search for


Analysis of stage III colon cancer with missing cause of death
Journal of the Korean Data & Information Science Society 2021;32:135-51
Published online January 31, 2021;
© 2021 Korean Data and Information Science Society.

Minjung Lee1

1Division of Economics & Information Statistics, Kangwon National University
Correspondence to: 1Associate professor, Division of Economics & Information Statistics, Kangwon National University, Chuncheon 24341, Korea. E-mail:

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2018R1D1A1B07041070).
Received December 1, 2020; Revised January 9, 2021; Accepted January 15, 2021.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
In the analysis of competing risks data, the cause of failure may be unknown or missing for some subjects. In such cases, excluding subjects with missing causes of failure from the analysis may lead to biased estimates or misleading inferences. In this paper, we studied methods to analyze competing risks data with missing cause of failure. We used multiple imputation methods to deal with missing causes of failure and estimated the regression parameters and cumulative incidence functions through competing risks modeling. We applied the cause-specific proportional hazards model with the multiple imputation methods to stage III colon cancer data obtained from the Surveillance, Epidemiology, and End Results program of the National Cancer Institute and estimated the effects of covariates on the hazards for death from colon cancer and for death from other causes and the cumulative incidence functions for death from colon cancer and for death from other causes for a patient with specific covariate values under the cause-specific proportional hazards model.
Keywords : Cause-specific proportional hazards model, competing risks data, cumulative incidence function, multiple imputation method.