Free scoping read. A written assessment of your manuscript in two business days, at no cost.

Start one
Contents

Academy · Research Data, Analytics & AI · Literature Search Strategy

The Comprehensive Literature Search as the Foundation of a Systematic Review: Database Coverage, Grey Literature and Documentation

A prospective study of 58 published systematic reviews found that searching MEDLINE alone missed as much as a third of the eligible evidence, whereas four core databases searched together recovered 98.3% of it. A recall gap of this size is large enough to alter the pooled result of a review.

Pubrica AcademyWritten by an evidence synthesis methodologist Reviewed bya systematic review data analyst Published10 October 2026 Last checked10 October 2026 Reading time8 minutes Sources10 peer-reviewed

A comprehensive literature search is the set of database, register, and grey-literature searches that a systematic review conducts to identify every study meeting its eligibility criteria, not only the studies that are most readily found. Its importance lies in the fact that a systematic review’s claim to being systematic, rather than a narrative summary of convenient sources, rests on the search being sufficiently exhaustive that any missing study is unlikely to be missing because of where the authors chose to look. A review’s pooled effect estimate is only as trustworthy as the completeness of the evidence base from which it is calculated, and the best-documented empirical study of database coverage shows that the choice of databases alone can shift a review’s recall of eligible studies from under 60% to over 98%.

Learning objectives

  • The measured size of the recall gap between searching one database and searching a validated four-database combination
  • Which specific databases that empirical study found necessary to reach near-complete recall, and what each one uniquely contributed
  • Why grey literature and registered trials require a separate search strategy rather than an afterthought

Common misconceptions

  • “Searching PubMed/MEDLINE thoroughly is enough to call a systematic review's literature search comprehensive.” Bramer and colleagues found that Embase alone identified 85.9% of the 1,746 known-relevant references across 58 systematic reviews, and that reaching 98.3% recall required Embase, MEDLINE, Web of Science, and Google Scholar together, with 292 references (17% of the total) findable in only one of the four databases.

Measured recall difference between one database and four

The clearest evidence that database choice changes a systematic review's results comes from a prospective study by Wichor Bramer and colleagues, who tested eight databases against 58 published systematic reviews containing 1,746 relevant references identified by means other than the databases themselves1. Searched alone, Embase recovered 85.9% of the known-relevant references, the strongest single-database performance in the study1. Adding MEDLINE to Embase raised recall to 92.8%, and adding either Google Scholar or Web of Science to that combination increased recall to 95.9%1. The full four-database combination, Embase, MEDLINE, Web of Science, and Google Scholar, reached 98.3% overall recall, with a 100% median recall per individual review1. The same study examined 200 recently published PubMed-indexed systematic reviews to establish how authors search in practice, and found that MEDLINE was searched in 97% of them, but Embase in only 61%, Web of Science in 35%, and Google Scholar in just 10%1. Based on the recall curves it had measured, the study estimated that this typical practice gives only a 40% probability of achieving over 95% recall and a 23% probability of 100% recall, implying that roughly 60% of published reviews likely missed more than 5% of the relevant literature available to be found1.

Rationale for searching four databases

The same study found that 292 of the 1,746 relevant references, about 17% of the total, were indexed in only one of the four core databases, with Embase alone contributing 132 of those uniquely findable references1. That figure directly addresses the objection that searching several databases is wasted effort because they largely duplicate one another: a meaningful share of the eligible evidence base is indexed in exactly one database and nowhere else, so omitting any one of the four core databases does not merely save search time but removes studies that no combination of the remaining three can recover. The study did report a genuine trade-off in effort: the full four-database search carried a number needed to read of 73 irrelevant records for every one relevant reference found, compared with 57 for Embase alone1. That additional screening burden is the cost of comprehensiveness, and it is the reason search breadth is a deliberate methodological trade-off that a review team makes and documents, rather than an assumption that additional databases can be added at no cost. The Cochrane Handbook's own chapter on searching for studies reflects the same logic, recommending that reviews of interventions search at minimum CENTRAL, MEDLINE, and Embase, supplemented by additional sources appropriate to the review's topic and resources3.

Grey literature, trial registers and citation chasing

A comprehensive search strategy extends past bibliographic databases to sources that are structurally invisible to them: trial registers, conference abstracts, theses, regulatory submissions, and other grey literature that often contains unpublished or not-yet-indexed results78. This is important because published and unpublished studies are not a random sample of all the research actually conducted; studies with null or unfavorable results are disproportionately likely to remain unpublished or to appear only in a grey-literature source, so a database-only search systematically biases a review’s evidence base toward positive findings5. An empirical study by Egger and colleagues found that excluding grey literature and non-English-language studies from meta-analyses materially changed some pooled effect estimates, which is the direct empirical basis for treating grey-literature searching as a required step rather than an optional supplement5. Checking the reference lists of included studies and of prior systematic reviews on the same question, a technique generally called citation chasing or snowballing, recovers additional eligible studies that a database search's indexing terms alone did not surface, and Royle and Milne's study of search methods used in Cochrane reviews found that reference-list checking of relevant systematic reviews specifically identified trials that the database searches themselves had missed6.

