Article type
Year
Abstract
Objectives: This workshop introduces individual participant data (IPD) meta-analysis in risk prediction research and the required statistical methodology. We will illustrate the implementation of IPD meta-analysis using case studies and example papers.
Description: The development and validation of prediction models is an important area in contemporary medical research. During the past few years, evidence synthesis and meta-analysis of individual participant data (IPD) has become increasingly popular, not only for intervention research but also for improving the development, validation and generalizability of prediction models. IPD meta-analyses (IPD-MA) provide unique opportunities to better develop and enhance the applicability of prediction models across (sub)populations and settings. There is, however, little guidance on how to conduct an IPD-MA aimed at developing and validating diagnostic and prognostic prediction models, and how to interpret their findings.
We will discuss how such types of IPD-MA differ from IPD-MA for assessing treatment effects and identify key advantages and challenges in IPD-MA of prediction models. We provide recommendations for the design of such IPD-MA including the selection of relevant studies. In small groups we will discuss statistical methods for handling between-study heterogeneity, and other issues regarding prediction model development and validation and illustrate this using various empirical examples across medical disciplines.
Description: The development and validation of prediction models is an important area in contemporary medical research. During the past few years, evidence synthesis and meta-analysis of individual participant data (IPD) has become increasingly popular, not only for intervention research but also for improving the development, validation and generalizability of prediction models. IPD meta-analyses (IPD-MA) provide unique opportunities to better develop and enhance the applicability of prediction models across (sub)populations and settings. There is, however, little guidance on how to conduct an IPD-MA aimed at developing and validating diagnostic and prognostic prediction models, and how to interpret their findings.
We will discuss how such types of IPD-MA differ from IPD-MA for assessing treatment effects and identify key advantages and challenges in IPD-MA of prediction models. We provide recommendations for the design of such IPD-MA including the selection of relevant studies. In small groups we will discuss statistical methods for handling between-study heterogeneity, and other issues regarding prediction model development and validation and illustrate this using various empirical examples across medical disciplines.