This repository contains the code used to generate the application for Causal survival analysis under competing risks using longitudinal modified treatment policies (Diaz, Hoffman, and Hejazi 2022). The data (data/derived/dat_final.rds) is synthetic, since the actual data set used contains protected health information.
The R/ folder contains code to create a descriptive alluvial plot, define super learning libraries for estimation, call lmtp::lmtp_tmle() and lmtp::lmtp_sdr() for every time point, estimate confidence intervals, and visualize the results, as described in the paper. Code written by Katherine Hoffman and Nima Hejazi.