Abstract
I have tried to bring out some of the important methodological problems found in examining the effectiveness of drugs used in the control of druginduced parkinsonism by referring mainly to studies in which I have taken part. I hope I have shown that the whole topic is far less well understood than is often assumed. The main points may be summarized as follows: there is doubt as to whether many of the drugs used in controlling drug-induced parkinsonism are really effective; the results of many studies are conflicting; many studies contain serious flaws in design; methods for assessing extrapyramidal signs are not well developed; we are ignorant of the way in which drug-induced extrapyramidal signs change spontaneously. There is a clear need for further research in this area to improve techniques of assessment, to provide basic information on drug-induced syndromes, and to rigorously examine the efficacy of the drugs used in controlling them.