IPDvariatie

IPDvariatie is a term used in the field of statistics and data synthesis to denote variation observed in individual participant data (IPD) across studies or within subgroups of IPD datasets. It captures differences in outcomes, covariate distributions, or effect sizes that remain after accounting for common factors.

In IPD meta-analysis, researchers collect raw data from multiple studies to estimate treatment effects more accurately than from aggregate data. IPDvariatie reflects the heterogeneity of effects at the participant level and can arise from genuine variation in biology, treatment response, measurement methods, or study design differences.

Methods for assessing IPDvariatie include hierarchical or mixed-effects models that partition variance into components attributable to between-study differences and within-study or subgroup differences. Quantifications such as tau-squared or variance components help determine whether a fixed-effect or random-effects approach is appropriate and guide subgroup analyses.

Implications: High IPDvariatie indicates that treatment effects differ across participants or studies, suggesting the need for tailored analyses, subgroup exploration, or cautious generalization. Low IPDvariatie supports pooled estimates assuming similar effects.

See also: Individual participant data meta-analysis; heterogeneity; random-effects model; meta-regression; hierarchical modeling.