Model-based imputation methods for estimating win statistics in survival data with non-informative censoring

The purpose of this study is to estimate win statistics (win ratio, win odds, and net benefit) under non-informative censoring for prioritized composite outcomes. Survival data with non-informative censoring, independent of the distribution of survival times, were considered, involving a binary treatment variable and prioritized composite outcomes. In this context, a model-based imputation (MBI) method for estimating win statistics is proposed. This method utilizes a conditional survival distribution derived from the Kaplan-Meier (KM) estimator to impute the survival times of censored individuals. The proposed method adjusts for censoring bias by imputing survival times using the KM estimator under non-informative censoring. Simulation results showed that the proposed method had less bias than both the unadjusted method and the inverse probability of censoring weighting (IPCW) method, while maintaining comparable empirical variance to IPCW. The simulation study demonstrated consistent results across varying sample sizes and censoring rates. MBI methods are proposed for two or more prioritized composite endpoints under non-informative censoring. These methods effectively adjust for bias caused by missing data due to censoring in prioritized composite outcomes while maintaining variability comparable to IPCW, making them useful for estimating win statistics.

Authors

Institutions

Publication Details

Journal
BMC Medical Research Methodology
Published
2026-11-14
DOI
https://doi.org/10.1186/s12874-026-03085-8
Primary Topic
Statistical Methods in Clinical Trials
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Model-based imputation methods for estimating win statistics in survival data with non-informative censoring

Hyonggin An, Yujin Jeong
BMC Medical Research Methodology
Statistical Methods in Clinical Trials
article

Model-based imputation methods for estimating win statistics in survival data with non-informative censoring

Hyonggin An, Yujin Jeong
article en

Abstract

The purpose of this study is to estimate win statistics (win ratio, win odds, and net benefit) under non-informative censoring for prioritized composite outcomes. Survival data with non-informative censoring, independent of the distribution of survival times, were considered, involving a binary treatment variable and prioritized composite outcomes. In this context, a model-based imputation (MBI) method for estimating win statistics is proposed. This method utilizes a conditional survival distribution derived from the Kaplan-Meier (KM) estimator to impute the survival times of censored individuals. The proposed method adjusts for censoring bias by imputing survival times using the KM estimator under non-informative censoring. Simulation results showed that the proposed method had less bias than both the unadjusted method and the inverse probability of censoring weighting (IPCW) method, while maintaining comparable empirical variance to IPCW. The simulation study demonstrated consistent results across varying sample sizes and censoring rates. MBI methods are proposed for two or more prioritized composite endpoints under non-informative censoring. These methods effectively adjust for bias caused by missing data due to censoring in prioritized composite outcomes while maintaining variability comparable to IPCW, making them useful for estimating win statistics.

BMC Medical Research Methodology
Korea University (KR)
Statistical Methods in Clinical Trials
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon