This review seeks to clarify and explicate an elusive concept: the appropriateness of diagnostic imaging. To ensure a common basis for discussion, several key components are articulated and defined. These include the diagnostic imaging procedure (DIP) itself, the subject (a patient), and the setting (a clinical scenario) in which the DIP is being considered. A review of the literature shows that appropriateness is a logical extension of empiric research, which has revealed substantial variation in the type and intensity of health services delivered to otherwise similar populations and communities in the United States. Against this background, the appropriate rate of a service in a population is transformed into appropriateness for an individual patient, which, when defined in terms of expected net health outcome, provides a conceptual link with the denominator of cost-effectiveness analysis. The complementary roles of clinical trials, technology assessment, decision-analytic modeling, and consensus methods in estimating appropriateness are compared and contrasted.