When studying the effects of a non-pharmacologic intervention, the choice of a control group is often difficult. In a study on the effectiveness of increased water intake on voiding dysfunction in elderly men, an unusual design was tested. This article addresses the internal validity and ethics of this design. The RCT had a 6-month follow-up period and was carried out among 141 elderly men with moderate lower urinary tract symptoms. The experimental group was given the instruction to drink more water, the control group received placebo medication. The participants were not informed that there was a 50% chance of receiving placebo. The prior expectations and preferences for the two study groups, blinding and participants’ views on the ethics of the trial were assessed. Prior to randomisation, patients had higher expectations for the experimental intervention, but there was no statistically significant difference in their preference. During the study period, two out of 71 patients in the control group unmasked the placebo. In general, both groups fully agreed with the informed consent procedure.