# Statistical Inference for Costs and Incremental Cost-Effectiveness Ratios with Censored Data

 dc.contributor Zhao, Hongwei dc.contributor Zhou, Lan dc.creator Chen, Shuai dc.date.accessioned 2014-09-16T07:28:19Z dc.date.accessioned 2017-04-07T20:00:04Z dc.date.available 2014-09-16T07:28:19Z dc.date.available 2017-04-07T20:00:04Z dc.date.created 2012-05 dc.date.issued 2012-07-16 dc.description.abstract Cost-effectiveness analysis is widely conducted in the economic evaluation of new treatment options. In many clinical and observational studies of costs, data are often censored. Censoring brings challenges to both medical cost estimation and cost-effectiveness analysis. Although methods have been proposed for estimating the mean costs with censored data, they are often derived from theory and it is not always easy to understand how these methods work. We provide an alternative method for estimating the mean cost more efficiently based on a replace-from-the-right algorithm, and show that this estimator is equivalent to an existing estimator based on the inverse probability weighting principle and semiparametric efficiency theory. Therefore, we provide an intuitive explanation to a theoretically derived mean cost estimator. In many applications, it is also important to estimate the survival function of costs. We propose a generalized redistribute-to-the right algorithm for estimating the survival function of costs with censored data, and show that it is equivalent to a simple weighted survival estimator of costs based on inverse probability weighting techniques. Motivated by this redistribute-to-the-right principle, we also develop a more efficient survival estimator for costs, which has the desirable property of being monotone, and more efficient, although not always consistent. We conduct simulation to compare our method with some existing survival estimators for costs, and find the bias seems quite small. Thus, it may be considered as a candidate for survival estimator for costs in a real setting when the censoring is heavy and cost history information is available. Finally, we consider one special situation in conducting cost-effectiveness analysis, when the terminating events for survival time and costs are different. Traditional methods for statistical inference cannot deal with such data. We propose a new method for deriving the confidence interval for the incremental cost-effectiveness ratio under this situation, based on counting process and the general theory for missing data process. The simulation studies show that our method performs very well for some practical settings. Our proposed method has a great potential of being applied to a real setting when different terminating events exist for survival time and costs. dc.identifier.uri http://hdl.handle.net/1969.1/ETD-TAMU-2012-05-10963 dc.language.iso en_US dc.subject Mean cost dc.subject Replace-from-the-right dc.subject Redistribute-to-the-right dc.subject Survival analysis dc.subject Survival estimator for costs dc.subject Incremental cost-effectiveness ratio dc.subject Confidence interval dc.title Statistical Inference for Costs and Incremental Cost-Effectiveness Ratios with Censored Data dc.type Thesis