Objective 　To calculate the typical time taken to complete protocols and reviews, to track how often reviews are updated and compare with Collaboration policies, to feed back any gaps in documentation of dates to individual Cochrane Collaborative Review Groups, and to suggest changes to presentation of reviews and to editorial processes.
Methods 　Data were extracted either from The Cochrane Library or a specially constructed database to calculate the age of reviews and protocols, and the frequency of updating of reviews. Issue 1, 2003, with 1 596 reviews and 1 200 protocols, was used as the index issue.
Results 　Median number of issues between a protocol and its completed review being published on The Cochrane Library is 5 (1.25 years). 65% of protocols have appeared on The Cochrane Library for no longer than two years, but the number of protocols more than two years old is probably increasing. One-third of reviews have been substantively updated, but generally only once and often within several months of the first publication of the review. The number of out-of-date reviews is probably increasing.
Conclusions 　While the stage between publishing a protocol and the completed review is usually completed in no longer than two years, the number of out-of-date reviews and protocols requires continuing attention. How up-to-date a review is depends on when the evidence base was last searched and when additional relevant evidence has been integrated into the review. This needs to be reflected in the information provided to readers of Cochrane Reviews and some alternative ways of presenting the components of this information are given. More accurate and complete reporting will also allow the Collaboration to track progress against policy.

Citation： Philippa Middleton. Life Cycle of Cochrane Reviews. Chinese Journal of Evidence-Based Medicine, 2004, 04(7): 488-492. doi: Copy