Dr Hazell's review of the use of stimulant medication in attention deficit hyperactivity disorder (ADHD) is clear, succinct and provides up to date information about the use of stimulant medication in this group of children. This said, there are a number of issues in the article which deserve comment and consideration.

Firstly, we are currently in the throes of an epidemic of ADHD in Australia. In my own practice, I hardly see a child at the moment in whom the question of ADHD is not raised. There is clear evidence of this in Table 2 of Dr Hazell's paper which shows that from 1990-1994 there was an 8fold increase in the prescription rate of dexamphetamine, and the use of methylphenidate more than trebled. There is little doubt that prescribing practices in Australia in the past have, with some notable exceptions, been relatively conservative, approximating those in the U.K. rather than those in North America, where prescription of stimulant medication is reported to be extremely common in some areas. There appears to be no satisfactory explanation for this sudden increase in prescribing, although ADHD and the use of stimulant medication has certainly had widespread publicity in the media.

Secondly, the diagnosis of ADHD sounds relatively simple, although 'the validity of ADHD as a distinct syndrome remains controversial'. If a child is seen as displaying 6 of the inattention items and 6 of the hyperactivity/impulsivity items, then she, or more likely he, meets the criteria for the diagnosis of ADHD. The examination of the items (see box on page 61) reveals the difficulty of making the diagnosis in what is clearly a continuum disorder rather than a categorical one. For example, the decision as to what constitutes 'the child often does not seem to listen when spoken to directly' is clearly a subjective judgement, as is 'the child often talks excessively'. The same could be said of virtually every other item. A parallel can be drawn with decisions about the diagnosis and treatment of many continuum disorders in physical medicine as well. For example, there is clearly a normal distribution of blood pressure in the population and the decision at which levels one should begin to treat 'hypertension' should rely on good epidemiological data regarding the morbidity associated with certain levels of hypertension, tempered with clinical judgement about the characteristics and needs of an individual patient. In the case of hypertension, there is clearly a pathological subgroup whose hypertension is secondary to an identifiable underlying disorder and for whom the consequences of not treating the disorder are potentially fatal. Unfortunately, or perhaps fortunately, no such underlying cause has, as yet, been established for ADHD. Nevertheless, there is clearly a group of children who benefit greatly from the prescription of stimulant medication and this has been repeatedly demonstrated in carefully conducted clinical trials. The danger is that this proven efficacy in carefully selected research samples may lead to over prescription through generalisation of the findings to children in the community.

Thirdly, Dr Hazell correctly points out that the short term adverse effects are relatively few and not particularly serious, but there is little mention of the long term implications of prescribing stimulant medication. It is widely stated that the development of tolerance is unusual and that children with ADHD do not become addicted to these drugs. This merits careful consideration, particularly when prescribing for adolescents, given that dexamphetamine, or, as it is more affectionately known by its street name, 'speed', is a common drug of abuse among adolescents and shift workers, and methylphenidate also has street value. The nonpharmacological management of ADHD using behavioural or other psychotherapeutic approaches deserves more consideration than it receives. Optimal management of ADHD should not rely on medication alone. This raises the importance of skills in child and family psychiatric assessment and in behavioural management among the clinicians who treat children and adolescents with ADHD.

Finally, given that '70% of children show a clinically significant response and the placebo response in these trials is 35-40%', perhaps more stringent criteria should be used before embarking on what is frequently a long term medication, whose more subtle effects on cognitive processes are not known at this time. As set out in Table 1, the prescription of stimulant medication is already restricted to paediatricians, neurologists and psychiatrists in most States and Territories, which seems desirable, given some of the issues raised above. However, in addition, one might require that the medication had been shown to be effective in a single blind controlled trial of placebo and variable doses of medication using a rating scale such as the Conner's Parent and Teacher Scales to demonstrate a response to the medication. Although this is somewhat cumbersome and requires the cooperation of parents, teachers and even pharmacists, it may be worth the effort, both from the point of view of the child, who might otherwise be on medication for considerable periods of time unnecessarily, and in terms of the long term costs of such medication to the family and the taxpayer.