Year: 2017
Type: dissertation
Abstract: Uncertainties in health utilities have received little attention in the literature. Health utility is a preference-based quality of life measure. It is commonly used to calculate quality-adjusted life years and to conduct cost-utility analysis. Under-estimation of the uncertainty of health utilities will lead to under-estimation of the uncertainty of the results of any economic evaluation that requires the use of health utilities. This thesis focuses on addressing uncertainty in health utilities in three areas: (1) the underestimation of uncertainties of health utilities derived from mapping algorithms and the derivation of methods to address it, (2) the challenge with prediction error in multi-attribute utility instrument valuation studies with small sample size and an emphasis of the use of shrinkage estimators as a potential solution, and (3) the underestimation of uncertainties of health utilities derived from multi-attribute utility instruments with an emphasis of the use of multiple imputation as a potential solution. For each issue, theoretical statistical explanations are provided and then followed by the proposed methods to address those issues. Simulation methods are also used to explore the usefulness of the proposed methods.
Source: TSpace (University of Toronto)
Author: Chan, Kelvin Kar-Wing
Institution: University of Toronto
Language: English
Subfield: Economics and Econometrics
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2774814066