Objectives: This workshop will introduce participants to individual participant data (IPD) meta-analysis in risk prediction research. We will discuss opportunities and challenges of combining IPD from multiple studies when developing or validating a prediction model. Subsequently, we will introduce statistical methodology and describe common software packages. Finally, we will illustrate the implementation of an IPD meta-analysis using case studies and example papers.
Description: Risk prediction models are commonly developed for predicting the presence (diagnostic models) or future occurrence (prognostic models) of a particular outcome. They aim to provide absolute outcome risks for distinct individuals based on multiple predictors such as subject characteristics, clinical history and physical examination items, or more complex clinical measures such as medical imaging and biomarker results. Although prediction models offer numerous opportunities to public health, their anticipated performance is often overoptimistic and their generalizability tends to be insufficient. During the past few years, IPD meta-analysis has become increasingly popular to address these issues, yet, little guidance exist on how to conduct meta-analysis of risk prediction studies.
In this workshop we will discuss why IPD meta-analysis offers unique opportunities to risk prediction research, and describe how multiple datasets can appropriately be combined for this aim. We will demonstrate how researchers can accommodate for heterogeneity across study populations. Finally, we will illustrate how researchers should interpret the achieved meta-analytical model performance and provide strategies for improving upon its generalizability.