| The most difficult and perplexing decision a person faces often involves his medical care. In such decisions, the best course of action cannot be clearly seen, and these decisions can be described as "opaque." Decision analysis is a technique that can make opaque decisions transparent. Unfortunately, decision analysis is too expensive for most patients and requires that patient preferences for medical outcomes be quantified. The objective of this dissertation is to increase the transparency of medical decisions by developing tools that make decision analysis feasible.;In this dissertation, three causes of opacity in medical decision making have been identified: the multiplicity of decision participants, patients' unfamiliarity with medical outcomes, and the lack of believability in automated analyses. While methods are proposed to reduce the opacity created by each of the three causes, the focus of the dissertation is on helping patients become more familiar with medical outcomes in order to ease the task of preference quantification.;This dissertation includes several concrete and specific contributions to the field of medical decision making. Three major contributions center on making patients more familiar with medical outcomes. First, a technique called outcome mapping has been developed which allows medical outcomes described physiologically to be cast in terms of their impacts on an individual patient's lifestyle. Second, outcome mapping has been implemented in a computer program called MAPPER, thus demonstrating the feasibility of the approach. Third, a study has been performed which indicates that the use outcome mapping affects assessed preferences for medical outcomes, increases patients' confidence in those assessments, and makes preference assessments easier.;Other contributions include a rule to identify the decision maker, and the implementation of an intelligent decision system that increases the believability of automated analyses by enriching the exchange of information between the user and the system.;By identifying the decision maker, by helping that decision maker become more familar with the outcomes he faces, and by providing believable automated decision analysis, it is possible to increase the transparency of medical decisions and gain the insight necessary to act in the face of difficult medical decisions. |
