Accounting for missing outcome data in meta-analysis: Part 2

Article type
Year
Authors
Chaimani A1, Mavridis D2, Higgins J3, White I4
1Paris Descartes University
2University of Ioannina
3School of Social and Community Medicine, University of Bristol, UK
4MRC Biostatistics Unit, Cambridge, UK
Abstract
Objectives: To understand the practical application of methods used to account for missing outcome data in a meta-analysis using an updated version of the metamiss command in Stata.

Description: This hands-on practical workshop is the second of two workshops on how to estimate meta-analytic treatment effects for dichotomous or continuous outcomes when these are missing for some of the randomized individuals. The workshop focuses on the technical application of the methodology assuming that all participants are familiar with the theory that will be presented in the first workshop (Part 1). We will present several examples of incorporating information about missing outcome data in the analysis and explore how the different assumptions impact on the summary estimates. We will run sensitivity analyses for the relationship between the outcome in the unobserved data and that in the observed data to evaluate how robust results are as we depart from the missing at random assumption. We will use examples involving both dichotomous and continuous outcome data. Throughout the workshop we will be using a new Stata command, called metamiss2, which is an extension of metamiss.

In order to benefit, participants will need to bring their laptops with Stata 12,13 or 14 installed.