[Distinguishing misconceptions between outcome prediction and treatment effect prediction models].

Prediction models are increasingly used to inform clinical decision-making. However, outcome prediction models (i.e., traditional prediction models) and treatment effect prediction models (i.e., prediction under interventions) are often confused in practice, which may mislead clinical decisions and lead to adverse clinical outcomes. Using typical study cases, this paper systematically clarifies common misconceptions and key differences between the two model types. It presents the latest advances in the development and application of treatment-effect prediction models, aiming to avoid misinterpretation and misuse, thereby enabling them to more accurately and appropriately inform clinical decision-making.