-- PeterBacchetti - 16 Jun 2012

In a study of how long participants survive or how long they go without experiencing a particular event of interest, fully describing how long they were observed is important. For those who experienced the event, follow-up is complete, regardless of when the event occurred, so follow-up time should be summarized for those who did not have the event while in the study (usually called censored observations). Often some of the study participants are lost to follow-up, resulting in shorter follow-up times than intended. These have greater potential to bias results, so describing them separately is usually warranted. The key elements to report are: For example:

Among 250 total subjects, 100 died during follow-up. Of the 150 not known to have died, 10 were lost to follow-up after a median of 2.5 years (range 0.4 to 5.1), and the remaining 140 were censored at the end of our follow-up period at 12/31/2007 with a median follow-up time of 3.7 years (range 2.0 to 6.9).