985; Wilens, Biederman, Geist, Steingard, & Spencer, 1993). However, electro- cardiography (ECG) should be carried out regularly and with each significant increase in dose, due to the risk of arrhythmias. Several children receiving tricy- clic antidepressants have died suddenly, although this rate may be no higher than the sudden death rate in the general population. Both parents and teachers have reported that behavioral symptoms improve as well with tricyclic antidepressants as with stimulants. Four studies, including two controlled trials, have included preschool children, and have shown significant im- provements. Similarly, eight studies that included ado- lescents (Gastfriend et al., 1985; Wilens et al., 1993) reported a similar degree of benefit.
Biederman, Baldessarini, Wright, Knee, and Har- matz (1989) reported one of the largest RCTs of de- sipramine in 62 children ages 6–17 years, of whom 43 had responded poorly to stimulants. Treatment was 6 weeks of either desipramine or placebo. Significant be- havioral improvements occurred at an average dose of 4.6 ± 0.2 mg/kg. Sixty-eight percent of desipramine- treated subjects were considered very much or much improved, compared with only 10% of those treated with placebo. In a second study, Biederman, Baldes- sarini, Wright, Keenan, and Faraone (1993a) reported that improvement was seen even in cases with comor- bid conduct disorder, depression, or anxiety. Cases with “pure” ADHD showed a trend toward a lesser placebo response and a greater difference between desipramine and placebo.
Possible benefits of the tricyclic antidepressants, com- pared with stimulants, are longer duration of action and improved response in children with comorbid anxiety and depression (McClellan, Rubert, Reichler, & Sylves-
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
218 W h at Wo r k s f o r W h o m ?
ter, 1990). Other advantages include the use of a single daily dose, lower risk of medication abuse, and lower risk of an afternoon rebound in hyperactive behavior.
Baseline blood pressure, pulse, and ECG, to rule out any preexisting conduction anomalies, should be re- corded before tricyclic antidepressants are prescribed. These parameters should also be monitored regularly during treatment and with each dose increase.
In summary, tricyclic antidepressants have been shown to be beneficial in the treatment of the primary symptoms of ADHD in 70% of children of all ages. Al- though children with “pure” ADHD are likely to show the most improvement with stimulant treatment, it may be preferable to use a tricyclic antidepressant in those with comorbid depression, anxiety, or aggression, es- pecially if the anxiety and/or depression worsens with
tABle 6.3. summary of stimulants and Atomoxetine
Stimulants Atomoxetine
Product names Methylphenidate: Ritalin, Equasym (IR or XL), Medikinet (IR or XL), Concerta XL
Dexamfetamine (Dexamphetamine): Dexadrine, Lisdexamfetamine, Adderall
Straterra
Dosage See Table 6.2 for details of slow-release preparations.
Methylphenidate: 10–60 mg/day
Dexamfetamine: 5–30 mg/day
0.5 mg/kg for the first week, then increase gradually over 4–6 weeks to 1.2 mg/kg
Onset of action Same day (see Table 6.2) 4 weeks for full effect
Route and form of administration
Oral tablet or capsule. Some can be sprinkled in food and drink. Methylphenidate skin patches are also available
Oral capsule; can be sprinkled in food or drink
Frequency of administration
Slow release: once a day
Immediate release: 2–6 times per day
Once a day; divide into two doses per day if side effects are seen with daily dosing
Indications ADHD with impairment together with behavioral strategies for parent and child, in addition to behavioral approaches in school
Second-line treatment if stimulants are ineffective, insufficient, or if side effects are seen
Substance misuse
Anxiety
Investigations prior to or during administration
Baseline and ongoing: height, weight, blood pressure and pulse
Others dependent on full medical history and physical examination
Baseline and ongoing: height, weight, blood pressure and pulse
Others dependent on full medical history and physical examination
Contraindications Substance misuse
Untreated epilepsy
Psychosis
Severe anxiety
Children who have not already tried stimulants if safe
Reluctance to take on a daily basis
Allergy to atomoxetine
Side effects Reduced appetite
Insomnia
Precipitation of seizures
Nausea and reduced appetite
Insomnia
Constipation
Suicidal ideation
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 219
stimulant therapy. Tricyclic antidepressants may also be indicated in cases in which there is a severe rebound with stimulant treatment. Careful monitoring of car- diac status with ECGs is essential, and advice must be given to parents/caregivers about the dangers of this medication in overdose.
Children with ADHD and a comorbid mood disorder are sometimes treated with the combination of a stimu- lant and an antidepressant. This practice seems to be more prevalent in the United States than in the United Kingdom. There is no good evidence to support this type of drug combination.
Nontricyclic Antidepressants
Barrickman, Noyes, Kuperman, Schumacher, and Verda (1991) reported an open-label trial of the selec- tive serotonin reuptake inhibitor (SSRI) fluoxetine in children and adolescents with ADHD. Sixty percent of subjects improved moderately, and there were few ad- verse reactions.
An earlier study (Zametkin, Rapaport, Murphy, Lin- noila, & Ismond, 1985) was the first to consider the use of the monoamine oxidase inhibitor (MAOI) group of antidepressants. This 12-week double- blind crossover study compared the effects of an MAOI with dexamfe- tamine in a sample of 14 boys ages 7.7–10.7 years with ADHD. Tranylcypromine sulphate or clorgyline was given twice a day in 5-mg doses. Dexamfetamine dos- age was 10 mg in the morning and 5 mg at lunchtime. Low- tyramine diets were required, because of the risk of a hypertensive crisis associated with disruption of tyramine metabolism by MAOIs, and blood pressure was measured weekly. There were no significant ad- verse reactions. The MAOIs and dexamfetamine im- proved disruptive behavior, as rated by both teachers and parents, to an equivalent degree. Attention also improved with both types of medication. There were no significant differences between the treatments. This small, well- designed study suggests that the newer, more selective MAOIs may provide a promising alter- native to stimulant treatment.
Akhondzadeh et al. (2003) investigated a type B MAOI, selegiline, which is metabolized to amphet- amine and methamphetamine stimulant compounds, in 28 children between ages 4 and 9 years who met criteria for a DSM-IV diagnosis of ADHD. In a small, 4-week RCT, they compared selegiline at doses of 5 mg/day (for children under 5 years of age) and 10 mg/day (for children over 5 years) with methylpheni- date. The two treatments showed equal efficacy and
side effects. Further studies of longer duration are re- quired.
Barrickman et al. (1995) reported a double- blind crossover study contrasting the effectiveness of bu- propion and methylphenidate in 15 subjects ages 7–17 years with ADHD. Bupropion is an antidepressant whose pharmacological profile is similar to that of the stimulants. According to parental reports, methylphe- nidate was significantly more effective than bupropion in improving attention. The two treatments did not dif- fer significantly in relation to improvement of conduct; both treatments significantly improved conduct above the baseline behavior.
Side effects of bupropion include skin reactions, which abate after stopping the drug. Bupropion lowers the seizure threshold at a frequency similar to that of the tricyclic antidepressants. Because it is a weak do- paminergic agent, there is a small risk that it may pre- cipitate a psychotic illness. Four such cases have been reported in the literature (mentioned by Barrickman et al., 1995).
