Nonparametric and Cox-Based Estimation of Average Treatment Effects in Competing Risks

Estimation of average treatment effects (ATE) of point interventions on time-to-event outcomes with K competing risks (K can be 1). The method uses propensity scores and inverse probability weighting for emulation of baseline randomization, which is described in Charpignon et al. (2022) .

causalCmprsk - Nonparametric and Cox-based Estimation of Average Treatment Effects in Competing Risks

The causalCmprsk package is designed for estimation of average treatment effects (ATE) of point interventions/treatments on time-to-event outcomes with K competing events (K can be 1). The method assumes that there is no unmeasured confounding and uses propensity scores weighting for emulation of baseline randomization.

The causalCmprsk package provides two main functions: fit.cox which assumes the Cox proportional hazards regression for potential outcomes, and fit.nonpar that does not make any modeling assumptions for potential outcomes.

Installation

The causalCmprsk package can be installed by

devtools::install_github("Bella2001/causalCmprsk")

Examples

The examples of how to use causalCmprsk package on real data can be found here.

References

• M.-L. Charpignon, B. Vakulenko-Lagun, B. Zheng, C. Magdamo et al., Causal inference in medical records and complementary systems pharmacology for metformin drug repurposing towards dementia, 2022, Nature Communications.