Abstract
Lahey et al. [1] commented that ‘no term in the history of childhood psychopathology has been subject to as many reconceptualizations, redefinitions, and renamings as the disorder referred to non-technically as hyperactivity’. Most recently, the DSM-IV has described three subtypes; Predominantly Inattentive, Predominantly Hyperactive/Impulsive; and Combined types of Attention Deficit Hyperactivity Disorder (ADHD). According to Hill and Taylor [2], the DSM-IV identifies about 4% of primary school age children, which is four to five times as prevalent as the ICD-10 condition of hyperkinetic disorder, which has essentially the same symptom profile, but requires more stringent criteria. Although Levy et al. [3] showed that the unitary DSM-III-R ADHD could be considered as a highly heritable continuum, recent investigations suggest that while the three DSM-IV subtypes show considerable genetic influences, different or additional genes may operate at the extremes of behaviour. The DSM-IV subtypes also differ in their comorbidities. Waldman et al. [4] utilized Australian Twin Study (ATAP) data to examine the causes of overlap and ADHD, Oppositional Defiant Disorder (ODD) and Conduct Disorder. They found considerable overlap in genetic and environmental influences in these three disorders. At a genetic level, ADHD and ODD were closely related (see MTA discussion below). Twin studies [5–7] have also shown that language and reading problems are differentially related to ADHD subtypes. Thus, instruments based on the DSM-IV (National Health and Medical Research Council (NH & MRC) report, 1997) should provide the most useful diagnosis of ADHD and related disorders.
The use of ‘objective’ tests such as the Continuous Performance test (CPT) to diagnose ADHD, remains controversial. This is partly because the tests have varied considerably in different laboratories, in terms of length, interstimulous intervals and norms. Most recently Conners [8] and colleagues reported a review of 20 studies that had an experimental and a control group, rank ordering them separately in terms of effect size of omission and commission errors. They found evidence of ceiling effects in those tasks, which produced a small number of errors. They have subsequently modified the Conners CPT to use 75% targets and 25% non-targets to produce a larger number of errors, and have reported specificity and sensitivity greater than 0.80.
Treatment
While the use of central nervous stimulant medications such as methylphenidate and dexamphetamine is accepted as the primary treatment for ADHD (NH & MRC Report on ADHD, 1997) questions remain about differing prevalence of medication use in the US versus UK and in different regions [9] in the US and Australia. Also, increased use of medication over the last decade [10] has raised concerns.
Jensen et al. [9] examined data obtained from 1285 children and their parents across four US communities. Analyses examined the frequency of ADHD diagnosis, the extent to which medications were prescribed, as well as the provision of other services (e.g. psychosocial treatments and school-based educational interventions). While 5.1% of children met full DSM-III-R ADHD criteria during the previous 12 months, only 12.5% of those meeting criteria were treated with stimulants during the previous 12 months. Some children who had been prescribed stimulants did not meet full ADHD criteria (8 of 16) but they manifested high levels of symptomatology. Jensen et al. concluded that while substantial over-treatment was not occurring, the relative paucity of school-based and/or psychotherapeutic services was notable. Trials in stimulant prescribing in NSW have recently been outlined in a NSW public health bulletin supplement [10].
Another issue of concern has been the trend in prescribing psychotropic medications to preschoolers. Zito and colleagues [11] reported a dramatic increase between 1991 and 1995. Prescription records revealed a 1.7–3.1% increase in methylphenidate prevalence in 2–4 year olds, with sizeable elevations for clonidine and antidepressants. Unresolved questions involved the long-term safety of psychotropics, particularly in light of earlier ages of initiation and longer durations of treatment, with possible impact on neurotransmitters.
The largest and most influential treatment study of ADHD was reported in 1999 by the MTA Cooperative Group (Multimodal Treatment Study of Children with ADHD) [12]. The multisite study assigned 579 children (aged 7–9.9 years) with ADHD combined type, to 14 months of medication (titration followed by monthly visits), intensive behavioural treatment (parent, school and child components), the above two combined and standard community care. For ADHD symptoms careful medication management was superior to behavioural treatment and to routine community care. Combined treatment did not yield greater benefits than medication for core ADHD symptoms, but may have provided modest advantages for non-ADHD symptoms (parentreported internalizing symptoms, oppositional/aggressive symptoms and Wechsler Individual Achievement Test reading achievement score).
While the study is of great interest, several caveats remain. First, Hyperactive/Impulsive and Inattentive subtypes were not included. Second, treatment did not include separate learning and reading components, other than those routinely carried out by schools. Thus, the study did not investigate treatment implications for the important Predominantly Inattentive subtype, the group most implicated in the increased use of stimulants in recent decades.
Another interesting finding is the beneficial effects of medication on Aggression – ODD symptoms with a reduction in mean score of 1.39 (SD = 0.92) to 0.65 (SD = 0.68). Levy et al. (in press) have suggested that the genetic overlap between ADHD and ODD reported by Waldman et al. [4] may, in part, account for this finding.
Hill and Taylor [2] have outlined an ‘auditable’ protocol for treating ADHD, which attends to comorbidity and side-effects. Their parent and teacher side-effect questionnaires appear on the Archives of Disease in Childhood website. More recently the American Academy of Child and Adolescent Psychiatry has published detailed practice parameters for the use of stimulant medications in the treatment of children, adolescents, and adults [13] using an evidence-based approach, detailed literature review and expert consultation.
A further issue relates to the use of slow-release stimulant medication. Slow-release methylphenidate and dexamphetamine preparations are freely available in the USA, but only prepared by a few individual pharmacies in New South Wales. The advantages are less need for administration during the school day, and more even coverage of symptoms, though clinical experience indicates a need for immediate release boosters. A possible disadvantage may be less sensitive control of the dose level. What of the future? Winsberg and Commings [14] found that homozygosity of the 10 – repeat allele of the dopamine transporter (DAT-1) gene was characteristic of a non-response to methylphenidate. This is the first pharmacogenomic study in child psychiatry, and may predict future approaches to diagnosis and treatment.