The scarcity of studies of treatment with nontricyclic antidepressants in ADHD makes it impossible to advise about their use, other than to comment that they are worth consideration if stimulants and tricyclic antide- pressants are contraindicated.
Clonidine
D. R. Palumbo et al. (2008) and Daviss et al. (2008) have reported a well- designed multicenter study of clonidine treatment in 122 children ages 7–12 years fulfilling criteria for a DSM-IV diagnosis of ADHD. In two successive 4-week titration periods, clonidine or matching placebo (in the first period) was compared with clonidine plus methylphenidate, methylphenidate on its own, or placebo (in the second period). The two titration periods were followed by an 8-week mainte- nance period. Clonidine showed significant benefits on Conners Parent Rating Scales and clinicians’ ratings of impairment (Children’s Global Assessment Scale). However, there was no benefit based on Conners Teach- er Rating Scales for clonidine treatment alone. At the end of the study, the mean ± SD doses of the medi- cations in each group were as follows: clonidine, 0.24 ± 0.11 mg/day; methylphenidate, 30.2 ± 18.9 mg/day; combined regimen, clonidine 0.23 ± 0.13 mg/day plus methylphenidate 25.4 ± 18.2 mg/day. There was one se- vere adverse reaction in the combined treatment group, with the development of a prolonged QTc interval and left ventricular hypertrophy on ECG, but no clinical
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
220 W h at Wo r k s f o r W h o m ?
symptoms or findings on echocardiography. Sedation was the main side effect with clonidine, but this de- creased after Week 8. The ESs were as follows: methyl- phenidate versus no methylphenidate, –0.49; clonidine versus no clonidine, –0.24; combined treatment versus placebo, –0.73.
In summary, clonidine appears to be as beneficial as the tricyclic antidepressants in the treatment of ADHD symptoms and should be a second- line treatment for those who cannot be treated with stimulants or atom- oxetine. The authors recommend an ECG before and at regular intervals during treatment, although a statement from a committee of pediatric cardiologists, accepted by the American Heart Association, stated that ECGs are not required (Gutgesell et al., 1999). The risk of a hypertensive rebound means that this treatment must not be stopped suddenly. Two children treated with methylphenidate and clonidine in the United States and one in Australia have died. Sudden cessation of cloni- dine may have been the cause, although this is unprov- en. Clonidine also frequently causes drowsiness. Al- though no definite recommendations can be made about the combination of clonidine and stimulant medication until there is more scientific information regarding the safety of their combined use, the rate of prescribing of clonidine, both separately and in combination with methylphenidate, has greatly increased (Swanson, Con- nor, & Cantwell, 1999; Swanson et al., 1996; Wilens, Spencer, Swanson, Connor, & Cantwell, 1999). Until more is known about the safety of this drug combina- tion, an ECG should be obtained at the start of treatment and with each dose change (Daviss et al., 2008).
Guanfacine
Guanfacine is related to clonidine but has a longer half- life and is less sedating. It is also more selective in its binding to the alpha- adrenergic receptors and does not bind to alpha1 receptors.
Biederman et al. (2008) have reported a placebo- controlled study of an extended- release formulation of guanfacine. A total of 345 patients ages 6–17 years, who met criteria for a DSM-IV diagnosis of ADHD, were randomly assigned to treatment with placebo or guanfa- cine at 2-, 3-, or 4-mg doses for a period of 8 weeks. All doses showed significant benefit by Week 5. There were more side effects in the treatment groups, with som- nolence correlating with higher doses. There were no significant ECG changes. However, when the response rate according to age was examined, adolescents failed
to show significant improvement, whereas younger children did. The authors hypothesize that adolescents would require a higher dose. The treatment ESs were 0.58 for a dose of 0.05–0.08 mg/kg, 1.19 for 0.09–0.12 mg/kg, and 1.34 for 0.13–0.17 mg/kg. These ESs are similar to those of other nonstimulant medications.
Faraone and Glatt (2010) recently reported a review of three clinical trials of the extended- release prepara- tion, including the study of Biederman et al. (2008). The authors concluded that longer term treatment is as- sociated with less sedation. In other words, persevering with the treatment, if possible, often leads to decreases in the side effect of somnolence.
Modafinil
Modafinil is a treatment used for narcolepsy. Amiri et al. (2008) summarized the current knowledge about this treatment. Its mechanism of action is not fully un- derstood, but it is thought to activate the hypothalamus by altering the balance of gamma- aminobutyric acid.
Greenhill et al. (2006a) reported a 9-week placebo- controlled study of modafinil in 128 children and adoles- cents ages 7–17 years, who fulfilled criteria for a DSM- IV diagnosis of ADHD. The dose range for modafinil was 170–450 mg once per day. There was significant improvement (p < .0001) with modafinil compared with placebo. Although there was a significantly higher rate of side effects (insomnia, headache, decreased appetite, weight loss) with modafinil, there was no increase in dropout rates compared with placebo.
Amiri et al. (2008) reported an RCT comparing modafinil (200–300 mg once a day) and methylphenidate (20–30 mg per day) in 60 children and adolescents ages 6–15 years, who fulfilled DSM-IV diagnostic criteria for ADHD. The treatment period was 6 weeks. There were no significant differences in treatment response or drop- out rates. Methylphenidate was associated with signifi- cantly more common side effects of decreased appetite and difficulty falling asleep compared with modafinil treatment. There was no increase in side effects associ- ated with modafinil relative to methylphenidate.
Carbamazepine
Silva, Munoz, and Alpert (1996) conducted a meta- analysis of carbamazepine use in patients with ADHD. They found reports of seven open studies involving a total of 189 patients, and three double- blind studies including a total of 53 patients. Overall, 71% of those
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 221
treated with carbamazepine in the controlled studies showed significant improvement, compared with 26% of those treated with placebo. There was an ES of 1.01 for the drug– placebo comparison in the double- blind studies. The meta- analysis confirmed that carbamaze- pine was significantly more effective than placebo in controlling ADHD symptoms. A similar response rate was found in the open studies.
Monitoring of the leukocyte count is required during carbamazepine treatment due to the risk of leukopenia. Liver function should also be tested regularly because of the risk of liver damage. Other side effects include rashes, ataxia, and drowsiness.
Antipsychotics
Antipsychotics should be used only if all the other groups of drugs we discussed earlier have been tried and the child is extremely disturbed. The child should be reassessed first to ensure that the diagnosis of ADHD is correct. Antipsychotics must be used only in the short term and at low dosage, because of the risk of side effects. On the whole, the risks are so great that antipsychotics should probably be avoided in ADHD treatment, although no good studies have evaluated the risks and benefits of this class of drug.
Other Medications
Davari- Ashtiani, Shahrbabaki, Razjouyan, Amini, and Mazhabdar (2010) have reported a 6-week RCT in Iran with 34 children ages 6–12 years, comparing methylphenidate and an anxiolytic drug, buspirone. Both treatments significantly reduced ADHD symp- toms from the second week of treatment. The outcome measure was teacher and parent ADHD rating scales. Side effects were also monitored. The maximum dose of methylphenidate was 60 mg per day, and that of bus- pirone, 45 mg per day. Although both drugs had a posi- tive effect and no significant side effects, methylpheni- date was significantly more effective than buspirone in terms of reductions in teacher- rated inattention scores. Buspirone was reported to have an anxiolytic effect, but no rating scale was used to assess this. There was no placebo comparison; therefore, the results need to be interpreted with caution.
