Session Details

PM5: Value of Information Analysis Using Linear Regression Metamodeling

(Event: 17th Biennial European Conference (ESMDM) - Leiden, The Netherlands)

Jun 10, 2018 2:00PM - Jun 10, 2018 5:30PM

Session Type: Short Course- PM 1/2 Day

Description

Background

Value of information analysis (VOI) is a key concept in decision analyses, because it informs sensitivity analysis, study design and resource allocation for further research. However, traditional methods of conducting VOI are often computationally demanding, particularly when computing the expected value of partial perfect information (EVPPI) and the expected value of sample information (EVSI). In this course, participants will learn how to compute various measures of VOI using a single dataset of probabilistic sensitivity analysis and an efficient regression metamodeling approach to compute EVPPI and EVSI using a Gaussian approximation approach in R.

Course Type

Half Day

Course Level

Advanced

Format Requirements

This course consists of lectures explaining the theory of VOI analysis interspaced with "hands-on" experience calculating various measures of value of information (VOI) analyses using linear regression metamodeling. Participants will work through structured examples using their own computers. Data sets and files needed for the course will be distributed during the course session. A basic level of experience with probabilistic sensitivity analysis and regression are required.

Overview

Value of information (VOI) quantifies the opportunity loss associated with choosing a suboptimal intervention based on current imperfect knowledge. In this course, participants will learn the theory and application of VOI to quantify the value of perfect, partial and sample information using a dataset of probabilistic sensitivity analysis and regression metamodeling.

Description & Objectives

The purpose of this course is to familiarize course participants with the concept and application of VOI using linear regression metamodeling and a Gaussian approximation approach. By the end of the course, the participants will gain hands-on-experience calculating various measures of VOI using an efficient approach, including:

Expected value of perfect information (EVPI): The value of eliminating all sources of parameter uncertainties in a model.
Expected value of partial perfect information (EVPPI): The value of eliminating uncertainty for one or a set of parameters accounting for possible correlation using linear regression metamodeling.
Expected value of partial sample information (EVSI): The value of collecting additional information on one or a set of parameters accounting for possible correlation for a given n using linear regression metamodeling and a Gaussian approximation approach.
Expected net benefit of sampling (ENBS): The benefit of collecting additional information after accounting for the cost of research for a range of sample sizes and alternative research study designs.
Optimal sample size (n*): The maximum number of patients in a research study design that provides the highest ENBS.

In addition, we will provide the R code to produce publication quality figures and tables for these measures.

Course Director

Fernando Alarid-Escudero
- Division of Health Policy and Management, University of Minnesota

Course Faculty

Hawre Jalal
- Department of Veterans Affairs

Session Fees

	Fee Type	Member Fee	Non-Member Fee
This session is free
	Early:	$158.00	$216.00
	Regular:	$158.00	$216.00
	Late:	$173.00	$241.00
This session is free
	Early:	$80.00	$80.00
	Regular:	$80.00	$80.00
	Late:	$91.00	$91.00