The excellent editorial by Morrison et al ^{Reference Morrison, Hutton, Shiers and Turkington1} strongly makes the case to shift current practice away from one in which service users are told that medication works and that they really must take it, towards one in which service users are presented with an accurate representation of the costs and benefits of antipsychotic medication and supported to make informed decisions about whether or not, for them, this is a option that appeals. This raises a very important challenge. How do we translate the information from this review of meta-analyses and double randomised controlled trials into something that will change the practice of front-line healthcare staff and be of direct use to service users? I am aware that despite similar conclusions being drawn with respect to antidepressant use for mild depression over 10 years ago ^{Reference Kirsch, Moore, Scoboria and Nicholls2} and even changes to National Institute for Health and Clinical Excellence guidelines about prescribing, ³ this has not led to a reduction in prescriptions of these drugs and I doubt very much they are now prescribed along with an accurate summary of exactly how much clinical benefit one can expect to see as a result of taking them. This is a plea that this excellent analysis is followed up by a strategy to ensure it has a direct impact on clinical practice as soon as possible.