I thank Katz et al for their insightful comments on this complex topic. I am familiar with their work but, as they suspected, unfortunately removed their 2004 reference from the final version of my editorial because of space restrictions imposed by the Journal. In fact, I reluctantly removed an entire section concerning the value of predicting improvement based on early response in certain psychopathological domains. Katz et al appear to be one of the few groups to examine the issue of differential response in various domains in sufficient detail (Reference Farabaugh, Mischoulon and FavaFarabaugh et al, 2005). As they recognise, the purpose of an editorial is not to provide an exhaustive review but a synopsis of studies of outstanding interest. Since submitting this editorial a year ago, colleagues and I have nearly completed a more thorough review of this topic, including the work of the San Antonio group and the parallel research that challenges the delayed onset of antipsychotics (Reference Agid, Kapur and ArenovichAgid et al 2003). I would very much welcome readers’ observations regarding the rapidity and measurement of onset of action of mood stabilisers and electroconvulsive therapy, which have received relatively little attention to date.