Sixty-two women consulting their general practitioner with symptoms of urinary tract infection were assigned at random to either an experimental group, who were given a set of pre-standardized instructions both verbally and in written form, or a control group, who received the same advice in verbal form only.On follow-up, the written advice was found to increase the amount of information remembered by the patients both about the medication and other behavioural advice, but to have no obvious effect on compliance with the course of antibiotics prescribed.Further research is needed, investigating different diseases, to substantiate these findings and help to decide whether patients in general practice should be given written advice.