|
Abstract:
|
Essential problems encountered in vaccine clinical trials are: finding to what extent the vaccine prevents the targeted disease, and establishing a correlation between disease protection and antibody responses as measured at a given time point after vaccination. In the past, to address these issues, models based on survival analysis approaches, in particular Cox proportional hazard regression model, were utilized. However, interpretation of Cox model parameters is limited. Therefore, there is a need to apply methods such as evaluations of explained variation and/or predictive accuracy. In our presentation, we focus on vaccine efficacy estimation based on time-to-event data, in particular on handling heterogeneity problems in such data. Some useful methods will be discussed.
|
