Survival analysis is the analysis of time-to-event data. Such data describe the length of time from a time origin to an endpoint of interest. For example, individuals might be followed from birth to the onset of some disease, or the survival time after the diagnosis of some disease might be studied. Survival analysis methods are usually used to analyse data collected prospectively in time, such as data from a prospective cohort study or data collected for a clinical trial. The time origin must be specified such that individuals are as much as possible on an equal footing. For example if the survival time of patients with a particular type of cancer is being studied, the time origin could be chosen to be the time point of diagnosis of that type of cancer. Equally importantly, the endpoint or event of interest should be appropriately specified, such that the times considered are well-defined. In the above example, this could be death due to the cancer studied. Then the length of time from the time origin to the endpoint could be calculated.One of the reasons why survival analysis requires ‘special’ techniques is the possibility of not observing the event of interest for some individuals. For example individuals may drop out of a study, or they might have a different event, such as in the above example death due to an accident, which is not part of the endpoint of interest. Another possibility is that there might be a time point at which the study finishes and thus if any individuals have not had their event yet, their event time will not have been observed. These incomplete observations cannot be ignored, but need to be handled differently. This is called censoring. Another feature of survival data is that distributions are often skewed (asymmetric) and thus simple techniques based on the normal distribution cannot be directly used.