July 19, 2022 - July 20, 2022
Back to all short courses
Multiple Imputation Methods for Addressing Missing Data in Health Economic Evaluation (Virtual)
LEVEL: Introductory
TRACK: Economic Evaluation
LENGTH: 4 Hours | Course runs 2 consecutive days, 2 hours each day
15:00PM–17:00PM British Summer Time (BST)
10:00AM–12:00PM Eastern Daylight Time (EDT)
14:00PM–16:00PM Coordinated Universal Time (UTC)
16:00PM–18:00PM Central European Summer Time (CEST)
Wednesday, 20 July 2022 | Course runs 2 consecutive days, 2 hours per day
15:00PM–17:00PM British Summer Time (BST)
10:00AM–12:00PM Eastern Daylight Time (EDT)
14:00PM–16:00PM Coordinated Universal Time (UTC)
16:00PM–18:00PM Central European Summer Time (CEST)
Click for time zone conversion
DESCRIPTION
Cost‐effectiveness analyses (CEA) conducted alongside both randomized and non-randomized studies are routinely used for informing healthcare decision making. However, these studies rarely succeed in collecting all the intended information about resource use, health outcomes and covariates. For example, patients are often lost to follow-up or fail to complete the quality of life or resource use questionnaires. While multiple imputation (MI) approaches for handling the missing data in CEA are increasingly available, recent reviews suggest that most published CEA handle missing data inadequately. Failing to appropriately address the missing data will lead to biased cost-effectiveness results, and potentially incorrect resource allocation decisions.
This course offers an in-depth description of MI methods for addressing the missing data in CEA. The course starts by providing an overview of the typical missing data mechanisms in CEA and the key principles of MI. The course will then describe more advanced approaches for appropriately handling complex distributions of cost-effectiveness endpoints, clustering data, and data that are missing not at random (MNAR). We will illustrate the MI methods in practice through a series of trial-based CEA examples, using the Stata software. Participants who wish to have hands-on experience must bring their personal laptops with Stata software installed. Familiarity with health economic evaluation is desirable, but the course assumes little or no familiarity with missing data or MI methods.
Manuel Gomes, PhD
Associate Professor of Health Economics
University College London
London, UK
Baptiste Leurent, PhD
Lecturer, Medical Statistics
University College London
London, UK
Basic Schedule:
Class Time: 2 hours daily