Use of the Win Ratio in Cardiovascular Trials
Abstract
Background: The primary outcome in many cardiovascular trials is a composite including non-fatal and fatal events. The time-to-first event analysis gives equal statistical weighting to each component event. The win ratio (WR), which takes into account the clinical importance and timing of the outcomes, has been suggested as an alternative approach.
Objectives: Compare the WR with the corresponding hazard ratios (HR) and 1/HR.
Methods: Cox PH models and WR.
Results: In the studied trials (n=16) the WR and HR differed only slightly. For example, in PARADIGM-HF (sacubitril/valsartan vs. enalapril), the primary-outcome of time-to-first HFH or CVD using Cox model gave a 1/HR=1.25 (1.12-1.41; z-score=4.8). Using WR for testing this composite in the hierarchical order of CVD and HFH gave a WR=1.27 (1.15-1.39; z-score=4.7), reflecting a similar effect of sacubitril/valsartan on CVD and HFH. In DIG (digoxin vs. placebo), the outcome of time-to-first HFH or CVD using Cox gave a 1/HR=1.18 (1.10-1.27; z-score=4.5). Using WR for testing this composite in the hierarchical order of CVD and HFH gave a WR=1.14 (1.05-1.20; z-score=3.1), reflecting a larger effect of digoxin on HFH than CVD. Several other trials and end-points including patient-reported measures were studied.
Conclusion: In 16 large cardiovascular outcome trials, HR and WR provided similar estimates of treatment effects. The WR allows the prioritization of fatal outcomes and the hierarchical testing of broader composite endpoints including patient-reported outcomes. In this way, the WR allows the incorporation of patient-centred and other outcomes while prioritizing the competing risk of death and hospital admission.
Origin : Files produced by the author(s)
Loading...