Table 6.4 summarizes information for clinicians regarding the use of tricyclic and nontricyclic antide- pressants, clonidine, guanfacine, modafinil, and carba- mazepine in treatment of ADHD.
Dietary Interventions
Exclusion Diets
It has been suggested that certain food additives may increase behavioral problems in normal and hyperac- tive children. A meta- analysis by B. Bateman et al. (2004) found that elimination diets reduced hyper- activity symptoms (standardized mean difference of 0.80). Children showed adverse behavioral responses to a range of substances, but most frequently chocolate, wheat, and dairy products, and the pattern was individ- ual to the child. This has been supported by Pelsser et al. (2011) in an RCT in the Netherlands and Belgium of one hundred 4- to 8-year-olds placed on a strict elimi- nation diet. This study is part of the Impact of Nutri- tion in Children with ADHD study. When challenged with their usual diet, 19 of 30 children (63%) in the sample relapsed. This study supports the finding that some children can benefit from changes to their diet. However, there is no good evidence for more rigorous exclusion diets. Contrary to popular belief, there is no evidence for the benefit of large doses of vitamins or herbal remedies.
Polyunsaturated Fatty Acids
Omega-3 and omega-6 fatty acids are “essential” fats that the body cannot make. The concentrations of these fatty acids are higher in the brain than in other parts of the human body. Benton (2007) reviewed the studies of supplementation with polyunsaturated fatty acids in ADHD and found no evidence that they are beneficial. In another review, B. M. Ross, Seguin, and Sieswerda (2007) reached the same conclusion. However, a recent RCT of 92 children ages 7–12 years reported a signifi- cant improvement in oppositionality and concentration after 15 weeks of omega-3 and omega-6 supplementa- tion among children who showed less hyperactivity and impulsiveness at baseline (Gustafsson et al., 2010). The intervention was not effective for those with more se- vere hyperactivity and impulsiveness, so that when the group of children was considered as a whole, the inter- vention was ineffective. On the basis of these findings, some children may benefit, but those with the most se- vere ADHD symptoms do not appear to do so. Larger studies are required. Transler, Eilander, Mitchell, and van de Meer (2010) have reviewed the findings from studies to date and recommend further research, be- cause it is unclear whether fatty acid supplementation is beneficial.
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
222
tA B
l e 6
.4 .
P ro
fil es
o f th
e o
th er
d ru
gs U
se d t
o tr
ea t
A d
h d
T ri
cy cl
ic a
nt id
ep re
ss an
ts N
on tr
ic yc
li c
an ti
d ep
re ss
an ts
C lo
n id
in e/
g u
an fa
ci n
e M
o d
afi n
il C
ar ba
m az
ep in
e
N am
es Im
ip ra
m in
e
D es
ip ra
m in
ea F
lu ox
et in
e (S
S R
I)
T ra
ny lc
y p
ro m
in e
su lf
at e
(M A
O I)
S el
eg il
in e
(M A
O I)
B up
ro pi
on
C lo
n id
in e
hy d
ro ch
lo ri
d e
G u
an fa
ci n
e
M o
d afi
n il
T eg
re to
l
D o
sa ge
Im ip
ra m
in e
7– 8
ye ar
s: 2
5 m
g
8 –1
1 ye
ar s:
2 5
–5 0
m g
≥1 1
ye ar
s 50
–7 5
m g
S ee
i nd
iv id
u al
d ru
g p
re sc
ri bi
ng g
u id
el in
es C
lo n
id in
e: 1
10 –2
6 0
µ
g /d
ay
G u
an fa
ci n
e: 0
.1 3
– 0.
17
m g
/k g
/d ay
2 0
0 –
4 0
0 m
g /d
ay , e
it h
er i
n a
si ng
le m
o rn
in g
d o
se o
r tw
o d
o se
s g
iv en
m o
rn in
g an
d m
id d
ay
5 –1
0 ye
ar s:
M ax
im u
m
4 0
0 –
6 0
0 m
g /d
ay . S
ta rt
a t
2 5
–5 0
m g
on ce
o r
tw ic
e a
d ay
10 –1
5 ye
ar s:
M ax
im u
m
0. 4
–1 g
/d ay
. S ta
rt a
t 50
–1 0
0 m
g on
ce o
r tw
ic e
a d
ay
O n
se t
of a
ct io
n 4
w ee
k s
fo r
fu ll
e ff
ec t
4 w
ee k
s A
p p
ro x
im at
el y
4 w
ee k
s S
am e
d ay
D ay
s to
w ee
k s
R o
ut e
an d
fo rm
of
a d
m in
is tr
at io
n O
ra l
ta bl
et s
O ra
l ta
bl et
s O
ra l
ta bl
et s
O ra
l ta
bl et
s O
ra l
ta bl
et s
o r
li qu
id
F re
qu en
cy o
f ad
m in
is tr
at io
n O
nc e
a d
ay a
t n
ig ht
S ee
i nd
iv id
u al
g u
id el
in es
2 –3
t im
es a
d ay
O nc
e o
r tw
ic e
a d
ay O
nc e
o r
tw ic
e a
d ay
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
223
In d
ic at
io n
s T
h ir
d -l
in e
tr ea
tm en
t
A d
d it
io n
al a
n x
ie ty
a nd
d
ep re
ss io
n
If s
id e
ef fe
ct s
o cc
u r
fr om
st
im u
la nt
s o
r at
om ox
et in
e
L ac
k of
b en
efi t
fr om
o th
er
tr ea
tm en
ts T
h ir
d -l
in e
tr ea
tm en
t w
h en
u
n ab
le t
o u
se s
ti m
u la
nt s
o r
at om
ox et
in e
F o
u rt
h -l
in e
tr ea
tm en
t if
no
th in
g el
se w
o rk
s u
nt il
m
o re
s tu
d ie
s co
m pl
et ed
F o
u rt
h -l
in e
tr ea
tm en
t
In ve
st ig
at io
n s
p ri
o r
to o
r du
ri ng
ad
m in
is tr
at io
n
E C
G a
t ba
se li
n e
an d
du ri
ng t
re at
m en
t A
ss es
s su
ic id
e ri
sk E
C G
p ri
o r
to c
om m
en ci
ng
tr ea
tm en
t an
d w
it h
ea ch
d
o se
i nc
re as
e
F u
ll m
en ta
l st
at e
as se
ss m
en t
C on
si d
er r
is k
of
d ep
en d
en ce
B lo
o d
p re
ss u
re a
nd p
u ls
e
L iv
er f
u nc
ti on
t es
ts
C om
pl et
e an
d d
if fe
re nt
ia l
bl o
o d
co u
nt
C on
tr ai
nd ic
at io
n s
C ar
d ia
c ab
no rm
al it
ie s
S u
ic id
e ri
sk a
nd
m ed
ic at
io n
u n
ab le
t o
b e
ke p
t sa
fe f
ro m
p at
ie nt
M an
ic p
h as
e D
ep re
ss io
n
A cu
te p
o rp
hy ri
a
B re
as t
fe ed
in g
C ar
d ia
c p
ro bl
em s
P sy
ch o
si s
D ep
re ss
io n
S ub
st an
ce m
is u
se
B on
e m
ar ro
w s
up p
re ss
io n
A cu
te p
o rp
hy ri
a
S id
e ef
fe ct
s D
ry m
o ut
h
S ed
at io
n
C ar
d ia
c ch
an ge
s
T re
m o
r
B eh
av io
r ch
an ge
s
W ei
g ht
g ai
n
G as
tr oi
nt es
ti n
al e
ff ec
ts
su ch
a s
n au
se a
R as
h an
d se
ve re
a ll
er g
ic
re ac
ti on
A n
x ie
ty
S u
ic id
al t
ho ug
ht s
S ed
at io
n
D ep
re ss
io n
D ry
m o
ut h
H ea
d ac
h e
G as
tr oi
nt es
ti n
al
d is
tu rb
an ce
s
A n
x ie
ty
In so
m n
ia
N au
se a
an d
vo m
it in
g
D ro
w si
n es
s
R as
h
B lo
o d
d is
o rd
er s
A g
g re
ss io
n
D ep
re ss
io n
P sy
ch o
si s
a D
es ip
ra m
in e
h as
b ee
n w
it h
d ra
w n
fr o
m u
se i
n th
e U
n it
ed K
in gd
o m
.