A comprehensive search is verifiable and reproducible only if every database, the exact search string run in each, the date it was run, and any limits applied are reported in sufficient detail that another reviewer could rerun the same search and retrieve the same records. The PRISMA 2020 statement requires this level of search documentation as a reporting item2, and its dedicated extension, PRISMA-S, was developed specifically because the single search-reporting item of the base checklist was insufficiently granular to allow readers to audit a review’s search independently4. The PRESS (Peer Review of Electronic Search Strategies) guideline, a peer-reviewed checklist designed for the review of a draft search strategy by a second, experienced search specialist before it is run, exists because search strategies routinely contain errors (missed synonyms, incorrect Boolean logic, or wrong field tags) that an independent reviewer detects more reliably than the original author alone9. Treating search documentation and peer review of the search strategy as a required, auditable step, rather than an undocumented internal process, converts the statement “we searched comprehensively” from an assertion into a claim that readers can verify.

Field-specific sources and updating of the search

The four-database core that Bramer and colleagues tested, Embase, MEDLINE, Web of Science, and Google Scholar, was validated across a broad mix of health-related topics, but their study tested eight databases in total and found that field-specific sources still added value for particular subject areas; a comprehensive search strategy must therefore be tailored to the review’s topic rather than treated as a single fixed formula applied identically regardless of subject1. A review on a nursing-practice question benefits from adding CINAHL, a review touching psychological or behavioral outcomes from adding PsycINFO, and a review on an agricultural or environmental exposure from adding a subject-specific index those four core sources do not cover well, because each of these specialist databases indexes journals and grey literature that the broader multidisciplinary databases index incompletely or not at all8. Deciding which, if any, field-specific sources to add is itself part of the documented search strategy, not an afterthought, and the Cochrane Handbook's guidance to supplement the minimum core with sources "appropriate to the review's topic" is deliberately written to require this topic-specific judgment rather than prescribing one universal list for every review regardless of subject matter3.

A second, often-overlooked aspect of comprehensiveness is timing: a search conducted early in a review’s life cycle can be months or, for a slow-moving review, well over a year out of date by the time the review is ready for submission, during which new eligible studies may have been published. The Cochrane Handbook specifically recommends rerunning the search close to the point of publication and screening any newly identified records before the review is finalized, precisely because a comprehensive search frozen at an early date may have become incomplete by the time readers see the published review3. For reviews on fast-moving topics, where new trials are published frequently enough that even a few months’ gap meaningfully changes the evidence base, documenting both the original search date and the date of any update search, and reporting whether the update search changed the review’s conclusions, provides readers with the information needed to judge how current the review’s evidence base was at publication, rather than leaving them to assume that a single reported search date covers the entire period up to print. The full Cochrane Handbook treats this update-search requirement as integral to the rest of its search-and-selection guidance rather than as an optional closing step that a review may omit once the main search is complete10.

Key points

Key points

Database selection alone can move a systematic review’s recall of eligible studies from under 60% to over 98%. The search strategy — the choice of databases and grey-literature sources and the documentation of the strategy — should therefore be treated as a methodological decision with a measurable effect on the conclusions of the review, not as an administrative step.

Practical application

Practical application

  1. Reviews should search, at minimum, Embase, MEDLINE, Web of Science, and a citation index or Google Scholar. This is the specific four-database combination shown to reach 98.3% recall in a controlled study against 58 published systematic reviews.
  2. Time should be allocated for grey literature and citation chasing in addition to database searching. Trial registers, conference abstracts and the reference lists of included studies and of prior reviews on the same question should be checked, since each recovers studies that database searching alone does not identify.
  3. The search strategy should be peer reviewed by a second experienced searcher before it is run. A structured checklist such as PRESS should be used, and every database, search string, date of search and limit should be documented so that the search can be audited and rerun.

Frequently asked questions

For a review aiming at comprehensive, publishable evidence synthesis, the empirical evidence indicates that it is not: MEDLINE alone was not even the strongest single database in the controlled comparison, and the standard recommendation of the Cochrane Handbook is to search at minimum CENTRAL, MEDLINE, and Embase together.

It improves recall of eligible studies, but each added database also adds screening burden, so the decision should be based on topic-appropriate sources and documented reasoning rather than on searching every available database regardless of relevance.

