01 · The Question
When Does a Finished Literature Search Need to Be Run Again?
You completed the search, screened the results, synthesized the literature, and moved on to writing. Then time passes.
A few weeks may not matter in a stable field. Several months can matter considerably in a rapidly developing one. A major study may appear, terminology may change, a database may add records, or an ongoing study you identified earlier may finally publish its results.
Updating a literature search is therefore not simply a matter of deciding that the original search has become "old." The more useful question is whether new evidence appearing since the last search could materially affect the accuracy, completeness, or conclusions of the work you are preparing to submit, publish, or rely upon.
03 · What You Need to Know
A Search Date Is a Boundary on the Evidence You Could Have Found
Your Literature Search Does Not Stay Current Automatically
A literature search captures what your information sources could retrieve when you searched them. Research published or indexed afterward was not available to that search.
This seems obvious, but it has an important consequence. A literature review can be methodologically sound when completed and become less current simply because the evidence base changes.
Cochrane's guidance on updating reviews makes this concern explicit: as new studies are completed, review findings may become out of date and potentially misleading to decision makers. The need to update therefore depends partly on whether new information could meaningfully alter what the review says.
There Is No Universal Expiration Date for a Literature Search
A search does not suddenly become invalid after three, six, or twelve months for every type of research.
The pace of evidence production varies enormously. A search concerning an emerging technology, new treatment, fast-changing policy, or rapidly expanding research topic can become dated quickly. A tightly focused historical or theoretical question may change much more slowly.
The type of review matters as well. A systematic review making claims about the current evidence base generally requires greater attention to search currency than a historical discussion of how a theory developed.
Search age
The amount of time that has passed since the literature was searched.
Search currency
How adequately that search still represents the evidence needed for the review at the point it will be submitted, published, or used.
A six-month-old search can be meaningfully outdated in one field and entirely adequate in another. Time matters because it creates an opportunity for the evidence to change, not because age alone determines quality.
Formal Review Standards May Specify How Recent the Search Must Be
If you are conducting a systematic review or another review governed by explicit methodological standards, those standards should guide the update decision.
For Cochrane intervention reviews, current MECIR conduct standards require authors to rerun or update searches for all relevant sources within 12 months before publication of the review or review update and screen the results for potentially eligible studies. The guidance states that six months is preferable.
Cochrane further recommends fully incorporating newly identified studies where feasible, while recognizing that authors sometimes need to balance incorporation against delaying publication.
These requirements apply specifically within that methodological context. They should not be converted into a universal rule that every thesis, narrative review, or research paper must rerun its search every six or twelve months.
The Faster the Field Changes, the Sooner You Should Consider Updating
Calendar time becomes more consequential when research is appearing rapidly.
Suppose you are reviewing generative artificial intelligence in education during a period of intense publication activity. A search conducted many months before submission may omit a substantial body of new evidence. The same interval may produce little change in a mature and narrowly defined literature.
Ask what has plausibly happened since the search. Have major studies been announced? Have conferences generated new research? Has a new technology or policy changed the phenomenon? Are journals publishing frequently on the topic? Were relevant ongoing studies close to completion?
The answers help determine whether a top-up search is likely to produce useful new evidence.
Major Changes in the Field Can Trigger an Update Even When Little Time Has Passed
Elapsed time is not the only trigger.
A landmark trial, major policy change, new technology, methodological development, public event, or influential publication may alter a literature quickly. If such a development is directly relevant to your research question, waiting for an arbitrary update interval makes little sense.
Similarly, if you discover that researchers have begun using terminology absent from your original strategy, you may need more than a simple date-limited update. The search strategy itself may require revision.
Updating a Search Does Not Necessarily Mean Starting From the Beginning
If the original search was sound and the review question has not changed, an update can often focus on records added since the previous search.
Cochrane notes that searches for review updates may naturally use limits based on the date records entered a database rather than publication date to reduce duplication. Its guidance for updating reviews also describes rerunning an existing strategy over the period following the most recent search as a possible starting point.
