| Cost-effectiveness analysis (CEA) has emerged to be a popular tool in health care for allocating scarce societal resources. In this dissertation, I study three aspects of CEA using the net health benefits (NHB) framework suggested by Stinnett and Mullahy (1998) as the basis of conducting CEA. This framework also has its underpinnings in welfare economic theory.; In the first piece, I use a model based on a family utility function with altruistic linkages to show that there can be direct and indirect welfare effects to all family members. I study how the spillover effects to the spouse in a two-person family should affect choice of treatments for prostate cancer. I then analyze treatment choices of prostate cancer patients using the linked SEER-Medicare database to test the predictions. I conclude that cost-effectiveness analyses may better reflect the full costs and benefits of medical interventions if they incorporate these family effects.; In the second piece, I introduce the concept of expected value of symmetry of information (EVSOI) as the expected costs of heterogeneity that represents the potential value of research to elicit information on heterogeneous parameters such as QOL weights. Such value of information analyses may help to quantify the importance of conveying private information of the individual patients to the physician that can facilitate the choice of optimal treatments for individual patients.; The third piece sets up a model of the natural history of schizophrenia to implement the NHB framework. It evaluates the cost-effectiveness of various pharmacological treatment algorithms used to treat schizophrenic patients. It also sets up a model where the ideas from the first two essays can be applied in future work.*; *This dissertation is a compound document (contains both a paper copy and a CD as part of the dissertation). The CD requires the following system requirements: Adobe Acrobat; Microsoft Office. |