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
224 W h at Wo r k s f o r W h o m ?
Iron Supplementation
Konofal et al. (2008) reported a small pilot study of 23 children ages 5–8 years, who met criteria for a DSM-IV diagnosis of ADHD and had serum ferritin (a measure of iron stores) levels below 30 ng/ml. None of the chil- dren was clinically anemic. The children were random- ized to receive iron supplementation or placebo. There was a significant improvement in children treated with ferrous sulphate at a dose of 80 mg/day. This study sug- gests that ferritin levels should be measured in children with ADHD. It is unclear whether the low serum fer- ritin levels were significantly associated with a history of poor diet.
Pycnogenol
Pycnogenol is a polyphenolic standardized extract of French maritime pine bark with antioxidant properties. Several case reports have suggested it is of benefit in ADHD. A group in Slovakia (Chovanova et al., 2006; Dvorakova et al., 2007) used pycnogenol supplemen- tation to reduce the damage to DNA that they found in children with ADHD, compared with controls. The authors explain that damage to DNA and other cellu- lar substances results from the production of oxygen- derived free radicals inside the cells, and presume that in ADHD, the cells are under particular stress from this process. The study was a placebo- controlled trial involving 61 children and adolescents ages 6–14 years. The authors did not report how the diagnosis of ADHD was made. They found a significant reduction in ADHD symptoms during the treatment (a period of 1 month), but relapse occurred 1 month after the treat- ment was stopped. Further research in this area may be worthwhile.
Homeopathy
A Cochrane review (Coulter & Dean, 2007) reported that there is currently little evidence for the efficacy of homeopathy in the treatment of ADHD.
Neurofeedback
Leins et al. (2007) reported a comparison study of a small (N = 38) number of children and adolescents ages 8–13 years, fulfilling criteria for a DSM-IV diagnosis of ADHD, who were randomized to two experimental groups that received two different neurofeedback train-
ing protocols (theta–beta frequencies and slow cortical potentials). The participants were blind to intervention type, but those providing the treatment were not. The aim was to assess for improvements in behavior and cognition. Both protocols led to improvements, which were sustained at 6-month follow- up. An earlier small study (Levesque, Beauregard, & Mensour, 2006) of 20 children, with a small control group of 5 children, also suggested that neurofeedback was beneficial, as did a study of 34 children ages 8–12 years (Fuchs, Birbau- mer, Lutzenberger, Gruzelier, & Kaiser, 2003). These studies need repeating with a larger sample size and a placebo group. In a review of neurofeedback stud- ies, Vernon, Frick, and Gruzelier (2004) emphasized the need for a standardized package of neurofeedback training if this potential treatment is to be effectively and consistently researched and implemented.
Melatonin for Sleep Problems
Van der Heijden, Smits, Van Someren, Ridderinkhof, and Gunning (2007) reported the first RCT of mela- tonin versus placebo treatment in 105 six- to 12-year- olds who fulfilled criteria for a DSM-IV diagnosis of ADHD and also had chronic sleep-onset insomnia. This was a well- designed study with several outcome measures. The doses of melatonin used were 3 mg for children weighing less than 40 kg and 6 mg for those weighing more than 40 kg. Actigraphy was used to measure sleep. Sleep logs were used to verify the ac- tigraphy recordings; these recorded sleep onset, sleep latency (time from lights out to sleep onset), wake up time, total time asleep, sleep efficiency, and moving time (the percentage of time spent moving during the presumed sleep period). The results confirmed the ben- efits of melatonin, with a significant improvement in time of sleep onset compared with placebo treatment. There was also a significant increase in the mean total time asleep with melatonin, and significant decreases in sleep latency and sleep restlessness.
An additional measure used by the authors, dim light melatonin onset (DLMO), the time at which the endogenous melatonin level starts to rise, is a mea- sure of the biological clock rhythm. Samples of saliva were obtained hourly by having the subjects chew on a cotton bud. Only dim light was permitted during the measurement. Children treated with melatonin showed a significant advance in DLMO compared with con- trols, and there was a linear relationship between the pre- and posttreatment levels, such that more delayed
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 225
DLMO levels at baseline were associated with greater improvements in sleep onset after melatonin treatment. There was no significant difference in adverse reactions between the placebo and melatonin treatments. None of the children needed treatment for the side effects, and none withdrew from the trial.
Psychosocial treatments
We recommend a useful review and summary of the literature by Barkley (2002), which suggests a program for psychosocial treatments and urges that these should be “maintained over extended periods of time.” Such an approach would require the commitment and col- laboration of everyone involved in the child’s care. The article describes components of behavioral parenting approaches and training of teachers in classroom man- agement. Several other authors have also reviewed the benefits of these interventions, and their findings are summarized below.
Behavioral Therapies
An individual behavioral therapy approach was a component of the multimodal treatment provided in the MTA study described earlier in this chapter. The behavioral approach consisted of a summer treatment program developed by Pelham and Hoza (1996). In the MTA study, the therapist who supervised the individual behavioral therapy with the child also provided par- ent training and school- based consultation. Although the impact of each of these individual aspects of the treatment cannot be separated out, the children who received the combined behavioral approach required less medication than those who did not receive it. Very few authors have reported good- quality studies of an individual behavioral approach in treatment of ADHD.
Pfiffner et al. (2007) reported posttreatment findings and 3- to 5-month follow- up of a controlled study of the response of the inattentive subtype of ADHD to a be- havioral program. Sixty-nine children ages 7–11 years were involved. The program, referred to as the Child Life and Attention Skills Program (CLAS), was com- pared to a wait-list control or a treatment- as-usual con- trol. The CLAS program consisted of parent training (eight to 10 sessions) and child skills treatment (eight to 10 sessions and added family sessions), which were manualized. There was a significant reduction (50%) in inattentive behaviors for the treated group versus the controls (p = .0004). Parent- and teacher- rated im-
provements in social functioning and organizational skills were significant. The authors reported high lev- els of parent, teacher, and child satisfaction with the program. However, this was a small sample, and most of the ratings were undertaken by parents and teachers who were not blind to the treatment group in which the child was placed, so it is possible there was some bias. Despite these shortfalls, this intervention seems prom- ising, although it targets inattentive children, who are often not highly represented in the clinic population. As in this study, they would be best referred by schools.
