rpsftm: An R Package for Rank Preserving Structural Failure Time Models.

Area of Science:

• Biostatistics
• Clinical Trials Methodology
• Survival Analysis

Background:

• Treatment switching in randomized controlled trials can bias efficacy estimates.
• Standard intention-to-treat analyses do not adequately adjust for participants changing treatments.
• Accurate estimation of treatment effects requires accounting for treatment switching.

Purpose of the Study:

• To present an R package implementing the rank preserving structural failure time model (RPSFTM).
• To provide a tool for adjusting survival outcome analyses for treatment switching.
• To enable estimation of causal treatment effects in the presence of treatment switching.

Main Methods:

• The rank preserving structural failure time model (RPSFTM) is employed.
• G-estimation is used to estimate the treatment effect (ψ).
• The method balances counter-factual event times between treatment groups.

Main Results:

• The RPSFTM method adjusts for treatment switching using randomization-based principles.
• It utilizes observed event times and treatment history for analysis.
• The R package facilitates the application of this statistical method.

Conclusions:

• The developed R package implements a robust method for handling treatment switching in survival data.
• Accurate causal treatment effect estimation is achievable with RPSFTM.
• This tool aids researchers in conducting more reliable clinical trial analyses.