When Should You Ask a Research Librarian to Review the Search?

When Is It Time to Have Someone Else Review Your Search?

You can design a competent database search yourself and still benefit from expert review. After spending hours developing terminology, adjusting Boolean logic, translating searches, and troubleshooting unexpected results, you become very familiar with your own strategy. That familiarity is useful, but it can also make assumptions and errors surprisingly difficult to see.

A research librarian or information specialist experienced in evidence searching can examine the strategy from a different perspective. They may notice a concept that has been translated too narrowly, a missing subject heading, an unnecessarily restrictive field, problematic Boolean grouping, an inappropriate filter, or database syntax that does not behave as you think it does.

The important question is not whether you are "bad at searching." Expert review is a quality-control step, particularly when the consequences of missing relevant evidence are substantial.

Ask for Review Before the Search Becomes Final

Search Review Is Methodological Quality Control, Not Just Technical Proofreading

For a systematic review, earlier involvement is usually better

If you are conducting a systematic review, waiting until the search is finished before contacting an information specialist can waste an opportunity.

The Cochrane Handbook recommends that review authors work closely with an experienced medical or healthcare librarian or information specialist from the protocol stage. Their contribution can extend beyond checking a final query. Depending on the project and available support, an information specialist may advise on which sources to search, help design database strategies, run searches, manage results, document the process, and contribute to search reporting.

This matters because some problems cannot be repaired elegantly at the end. If the wrong databases were selected, the concepts were structured poorly, or important terminology was overlooked from the beginning, proofreading the final Boolean syntax will not fully solve the problem.

Design assistance and peer review are related but different

It is useful to distinguish collaboration on the search from independent review of it.

The same project may benefit from both. Having an information specialist develop a search does not make quality control irrelevant, just as an experienced researcher can still benefit from manuscript peer review.

PRESS provides a structured framework for reviewing searches

The Peer Review of Electronic Search Strategies (PRESS) guideline was developed specifically for peer review of search strategies used in systematic reviews, health technology assessments, and other evidence syntheses.

The PRESS 2015 evidence-based checklist retained six major areas for review:

• translation of the research question;
• Boolean and proximity operators;
• subject headings;
• text-word searching;
• spelling, syntax, and line numbers;
• limits and filters.

This breadth is important. Search review is not merely a hunt for missing parentheses. A syntactically perfect strategy can still represent the research question poorly.

Ask for review when the concepts themselves are difficult to search

Some research questions translate naturally into searchable concepts. Others do not.

Expert review becomes particularly useful when a concept is difficult to define, crosses disciplinary boundaries, has extensive synonyms, depends on proximity relationships, or cannot be represented neatly through established controlled vocabulary.

If a concept seems too broad to search effectively, an information specialist can help determine whether the difficulty lies in terminology, search structure, or the conceptual scope itself. They cannot decide your substantive research question for you, but they can expose where its translation into searchable concepts becomes unstable.

Emerging and unstable terminology is a good reason to seek another perspective

A newer research area may use several competing expressions while lacking mature subject headings. An interdisciplinary topic may use entirely different vocabulary across fields.

When the topic has no stable terminology, another experienced searcher may identify synonyms, historical expressions, controlled-vocabulary terms, or disciplinary language that your initial literature sample did not reveal.

Similarly, if the concept is too new for a dedicated subject heading, expert input can help determine how existing controlled vocabulary and free-text terminology should be combined without pretending that a neighboring heading is an exact substitute.

Unexpected retrieval is another useful trigger

If your search retrieves enormous amounts of irrelevant material, misses papers you know should be found, or behaves in ways you cannot explain, review can help distinguish conceptual problems from technical ones.

A librarian may inspect whether irrelevant retrieval is being driven by particular terms, fields, or operators. Conversely, a known paper that disappears may reveal missing terminology, inappropriate field restrictions, indexing issues, or an overly restrictive concept block.

You do not necessarily need expert review every time one keyword misbehaves. The case becomes stronger when the cause remains unclear after systematic troubleshooting or when the consequences of missing evidence are substantial.

Database translation is a particularly error-prone stage

A search that works in one database cannot usually be copied literally into another. Controlled vocabularies, fields, truncation, wildcards, proximity operators, filters, and other search functions vary across platforms.

If a PubMed strategy fails when translated elsewhere, an information specialist familiar with the destination database may identify where conceptual equivalence has been lost.

Likewise, dramatically different result counts across databases may be perfectly legitimate, but they can also reveal mistranslated subject headings, fields, or operators. An experienced searcher can help separate those explanations.

Ask for help before repeated revisions turn into accumulated complexity

There is a point at which self-troubleshooting produces diminishing returns. You fix one problem, create another, add an exception, and eventually struggle to remember why several parts of the query exist.

If the strategy has become difficult to explain or maintain, outside review can be more productive than another layer of corrective logic.

The same applies when repeated revisions are no longer producing understandable improvements. At that stage, a reviewer may see a structural problem that is difficult to recognize from inside the revision history.

Peer review should happen while changes can still be made

For systematic reviews, Cochrane strongly recommends peer review before searches are run. The PRESS guideline development work also found that surveyed experts generally favored reviewing the MEDLINE strategy after it had been prepared but before translation into additional databases.

That timing has a practical advantage. If the primary strategy contains a conceptual or technical error, correcting it before translation prevents the same problem from being reproduced across several databases.

Reviewing only after screening has begun is still better than ignoring a discovered problem, but remediation becomes more expensive because searches may need to be rerun, records re-exported, duplicates managed again, and screening supplemented.

A librarian cannot certify that your search found everything

Expert review improves quality control, but it does not turn literature searching into a guarantee of complete retrieval.

Relevant studies may still be difficult to identify because of database coverage, indexing, terminology, publication practices, or the limitations of bibliographic searching itself. Search quality should therefore not be reduced to a claim that an expert has "validated" the strategy as exhaustive.

A more defensible interpretation is that the strategy has undergone informed methodological scrutiny and that identifiable weaknesses have been addressed where possible.

Give the reviewer enough context to do meaningful work

A reviewer needs more than the final search string. Provide the research question, protocol or eligibility criteria where available, the database and platform, the complete strategy, important limits or filters, and several known relevant papers if you have them.

Also explain unresolved concerns. If one term creates substantial noise or one concept consistently loses relevant papers, say so. This allows the reviewer to examine both the formal strategy and the behavior that prompted concern.

Bring in Expert Review Before Translating a Difficult Search

The reviewer did not replace the team's substantive judgment about what belonged in the review. The value of the review was in testing whether those substantive decisions had been translated faithfully into information-retrieval logic.

Common Misconceptions About Librarian Search Review

Match the Level of Search Review to the Consequences of Getting It Wrong

Not every exploratory database search requires formal peer review. If you are looking for a few papers to understand an unfamiliar topic, iterative searching may be entirely appropriate. The case for expert involvement becomes stronger as comprehensiveness, reproducibility, and the consequences of missed evidence become more important.

If formal peer review is used, document it. Cochrane points to PRISMA 2020 and PRISMA-S guidance recommending that authors describe how the strategy was developed and, where applicable, how it was peer reviewed and which tool, such as PRESS, was used.

Is It Time to Ask a Research Librarian?

Questions About Research Librarians and Search Review

Do Not Reserve Expert Search Review for When Everything Has Gone Wrong

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.