A Kaplan-Meier curve shows the proportion of patients (or animals) who have not yet had an event, such as
death or relapse, over time. It is the standard figure for survival data.
How to read it
Each drop is one or more events at that time.
Tick marks are censored subjects: they left the study or the study ended before they had
an event. They count until that time but not after.
Median survival is the time where the curve crosses 0.5.
Data you need
One row per subject with three columns: group, time to event or last follow-up, and status (1 = event,
0 = censored).
Comparing groups
The log-rank test checks whether the curves differ. Report its p-value on the plot. For effect size, a Cox
model gives a hazard ratio with a confidence interval.
Making a clean figure
Use a step line for each group, with censoring ticks.
Add a number-at-risk table below the x-axis if your journal expects it.
Start the y-axis at 0 and end it at 1 (or 0–100%).