Searching Grey Literature and Other Sources

Overview

Bibliographic databases are essential but incomplete. Relevant evidence may appear in trial registries, dissertations, conference proceedings, government reports, health-technology assessments, organisational documents, preprints or records that are difficult to retrieve through text searching. Supplementary methods such as citation searching and expert contact can identify additional sources.

“Grey literature” broadly refers to material produced outside conventional commercial or academic publishing channels. Grey status does not determine methodological quality. A government report may be rigorous, while a journal article may be poorly conducted. Every source must be appraised according to its design, provenance, methods and relevance.

Why this matters

Published journal literature may over-represent studies with statistically significant, favourable or novel findings. Relying only on indexed publications can therefore distort the evidence base. Registry entries may reveal completed but unpublished studies, changed outcomes or ongoing work. Reports and theses may contain detailed methods or local evidence not available in journal articles.

Searching beyond databases also creates challenges. Websites change, search engines personalise or rank results, grey documents may disappear, metadata may be incomplete and duplicate reports may describe the same underlying study. The team must plan sources, document actions and verify records rather than collect material indiscriminately.

Choosing supplementary sources

The additional sources should follow from the question and eligible evidence types.

Trial and study registries

Registries can identify ongoing, completed and unpublished studies. They may provide registration dates, planned outcomes, intervention details, recruitment status and sponsor information. For intervention reviews, search international and national trial registries where relevant. Registry records should be linked to publications using identifiers, investigators, sample sizes, dates and intervention details.

Theses and dissertations

Dissertations may contain detailed methods, null findings or locally relevant evidence. They can be located through institutional repositories, national thesis services and multidisciplinary dissertation databases. The final degree, institution, year and stable source should be recorded.

Conference proceedings and abstracts

Conference sources may reveal emerging or unpublished studies, but abstracts usually provide limited methods and results. Search discipline-specific conference archives, Embase conference records, society websites and cited proceedings where relevant. Seek a full report or contact authors before relying on an abstract for detailed conclusions.

Government and organisational reports

Policy evaluations, surveillance reports, technical guidance and programme assessments may be published by ministries, public-health agencies, non-governmental organisations and professional bodies. Search organisation websites, document repositories and targeted web queries. Verify the issuing body, publication date, version and whether the document was superseded.

Health-technology assessment and guideline sources

Health-technology assessment agencies and guideline developers may publish evidence reports, search strategies and evidence tables. These can identify studies and contextual information, although the review team must still assess date, scope and methods.

Preprints

Preprints can provide rapid access to research before peer review. They should be labelled as preprints, checked for later journal publication and interpreted cautiously because content may change. The protocol should state whether preprints are eligible and how updated versions will be handled.

Regulatory and safety sources

Questions about medicines, devices or harms may require regulatory assessments, safety communications or clinical study reports. Access and terminology vary, and specialist guidance may be needed.

DECISION GUIDE

Match supplementary sources to the review question

• Intervention effectiveness: trial registries, conference proceedings, regulatory sources and citation searching.
• Policy or service evaluation: government, agency and organisational reports.
• Local or regional evidence: institutional repositories, regional indexes and ministry websites.
• Emerging evidence: preprint servers, registries and recent conference sources.
• Qualitative or complex topics: theses, citation searching, expert recommendations and targeted website searches.
• Harms: regulatory documents, trial registries and broader study-design sources.

Do not search every possible source by default. Pre-specify the sources most likely to contribute relevant evidence.

Searching the web systematically

A web search should be planned rather than conducted as an undocumented series of clicks. Record:

• search engine or website;
• exact query or terms;
• date searched;
• filters or site restrictions;
• number of result pages or records screened;
• stopping rule;
• documents retained; and
• archived link or downloaded file where permitted.

Targeted syntax such as site:gov or a specific organisation domain can improve focus. Search several plausible phrasings because web indexing and ranking differ from bibliographic databases. CADTH’s Grey Matters tool provides a structured list of health-related grey-literature sources and encourages systematic documentation [1].

Grey-literature search log

illustrative

Table 1. Example supplementary-search record

Source	Search or navigation method	Date	Stopping rule	Records retained
WHO ICTRP	Gestational diabetes AND postpartum	20 Jul 2026	All returned records screened	4
Ministry website	Site search plus targeted Google query	21 Jul 2026	First 10 result pages and all programme-report sections	3
University repositories	Keywords and subject browsing	22 Jul 2026	All records to 2026 screened	6
Conference archive	Meeting years 2018–2026	23 Jul 2026	All relevant abstract categories screened	5

Note: The values are illustrative. Search limits should be justified for the actual review.

Citation searching

Citation searching starts from relevant seed references.

• Backward citation searching retrieves references cited by a seed article.
• Reference-list checking is manual inspection of the seed article’s bibliography.
• Forward citation searching retrieves later records that cite the seed article.
• Iterative citation searching repeats the process using newly included records.

The TARCiS statement standardises terminology and provides recommendations on when, how and how transparently to report citation searching [2]. Citation searching is generally supplementary; it should not replace comprehensive database searching when completeness is required.

Citation indexes differ in coverage. When feasible, using more than one source can improve retrieval. Record the citation index, date, direction, seed references, number of iterations and screening process.

Contacting experts and authors

Contact may be justified to:

• locate an inaccessible report;
• clarify whether several records describe one study;
• obtain current status of a registered study;
• identify a full report behind a conference abstract; or
• ask about missing outcome information.

Use a standard, professional message and retain a contact log. Expert recommendations should supplement, not replace, planned searching. Authors may remember familiar or positive studies more readily than unpublished or contradictory work.

