Semiparametric Regression Analysis of Mixed Recurrent Event and Panel Count Data with Multiple Causes of Failure

Recurrent event data and panel count data are common in survival studies. There are situations in which some of the study individuals may be followed continuously during the study period, while the others may only be assessed at a series of discrete time periods. As a result, a data structure that combines recurring event and panel count data for a single study arises. This article describes the regression problem for analyzing such mixed type recurrent event data based on the mean function of the underlying recurrent event process with multiple failure modes. We introduce a proportional cause-specific mean model for multiple causes of failure. The estimators of the regression parameters and the baseline cause-specific mean function are derived, and their asymptotic properties are studied. The finite sample behaviour of the suggested estimators is evaluated using simulation studies. Finally, a real data is used to illustrate the proposed techniques.