Adjusted win ratio using the inverse probability of treatment weighting
- 10 November 2023
- journal article
- research article
- Published by Taylor & Francis in Journal of Biopharmaceutical Statistics
- Vol. 35 (1) , 21-36
Abstract
The win ratio method has been increasingly applied in the design and analysis of clinical trials. However, the win ratio method is a univariate approach that does not allow for adjusting for baseline imbalances in covariates, although a stratified win ratio can be calculated when the number of strata is small. This paper proposes an adjusted win ratio to control for such imbalances by inverse probability of treatment weighting (IPTW) method. We derive the adjusted win ratio with its variance and suggest three IPTW adjustments: IPTW-average treatment effect (IPTW-ATE), stabilized IPTW-ATE (SIPTW-ATE) and IPTW-average treatment effect in the treated (IPTW-ATT). The proposed adjusted methods are applied to analyse a composite outcome in the CHARM trial. The statistical properties of the methods are assessed through simulations. Results show that adjusted win ratio methods can correct the win ratio for covariate imbalances at baseline. Simulation results show that the three proposed adjusted win ratios have similar power to detect the treatment difference and have slightly lower power than the corresponding adjusted Cox models when the assumption of proportional hazards holds true but have consistently higher power than adjusted Cox models when the proportional hazard assumption is violated.Keywords
Funding Information
- Medical Research Council (MR/S014357/1)
This publication has 30 references indexed in Scilit:
- Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studiesStatistics in Medicine, 2015
- Variance reduction in randomised trials by inverse probability weighting using the propensity scoreStatistics in Medicine, 2013
- Inverse probability weighting for covariate adjustment in randomized studiesStatistics in Medicine, 2013
- A tutorial on propensity score estimation for multiple treatments using generalized boosted modelsStatistics in Medicine, 2013
- The win ratio: a new approach to the analysis of composite endpoints in clinical trials based on clinical prioritiesEuropean Heart Journal, 2011
- An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational StudiesMultivariate Behavioral Research, 2011
- 6. A Diagnostic Routine for the Detection of Consequential Heterogeneity of Causal EffectsSociological Methodology, 2008
- Predictors of mortality and morbidity in patients with chronic heart failureEuropean Heart Journal, 2005
- Stratification and weighting via the propensity score in estimation of causal treatment effects: a comparative studyStatistics in Medicine, 2004
- Effects of candesartan on mortality and morbidity in patients with chronic heart failure: the CHARM-Overall programmeThe Lancet, 2003