When is an old systematic review worth updating?

Systematic reviews go out of date. Studies of how quickly this happens have found that a substantial minority of reviews show a signal that their conclusions need updating within a few years of publication, and that the time varies widely by field. There is no age at which a review becomes obsolete. What matters is whether new evidence could change the answer.

An update is a legitimate and efficient project: the question, the protocol and the search strategy already exist, and the novelty comes from the new studies. It is also one of the four kinds of novelty SynthGap looks for in a topic.

Signals that an update is worth doing

New randomized trials or large studies have been published since the last search date, especially if they are large or contradict the review.
The review's conclusion was uncertain or based on few studies, so one new study could change it.
Practice has changed: new drugs, new devices, new definitions of the condition or new standard comparators.
The topic is still an active question: guidelines or clinicians still cite the review, or are uncertain about the answer.
The review used methods that are now outdated, for example no risk of bias assessment, no certainty grading, or a pooling method that is no longer recommended.

How to check what has changed

Note the date of the review's last search, not its publication date. Evidence published after the search date is what an update would add.
Search PubMed, trial registries and a citation index for studies that cite the review, or that match its question, and that appeared after the last search.
Count the new eligible studies and estimate their weight compared with the original ones.
Check whether anyone has started an update: PROSPERO, the Cochrane Library (where Cochrane reviews have a status) and the original authors.
Contact the original authors. They can share the protocol and the search strategy, and may want to join.

Let SynthGap run the first pass

Enter an area. SynthGap proposes candidate topics, checks each against existing systematic reviews, and lists the primary studies behind it. Start on the Free plan: no card needed, and every run shows every topic it finds.

Update, new review, or living review?

Update: keep the original question and methods and add the new evidence. It is the simplest, and the most credible if the original methods are sound.
New review: choose it when the question or the methods need to change substantially, for example a new population or outcome.
Living systematic review: continually updated as new evidence appears. It suits fast-moving questions where decisions depend on it, but it needs an ongoing team and a plan to maintain it.

Reporting an update

Say that the review is an update, cite the original, give the new search dates and report how many studies were new. PRISMA 2020 asks you to report the dates of the searches and any changes from the registered protocol. Showing the original and updated results side by side lets readers see what changed.

Questions

How old is too old for a systematic review?

There is no fixed age. A review from several years ago in a field with no new trials may still be current, while a review from last year can be out of date if a large trial has just reported. The date of the last search and the amount of new evidence matter more.

Do I need the permission of the original authors?

Not for an independent update based on published methods, but it is good practice to contact them. For updates of Cochrane reviews, the Cochrane editorial process applies.

Is an update novel enough to publish?

Journals differ, but an update that adds substantial new evidence and changes or confirms an important conclusion is publishable, particularly if it improves the methods.

Keep reading

General methodological guidance, not medical advice. SynthGap's suggestions are AI-generated from the published literature, so verify them before registering a protocol.