A Reviewer Asks for Newer Evidence After Your Search Is Closed

Your manuscript is submitted, your search log is archived, and then a reviewer writes: the evidence base looks dated. Perhaps the review was screened months ago, or the reviewer points to a recent paper you did not capture. The search is "closed" in the sense that you already ran it, screened records, and wrote the synthesis. The task now is not to start over from a blank page. It is to reopen the search in a controlled way, document exactly what changed, and deliver a revised evidence base that a reader could audit.

This article walks through that repair. It uses a systematic-review manuscript as the running example, but the same decisions apply to a grant application, a lab handoff, or any project where a coauthor needs to see how a conclusion was reached. The goal is a corrected deliverable: an updated search, a dated record of the rerun, and a manuscript that states plainly what the new evidence does and does not change.

A closed search is a documented snapshot, not a permanent claim about the literature. Cochrane's guidance on searching for and selecting studies treats the search as something planned from eligibility criteria, run across multiple sources, and documented so the process can be followed. That same guidance recommends rerunning searches before publication. A reviewer asking for newer evidence is therefore asking you to do something the method already anticipates.

The practical problem is that your original search was built around a date, a set of databases, and a screening decision log. When you reopen it, you need to preserve the original snapshot while adding a new one. If you simply rerun a query and overwrite your notes, you lose the trail that made the first search credible.

Before running anything, establish what you actually have. The answer determines how much work the repair requires.

Check whether you can produce each of the following:

• The exact query strings, including any subject headings and free-text terms, and the date each was run.
• The databases and other sources searched, with the date limits applied.
• The number of records retrieved and the number removed as duplicates.
• The screening decisions, at minimum at the title-and-abstract stage, and the reasons records were excluded at full text.
• The version of the manuscript the reviewer commented on.

If any of these are missing, reconstruct them from your search details and screening files before rerunning. A rerun layered on an undocumented original is hard to defend.

Reviewer comments about currency usually fall into one of three shapes, and each implies a different repair:

1. A date gap. The search stopped at a date that now looks old. The repair is a date-limited rerun from the original end date to today.
2. A missing source. The reviewer names a paper or a journal you did not search. The repair is to check whether that source was in scope and, if so, to add it and document the addition.
3. A missing concept. The reviewer believes a relevant term or method was not captured. The repair is to revisit the eligibility criteria and the term list, then rerun with the added terms and record the change.

Naming which shape applies keeps the response proportionate. A date gap does not require rebuilding the entire strategy.

A rerun that tries to redo everything invites inconsistency. A rerun that changes too little may not satisfy the reviewer. Set the scope explicitly and write it down before you search.

Cochrane's guidance starts search planning from eligibility criteria. If the reviewer is asking about currency, the population, intervention, and outcome definitions in your protocol should stay as they were. Changing them mid-rerun turns a currency update into a new review, which is a different deliverable.

Use the end date of your original search as the start of the new window, so the two searches meet without a gap. If your original search used a date limit rather than a run date, state which one you are anchoring to. Record the new run date.

If the reviewer names a database or journal, check it against your original source list. Adding a source is a defensible response when the source covers your question and was omitted. Document the addition and the reason. If the source was deliberately excluded, say so in your response letter rather than silently adding it.

One of the most common gaps in a reopened search is that the query no longer matches what was actually run. PubMed's search details show how terms were translated and combined, which is exactly the record you need to reproduce a query faithfully. Use that view to confirm the final logic rather than retyping from memory.

When you add terms, keep the original free-text and subject-heading structure. Cochrane's guidance emphasizes combining free-text terms with subject headings; a rerun that drops one half of that combination is not comparable to the original. If you are searching through a scholarly search interface, run the same concept blocks you used before and note any interface-specific syntax differences.

For a fresh pass over the literature, WisPaper search provides a scholarly search interface where you can run the updated concept blocks and collect candidate records. Keep your query strings in your own log regardless of the interface, so the record survives any change of tool.

