Abstract

A questionnaire describing six clinical scenarios was mailed to urologists (in Britain, Canada, and the United States) and to radiation and medical oncologists in the United States, who practice genito-urinary (GU) oncology. In each scenario, the surgeon or physician was asked to consider himself as a patient with bladder, prostate, or kidney cancer, and to select his own treatment. Accompanying each scenario were one or two clinical trials for which the physician would be eligible. He was asked to state if he would agree to be randomized in each trial, and if he refused, to state his reasons. We found that (1) there were major differences of opinion about management for each scenario; (2) choice of treatment was influenced more by specialty training or geographic location than by the results of previous clinical trials (which are available to all); (3) British urologists tended to be less aggressive than their North American colleagues, with respect to the use of radical surgery and chemotherapy; (4) acceptance of clinical trials ranged from 3% to 60%; and (5) agreement to clinical trials was quite poor even when they were designed to compare the most popular options for management. This physician surrogate method is a valuable tool in assessment of the degree of consensus amongst expert physicians and in the determination of whether new clinical trials address important areas of controversy.