Tracing expert thinking in clinical trial design

This paper describes a method for the analysis of the clinical trial design process used by experts. With this procedure, the scientific ideas and their sources can be identified and related to the clinical trial protocol actually prepared by the experts. An example is given using the work of the Intergroup Rhabdomyosarcoma Study Committee (IRS). That committee has been the primary contributor of information dealing with the treatment of rhabdomyosarcoma inchildren. The IRS-III protocol was used in this analysis of expert behavior because the protocol was adopted by the leading pediatric oncology clinical trial groups in North America and Europe. The analysis showed that the experts rely heavily, for much of the design, on ideas presented in numerical displays in published documents. Further, those aspects of the design which are innovative can be traced and better understood by applying the new procedure.