Systematic reviews and meta-analyses are an important cornerstone of contemporary evidence-based medicine. The large majority summarize published aggregate data, but it is increasingly common that individual participant data (IPD) are obtained from primary studies. As a result, new opportunities arise and more advanced statistical methods are needed to properly analyze the available data. In this course, we discuss how a meta-analysis involving IPD may help to identify sources of heterogeneous treatment effects, to investigate the accuracy of diagnostic tests, to develop clinical prediction models and to externally validate such models. We place particular emphasis on statistical methods for dealing with between-study heterogeneity, and discuss how to interpret corresponding results. The course consists of plenary presentations, small-group discussions, reading assignments, and computer exercises.
This course will introduce participants to the fundamental statistical methods and principles for evidence synthesis and meta-analysis when IPD (Individual Participant/Patient Data) are available from multiple related studies. The course will consider continuous, binary and time-to-event outcomes, and both fixed-effect and random-effects meta-analysis models. Day 1 will focus mainly on the general rationale and advantages of IPD meta-analysis. Day 2 will focus on the synthesis of IPD from randomised trials of interventions, where the aim is to quantify a treatment effect (usually in the presence of between-study heterogeneity) or to identify treatment effect modifiers (interactions) for stratified medicine. Day 3 will focus on key differences and potential objectives of IPD meta-analysis of observational studies, where the aim is to identify prognostic factors or to develop/validate a risk prediction (prognostic) model. Day 4 will focus on statistical methods for developing and validating risk prediction models in IPD meta-analysis. On Day 5, students will prepare a protocol for a case study and discuss this with their peers.
The key messages will be illustrated with real examples throughout, and participants will conduct a variety of IPD analyses within R, to practise the key methods and reinforce the learning points. By the end of the course, you should be able to:
- Explain the rationale for performing an individual participant data meta-analysis (IPD-MA)
- Understand the advantages, limitations and key characteristics of IPD-MA in intervention, diagnostic and prognostic research
- Understand the relevance of between-study heterogeneity, and be familiar with statistical methods for investigating and reporting this
- Be familiar with statistical methods for summarizing relative treatment effects and exploring the presence of treatment-covariate interaction
- Be familiar with statistical methods for developing and validating clinical prediction models using IPD from multiple studies or settings
- Be familiar with statistical methods for investigating and comparing diagnostic test accuracy using IPD
- Interpret and critically appraise the results from an IPD-MA
Meta-analysis of individual participant data: rationale, conduct, and reporting. BMJ 2010:340;c221.
Get real in individual participant data (IPD) meta-analysis: a review of the methodology. Res Synth Methods 2015:6;293-309.
We expect participants to have a basic knowledge about the principles of intervention research, diagnostic research, prognostic research, systematic reviews and meta-analysis. Some basic knowledge of R is helpful (but not required).