Abstract
Semi-competing risks refer to the setting where primary scientific interest lies in estimation and inference with respect to a non-terminal event, the occurrence of which is subject to a terminal event. In this paper, we present the R package SemiCompRisks that provides functions to perform the analysis of independent/clustered semi-competing risks data under the illness-death multi-state model. The package allows the user to choose the specification for model components from a range of options giving users substantial flexibility, including: accelerated failure time or proportional hazards regression models; parametric or non-parametric specifications for baseline survival functions; parametric or non-parametric specifications for random effects distributions when the data are cluster-correlated; and, a Markov or semi-Markov specification for terminal event following non-terminal event. While estimation is mainly performed within the Bayesian paradigm, the package also provides the maximum likelihood estimation for select parametric models. The package also includes functions for univariate survival analysis as complementary analysis tools.
Highlights
Semi-competing risks refer to the general setting where primary scientific interest lies in estimation and inference with respect to a non-terminal event, the occurrence of which is subject to a terminal event (Fine et al, 2001; Jazicet al., 2016)
We offer three flexible multi-state illness-death models for the analysis of semi-competing risks data: accelerated failure time (AFT) models for independent data; proportional hazards regression (PHR) models for independent data; and PHR models for cluster-correlated data
We categorize the approaches based on the analysis type, the survival model (AFT or PHR), data type, accommodation to left-truncation and/or interval-censoring in addition to right-censoring, and statistical paradigms
Summary
Semi-competing risks refer to the general setting where primary scientific interest lies in estimation and inference with respect to a non-terminal event (e.g., disease diagnosis), the occurrence of which is subject to a terminal event (e.g., death) (Fine et al, 2001; Jazicet al., 2016). We offer three flexible multi-state illness-death models for the analysis of semi-competing risks data: accelerated failure time (AFT) models for independent data; proportional hazards regression (PHR) models for independent data; and PHR models for cluster-correlated data These models accommodate parametric or non-parametric specifications for baseline survival functions as well as a Markov or semi-Markov assumptions for terminal event following non-terminal event. From a Bayesian perspective for models (8)-(10), we can adopt either a parametric Weibull or nonparametric PEM specification for baseline hazard functions h0g with their respective configurations of prior distributions analogous to those outlined in Section PHR models for independent semicompeting risks data. It provides two auxiliary functions, initiate.startValues_HReg and initiate.startValues_AFT, that can be used to generate initial values for Bayesian estimation; simID and simSurv functions for simulating semi-competing risks and univariate survival data, respectively; five covariates and parameter estimates from CIBMTR data; and the BMT dataset referring to 137 bone marrow transplant patients
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
