RCTrep: An R Package for the Validation of Estimates of the Average Treatment Effect

Abstract

Despite the recent development of numerous methods aiming to estimate individual-level treatment effects based on observational data, assessing the validity of these estimates remains challenging. It is often unclear whether the observational data meet the assumptions imposed by a method. Additionally, there is often great flexibility in model choice when implementing a given method. This article introduces the R package RCTrep, designed for easy assessment of the validity of estimates of the average treatment effect obtained from observational data. This is achieved by a) making it easy to obtain and visualize estimates derived using a large variety of methods, and b) ensuring that these estimates are easily compared to a gold standard on population and subpopulation levels. RCTrep outlines a four-step workflow, namely, set-selection, estimation, diagnosis, and validation. The package provides a simple dashboard to review the obtained results. This article serves as a user guide for researchers aiming to leverage the potential of observational data to inform personalized treatment.

Value of Real-World Evidence for Treatment Selection: A Case Study in Colon Cancer

“…All the estimation was performed by RCTrep R package. 27 In the second step, we generate a decision tree on the basis of the estimates of 5-year OS improvement of ACT and a threshold (Data Supplement, Details of the tree-fitting, and the way in which the exact numerical threshold is set). The optimal decision for each patient is derived on the basis of the probability of the estimate of 5-year OS improvement larger than the threshold.…”