New records should go through the same staged screening as the original. Cochrane's guidance describes screening in stages and linking multiple reports of one study, which matters here: a recent paper may be a follow-up report of a study you already included, not a new study.

Apply the same inclusion and exclusion rules. For each excluded record, record a short reason. For each record moved to full text, record the identifier.

Before you add a "new study" to your synthesis, check whether it is another report of an included study. Counting reports as separate studies inflates your evidence base and is exactly the kind of error a careful reviewer will catch.

Cochrane's guidance includes deduplicating records. Your rerun will overlap with the original set at the boundary date. Remove those duplicates and record how many you removed, so the count of genuinely new records is transparent.

Once screening produces a small set of new included studies, keep them separate from the original set until you have decided how they affect the synthesis. A separate folder or collection makes the delta visible.

If you want the new records and the original included papers in one place, My Library lets you save or upload papers and organize them. Keeping the rerun records in a distinguishable group means a coauthor can see precisely what arrived after the original search closed.

When you need to interrogate the new papers against your existing synthesis, Library QA supports asking questions based on papers uploaded to My Library or added from search results. That is useful for checking whether a new study reports the outcome your synthesis depends on, before you commit to a revision.

This is the substantive step, and it should be recorded as a decision, not left implicit in a revised paragraph. For each new included study, decide one of the following and write it down:

• No change to the conclusion. The study is consistent with the existing synthesis. Note it as supporting evidence.
• Strengthened conclusion. The study adds precision or a new population consistent with the direction of the existing findings.
• Qualified conclusion. The study introduces a subgroup, setting, or outcome that narrows the scope of your claim.
• Reversed or contested conclusion. The study conflicts with the existing synthesis. This requires the most careful handling and a clear statement in the manuscript.

Record the decision and the reason next to each study. This decision log is the artifact that lets you answer the reviewer precisely and lets a coauthor follow your reasoning.

Revise the methods section to state the original search date, the rerun date, the sources searched in each pass, and the number of new records screened and included. Revise the results to report the new included studies. Revise the discussion only where the decision log says the conclusion changed.

In the response letter, describe the rerun as a dated event: what you searched, when, what you found, and what changed. Reviewers respond well to a specific account of the delta. A vague statement that the literature was updated is harder to evaluate than a short table of new records and their effect on the synthesis.

If your reference list grew during the revision, check the new entries before submission. TrueCite screens BibTeX citations against real academic databases to flag hallucinated references, which is a useful check when references have been added under time pressure.

The rerun is ready to submit when you can answer yes to each of these:

• The original search is still documented, and the rerun is documented separately with its own run date.
• The new window starts at the original end date, with no unexplained gap.
• Every new record has a screening decision and a reason.
• Linked reports of already-included studies are identified rather than counted as new studies.
• Each new included study has a recorded decision about its effect on the synthesis.
• The methods, results, and response letter all report the same numbers.
• The reference list has been checked for entries that do not correspond to real records.

If any item is missing, the gap is in the trail rather than in the search, and it is worth closing before you resubmit.

Three habits make the next reviewer request much cheaper.

First, treat the search as a dated artifact from the start. Record query strings, sources, run dates, and record counts as you go, not after a reviewer asks. PubMed's saved searches and search details support this, and Cochrane's guidance on documenting the process describes the same discipline.

Second, plan for a rerun before publication. Cochrane's guidance recommends rerunning searches before publication, so build the rerun into your timeline rather than treating it as an emergency.

Third, keep the delta visible. Whether the project ends in a manuscript, a lab handoff, or a new review, the person who inherits it needs to see what changed and why. A decision log next to the new records does that work.

For longer projects where the rerun is one part of a larger literature task, Scholar Agent supports literature review and related research tasks. If the project is organized around a shared workspace, Research Projects is where a new project can be created so the rerun records sit alongside the rest of the work.