ABSTRACT
ABSTRACT The evaluation of a diagnostic test addresses three major questions: (a) how accurate is the test in diagnosis and/or prediction?; (b) what is the effect of the test on subsequent healthcare decisions and choices?; and (c) what is the effect of the test on patient outcomes? Among these three questions, the assessment of accuracy has received the most extensive attention in the literature. However, comparative effectiveness research calls for the comparison of the impact of different diagnostic tests on subsequent care and patient outcomes. This is a challenging task because the main function of tests is to provide information, which is subsequently incorporated into further diagnostic and therapeutic decision making. Thus, the impact of tests on patient outcomes is mediated by the intervening decisions about medical care. In this chapter, we discuss the methodologic challenges in the evaluation of the downstream consequences of tests, emphasizing the potential role of studies with observational data and simulation modeling.
