JSM 2004 - Toronto

Abstract #301775

This is the preliminary program for the 2004 Joint Statistical Meetings in Toronto, Canada. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2004); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2004 Program page
Activity Number: 409
Type: Contributed
Date/Time: Thursday, August 12, 2004 : 8:30 AM to 10:20 AM
Sponsor: Section on Health Policy Statistics
Abstract - #301775
Title: A Comparison of Methods Used for Constructing Confidence Intervals for Cost-effectiveness Ratio
Author(s): Ming-Yu Fan*+ and Xiao-Hua A. Zhou
Companies: University of Washington and University of Washington
Address: 1959 NE Pacific St., Seattle, WA, 98195-6490,
Keywords: cost-effectiveness ; simulation ; bootstrap
Abstract:

The incremental cost-effectiveness ratio (ICER) has been well accepted as a measure for cost-effectiveness in health sciences. Due to the skewed distribution nature for costs and estimated ICER, however, constructing an appropriate confidence interval for ICER is a challenge. Many methods have been proposed, yet systematic comparisons are rarely found in the literature. Also, there has been a conflict recommendation on which method is better. Polsky et al. (1997) and Briggs et al. (1999) have performed simulation studies to compare four and eight methods, respectively. However, their simulations only considered a selected number of available approaches and a very limited type of distributions for cost and effectiveness outcomes. We have conducted an extensive simulation to compare most commonly used as well as some new methods by generating different types of skewed samples. The methods are also compared across wide range of parameters to incorporate the variation observed in real data. The result suggests that bootstrap-t method is consistently better than other bootstrap or normal-theory-based methods for non-normal data. For normal data, all methods yield desirable result.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2004 program

JSM 2004 For information, contact jsm@amstat.org or phone (888) 231-3473. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2004