The 130th Annual Meeting of APHA

Abstract #38708

Jeffrey S. Hoch, PhD, Department of Epidemiology and Biostatistics, University of Western Ontario, UWO School of Medicine and Dentistry, Kresge Building, London, ON N6A 5C1, Canada, (519) 661-2111 x86270, jeffhoch@biostats.uwo.ca and Jeffrey D. Blume, PhD, Center for Statistical Sciences, Brown University, 167 Angell Street, Room 009, Brown University, Box G-H, Providence, RI 02912.

The goal of economic evaluation (e.g., cost-effectiveness analysis) is to provide an estimate of the relative trade-off between additional costs and outcomes. In essence, the exercise is a search for evidence about a novel technology or treatment's value for money in relation to an appropriate alternative. Recent work has considered this search for evidence in the framework of a net benefits regression approach. With this regression approach, the focus of an economic evaluation involves the treatment indicator's coefficient (used to estimate incremental net benefits) and p/2 its associated one-sided p-value (used to test the null hypothesis that treatment is not cost-effective). This frequentist analysis can be supplemented by using the quantity 1 - p/2 to construct the y-axis of the Bayesian cost-effectiveness acceptability curve. In this paper we briefly introduce the likelihood paradigm and discuss how it can be used to measure statistical evidence of cost-effectiveness. Likelihood methods have many advantages (e.g., a low probability of observing strong misleading evidence, no penalty for repeatedly reexamining one's data, and no reliance on a prior to arrive at a conclusion about the evidence of a treatment's cost-effectiveness). These advantages are explored using data from an economic evaluation of a Program in Assertive Community Treatment (PACT).

Learning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to

• Describe how to measure evidence of cost-effectiveness.
• Apply likelihood methods to economic evaluations using the net benefit regression approach.
• Discuss the advantages of using the likelihood methods for economic evaluation.

Keywords: Economic Analysis, Statistics

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: none
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.