This paper presents some results of a Swedish survey to collect the opinions of experts on drugs and driving about various methodological issues. Its results will provide a basis for guidelines aiming to improve the quality of future research. The survey was especially concerned with the methodology of research to assess the effects of drugs on driving performance. 87 questionnaires were sent out, mostly to the USA and the UK, but also to Europe and Australia; of the 40 that were completed, 11 replies came from the UK, 19 from other European countries, 8 from the USA, and 2 from Australia. The questionnaire addressed 13 major topics, from three interdependent problem areas: (1) the power of the test; (2) interpretation of results; (3) generalisation of conclusions. A proposition was made about each topic, and experts were asked whether they agreed (most of them usually did), and how they would qualify their opinions. The 13 topics were: (1) age and gender of subjects; (2) how far healthy volunteers or patients should be used; (3) drug doses; (4) durations of drug treatments; (5) control conditions; (6) choice of performance tests; (7) reliability; (8) validity; (9) laboratory tests, simulators, and actual driving tests; (10) study purpose and hypothesis; (11) sample size; (12) power of statistical tests; (13) transfer of all test scores to a common scale.
Abstract