Abstract Number:
3327
Submission Type:
Contributed Abstract
Contributed Abstract Type:
Speed
Participants:
Hyejeong Jang (1), Dipenkumar Modi (2), Seongho Kim (1)
Institutions:
(1) Wayne State University, Detroit, MI, (2) Karmanos Cancer Institute, Detroit, MI
Co-Author(s):
First Author:
Presenting Author:
Abstract Text:
We investigate the impact of unequal censorings on the comparison of survival distributions by evaluating seven statistical tests: log-rank (LR), Gehan-Breslow generalized Wilcoxon (GB), Tarone-Ware (TW), Peto-Peto (PP), modified Peto-Peto (mPP) tests, a combined test using LR and TW methods, and an all-combined test. Through 1,000 simulations, we evaluate the performance of these in comparing two survival curves, considering the size and power of the tests across four types of censoring: overall, early, middle, and late censoring periods. We explore different scenarios with sample sizes of 20 and 50 for each group and varying levels of censoring (no censoring, 10%, and 25%). Additionally, we assess the proportional hazard assumption is also evaluated during the simulations. Regarding overall censoring, both LR and all-combined tests demonstrated the highest power. Early censoring appeared to minimally impact the size and power of all seven tests. However, LR and combined tests exhibited greater power overall, while GB showed the least power in the middle censoring scenario. In cases of late censoring, the all- combined tests displayed the highest power.
Keywords:
Log-rank test |Gehan-Breslow generalized Wilcoxon test|Tarone-Ware test|Peto-Peto test|Modified Peto-Peto test|Survival analysis
Sponsors:
Biometrics Section
Tracks:
Survival Analysis
Can this be considered for alternate subtype?
No
Are you interested in volunteering to serve as a session chair?
No
I have read and understand that JSM participants must abide by the Participant Guidelines.
Yes
I understand that JSM participants must register and pay the appropriate registration fee by June 1, 2024. The registration fee is non-refundable.
I understand