Methylphenidate for ADHD among children and adolescents: a meta-analysis

This meta-analysis on the efficacy and side-effects of methylphenidate to treat persons who have been diagnosed with ADHD expresses serious doubts about the quality of the scientific evidence for its efficacy. There is an increased risk of non-serious side-effects, which lead to deblinding, but the studies were not designed to assess differences in side-effects. A striking finding is that only a fraction of the studies of methylphenidate that have been conducted were deemed eligible for inclusion in the meta-analysis, which means that the majority of the studies are of poor to very poor quality. The average duration of the studies was 75 days, whereas in practice children are treated for longer periods, and the meta-analysis does not offer any data on long-term side-effects.
In other respects too, the authors’ conclusions are in line with the conclusions presented in Gebu 2012; 46: 121-129. There is a need for large-scale studies into the effect of non-medical interventions to treat ADHD. And there is an urgent need for well-designed randomised double-blind studies into the efficacy of methylphenidate. The high risk of deblinding in research into psychopharmaceuticals constitutes a problem in the interpretation of the findings. The authors’ proposal to improve the quality of the randomised double-blind study design by using active placebos (a.k.a. nocebos) to minimise the risk of deblinding of participants and observers is an interesting one and deserves to be implemented. A nocebo causes the same perceptible side-effects, while being expected not to have the intended favourable effect. A comparable study using nocebos with atropine in the treatment of depressive disorder found no difference between the effects of an antidepressant and those of placebo (Gebu 2002; 36: 51-59).
In view of the serious doubts about the efficacy of methylphenidate for the treatment of behaviours labelled ADHD, non-medical treatment should be preferred.
Once more, we must conclude that it is difficult to set up well-designed and sound scientific research into medications in paediatric psychiatry. Could this be related to the problem of the ill-defined boundary between normal and abnormal behaviour, which moreover is variable and age-related?


  1. Diagnostic and Statistical Manual of mental disorders. Washington DC: American Psychiatric Association, 2014.
  2. Storebø OJ, et al. Methylphenidate for children and adolescents with attention deficit hyperactivity disorder (ADHD). Cochrane Database Syst Rev 2015: CD009885.

*The literature refers to the Dutch text