Comparing Strategies for Modeling Competing Risks in Discrete-Event Simulations: A Simulation Study and Illustration in Colorectal Cancer

Abstract

Background. Different strategies toward implementing competing risks in discrete-event simulation (DES) models are available. This study aims to provide recommendations regarding modeling approaches that can be defined based on these strategies by performing a quantitative comparison of alternative modeling approaches. Methods. Four modeling approaches were defined: 1) event-specific distribution (ESD), 2) event-specific probability and distribution (ESPD), 3) unimodal joint distribution and regression model (UDR), and 4) multimodal joint distribution and regression model (MDR). Each modeling approach was applied to uncensored individual patient data in a simulation study and a case study in colorectal cancer. Their performance was assessed in terms of relative event incidence difference, relative absolute event incidence difference, and relative entropy of time-to-event distributions. Differences in health economic outcomes were also illustrated for the case study. Results. In the simulation study, the ESPD and MDR approaches outperformed the ESD and UDR approaches, in terms of both event incidence differences and relative entropy. Disease pathway and data characteristics, such as the number of competing risks and overlap between competing time-to-event distributions, substantially affected the approaches’ performance. Although no considerable differences in health economic outcomes were observed, the case study showed that the ESPD approach was most sensitive to low event rates, which negatively affected performance. Conclusions. Based on overall performance, the recommended modeling approach for implementing competing risks in DES models is the MDR approach, which is defined according to the general strategy of selecting the time-to-event first and the corresponding event second. The ESPD approach is a less complex and equally performing alternative if sufficient observations are available for each competing event (i.e., the internal validity shows appropriate data representation).

Comparing Modeling Approaches for Discrete Event Simulations With Competing Risks Based on Censored Individual Patient Data: A Simulation Study and Illustration in Colorectal Cancer

“…The code for implementation in R 14 is provided online: https://personex.nl/research/competingrisks/. Modeling approaches using event-specific probabilities and distributions (ESPD) or a multimodal distribution and regression model (MDR), which have been recommended for uncensored IPD, 11 are not presented here because their implementation was considered too cumbersome for censored IPD (see Discussion for a more detailed discussion).…”

“…10 Modeling approaches corresponding to these strategies have been proposed and compared when informed by "uncensored" IPD generated according to mixtures of event-specific distributions (ESDs). 11 A general recommendation was made to sample the time to event based on a multimodal time-to-event distribution and then to determine the corresponding event second. An approach that first selects the event to occur and then the corresponding time to event also showed good performance with an easier implementation.…”

Quality-Adjusted Life-Year Losses Averted With Every COVID-19 Infection Prevented in the United States

“…We used an event-specific probability and distribution approach to simulating competing risks, where we first simulated the proportion of cohort belonging to each competing risk and then applied the subhazard of the event to model the time of incidence for the event. 18 , 19 Where evidence exists, we stratified the probabilities and hazard estimates by age groups ( Table 1 ). The infected person was followed until death or recovery.…”

Lifetime Health and Economic Outcomes of Active Surveillance, Radical Prostatectomy, and Radiotherapy for Favorable-Risk Localized Prostate Cancer

“…15 Because distributions had to be modeled based on evidence from different sources, the competing risks of PFS, MFS, and BMS were modeled by simulating the first event to occur. 16 OSmet was modeled by pooling replicated individual patient data from the CHAARTED and GETUG15 trials and fitting separate distributions for patients with low-or high-volume disease and for ADT with or without docetaxel. 17 This approach acknowledged that patients with previous local treatment have different outcomes compared with those with de novo metastatic disease.…”