Evidence-appraisal glossary
Integrated Discrimination Improvement
Integrated discrimination improvement summarizes how much adding a new predictor to a model widens the gap in predicted risk between people who do and do not have the outcome. It complements net reclassification improvement.
Also called: IDI.
It compares the average predicted probabilities for those with and without events, before and after a marker is added, capturing improvement in separation across all thresholds at once. Like net reclassification improvement, it can look impressive even when the model is poorly calibrated or clinically unhelpful, so it should not stand alone. Decision curve analysis and net benefit better answer whether a new marker actually helps decisions.
This is a plain-language methodology definition for reading research. It is general education, not medical advice.