Regression Modeling for Recurrent Events Possibly with an Informative Terminal Event Using R Package reReg

Area of Science:

• Biostatistics
• Survival Analysis
• Computational Biology

Background:

• Recurrent event data analysis is crucial in fields like medicine and public health.
• Existing methods may not adequately handle informative terminal events or complex censoring mechanisms.
• A need exists for flexible, integrated tools for recurrent event regression.

Purpose of the Study:

• To introduce the reReg R package for comprehensive recurrent event data analysis.
• To provide a unified regression framework for recurrent events and informative terminal events.
• To offer practical tools for modeling, visualization, and simulation in survival analysis.

Main Methods:

• Utilizes a general scale-change regression model encompassing Cox-type, accelerated rate, and accelerated mean models.
• Accommodates informative censoring using subject-specific frailty without parametric assumptions.
• Allows distinct regression models for recurrent and terminal event processes.

Main Results:

• The reReg package offers a flexible and unified approach to recurrent event analysis.
• It effectively handles informative terminal events and censoring.
• Includes tools for data visualization and simulation to aid analysis.

Conclusions:

• The reReg R package is a valuable, user-friendly tool for advanced recurrent event analysis.
• It enhances the ability to model complex survival data in biomedical and public health research.
• Facilitates robust statistical inference for recurrent events with informative censoring.