The exact procedure depends on the database because indexing dates, publication dates, and search fields are not interchangeable. Search documentation from the original review is therefore extremely valuable.
| Situation |
Likely Search Response |
| Same question, same terminology, short interval |
Run an appropriately designed top-up or update search |
| New terminology has emerged |
Revise the strategy and consider whether older records should also be searched using the new terms |
| Eligibility criteria have changed |
Reassess the original strategy because previously irrelevant records may now qualify |
| Important new information source becomes relevant |
Search that source for the period necessary to provide adequate coverage |
| Major new study or development appears |
Update promptly and investigate related new evidence |
| Long delay between search and submission or publication |
Rerun or update appropriate searches before finalizing the work |
Changes to the Research Question May Require More Than an Update
A top-up search assumes that the original search still addresses the question you are asking.
If your population, intervention, exposure, outcome, phenomenon, eligibility criteria, or conceptual scope has changed, simply searching the most recent months may leave older relevant literature undiscovered.
Cochrane's update guidance specifically notes that important changes to PICO elements or eligibility criteria, as well as new terminology, may require portions of the search to be rerun back to the earliest records.
In other words, sometimes you need an updated search. Sometimes you need an updated search strategy.
Database Changes Can Also Matter
Databases themselves change. Controlled vocabulary evolves, interfaces change, indexing practices are revised, and search syntax can be modified.
Cochrane therefore recommends consulting an information specialist when updating a systematic review because an older strategy may require amendments even if the underlying review question remains unchanged.
For a thesis or less formal review, the same principle applies at a smaller scale: do not assume that copying and pasting a search string from a year ago will always reproduce an equivalent search today.
An Update Search Is Not Complete Until You Examine What It Found
Running the query again is only retrieval. The new records still need to be screened and, when relevant, incorporated into the work.
For Cochrane intervention reviews, updated results must be screened for potentially eligible studies. Ideally, newly eligible studies should be fully incorporated. The methodological details will differ for other review types, but the logic remains: finding new literature and then ignoring it does not make the review current.
A New Study Does Not Automatically Require Rewriting Everything
An update may retrieve new evidence without materially changing the review.
A new study might reinforce the existing conclusion, improve confidence in an interpretation, extend evidence to another population, or simply provide another example of a well-established pattern. Another study might substantially alter an effect estimate, expose a contradiction, or undermine a central claim.
Assess what the new evidence does to the synthesis rather than assuming that every newly published paper requires a complete reconstruction.
Search Updates and Full Review Updates Are Different Decisions
You may run a search to determine whether meaningful new evidence exists without immediately undertaking a complete new edition of a review.
Cochrane distinguishes a scoping search used to assess whether an update may be needed from a full update in which new search results are incorporated. More broadly, this illustrates a useful distinction: checking for new evidence and fully updating a synthesis are related but separate activities.
Document the Update
Record when the update was run, which sources were searched, what date range or other limits were used, whether the strategy changed, and how the new records were handled.
This documentation helps readers understand the currency of the evidence and makes later updates considerably easier. Cochrane requires sufficient documentation for searches to be reported correctly and reproduced to the extent possible.
Watch Out
Do not change the search strategy during an update without documenting why. A modified strategy may retrieve different records because the evidence base changed, because the database changed, or simply because you changed the search. Those possibilities should not be silently conflated.
04 · A Practical Example
Updating a Search Before a Dissertation Is Submitted
Hypothetical Example
A Literature Search Completed Eight Months Ago
Suppose a doctoral researcher completed the main literature search eight months ago for a dissertation examining university students' use of generative AI. The literature review has since been written, but the dissertation is only now approaching submission.
Assess the field The researcher recognizes that the topic has continued developing rapidly and that many new studies are likely to have appeared since the original search.
Review the original strategy The researcher checks the databases, terminology, search dates, and saved search strings used previously.
Check for terminology changes New terms and more specific descriptions of generative AI use have become common. The researcher determines whether the strategy needs expansion rather than simply rerunning it unchanged.
Run the update Appropriate searches are rerun for the period since the previous search, with necessary adjustments documented.
Screen the new literature Several new studies reinforce existing themes, while two introduce findings about assessment design that require qualification of part of the literature review.
Revise selectively The researcher updates the relevant synthesis, citations, discussion of the evidence gap, and any claims that were framed as describing the current state of research.
The eight-month interval alone did not prove that an update was required. The combination of elapsed time, rapid publication activity, and the dissertation's claims about a fast-changing evidence base made updating prudent.
