What is a clinically important level of improvement in symptoms of attention-deficit/hyperactivity disorder?

Objective: To compare the desired and actual reduction in scores on a parent reported behaviour rating scale in a naturalistic sample of children and adolescents who had been treated with psychostimulant medication, referenced to global ratings of treatment benefit.
Method: Forty-five parents reporting poor global response to psychostimulant treatment, 44 reporting moderate response, and 49 reporting a high response retrospectively completed Conners rating scales describing their child prior to treatment, the child currently, and how the parent hoped the child would be following treatment.
Results: Percentage actual improvement in behaviour rating scales from baseline ranged from around 25% for the poor responders to above 50% for the high responders. Desired improvement was above 50%, with no significant difference between the groups on level of expectation.
Conclusions: Percentage cut points used to indicate clinical improvement reported in previous controlled trials of psychostimulant medication are probably too low, and could lead to an overestimate of treatment effect. Expectation of treatment benefit is unlikely to contribute to variation in treatment response.