Progress in this area has been summarized in a re- view by the American Academy of Child and Adoles- cent Psychiatry (AACAP; Pliszka & AACAP Work Group on Quality Issues, 2007). Although behavioral approaches are less effective than medication in reduc- ing the primary symptoms of ADHD, they have been shown to improve the targeted behaviors to a degree, in addition to improving social skills and academic per- formance.
As summarized in the AACAP review (Pliszka & AACAP Work Group on Quality Issues, 2007), the main shortcomings of behavioral management in ADHD are that improvements in behavior tend not to be sustained over time and they generalize poorly to situations other than those in which the training oc- curred. Booster sessions, and training in the setting where the behavioral improvement is required, should help to remedy these problems.
In summary, behavior therapy alone is less effective than stimulant medication. Combining behavior ther- apy with a low dose of stimulants, however, may lead to sufficient behavioral improvement, so that a higher dose of medication is not required. Behavioral therapy is most likely to lead to improvements in on-task be- havior and a reduction in disruptive and rule- breaking behavior. Improvements in academic performance usu- ally require medication. It is likely that an individual behavioral approach needs to be combined with parent training and school consultation; this is the first-line in- tervention for mild and moderate ADHD recommend- ed by the NICE (2008) guidelines for the treatment of ADHD.
Cognitive‑Behavioral Therapy
Cognitive- behavioral therapy (CBT) combines the techniques of behavioral management with training in problem solving and self- monitoring. Several studies have failed to confirm benefits from CBT. R. T. Brown,
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
226 W h at Wo r k s f o r W h o m ?
Wynne, and Medenis (1985) reported a study in which 30 boys between 6 and 12 years of age, with a diagno- sis of ADD-H, were randomly assigned to one of three treatment groups: methylphenidate therapy, cognitive training, or both interventions combined. There was a no- treatment control group of 10 children on the wait list (nonrandomly assigned). The treatment lasted 12 weeks and was provided as two 1-hour sessions a week (24 sessions in total). The methylphenidate group im- proved significantly compared with the cognitive train- ing and no- treatment groups. There was no significant improvement on academic measures, with the excep- tion of the Durrell Subtest of Listening Comprehen- sion. The combination of methylphenidate and CBT had no significant advantage over methylphenidate treatment alone. CBT alone produced some improve- ment in attention, but the benefits were not as large as with methylphenidate.
As with behavioral therapy, there is a problem with teaching children to generalize any problem- solving strategies they master to different settings and in get- ting them to use them spontaneously. The view of ex- perts in the field (Abikoff, 1991; Pliszka & AACAP Work Group on Quality Issues, 2007) is that CBT is only occasionally beneficial for children with ADHD, and even then is less efficacious than medication.
In summary, CBT appears to be less effective than medication in treating the primary symptoms of ADHD; however, the majority of studies in the field so far have been limited in design. CBT has no advan- tages over behavioral therapy in relation to academic performance, and medication appears to be preferable to these psychological approaches. CBT may improve self- control, although most studies to date have failed to confirm this.
Parent Training
A course of parent training normally comprises six to 12 sessions and is best carried out when the child is prepu- bertal. Children with more severe behavioral problems will require treatment with medication first. Booster training sessions are usually necessary. Sonuga-Barke, Daley, Thompson, Laver- Bradbury, and Weeks (2001) reported an RCT of parent training in 78 preschool children. The parent training comprised a structured, manualized, 8-week program of weekly home visits by a specially trained health visitor therapist who provid- ed one-to-one advice about reducing defiant behaviors and increasing attention and organizational skills. The
preschool- age children were assessed to have ADHD behaviors. Parent training was compared with a wait- list control group and a parent counseling and support group. The authors reported that ADHD behaviors were significantly reduced, and mothers’ sense of well- being was increased in the parent training group rela- tive to the other two groups. The beneficial effects of the intervention were maintained 15 weeks posttreat- ment. Importantly, this study showed that primary care staff can be trained to provide the intervention.
K. Jones, Daley, Hutchings, Bywater, and Eames (2007) reported on use of the Incredible Years parent training program (see also Chapter 4) for preschool children showing signs of developing ADHD. Use of the program led to a significant improvement in paren- tal reports of problem behaviors (52%), compared to wait-list controls (21%). These results are similar to those of Sonuga-Barke et al. (2001), who reported that 53% of participants showed significant improvement, and Bor et al. (2002), who reported 80% improvement.
Webster- Stratton, Reid, and Beauchaine (2011) have reported similar results from an RCT of a combined parenting and child program intervention based on the Incredible Years program. The study included 99 chil- dren diagnosed with ADHD (hyperactive or combined subtypes) ages 4–6 years. The intervention comprised a 20-week program of weekly, 2-hour sessions for parents with a separate “Dinosaur School” children’s program running in parallel. Mothers and independent observ- ers reported significant improvements in the children’s behavior; mothers, but not fathers, reported improved parenting behaviors. The authors did not seek teachers’ reports. Another weakness is that because the control group was a wait-list control, there may have been some bias in the mothers’ reports, although the independent observers’ reports agreed with the mothers’ reports. Overall, this hopeful finding confirms the benefits of a group-based parenting intervention with a program encouraging children with their social development. Longer-term follow- up is required in future studies to confirm whether the benefits persist.
van den Hoofdakker et al. (2007) have reported a useful study of how behavioral parent training (BPT) can be incorporated into routine clinical care (RCC). In this study from the Netherlands, 94 children be- tween ages 4 and 12 years, who met diagnostic criteria for DSM-IV ADHD, were randomly assigned to RCC alone or RCC with additional BPT. RCC in both groups may have included medication. Follow-up was conduct- ed 25 weeks after the cessation of the BPT program.
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 227
The BPT was a manualized program that comprised twelve 2-hour sessions of group parent training ad- ministered by two psychologists. There was no differ- ence between the two groups with respect to outcome of ADHD symptoms. However, at follow- up, there was a significant difference in favor of BPT plus RCC for individually targeted externalizing behavior problems in the children. The combined treatment also led to a greater reduction in internalizing symptoms.
There was no significant difference between the treatment groups in the dose of stimulant medication required, but children in the RCC-only group were sig- nificantly more likely to be prescribed an additional treatment (clonidine, risperidone, or citalopram in this study). Interestingly, families in the RCC-only group had significantly more contact with clinical staff. There was no benefit from BPT in relation to parental stress; this finding is similar to that reported from the MTA study (discussed earlier in this chapter). The authors suggest that parental stress levels may already have de- creased significantly with the initial RCC. They recom- mend further studies to clarify which families require the BPT intervention.
