Decision making in diagnostic radiology

Radiology can be more interesting if the principles of decision making are understood and used to the advantage of the radiologist, referring clinician and patient. This article seeks to revise the principles that radiologists use intuitively. Particular emphasis is placed on the importance of the pre-test probability, as it can be shown to: (i) vary with the referral pool of patients; (ii) be a major determinant of the amount of additional information gained from the test; (iii) alter the sensitivity and specificity of the test; and (iv) determine largely the significance of a positive or negative test result. An investigation helps a clinician by providing information that will move the probability of disease above an action threshold or below its exclusion threshold. If this does not occur as a result of one study, another is selected. This article also describes the factors involved in such a decision tree analysis, as well as discussing the reasons for both selecting a particular examination and deciding when a test should not be performed.