Confidence Intervals for Median Cost Estimation with Censored Data

• Activity Number: 230
• Type: Topic Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Hongwei Zhao*+ and Chen Zuo and Heejung Bang
• Companies: Texas A&M Health Science Center and Renmin University of China and Weill Cornell Medical College
• Keywords: Cost analysis ; nonparametric inference ; quantile estimation ; survival analysis

Abstract

Medical cost estimation has been performed increasingly to evaluate economic impact associated with a new procedure. While the mean cost is usually the preferred measure since the total cost can be derived from it, the median value of the cost distribution can also be of interest since it reflects a central or typical cost value, and is less influenced by the extreme values of costs. Many cost data are collected along with (prospective) clinical trials nowadays, as a result, the cost data are often censored. In this talk, we will present methods for constructing confidence intervals for the median cost from one sample, and the ratio and difference of the medians from two samples. Simulation results will be presented together with an example from a clinical trial of cardiovascular disease.