There has long been considerable individual variation in the regimes for managing acutely disturbed patients, in terms of the drugs used, routes of administration and the dosages given. We all tend to learn from ‘accepted’ practice in the units in which we train, and the lack of good-quality controlled trials, published in leading and accessible journals, compounds the problem. Given the importance of this issue, when concerns about sudden death in the context of high dose regimes are generating considerable publicity (for example, MIND is calling for a blanket ban on dosages greater than those given in the BNF), the need for rigorous research has never been greater. Guidelines have, however, become established in a number of units, and this clear summary by Macpherson et al is therefore to be welcomed.