Can Predictive Modeling Inform the Selection of Time Zero for Target Trial Emulations? An Empirical Study of Atorvastatin Initiation in Medicare Beneficiaries
This study demonstrates that empirically identifying strong predictors of atorvastatin initiation, such as recent hospitalizations for cerebral or myocardial infarction, can effectively guide the selection of valid "time zero" events for target trial emulations in real-world data, thereby mitigating channeling bias and residual confounding.