Verifying and linking records

A source should be verified before inclusion. Check:

• authors or responsible organisation;
• title and publication year;
• persistent identifier or stable location;
• study design and participants;
• relationship to other reports;
• version or publication status;
• whether corrections, retractions or final publications exist; and
• whether the source contains sufficient information for the planned appraisal and synthesis.

Stopping decisions

Unlike database searches, web and organisational searching may have no natural endpoint. Predefine pragmatic rules such as:

• screen all results from a registry query;
• examine specified website sections;
• screen the first stated number of web-result pages;
• search a defined list of organisations;
• stop iterative citation searching when a full iteration yields no new eligible records; or
• stop expert contact after a defined number of attempts.

Stopping rules do not guarantee completeness. They make the limitations transparent and reproducible enough for appraisal. Current Cochrane and PRISMA-S guidance also requires supplementary sources and search methods to be planned and reported transparently [3,4].

PROTOCOL CHECKLIST

Plan non-database searching

• Identify which grey-literature types are eligible.
• Select sources linked to the review question.
• Define search queries, years and website sections.
• Define seed references and citation-searching methods.
• State how preprints and conference abstracts will be handled.
• Create procedures for linking multiple reports of one study.
• Define contact and stopping rules.
• Save documents, metadata and search logs securely.
• Plan how unstable URLs and updated versions will be checked.

Grey literature and publication biasMMed extension

Grey-literature searching can reduce dependence on journal publication, but it does not automatically remove publication bias. Unpublished studies may remain undiscoverable, registries may be incomplete and grey sources may selectively report outcomes. The review should describe the search and evaluate the remaining risk rather than claim that bias has been eliminated [5,6].

Worked example

WORKED EXAMPLE

Supplementing the postpartum follow-up search

Question: What interventions improve completion of postpartum glucose testing after gestational diabetes?

Supplementary plan:

1. Search international trial registries for completed, ongoing and unpublished interventions.
2. Search Embase conference records and relevant diabetes and obstetric conference archives.
3. Search health-ministry, diabetes-association and maternity-quality websites for programme evaluations.
4. Search theses repositories for intervention and implementation studies.
5. Conduct backward and forward citation searching from all included studies and recent relevant reviews.
6. Link registry, abstract and article records using investigators, sample sizes, intervention details and dates.
7. Contact investigators where a completed registry has no accessible results.

Interpretation: A conference abstract can establish that a project existed, but insufficient reporting may prevent risk-of-bias assessment or effect estimation. The review records this limitation instead of treating all sources as equivalent.

Appraising supplementary searching

APPRAISAL CHECKLIST

Questions for grey-literature and supplementary searches

• Were the selected sources appropriate for the question?
• Were search dates, queries, navigation routes and limits recorded?
• Were citation-searching seed records and directions reported?
• Were preprints, abstracts and unpublished data clearly labelled?
• Were multiple reports linked to one underlying study?
• Were source authenticity, version and later publication checked?
• Were stopping rules stated?
• Did the review acknowledge what may still have been missed?

Suggested protocol or reporting wording

Common mistakes

Knowledge check

Why might a trial registry be useful even when a journal article is available?

Answer and reasoning

The registry can show prospectively registered outcomes, recruitment status, protocol details and dates. Comparing it with the publication may reveal changes or provide information not reported in the article. The records should be linked as reports of one study.

What is forward citation searching?

Answer and reasoning

It retrieves later records that cite a selected seed reference. It differs from backward citation searching, which retrieves references cited by the seed record.

Can a conference abstract be included in a review?

Answer and reasoning

It can be eligible if the protocol allows it, but the information may be insufficient for appraisal or synthesis. The team should search for a full report, registry entry or later publication and state any remaining uncertainty.

Why are stopping rules needed for website searching?

Answer and reasoning

Website and general web searches may produce changing or effectively unlimited results. A stated stopping rule makes the process more transparent and allows readers to judge the practical limits and risk of missed evidence.

Key take-home messages

SUMMARY

Key take-home messages

• Databases do not contain all relevant health evidence.
• Select grey-literature and supplementary sources according to the review question.
• Document web queries, sources, dates, result limits and stopping rules.
• Use citation searching as a supplementary method and report its direction and seed records.
• Verify source versions and link multiple reports to the same study.

References

1. Canadian Drug Agency. Grey Matters: a practical tool for searching health-related grey literature. Ottawa: Canada’s Drug Agency. https://www.cda-amc.ca/grey-matters-practical-tool-searching-health-related-grey-literature
2. Hirt J, Nordhausen T, Fuerst T, Ewald H, Appenzeller-Herzog C, TARCiS Study Group. Guidance on terminology, application, and reporting of citation searching: the TARCiS statement. BMJ. 2024;385:e078384. https://doi.org/10.1136/bmj-2023-078384
3. Lefebvre C, Glanville J, Briscoe S, Featherstone R, Littlewood A, Metzendorf M-I, et al. Technical Supplement to Chapter 4: Searching for and selecting studies. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al., editors. Cochrane Handbook for Systematic Reviews of Interventions. Version 6.5. Cochrane; 2024.
4. Rethlefsen ML, Kirtley S, Waffenschmidt S, Ayala AP, Moher D, Page MJ, et al. PRISMA-S: an extension to the PRISMA Statement for Reporting Literature Searches in Systematic Reviews. Syst Rev. 2021;10:39. https://doi.org/10.1186/s13643-020-01542-z
5. Paez A. Gray literature: an important resource in systematic reviews. J Evid Based Med. 2017;10(3):233–240. https://doi.org/10.1111/jebm.12266
6. 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. Syst Rev. 2015;4:138. https://doi.org/10.1186/s13643-015-0125-0