Corcoran and Dattalo (2006) reported a meta- analysis of studies involving parents in treatments for ADHD. They found that only 16 studies met the re- quired inclusion criteria. In these studies, involving parents had significant benefit only for academic per- formance (ES = 0.8204), the child’s family functioning (0.67), and internalizing problems (0.635). Teacher- reported ESs were the largest (0.75) and child- reported ESs were the smallest (0.11). The ES for ADHD symp- toms was 0.397; that for externalizing problems was 0.361. Social competence showed the lowest ES of 0.071.
In the United Kingdom, NICE (2008) has similarly reviewed the literature and concluded that parent train- ing based on accepted programs for the treatment of conduct disorder (see Chapter 4) is beneficial. They suggest that, apart from some additional psychoeduca- tion around the diagnosis of ADHD, not much needs to be changed in these manualized and well- researched programs. The NICE Guidance Group also recom- mends that parent training, with a group treatment approach, and possibly individual behavioral therapy, should be the first treatment of choice in mild to moder- ate ADHD.
Adding parent training in the home setting to medi- cation treatment has also been shown to be beneficial (Tutty, Gephart, & Wurzbacher, 2003). In this RCT
there was an improvement in ADHD symptoms and the use of parenting discipline strategies. However, there was no improvement in teacher ratings or in the child’s ability to pay attention in the treatment group com- pared with the controls.
Social Skills Training
Children with ADHD unquestionably show social defi- cits that are nearly always impairing, but the basis of these difficulties is not yet fully understood. Attempts have been made to correct them by social skills train- ing (reviewed by Cousins & Weiss, 1993), which com- bines many of the behavioral and cognitive- behavioral approaches discussed earlier. Unfortunately, several studies have shown that despite improvements in be- havior observed by parents, teachers, and research- ers, they often are not perceived by peers; this has led several authors to suggest the involvement of peers in social skills treatment programs. This is an important area for further research, because impaired social skills appear to be the most disabling and persistent deficit in children with ADHD (Hechtman, Weiss, Perlman, & Amsel, 1984).
Interventions with Teachers
Miranda, Presentacion, and Soriano (2002) reported an RCT of an approach whereby teachers in Spanish schools were trained to help pupils with self- evaluation techniques and with implementing a token economy. The study involved 50 children ages 8–9 years, fulfill- ing criteria for a DSM-IV diagnosis of ADHD. There was a reduction of hyperactive and impulsive behaviors and a significant improvement in self- control. However, most of the children still required other interventions, such as medication.
Multimodal Interventions
There has been increasing interest in multimodal thera- pies for ADHD, because treatments of any kind rarely produce a complete “cure” with generalization of be- havioral improvements to all settings (Pelham & Mur- phy, 1986). There have been few well- designed stud- ies in this area, and the conclusions have been mixed. Majewicz- Hefley and Carlson (2007) provided a very helpful review of combined psychosocial and pharma- cological treatments. This rigorous meta- analysis re- quired qualifying studies to have adequate quantitative
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
228 W h at Wo r k s f o r W h o m ?
data to enable calculation of ESs. It described outcomes in terms of the following five criteria: inattention, hy- peractivity, impulsivity, social skills, and academ- ics. Out of a possible 26 articles describing combined treatments, eight studies adequately met the inclusion criteria. The meta- analysis of psychological and phar- macological interventions confirmed that combined treatments had large ESs on the core features of ADHD (1.27 for inattention and hyperactivity, and 0.91 for im- pulsivity), as well as the peripheral feature of social skills (0.90). It found a small ES for academics (0.19). All the studies employed behavioral management as one of the comparison groups; one also included social skills training.
A major problem is that all of these studies followed the children for only short periods. One long-term follow- up study of a large sample of boys, which, un- fortunately, lacked a control (Satterfield, Satterfield, & Cantwell, 1981; Satterfield, Satterfield, & Schell, 1987), found that combined treatments were more ef- ficacious than medication alone when the subjects were ages 14–21 years. The treatments included par- ent training, group therapy, psychotherapy, and educa- tional interventions. Choice of the combination of treat- ments depended on the needs of the child and family, very much like the situation in clinical practice. The sample, which comprised boys ages 6–12 at the start of the study, was divided into two groups: one receiving only drug treatment and the other receiving multimodal therapy. The children who continued with multimodal therapy for 2–3 years responded best, with significantly lower levels of delinquency than the medication- only group. These findings support the use of multimodal therapy and long-term treatments for this group of chil- dren. However, in this study, allocation to therapy was not randomized, and the treatments were not standard- ized. Although the complexity and variation of cases means that this is often the reality of case management in clinical practice, these results must be interpreted with caution.
Family Therapy
There have been very few outcome studies of family therapy in this group of children. Bjornstad and Mont- gomery (2005) reviewed the literature and concluded that two studies addressed this issue. Both used a structural behavioral approach. They concluded that a manualized family therapy approach may be as useful as usual medical management, but not more effective.
Family therapy per se may not be the most efficacious treatment for ADHD, but a systemic approach to the impact of the disorder on family and school functioning is important (Bernier & Siegel, 1994).
Psychodynamic Therapies
There have been no trials of the use of psychotherapy in children and adolescents with ADHD, but the relatively poor insight of these children (Pelham et al., 1993) sug- gests that it is unlikely to be particularly beneficial.
Consultation between schools and mental health Professionals
Consultation with schools and teachers by mental health professionals is crucial. Social workers also need the opportunity to discuss children with whom they are involved. There is a need for clinical practice guidelines regarding the management of children with ADHD, because the problem is so highly prevalent and professionals from many different backgrounds become involved in case management. A clinical ex- ample, reported by Langberg et al. (2008), concerns data from the MTA study (discussed in detail earlier in this chapter). In this study, the severity of symptoms in the ADHD group declined with increasing age as a result of developmental changes. However, this de- cline temporarily slowed over the year that coincided with the children’s transition to middle school at age 11–12 years. It therefore seems important to advise children, their caregivers, and teachers that additional support should be made available in school during the transition year. It may be possible to reduce some of this support the following year as the ADHD symptoms again resume their developmental decline. This is the first study to describe clear evidence for the disruptive impact of transition, a problem that caregivers, teach- ers, and clinicians have been describing for some time.
sUmmArY
The evidence suggests that ADHD is a disorder that ex- ists along a continuum and has a heterogeneous etiol- ogy, and its severity relates to the long-term outcome. Depending on the definition, its prevalence is 1–5% of school- age children. Genetic factors are now known to be important, and the fact that families may therefore have several members with ADHD (including one or
EBSCOhost - printed on 9/16/2020 12:59 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
228 W h at Wo r k s f o r W h o m ?
data to enable calculation of ESs. It described outcomes in terms of the following five criteria: inattention, hy- peractivity, impulsivity, social skills, and academ- ics. Out of a possible 26 articles describing combined treatments, eight studies adequately met the inclusion criteria. The meta- analysis of psychological and phar- macological interventions confirmed that combined treatments had large ESs on the core features of ADHD (1.27 for inattention and hyperactivity, and 0.91 for im- pulsivity), as well as the peripheral feature of social skills (0.90). It found a small ES for academics (0.19). All the studies employed behavioral management as one of the comparison groups; one also included social skills training.
