In radiographic evaluation of the results of endodontic therapy the development or persistence of periapical radiolucencies often serve as criterion for therapeutic failure. However, using three endodontists and three oral radiologists as observers the present study found consensus only in 27% of cases classified as having periapical radiolucencies. By applying signal-detection theory to periapical radiographic diagnosis, the variations among observers were explained by their adoption of different criteria of periapical disease resulting in different positions on a receiver-operating-characteristic (ROC) curve. The potential of reducing the interexaminer variation through two different calibration programs was investigated. Even though both methods resulted in increased examiner agreement the benefits seemed limited. It was argued that this is due to the complex structure of the decision making process. Among a group of well-experienced general practitioners substantial variations in attitudes to treatment of periapical lesions in endodontically treated teeth was demonstrated. From decision theory and endodontic teaching paradigms a hypothesis was evolved explaining the variations in treatment with variations in subjective probabilities of disease and expected complications. However, this hypothesis could not be confirmed in groups of endodontists and general practitioners. Instead the decision makers seemed to rely on a limited number of heuristic principles. From a normative point of view clinical management of periapical lesions in endodontically treated teeth was approached by using formal decision analysis. Temporal and logical display of decision alternatives, values of probabilities and utilities of the different outcomes are the features of such an analysis. In the concluding part of the study this method was used to calculate optimal decision strategy for an endodontic recall program. It was demonstrated that patients should be examined one year after endodontic treatment. Patients with signs of periapical disease are reexamined after a further three years. At this moment a mild suspicion of periapical pathosis will lead to retreatment of a tooth.