Generalized win fraction regression for composite survival endpoints

2026-04-05

AI Summary

This study addresses the challenge of modeling composite time-to-event outcomes with prioritized components in the presence of right censoring and tied event times. The authors propose a generalized win ratio regression framework that models the conditional win ratio through a specified link function—such as logit, probit, or identity—and handles right-censored data via inverse probability censoring weighting. Innovatively, under the logit link, the log-odds of the win ratio is introduced as the effect measure, with ties conservatively treated as non-wins. The method reduces to the proportional win ratio model in the absence of ties and unifies several existing estimation approaches. Large-sample properties of the estimators are established using M-estimation theory and sandwich variance estimation. Simulations demonstrate robust performance across varying censoring rates, and the approach is successfully applied to a reanalysis of the HF-ACTION clinical trial.

Abstract

We propose a generalized win fraction regression framework for prioritized composite survival outcomes. The framework models the conditional win fraction through a chosen link function (including identity, logit, or probit), thereby accommodating multi-component time-to-event endpoints within a unified regression structure. To handle right censoring, we construct inverse-probability-of-censoring-weighted estimating equations that target the win fraction as if censoring were absent. Under the identity link, regression parameters characterize covariate associations on the natural win fraction scale. Under the logit link, they characterize the log odds of winning -- a new and complementary effect measure that treats ties as failures to win, imposing a more conservative standard than the win ratio or win odds. When there are no ties, the logit win fraction model reduces to proportional win fraction regression; moreover, the unweighted version of our estimating equations numerically coincides with the proportional win fraction point estimator regardless of ties. We establish large-sample properties of the proposed estimators and derive a consistent sandwich variance estimator that accounts for uncertainty from the estimated censoring weights. Extensive simulations examine finite-sample performance across link functions and censoring rates, and our method is illustrated through a reanalysis of the HF-ACTION clinical trial.

Problem

• composite survival endpoints
• win fraction
• right censoring
• regression modeling
• time-to-event data

Innovation

• win fraction regression
• composite survival endpoints
• inverse-probability-of-censoring weighting
• logit link
• sandwich variance estimator