Because published, peer-reviewed studies are not a representative sample of all research conducted on a question; studies with null or unfavorable results are more likely to remain unpublished or to appear only in grey literature, so omitting these sources biases a review’s evidence base toward positive results.

Citation chasing means checking the reference lists of included studies and of prior reviews on the same topic; it has been shown to recover eligible studies that database search terms did not surface, because indexing and keyword assignment are imperfect.

Every database searched should be reported, together with the exact search string used in each, the date it was run and any limits applied; where available, a second search specialist should review the draft strategy using a structured checklist such as PRESS before it is finalized.

References

  1. Bramer WM, Rethlefsen ML, Kleijnen J, Franco OH. Optimal database combinations for literature searches in systematic reviews: a prospective exploratory study. Systematic Reviews. 2017;6:245. read 10 October 2026. View on Systematic Reviews
  2. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. read 10 October 2026. View on The BMJ
  3. Lefebvre C, Glanville J, Briscoe S, et al. Chapter 4: Searching for and selecting studies. In: Higgins JPT, Thomas J, Chandler J, et al (eds). Cochrane Handbook for Systematic Reviews of Interventions version 6.4. Cochrane, 2023. read 10 October 2026. View on Cochrane Training
  4. Rethlefsen ML, Kirtley S, Waffenschmidt S, et al. PRISMA-S: an extension to the PRISMA statement for reporting literature searches in systematic reviews. Systematic Reviews. 2021;10:39. read 10 October 2026. View on Systematic Reviews
  5. Egger M, Juni P, Bartlett C, Holenstein F, Sterne J. How important are comprehensive literature searches and the assessment of trial quality in systematic reviews? Empirical study. Health Technology Assessment. 2003;7(1). read 10 October 2026. View on NIHR Health Technology Assessment
  6. Royle P, Milne R. Literature searching for randomized controlled trials used in Cochrane reviews: rapid versus exhaustive searches. International Journal of Technology Assessment in Health Care. 2003;19(4):591-603. read 10 October 2026. View on Cambridge University Press
  7. Paez A. Gray literature: An important resource in systematic reviews. Journal of Evidence-Based Medicine. 2017;10(3):233-240. read 10 October 2026. View on Journal of Evidence-Based Medicine
  8. Godin K, Stapleton J, Kirkpatrick SI, Hanning RM, Leatherdale ST. Applying systematic review search methods to the grey literature: a case study examining guidelines for school-based breakfast programs in Canada. Systematic Reviews. 2015;4:138. read 10 October 2026. View on Systematic Reviews
  9. McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. PRESS Peer Review of Electronic Search Strategies: 2015 guideline statement. Journal of Clinical Epidemiology. 2016;75:40-46. read 10 October 2026. View on Journal of Clinical Epidemiology
  10. Higgins JPT, Thomas J, Chandler J, et al (eds). Cochrane Handbook for Systematic Reviews of Interventions version 6.4. Cochrane, 2023. read 10 October 2026. View on Cochrane Training

Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.

How to cite this article

Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.

Pubrica Academy. (2026, October 10). The Comprehensive Literature Search as the Foundation of a Systematic Review: Database Coverage, Grey Literature and Documentation. Pubrica. https://pubrica.com/academy/comprehensive-literature-search-in-systematic-review/

This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.

Comprehensive Search Readiness Checklist

This checklist should be completed before a systematic review’s search strategy is finalized.

[ ] Searched at minimum Embase, MEDLINE, and a citation index (Web of Science and/or Google Scholar)
[ ] Searched at least one trial register relevant to the review topic
[ ] Searched grey-literature sources (conference abstracts, theses, regulatory submissions) where relevant
[ ] Checked reference lists of all included studies and of prior reviews on the same question
[ ] Had a second, experienced searcher peer-review the draft search strategy
[ ] Documented every database, exact search string, date run, and limits applied

Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.

How Pubrica can help

Pubrica’s evidence synthesis team designs and peer-reviews multi-database search strategies, including grey-literature and register searches, so that the evidence base of a systematic review is defensible against the recall gap documented in controlled comparisons.

Find something else

Search the Pubrica Academy

Access is free and requires no account. Enter a topic, or open the full library to filter by category, sub-topic or resource type.

PRESS-style peer review of search strategies

A scoping assessment is free and takes two business days. It provides a written assessment of what the work requires and which specialist would undertake it.

Get a free quote

Regulations

Check your project against this rule

  1. 1Your work
  2. 2Your details
  3. 3Send
What do you need?

Material you send is held under signed confidentiality, with access limited to the assigned team and encrypted transfer. It is not used to train anything and is not passed to third-party tools. Ask for the agreement before you send anything you would rather cover first.