The Essentials of Biostatistics for Physicians, Nurses, and Clinicians

10.1 Time-to-Event Data and Right Censoring 159

determination of a particular adverse event that the treatment is sus-
pected to cause. The term survival analysis came about because it was
originally used when mortality was the outcome, but it can be used for
time - to - event data for any event. More generally, the curve does not
necessarily have to be a function of time. It is even possible for time
to be replaced by a variable that increases with time, such as the cost
of a worker ’ s compensation claim where the event occurs when the
claim is closed.

10.1 TIME - TO - EVENT DATA AND

RIGHT CENSORING

What characterizes survival data is that some patients have incomplete
results. In a particular study, there is a time at which the study ends
and the data must be analyzed. At that point, some of the patients may
not have experienced the event (either because they will never have the
event or because the event will occur some time later). The data for
these patients should not be thrown out because that would (1) ignore
valuable information about the time to event, since these patients time

Figure 10.1. A typical survival curve.

S(t): Survival at time t

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

(^01234567)
Years
Survival Probability
S(t)