Analysis of complex decision-making processes in health care: cognitive approaches to health informatics.

Decision making by health care professionals is often complicated by the need to integrate ill-structured, uncertain, and potentially conflicting information from various sources. In this paper cognitive approaches to the study of decision making are presented within the context of a variety of complex health care applications. In recent years it has become increasingly accepted that in order to build information systems that can support complex decision making it will be necessary to more fully understand human decision-making processes. Methodological approaches are described that aim to explicate the decision making and reasoning skills of subjects as they perform activities involving the processing of complex information. The paper begins by presenting the theoretical foundations for cognitive analyses of decision making, including discussion of major approaches to the study of decision making in a range of real-world domains, including medicine. Applications of cognitive approaches are then illustrated, including a description of a study in which subjects were asked to "think aloud" in providing treatment decisions for complex medical cases. The resulting protocols were then analyzed for subjects' use of decision strategies and problems in reasoning. Extension of cognitive approaches to the study of group decision-making processes is also described. Recent approaches are discussed which borrow from advances in the study of human-computer interaction and which utilize video analysis of decision-making activities involving information technologies. Using these approaches it has been found that health care information systems, such as computerized patient record systems, may have inadvertent effects on human decision making. Implications of a cognitive approach to improving our understanding of complex decision making are discussed in the context of developing appropriate computer-based decision support for both individuals and groups.