A major problem is that all of these studies followed the children for only short periods. One long-term follow- up study of a large sample of boys, which, un- fortunately, lacked a control (Satterfield, Satterfield, & Cantwell, 1981; Satterfield, Satterfield, & Schell, 1987), found that combined treatments were more ef- ficacious than medication alone when the subjects were ages 14–21 years. The treatments included par- ent training, group therapy, psychotherapy, and educa- tional interventions. Choice of the combination of treat- ments depended on the needs of the child and family, very much like the situation in clinical practice. The sample, which comprised boys ages 6–12 at the start of the study, was divided into two groups: one receiving only drug treatment and the other receiving multimodal therapy. The children who continued with multimodal therapy for 2–3 years responded best, with significantly lower levels of delinquency than the medication- only group. These findings support the use of multimodal therapy and long-term treatments for this group of chil- dren. However, in this study, allocation to therapy was not randomized, and the treatments were not standard- ized. Although the complexity and variation of cases means that this is often the reality of case management in clinical practice, these results must be interpreted with caution.
Family Therapy
There have been very few outcome studies of family therapy in this group of children. Bjornstad and Mont- gomery (2005) reviewed the literature and concluded that two studies addressed this issue. Both used a structural behavioral approach. They concluded that a manualized family therapy approach may be as useful as usual medical management, but not more effective.
Family therapy per se may not be the most efficacious treatment for ADHD, but a systemic approach to the impact of the disorder on family and school functioning is important (Bernier & Siegel, 1994).
Psychodynamic Therapies
There have been no trials of the use of psychotherapy in children and adolescents with ADHD, but the relatively poor insight of these children (Pelham et al., 1993) sug- gests that it is unlikely to be particularly beneficial.
Consultation between schools and mental health Professionals
Consultation with schools and teachers by mental health professionals is crucial. Social workers also need the opportunity to discuss children with whom they are involved. There is a need for clinical practice guidelines regarding the management of children with ADHD, because the problem is so highly prevalent and professionals from many different backgrounds become involved in case management. A clinical ex- ample, reported by Langberg et al. (2008), concerns data from the MTA study (discussed in detail earlier in this chapter). In this study, the severity of symptoms in the ADHD group declined with increasing age as a result of developmental changes. However, this de- cline temporarily slowed over the year that coincided with the children’s transition to middle school at age 11–12 years. It therefore seems important to advise children, their caregivers, and teachers that additional support should be made available in school during the transition year. It may be possible to reduce some of this support the following year as the ADHD symptoms again resume their developmental decline. This is the first study to describe clear evidence for the disruptive impact of transition, a problem that caregivers, teach- ers, and clinicians have been describing for some time.
sUmmArY
The evidence suggests that ADHD is a disorder that ex- ists along a continuum and has a heterogeneous etiol- ogy, and its severity relates to the long-term outcome. Depending on the definition, its prevalence is 1–5% of school- age children. Genetic factors are now known to be important, and the fact that families may therefore have several members with ADHD (including one or
C o p y r i g h t 2 0 1 5 . T h e G u i l f o r d P r e s s .
A l l r i g h t s r e s e r v e d . M a y n o t b e r e p r o d u c e d i n a n y f o r m w i t h o u t p e r m i s s i o n f r o m t h e p u b l i s h e r , e x c e p t f a i r u s e s p e r m i t t e d u n d e r U . S . o r a p p l i c a b l e c o p y r i g h t l a w .
EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 9/16/2020 1:10 AM via LA TROBE UNIVERSITY AN: 826932 ; Peter Fonagy, David Cottrell, Jeannette Phillips, Dickon Bevington, Danya Glaser, Elizabeth Allison.; What Works for Whom?, Second Edition : A Critical Review of Treatments for Children and Adolescents Account: s3094162.main.ehost
Attention‑Deficit/Hyperactivity Disorder 229
both parents) may have an impact on their response to interventions and engagement with services. The disorder is often persistent throughout childhood and adolescence, and may require continuing treatment in adulthood. The literature shows that there are often high rates of diagnoses of comorbid conduct disorder, attachment disorder, mood disorder, and specific learn- ing disabilities. Early diagnosis and treatment of ADDs improves the prognosis. ADDs are particularly persis- tent in the presence of a family history, psychosocial adversity, and comorbidity.
Physical treatments
• There is limited evidence (due to a limited number of studies and small sample sizes) that favors the use of stimulant medication to improve the working memo- ry of children and adolescents with ADHD.
• There is no evidence that prosocial behaviors in- crease with stimulant or atomoxetine use, although aggression and oppositionality are significantly re- duced.
• There is conflicting evidence on the impact of stimu- lants on academic performance in the absence of school and behavioral interventions.
• There is no evidence of serious, irreversible side ef- fects from stimulant treatment. However, two studies have confirmed a small reduction in actual relative to predicted height after long-term (more than 2 years) use at higher doses. One study that explored the use of drug holidays reported that they protect against the height reduction. There is conflicting evidence re- garding the increased risk of depression and suicidal ideation with stimulants and atomoxetine.
• There is strong evidence for the use of tricyclic anti- depressants in children and adolescents with ADHD, with or without anxiety, depression, or aggression, if stimulants and atomoxetine have been ineffective or caused side effects. Careful monitoring of cardiac status is required.
• There is limited evidence (due to the number of stud- ies) for the use of nontricyclic antidepressants in chil- dren and adolescents with ADHD.
• There is limited evidence (due to the number of stud- ies) for the use of clonidine, with or without stimu- lants, in some children with ADHD who have not responded to stimulants or atomoxetine alone.
• There is limited evidence (due to the number of stud- ies) for the use of guanfacine in some children with ADHD who have not responded to stimulants or ato-
moxetine, if they can tolerate the side effect of som- nolence, which is common in the first few weeks of treatment.
• There is limited evidence (due to the number of stud- ies) for the use of modafinil in some children with ADHD who have not responded to stimulants or ato- moxetine.
• There is limited evidence (due to the small sample sizes of studies) for the use of carbamazepine in some children with ADHD who have not responded to stimulants or atomoxetine.
• There is no evidence for the use of buspirone in the treatment of ADHD in children and adolescents, be- cause there has been no comparison with placebo treatment as yet.
• No studies have investigated whether there is im- proved effectiveness of antipsychotics or tricyclic an- tidepressants in combination with stimulants.
• There is no evidence as to who will respond well to medication for ADHD, and no evidence about when to stop medication.
ADHD and Comorbid Difficulties
• There is strong evidence for the use of stimulants or atomoxetine in children and adolescents with impair- ing ADHD (including inattention, hyperactivity, and impulsivity in the classroom), irrespective of age, and with any comorbid condition except anxiety and depression.
• There is strong evidence that treating ADHD with stimulants and other forms of treatment reduces the risk of drug misuse in adolescents with ADHD. Evi- dence suggests that medication should not be with- held when treating ADHD in substance- misusing youth.
• There is conflicting evidence (with some studies showing benefit and others not) for the use of stimu- lants in children and adolescents with ADHD and comorbid anxiety. Some studies have shown that pro- viding a behavioral intervention as well as the stimu- lant treatment confers additional benefit.
• There is limited evidence (one RCT) supporting the use of atomoxetine in children and adolescents with ADHD and comorbid anxiety.
