Professor Jonathan Green, Professor of Child and Adolescent Psychiatry, University of Manchester comments on a new Cochrane review of evidence for the use of medicines for attention deficit hyperactivity disorder (ADHD):
“It would be easy to draw the wrong overall conclusions from this Cochrane review.
Despite the fact that generally speaking the studies find a positive effect of methylphenidate, the Cochrane reviewers define the trials as having a ‘high risk’ of bias - and this leads them to say that the evidence is weak.
The basic issue that this review identifies is the poor design of studies to date in this field - specifically that the outcome measures, mainly parent and teacher report behaviour, may not have been made completely blind to whether the child was having the active medication or placebo. This leads to the Cochrane review of 'high risk' of bias.
This problem is intrinsic to many psychological treatment studies since it is difficult to find ‘objective’ measures of the disorder in question outside parent and teacher report. However until recently clinical scientists and trialists themselves in this field have been insufficiently focused on getting this aspect of trial design right (this has partly been a matter of culture within clinical trials in this field, partly that there are real challenges in finding appropriate objective measures).
My own view is that clinical scientists in this area should put more effort into finding valid objective measures of ADHD symptoms for use in trials, rather than concentrating on the use of so-called 'nocebos', as suggested in the report
It would therefore be wrong to draw the conclusion from this review that methylphenidate is ineffective. In fact clinical level evidence strongly supports the effectiveness of methylphenidate for many children with ADHD, and this is supported in the trial evidence (albeit with their design weaknesses as above).
The evidence for the positive effect of methylphenidate on ADHD symptoms is for instance much stronger than for psychological treatment.
My own view is that clinical scientists in this area should put more effort into finding valid objective measures of ADHD symptoms for use in trials, rather than concentrating on the use of so-called 'nocebos', as suggested in the report – a use that I think will be fraught with difficulty; partly because the incidence of unwanted-effects from methylphenidate is so variable and unpredictable.”
Read the full review, ‘Methylphenidate for children and adolescents with attention deficit hyperactivity disorder (ADHD)’ here.