I'm a beginner in R and want to analyze KM graph. What does this plot signify?
Is there is worst prognosis at mid expression levels? And what does patient number at risk given at the bottom represent?
The lengths of the segments parallel to the x-axis represent event-free duration, i.e. survival duration. Each segment ends with an event (e.g. death). The y-axis represents the cumulative probability of surviving for a given time. For the example curve shown here, if you're in the blue group your probability of surviving for 500 time units is ~37% and if you're in the red group, it's a bit less than 75%. The number at risk is the number of surviving individuals. This is used to calculate the probabilities for each interval (as #surviving after the event/#surviving in the interval before the event). This gives a sense of scale, for example note that the many small steps in the red curve reflect the higher number of individuals involved (i.e. one death out of 42 changes the probabilities less than one out of 4 individuals). Because of this, while the gap between the red and blue curves is the quantity of interest, you can't rely on visual inspection only and a statistical test is required to decide whether the difference is statistically significant. I suggest you read some of the literature on the subject.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
