|
Abstract:
|
It is common to see the endpoint is the composite of recurrent events with a terminal event in a clinical trial study. A simulation study is conducted to investigate a range of recurrent event methodologies including the Prentice, Williams, Peterson (PWP), Wei, Lin, Weissfeld (WLW), Anderson-Gill, negative binomial and unmatched win ratio analyses were investigated and the pros and cons of each of these will be discussed. Simulations were performed to determine the power of the different methodologies. Simulating recurrent event data from a composite endpoint requires a substantial number of complex assumptions e.g. the event rates and hazard ratios for each component and each recurrence of the event, the dependence between events of the same type and different types.
|
