Members of Cochrane and others have developed techniques to manage Type I and II errors that can occur over time by updating and repeating meta-analyses. Cochrane has provided funding for a review of these techniques undertaken by Mark Simmonds at York University and other members of the Statistical Methods Group. Two principle techniques emerge as controlling for these errors: Trials Sequential Analysis and Sequential Meta-analysis. Further detailed information is available (see full statement below]). Some review authors have used these techniques, we do not currently encourage or discourage their use at this point.
Cochrane Scientific Committee recommendation (Full statement, July 2017):
Further technical examination of these two approaches is required before the Committee can decide whether there is a preferred approach or whether the methods provide added value to managing random error. An expert panel established will discuss further and report back to the Committee before arriving at a final decision.
Expert panel consensus statement (Full statement, December 2018):
The Expert Panel recommends against the use of sequential methods for updated meta-analyses in most circumstances within the Cochrane context. They should not be used for the main analyses, or to draw main conclusions.
If you have any questions, contact the Cochrane Methods Team (email@example.com).