The origin is the start of treatment.ĭata can be either right, left or interval censored. This data consists of survival times of 228 patients with advanced lung cancer. To illustrate time-to-event data and the application of survival analysis, the well-known lung dataset from the ‘survival’ package in R will be used throughout. This type of data can be analysed with a set of techniques known as ‘Survival Analysis’. The key attribute of this kind of missing data is that the time to the event is partly known – it is at least as long as the event-free period observed for the subject – in other words, it is censored. In a clinical trial, the origin might be randomisation or the start of a treatment, and the endpoint of interest might be disease diagnosis, the occurrence of an adverse event, disease progression, or even death. In this blog we focus on techniques for dealing with this, known as ‘Survival Analysis’.Ĭensoring occurs in time-to-event data (the time from a defined origin until the event of interest), when the event has not been observed (i.e.

In our last blog we discussed Handling Missing Data in Clinical Trials, and mentioned a kind of missing data known as ‘censoring’.

Rules of survival summary