Basic and advanced statistical methods for meta-analysis
The Meta-analysis short course for health professionals is designed to provide an overview of different meta-analysis methods and common problems encountered with extracting data. Basic and advanced methods which can be used to combine data from various study types will be covered using statistical software such as R and Stata.
Topics covered will include standard methods for intervention comparisons, approaches which can be used for combining different summary measures, subgroup analyses and methods to investigate heterogeneity, as well as advanced methods for diagnostic accuracy, individual patient data and network meta-analysis.
The last date for receipt of complete applications is 5pm Friday 7 March 2025. Regrettably, late applications cannot be accepted.
The overall aims of this module are to enable students to:
- Be able to plan, carry out and interpret meta-analysis of different study designs for questions in evidence based healthcare.
- Extract data in different formats and deal with missing data
- Use statistical software to perform meta-analysis (such as Stata or R)
- Use methods to explore heterogeneity and appropriately use fixed and random effects, subgroup analysis, sensitivity analysis and meta-regression
- Carry out and interpret diagnostic accuracy meta-analysis and network meta-analysis
- Understand the advantages and limitations of an individual patient data meta-analysis
The course will cover the following topics:
- Introduction to different meta-analysis methods and the advantages or disadvantages of each
- Common problems encountered with data extraction
- Heterogeneity, fixed and random effects, meta-regression, unit of analysis, follow-up and cross-over studies
- Approaches to meta-analysis of different study designs
- Diagnostic accuracy meta-analysis
- Introduction to meta-analysis using statistical software (such as Stata and R)
- Network meta-analysis
- Introduction to individual patient data meta-analysis
Teaching will include a combination of presentations by tutors, full group and small group practical sessions.
This module assumes the students have at least some familiarity with the following:
- Distinction between continuous and categorical variables
- Interpretation of p values and statistical significance
- Interpretation of confidence intervals
- Concept of the Normal distribution
- Risk ratio, odds ratio and linear and logistic regression
If you do not have these pre-requisites, the online module Introduction to Statistics for Health Care Research is recommended.