Cox regression time-to-event interpretation

For a course I need to superficially understand a statistical approach in a research paper, available at:

Vericiguat in Patients with Heart Failure and Reduced Ejection Fraction (

In this paper Cox regression was used for time-to-event analysis and to calculate hazard ratio. My understanding of this method is the primary outcome is the variable and time is constant for the two treatment arms. Cox regression allows all data to be used regardless of subject completing the study, then Kaplan Meier curves allow for visual depiction of the data. The hazard ratio of 0.9 (95% CI, 0.82-0.98) for the composite outcome indicates a 10% decrease in relative terms compared to placebo, with statistical significance.

Does this seem accurate on a surface level? Appreciate any replies.



Less is more. Stay pure. Stay poor.
"Regardless of subjects completing the study", do you mean regardless of complete followup or lack of being followed long enough for outcome to occur?

Pretty good description!
Thanks for the reply. I worded this poor. One limitation to Cox Regression is subjects may not have been followed long enough to experience the outcome. If a subject doesn't complete the trial, the time to event analysis allows comparisons between survivors in each group at various time points.

I hope I am thinking of this correctly.


Less is more. Stay pure. Stay poor.
Yes the phrasing is, "censor". Most of the time it is right censoring, but of interest there is a type also called left censoring!