Proportionality of hazards was assessed for each variable and Schoenfeld residuals were visually inspected for potential time–variant biases. Our assessment of the proportionality of hazards assumption and visual inspection of Schoenfeld residuals showed that none were significant based on a p value threshold of 0·05.
The proportional hazards assumption was confirmedby residual plots
We examined the proportional hazards assumption by testing statistical significance of interactions between follow-up time and exposures.
We used the Schoenfeld residual test to verify the assumption of proportional hazards in the Cox analysis, which was fulfilled for the end points of death from any causeand further bleeding.
We examined the assumption ofproportional hazards by using a Wald test of the interaction between treatment status and time.
To assess the validity of the proportional hazards assumption, the assumption was assessed by log-minus-log-survival function and found to hold. To confirm the assumption of proportionality, time-dependent covariate analysis was used.