These pages use javascript to create fly outs and drop down navigation elements.

2008 HSR&D National Meeting –  Implementation Across the Nation: From Bedside and Clinic to Community and Home

Printable View
National Meeting 2008

2016 — Statistical Methods for the Analysis of Health Care Costs

Zhou XA (Seattle VA Medical Center)

Workshop Objectives:
To assure that health services researchers in VA HSR&D adequately prepare and plan for their studies, in this workshop we will discuss statistical methods for appropriate analyses of health care costs. Specifically, the purpose of this workshop is to discuss the special problems inherent in studies on health care costs, including skewed and heteroscedastic nature of the distribution of health care costs as well as the occurrence of zero cost values. Without taking these special features into consideration, a statistical method may lead to the wrong conclusion. Some newly developed statistical methods that are appropriate for handling these features of health care costs will be discussed.

Activities:
We will present important statistical methods for handling some common statistical issues in health services research, and describe available software with illustrations on how to implement these methods. Case studies will be presented that illustrate how these methods are employed in practice. The strengths and limitations of potential approaches will also be described, enabling participants to intelligently apply these techniques to their own studies.

Target Audience:
The workshop is intended for health services researchers and statisticians who are interested in understanding the basic issues in the analysis of health care costs. Those interested in understanding how more complex statistical methods are used in implementing predicting health care costs will also benefit from workshop participation.

Assumed Audience Familiarity with Topic:
The workshop is designed for the broad health services research audience with low to moderate levels of statistical understanding required.