Biology, images, analysis, design...
Use/Abuse Principles How To Related
"It has long been an axiom of mine that the little things are infinitely the most important" (Sherlock Holmes)

Search this site




There are three types of censoring: right, left and interval.
  1. Right censoring is the most common especially in experimental work such as randomised trials. It occurs when the event does not occur within the observation period. It is only known that the period of survival is longer than the period of the study. This may occur because (a)the individual survives up till the end of the study period - this is sometimes called administrative censoring or (b)the individual withdraws from the study.
  2. Left censoring occurs when it is only known that the time to the event is less than a certain period. This can occur when a test is taken to establish the presence or absence of a disease but the person became infected at some indeterminate period prior to the test.
  3. Interval censoring occurs when the exact time of an event is unknown but it is known to have occurred within a particular period of time.


It is usually necessary to assume that censoring is independent of future survival. In other words there is no difference between the subsequent survival of censored individuals compared to uncensored individuals.

This is probably a safe assumption for administrative censoring. However for withdrawals this may not be the case. If the withdrawal is related to outcome then estimation of the survival time will be biased upwards.