• There is strong evidence for the use of stimulants in children and adolescents with ADHD and comorbid aggression, including ODD and conduct disorder.
• There is strong evidence for the use of stimulants in children and adolescents with ADHD and comorbid
EBSCOhost - printed on 9/16/2020 1:10 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
230 W h at Wo r k s f o r W h o m ?
generalized learning disability. However, the more severe the learning disability, the less effective the medication will be.
• There is conflicting evidence (one study showing benefit and another no benefit) for the use of stimu- lants in children and adolescents to improve the out- come of specific learning disabilities.
• There is strong evidence for the use of stimulants in children and adolescents with ADHD and comorbid epilepsy, as long as the epilepsy is well controlled be- fore initiating stimulant treatment.
Dietary Interventions
• There is limited evidence (due to a small number of studies) for the use of an elimination diet in individ- ual children if they appear sensitive to certain foods, but there is no evidence to support the blanket exclu- sion of additives and colorings in children’s diets.
• There is no evidence (due to the lack of RCTs) for the use of omega-3 and omega-6 fatty acids, iron supple- mentation, or pycnogenol in the treatment of ADHD in children and adolescents.
Other Treatments
• There is no evidence (due to the lack of RCTs) for the use of homeopathy or neurofeedback in the treatment of ADHD in children and adolescents.
Psychosocial treatments
• There is strong evidence for the use of a behavior- al parenting approach plus advice to the child and teaching staff in treating children and adolescents with mild ADHD.
• There is strong evidence for the use of a multimodal approach for children and adolescents with impairing ADHD. This comprises a parent behavioral group approach, a child behavioral approach, and a school consultation and behavioral approach, together with medical treatment with frequent initial monitoring and titration of medication by a doctor with specialist knowledge of ADHD.
• There is strong evidence that behavior therapy on its own is less effective than intensely monitored stimulant medication, but it can prevent a need for higher doses of medication. Behavior therapy improves on-task behav- ior and reduces disruptive behavior, but studies have reported little generalization across settings.
• There is conflicting evidence from a small number of studies (some showing benefit, others not) to support the use of CBT; the behavioral component may be the effective component.
• When intense medication monitoring is unavailable and there is the possibility of only infrequent ap- pointments (standard medical care) in children and adolescents with impairing ADHD, there is strong evidence for the use of either the behavioral approach (including approaches aimed at the parents, the child, and school consultation, as described above) or stan- dard medical care.
• There is some evidence (small number of studies) that social skills interventions produce no significant benefit in peer relationships.
• There is no evidence (no studies) for the effectiveness of systemic or psychodynamic therapies.
imPlicAtioNs
service Provision
• The outcome of attention deficit problems, which are common, is poor if they are untreated. However, ef- fective treatments exist and can significantly improve quality of life. Therefore, adequately resourced ser- vices should be made available across the lifespan.
• Intervention should commence earlier rather than later in order to reduce the risk of development of co- morbid problems, including depression and low self- esteem, as well as learning difficulties and conduct disorders.
• Multiagency training is required to enable early and accurate recognition of ADHD and to promote refer- ral to appropriate services.
• Given that evidence supports a multimodal (and mul- tiprofessional) approach, clear pathways to assess- ment and treatment services need to be identified so that resources can be most efficiently allocated.
• Services should be organized so that adequately trained staff members can assess and treat patients and monitor their treatment over long periods, con- tinuing throughout the school years and into adult- hood. There should be medical input from a psy- chiatrist who treats children/adolescents and/or a pediatrician with specialist training in ADHD.
• There needs to be a protocol for transition from child to adult services, and provision of adequately trained staff members who are able to work with adults af- fected by ADHD.
EBSCOhost - printed on 9/16/2020 1:10 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
Attention‑Deficit/Hyperactivity Disorder 231
Clinical
• Assessment should look beyond the core symptoms of ADHD to include associated psychiatric comor- bidities, as well as physical and educational function- ing and social contexts. It is important to exclude other reasons for problem behavior.
• If a child or adolescent is only mildly impaired by ADHD, the recommended treatment is a behavioral parenting approach and advice to the child and the teaching staff.
• If diagnostic criteria are met, and the behavior prob- lems are pervasive and severe in at least two different types of setting, then a trial of medication is indicat- ed as the first line of intervention, but with the addi- tional behavioral (parent and individual) and school interventions put in place as soon as possible.
• As assessment under the age of 6 years is likely to be less reliable, medication normally should not be used routinely as the first-line treatment in this age group. However, evidence does support the effective- ness of stimulant medication for the core symptoms of ADHD in 4- to 5-year-olds.
• Medication should still be given when there are co- morbid problems such as conduct disorders, autistic spectrum disorders, learning disabilities, Tourette syndrome, anxiety, depression, substance use dis- order (in such cases, modified- release stimulants or atomoxetine should be used), and epilepsy (when stable). Interventions for the comorbid conditions need to be available within the same service, or there needs to be good communication between services treating the comorbid conditions.
• As it is not possible to predict which dose will be ef- fective, dosage should be increased within safe limits until an effect is achieved. Effective monitoring of dose titration is needed to minimize adverse effects of drug treatment. In the MTA study, this involved monthly monitoring appointments.
• There should be annual trials of drug holidays to see whether stimulant medication is still required.
• In the case of high-dose stimulant medication, regu- lar drug holidays are recommended to prevent growth retardation.
• Stimulants may impact on a range of symptoms.
There is no evidence that the trial of other drugs dur- ing initiation of stimulant treatment is particularly helpful. If there is no, or only partial, resolution of symptoms with stimulants, other medication should be considered, such as discontinuing the stimulants and replacing them with atomoxetine, tricyclic anti- depressants, clonidine, or SSRI antidepressants. Evi- dence is lacking in either direction for augmentation of stimulant medication rather than replacement.
• If stimulants aggravate emotional problems (anxiety or depression), then tricyclic antidepressants should be considered instead.
• If there are marked adverse reactions to specific foods, then evidence suggests considering a trial of exclusion of that particular food or food type. If spe- cific reactions have not been noticed, then evidence suggests that the effort and restrictions involved in imposing a selective exclusion diet are not warranted.
• There is evidence that additional educational input is required to help children with delayed attainments to catch up, but there is no evidence for specific educa- tional approaches.
research
• Given the significant proportion of children whose ADHD does not respond to methylphenidate, fur- ther studies exploring pharmacological augmentation strategies should be undertaken.
• While it is known that methylphenidate and other medications are effective, it is less clear for how long treatment should continue. Systematic outcome stud- ies of longer term medication and its withdrawal are required.
• Given the evidence for better prognosis following early diagnosis and treatment, studies should focus on evaluating early recognition protocols and early interventions.
• Research is needed to examine effective educational interventions that address the educational difficulties experienced by many children with ADHD.
• Given the impact of the core symptoms of ADHD on families, further research into systemic family therapy interventions is required.
EBSCOhost - printed on 9/16/2020 1:10 AM via LA TROBE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use
- PeterFonagyDavi_2015_6AttentionDeficitHype_WhatWorksForWhomSecon
- PeterFonagyDavi_2015_Summary_WhatWorksForWhomSecon