The following material was contributed by Jenny Lyon (IPS)


The clinical and cost effectiveness of methylphenidate for hyperactivity in childhood: Version 2 - for consultation

Authors J Lord and S Paisley

Correspondence to Appraisals Group, NICE
90 Long Acre
London WC2E 9RZ

Date completed July 2000

Expiry date July 2003

Please note that:

· the following summary has been produced by IPS and not by NICE

· NICE's responsibility is for England and Wales and not Scotland

· Section 4 (Economic Evidence) has not been included in this summary

Diagnostic criteria
The report contains the DSM IV and ICD 10 diagnostic criteria for Attention Deficit / Hyperactivity Disorder (AD/HD) and Hyperkinetic Disorder (HD) respectively, and notes that HD criteria are more stringent on two accounts:

AD/HD only requires significant signs* in either attention or hyperactivity/ impulsiveness for a diagnosis, whereas HD requires signs in both. As a result, a child can be diagnosed as AD/HD predominantly hyperactive / impulsive, AD/HD predominantly inattentive or AD/HD combined;

AD/HD only requires that signs do not occur exclusively during the course of PDD or psychotic disorders and that they cannot be better accounted for by other diagnoses such as anxiety or depression : ICD excludes a diagnosis if a child has a PDD.

Diagnostic Validity
A number of factors make a diagnosis of HD or AD/HD problematic:
· hyperactivity is normally distributed and the cut-off point between normal and abnormal children is essentially arbitrary, but there is evidence that the presence of signs of inattention, impulsiveness and hyperactivity, does define a distinct group of children;

· the aetiology of AD/HD is still unclear. There is increasing evidence that it is a brain disorder with a genetic component although further research is needed to establish this. Research suggests a range of risk factors including psycho-social along with genetic factors;

· co-morbid disorders are common including ODD CD, learning disorders, anxiety and depression and tic disorders and Tourette's Syndrome;

· there is debate about whether patterns of hyperactive/impulsive/inattentive behaviour are best characterised as a homogenous disorder (as suggested by a HD diagnosis) or rather as two or more problem complexes (as in the subtypes of AD/HD).

Estimates vary widely within and between countries (from 1.0% to 17.8%). Differences are probably due to:

· different diagnostic criteria;
· different methods of assessment;
· cultural differences in the interpretation of behaviour;
· true variation due to differences in underlying risk factors

* "signs" is used rather than symptoms because symptoms refer to subjective reporting and "signs" to external observations.

· ODD: 30%
· CD: 28%
· anxiety: 26%
· depression: 18%
· learning disabilities 12%

Natural History
Prospective studies of school-age children have shown persistence of symptoms at least into early adolescence. Research in this area is problematic because of changing diagnostic criteria and differences in study design. However, most patients (70% to 80%) continue to show symptoms into adolescence and to meet the diagnostic criteria. As adults 60% of patients still show symptoms though fewer meet diagnostic criteria.
HD and AD/HD have been linked to increased risk of poor academic achievement, emotional and social problems, unemployment, criminality and substance abuse in later life.

Guidelines and Management
All 6 identified published guidelines, world-wide, acknowledge the following problems re reaching consensus on the management of AD/HD:

· differing diagnostic criteria;
· under and over diagnosis leading to possible over-prescription of medication;
· different licensing regulations for medication.

Taking the above into account, assessment should include;

· asking the parents about the child's behaviour;
· direct observation in the consulting room
· rating scales (never as the sole basis of diagnosis)
· if appropriate, physical examination tests to rule out other possible causes of hyperactivity or inattention such as hearing loss, epilepsy, thyroid disorders, fragile x or side effect of drug treatments
· assess the child more than once
· note that assessment parameters might differ for different age groups

Non-Drug Therapy
· behavioural approaches for child and sometimes for parents and/or whole family
· educational interventions
· dietary interventions - often difficult to implement and there is no evidence linking diet with hyperactivity. Such interventions should not be used routinely.

Drug Therapy
· pharmacological interventions generally recommended only as an adjunct to non-drug interventions
· medication remains controversial largely because of concerns over long-term safety and the risk of abuse
· opponents argue that medication is sometimes used for the benefit of parents and teachers

methylphenidate (Ritalin and Equasym)
· should not be used for all cases of AD/HD
· only be used after a detailed history and evaluation
· use depends on severity of symptoms and appropriateness of child's age
· not indicated for use in children less than 6 years of age
· should usually be discontinued during or after puberty

Dexamphetamine (Dexedrine)
· contraindications and side effect similar to MPH

· available in England only on a named patient basis due to concerns over liver toxicity

Tricyclic antidepressants (amitriptyline, imipramine and nortriptyline)
· licensed for use in nocturnal enuresis in children
· sometimes prescribed with AD/HD where anxiety is also present

Other drugs
· a number of other drugs are occasionally used, including Clonodine, monoamine-oxidase inhibitors and selective serotonin-reuptake inhibitors and beta-blockers. Concerns have been expressed about the lack of safety and efficacy data to support the use of these drugs in children

Outcome Measures
There are 3 main methods of measuring outcome: direct observation; psychometric testing; rating scales and questionnaires.

A review of rating scales found that the Conners' Rating Scale contained scales - particularly the teacher scale and the hyperactivity index of the 39 item scale (CTRS-39-HI), that were highly effective at discriminating between referred AD/HD children and normal controls*.

*there is no direct reference in the report re the differential validity of questionnaires and rating scales, i.e. can they discriminate between children with AD/HD and children with other areas of dysfunction?

Also good evidence for the validity and reliability for the Attention Deficit Disorders Evaluation Scale (ADDES), the Behaviour Assessment System for children (BASC) and some versions of the Child Behaviour Checklists (CBCL).

Clinical Effectiveness

Is MPH more effective than no treatment (placebo)?
There is evidence from meta-analysis of placebo-controlled studies that MPH is efficacious at improving core ADHD symptoms, at least in the short term while children continue to take medication. There is some evidence of improvements across other outcome dimensions.

Is MPH more or less effective than Dexamphetamine?
There are few randomised comparisons. The existing evidence is of relatively poor quality and gives inconsistent results.

Is MPH more or less effective than tricyclic antidepressants?
There is insufficient evidence to judge.

Is MPH more or less effective than non-drug treatments?
There is little evidence of the relative effectiveness of MPH to behavioural interventions. The studies that do exist are of relatively poor quality, but suggest that MPH is more effective over the medium and short term.

Do non-drug interventions add to the effectiveness of MPH?
There is insufficient evidence to support the superiority of MPH combined with a behavioural intervention over MPH alone. The evidence is of relatively poor quality and most comparisons fail to detect any significant outcome, although some findings in favour of combined treatment have been reported.

Does MPH add to the effectiveness of non-drug interventions?
There is evidence, some of relatively good quality, which suggests that the addition of MPH to behavioural treatment is beneficial. Improvements in short and medium outcomes were observed across a number of dimensions.

©2000 IPS
updated 3dec00 1600 rl