Modeling Recurrent Events: A Tutorial Based on Relapse and Remitting Episodes during Medication-Assisted Treatment...

Area of Science:

• Biostatistics
• Epidemiology
• Clinical Trials

Background:

• Recurrent events are common in health and medical studies, where participants may experience outcomes multiple times.
• Analyzing recurrent event data requires specialized statistical methods beyond standard survival analysis.

Purpose of the Study:

• To provide a comprehensive tutorial on advanced statistical frameworks for analyzing recurrent events data.
• To demonstrate the implementation and interpretation of these methods using real-world clinical trial data.
• To guide researchers in selecting the most appropriate method based on their specific research objectives.

Main Methods:

• Discussion of four general frameworks: extended Cox, parametric survival, longitudinal, and multistate models.
• Detailed implementation guidance including dataset structure, R code, and results interpretation.
• Application of methods to data from the CTN-0051 study on opioid use disorder treatments.

Main Results:

• The tutorial illustrates how to estimate adjusted effects, make individual-level predictions, and model complex disease state transitions.
• Comparison of the strengths and limitations of each analytical framework.
• Recommendations for choosing the optimal method based on primary research goals.

Conclusions:

• This tutorial enhances the accessibility of advanced statistical methods for recurrent events data analysis.
• Researchers can improve their utilization of recurrent events data through understanding and applying these techniques.
• Proper analysis of recurrent events is crucial for drawing accurate conclusions in health research.