A class of accelerated means regression models for recurrent event data.

Area of Science:

• Biostatistics
• Survival Analysis
• Statistical Modeling

Background:

• Recurrent event data analysis is crucial in medical research.
• Existing models like proportional means and accelerated failure time have limitations.
• Flexible modeling of covariate effects on event rates is needed.

Purpose of the Study:

• To propose a general class of accelerated means regression models for recurrent event data.
• To offer a flexible framework for analyzing covariate effects on mean functions of counting processes.
• To develop robust inference and model checking procedures.

Main Methods:

• Development of a general class of accelerated means regression models.
• Utilizing estimating equation approaches for parameter inference.
• Establishing large and final sample properties of proposed estimators.
• Implementing graphical and numerical model checking procedures.

Main Results:

• The proposed models flexibly incorporate covariate effects on mean functions.
• The class includes proportional means, accelerated failure time, and accelerated rates models as special cases.
• Valid statistical inference and model checking methods are established.
• The model's utility is demonstrated with chronic granulomatous disease data.

Conclusions:

• The general accelerated means regression model provides a flexible and powerful tool for recurrent event data analysis.
• The developed methods ensure reliable parameter estimation and model validation.
• This approach enhances the understanding of disease progression and treatment effects in clinical studies.