Clinical decision rules in radiology.

Clinical decision rules to guide physicians in the appropriate use of diagnostic imaging studies need to be developed, in part, to control the overutilization of imaging studies. These rules need to be evidence based. Randomized control trials can be designed to assess the value of imaging studies and interventional procedures at levels five and six, patient outcome and societal efficacies, in Fryback and Thornbury's hierarchical model of efficacy. Results of these trials can be used to develop clinical decision rules. However, the efficacy of most diagnostic imaging studies can be assessed better at levels three and four, diagnostic thinking and therapeutic efficacies. Studies can be designed to determine clinical situations in which imaging studies will be of no value or to determine clinical criteria for the use of imaging studies. Systematic reviews of the literature and other techniques of analysis of evidence, such as decision trees, also can be used to develop evidence-based clinical decision rules for the use of diagnostic imaging procedures.