ANALYSING DECISION UNCERTAINTY TO INFORM FUNDING AND RESEARCH DECISIONS

Overview: This course will teach participants how to undertake uncertainty analysis around the cost-effectiveness of health technologies. Participants will be taught how decision uncertainty can be explored in a model based cost-effectiveness analysis using deterministic and probabilistic sensitivity analysis and value of information analysis. Participants will also learn how uncertainty analysis informs decisions about whether to reimburse a health technology and whether further research to support the use of a technology is worthwhile.

Background: This course will teach participants how to undertake: (i) deterministic and probabilistic sensitivity analysis; and (ii) value of information analysis, including expected value of perfect information, expected value of partial perfect information, and expected value of sample information. Participants will also learn (iii) how to present the results of sensitivity analyses, including cost-effectiveness acceptability curves and (iv) how uncertainty analysis can be used to inform reimbursement and research decisions for health technologies. Participants will undertake and present the results of deterministic sensitivity analysis using an Excel based Markov model. They will learn how to make that model probabilistic and learn how to present the results of probabilistic sensitivity analysis. Participants will be taught how the cost of uncertainty can be demonstrated using expected value of information analysis, and will see how value of information analysis can be used to support a case for further research.

Format Requirements: The format for the course is a series of lectures interspersed with guided practical exercises. Participants will require the use of a laptop installed with Microsoft Excel. This course assumes that participants will have experience in conducting cost-effectiveness analysis and using Markov models.

Description and Objectives: This course will focus on how, analytically, to explore decision uncertainty and its consequences. This course aims to explain the analytic methods that underlie explorations of decision uncertainty, including cost-effectiveness acceptability curves, cost-effectiveness planes and value of information analysis. The course begins with a description of multi- or uni-variate sensitivity analyses and their limitations. The main focus will be on implementing probabilistic sensitivity analyses and assessing the consequences of uncertainty using value of further research analyses. There will be a practical component to this course, where the methods outlined will be put in practice. Practical and policy relevant examples will be provided that show how the types of analyses described in the course have been used in a real world setting.

Objectives:

• Participants will learn how and why uncertainty can affect reimbursement decisions for health technologies. 
• They will learn how to undertake deterministic sensitivity analysis and how to make a model probabilistic using an Excel based Markov model, and how to present the results of deterministic and probabilistic sensitivity analysis. 
• Participants will learn how the cost of uncertainty can be demonstrated using expected value of information analysis, and will see how value of information analysis can be used to support a case for further research.