Event

Standard (pairwise) meta-analysis enables the statistical synthesis of data from multiple studies to obtain a single combined estimate of a treatment effect. Network meta-analysis is an extension to pairwise meta-analysis that allows for the comparison of multiple competing interventions that form a connected network of studies. Many of the methods for pairwise and network meta-analysis are commonplace but new methods continue to emerge. For example, new methods continue to be proposed for synthesizing data for rare outcomes, and network meta-analysis of multi-component interventions.

The workshop is divided into two days. Interested persons can register for one or both days.

Day 1 of the two-day workshop, will focus on methods for pairwise meta-analysis. We will begin with an introduction to pairwise meta-analysis models (common, fixed- and random-effects), including different estimators for heterogeneity in random-effects models, and provide participants with the skills to be able to identify, quantify, and investigate heterogeneity through subgroup analysis and meta-regression. We will then focus on special topics, including pairwise meta-analysis of rare outcomes.

Day 2 will focus on network meta-analysis methods. We will begin day 2 with an introduction to network meta-analysis models, fitting the models, and a discussion of the key assumptions that underpin network meta-analysis. We will cover how to assess inconsistency in a network, and present the results of a network meta-analysis. We will then focus on special topics, including component network meta-analysis.

In the lectures, methods and concepts will be introduced. Practical sessions will provide participants with an opportunity to discuss the concepts and implement the methods using the statistical package Stata or R.

Prerequisites and intended audience:

Day 1: The intended audience is quantitative epidemiologists and applied statisticians working in health research.  It will be assumed that participants will have a working knowledge of the statistical package Stata version 14 or above or R statistical software, and have training or experience of systematic reviews, introductory statistics and multivariable regression.

Day 2: The audience will be assumed to be familiar with the statistical methods for pairwise meta-analysis, and multivariable regression, for which day 1 attendance will be sufficient. For Day 2, it will be assumed that participants will have a working knowledge of the statistical package R, and have training or experience of pairwise meta-analysis, introductory statistics and multivariable regression.

Dates: Tuesday 17th and Wednesday 18th February

Registration prices: Standard $1020, Student $740 (for both days) OR Standard $540, Student $390 for single-day attendance

Registration is now open via Trybooking