学术报告（张一驰 2025.1.9）
Addressing Post-treatment Selection Bias in Comparative Effectiveness Research with an Application to Electronic Health Records Data
Abstract: To examine methodologies that address treatment switching and informative censoring after treatment assignment, six different approaches were evaluated under a comparative effectiveness framework: intention-to-treat, as-treated, intention-to-treat with censor-weighting, as-treated with censor-weighting, time-varying exposure, and time-varying exposure with censor-weighting. Marginal structural models were employed to address time-varying exposure, confounding, and possibly informative censoring. These methodologies were applied to an electronic health records dataset of adult patients who were hospitalized with acute coronary syndrome and treated with either clopidogrel or ticagrelor. The effectiveness endpoint was the occurrence of death, myocardial infarction, or stroke. Hazard ratio estimates were similar to that obtained from the randomized study of Platelet Inhibition and Patient Outcomes (PLATO).
