Advances and applications of the Win Ratio in clinical studies

Since its introduction in 2012 by Pocock et al., the win ratio (WR) has become increasingly popular and has been used to analyze the results of multiple clinical trials. This approach is attractive because it combines information from multiple endpoints and takes into account their relative importance. However, it has been shown that win ratio estimate can be biased when the data are censored or missing. Unfortunately, in practical applications, these issues are often ignored and more ties often result from subjects with censoring or missing data than those without. Although some approaches have been proposed to deal with censoring for time-to-event endpoints, their implementation in practice is lagging and this is still an active area of research. In this presentation, we demonstrate that when all endpoints are time-to-event, the win ratio parameter depends on the length of follow up. Furthermore, in the presence of censoring, traditional WR estimator as described by Pocock et al. estimates a new parameter - a weighted average of win ratios at multiple lengths of follow up. Recognizing that the hypothesis testing in practice is often based on this new parameter, we show how to evaluate the sample size requirements in a trial design where the traditional WR estimator is used for primary analysis, under the assumption of administrative censoring due to uniform enrollment. 

The win ratio method, used to analyze composite endpoints in clinical trials, has gained substantial popularity in recent years because of its ability to prioritize components of the composite outcome. Despite gaining popularity and being extended by solving some of its issues, little work has been done to incorporate covariate information into the win ratio. In this article, we extend the win ratio method by incorporating weights to each win or loss based on the distance between the compared pair using their covariate values. This approach aims to improve the power of the original win ratio when the covariates used for computing the weights are associated with the components of the composite outcome. Through detailed simulation studies and real data analyses, we demonstrate the utility of our proposed method. In general, our simulation studies reveal that the proposed method is more powerful in detecting the difference between the treatment and control groups when the covariates used to calculate the weights are associated with the outcomes, and it performs very similarly to the original method when there is no such association. 

The Win Ratio (WR) method, introduced by Pocock et al. in 2011, provides a powerful statistical approach for analyzing composite outcomes by prioritizing components based on their clinical significance. With its growing application across various medical fields, understanding the WR method and its extensions has become increasingly important for researchers and clinicians. Dr. Wei will introduce the implementation of the WINDSURFER (Win ratio analysis to Determine a strategy of non-invasive SUpport for Respiratory Failure in the EmeRgency Department) trial using WR method. The considerations and challenges encountered during the study design of this trial will be shared. 

The win-ratio, a term introduced by Pocock et. al. (2012), has become a popular approach to the statistical analysis of controlled clinical trials with multiple prioritized point outcomes. Briefly, comparisons of the time to the primary outcome (say mortality) are made between each subject in the active treatment group and each subject in the control group with indeterminacies resolved where possible by comparisons of times to secondary outcomes. This approach usually gives similar results to the conventional analysis based on the time to the first event, but it has the conceptual advantage of basing the comparison on the more important outcome when the results for the two outcomes differ. See Oakes (2025) for further discussion of the properties of the win-ratio statistic. As noted by Finkelstein and Schoenfeld (2018), the changing composition of the win-ratio statistic over time makes it hard to interpret interim analyses of accumulating data or to develop rules for interim stopping. Oakes (2016) proposed a curtailed win-ratio statistic, based on data accumulated on each subject up to a prespecified window (say one-year of follow-up). This approach avoids the difficulty and allows conventional sequential stopping rules to be used. 

The win ratio (WR) is a method for comparing composite endpoints. Originally used in randomized clinical trials, there is interest in applying this approach to observational studies with multiple clinically relevant endpoints. The unmatched WR approach compares all participants in one treatment arm to all participants in another arm and declares a 'winner' according to a hierarchy of potential outcomes for each pairwise comparison. The WR is then estimated as the ratio of 'winners' in the treatment arm to 'winners' in the control arm, excluding any hierarchical 'ties'. While this provides unbiased estimates in settings where there is balance in groups on measured and unmeasured confounders, in observational settings, confounding related to non-random assignment of treatment must be considered. Methods to address potential covariate imbalance include matching, stratification, and inverse probably of treatment weighting, however, each has potential limitations such as limiting sample size and representativeness. We explore use of the WR in an observational setting using a retrospective study comparing composite endpoints in Hospital-acquired/Ventilator-Associated Pneumonia by prescribed antibiotic treatment. We examine imbalance in measured confounding; characterize how each method affects the WR estimate; and discuss implications for their use in observational studies.