| GROUP | PRIMARY RESPONSIBILITY | NOTES |
|---|---|---|
| Librarian | ✔ | Librarian performs the searches as part of the fee-based service. |
| ES/SR Team | Important: If a Librarian is not part of the ES/SR Team, then the ES/SR Team is responsible for search tasks 5a. – 5f. |
The literature search is essential – it generates the data available for analysis. The other tasks of the systematic review process (i.e., screening, appraisal, data extraction, and synthesis) are dependent on the identification of eligible studies. As a result, the literature search must be comprehensive, objective, and reproducible to identify as many relevant studies as possible to minimize bias (i.e., citation bias, publication bias, language bias, etc.). (Higgins JPT, 2020; Rethlefsen et al., 2021)
As a ES/SR Team Member/co-author, the Librarian:
It is important to locate and identify all relevant studies using a wide range of resources and search methods for the systematic review, scoping review and other select evidence synthesis types. Developing the search strategy is different from conducting the search. A comprehensive search strategy will help to reduce the risk of bias by identifying as much relevant evidence as possible. (Tsafnat et al., 2014; Higgins JPT, 2020)
A well-defined ES/SR question is essential to develop a comprehensive search strategy. As a member of the ES/SR team, the Librarian:
There is no single database that is able to provide a complete and accurate list of all studies that meet the various ES/SR guidelines’ criteria due to the differences in the articles included, controlled vocabulary/subject headings, syntax, and the indexing methods used between databases.
As noted in the respective ES/SR guidelines, for a search to be comprehensive, it is important to perform a search using a combination of controlled vocabulary/subject headings and keywords in multiple databases, grey literature, and other resources – the searches must be reproducible. This approach helps to minimize selection bias, publication bias and language bias and results in more reliable estimates of effects and uncertainties. The Technical Supplement to Chapter 4: Searching for and selecting studies to the Cochrane Handbook for Systematic Reviews of Interventions Version 6.3, includes additional information on bibliographic databases other than MEDLINE, Embase, and Cochrane CENTRAL. (Higgins JPT, 2020; Rethlefsen et al., 2021)
Sources to search include:
Bibliographic Databases – Searches of health-related bibliographic databases are usually the most efficient way to identify an initial set of results.
Database selection should be guided by the type of ES/SR question. If the ES/SR question is specialized, multi-disciplinary or involves new technologies, specialty databases (i.e., CINAHL, PsycINFO, ERIC). multidisciplinary (i.e., Scopus, Web of Science), and other resources, including the grey literature, may need to be identified and searched (Higgins JPT, 2020). Additionally, it is recommended to review the requirements or recommendations specified in the various guidelines and recommended best practices for the type of review being conducted.
The three bibliographic databases generally considered the most important sources to search for biomedical studies are PubMed/MEDLINE, Embase and the Cochrane Central Register of Controlled Trials (Higgins JPT, 2020). It is recommended to set up an account in each database so that you can save your search strategy. For Cochrane Central, you need to use all lowercase characters for your email address/login.
Other sources to be considered by the SR Team:
ES/SR guidelines require the primary investigator/project lead/researcher to search multiple databases. Even if the database is on the same platform (i.e., OVID, Elsevier, EBSCO, etc.) or its source of records is the same (i.e., National Library of Medicine is the source for MEDLINE records on applicable platforms, i.e., PubMed, Ovid MEDLINE, Web of Science MEDLINE, EBSCO MEDLINE, etc.), the database must be searched using its specialized search syntax. An Ovid MEDLINE search cannot be simply cut and pasted into Ovid Embase. Likewise, a PubMed search cannot be simply cut and pasted into Ovid MEDLINE. The search strategy must be 'translated' between databases to take into account the differences in the articles included, controlled vocabulary/subject headings, syntax, and the indexing methods.
The Library’s Database Comparison Chart summarizes key features, search syntax, search tips, proximity & adjacency operators, and common field tags of popular bibliographic databases. This tool is helpful when translating a search from one database to another database.
As a member of the ES/SR Team, the Librarian translates the approved master strategy in each of the remaining databases following the approach detailed in Task 5b. Before conducting each translated search, the Librarian submits the translated search strategy to the ES/SR Team to review/approve.
The intent of PRISMA-S (Preferred Reporting Items for Systematic reviews and Meta-Analyses literature search extension) is to complement the PRISMA 2020 Statement and its extensions. The checklist provides the ES/SR Team the framework that helps ensure transparency and maximum reproducibility of the search component of the ES/SR. It covers multiple aspects of the search process for systematic reviews and is intended to guide reporting, not conduct, of the search. Rethlefsen et. al.'s article includes explanation and elaboration for each checklist item. (Rethlefsen, 2021)
For the other ES/SR in this Guide, review the related search guidelines and reporting guidelines at Evidence Synthesis – Guidelines and Resources and/or use the PRISMA-S to the extent possible, and adapt it appropriately.
Recording the flow of citations through the systematic review process is a key component of the PRISMA Statement. It is helpful to include the dates searched and identify how many records were identified within each database and additional source. Knowing the number of records from each source also helps with reproducibility. (Reflethsen, 2021)
For the other ES/SR in this Guide, review the related search guidelines and reporting guidelines at Evidence Synthesis – Guidelines and Resources and/or use the PRISMA-S to the extent possible, and adapt it according.
EndNote is the UT Southwestern subscribed citation management program from Clarivate. This citation management program saves time: you can consolidate and organize the citations from your multiple searches, de-duplicate the results and automatically format in-text citations and bibliographies when drafting your manuscript.
The Librarian searches multiple databases and forwards results from each searched database to the ES/SR Team as EndNote ready files and, upon request, Microsoft Word documents.
Exporting Search Results – Recommended Settings by Database:
Grey literature (or gray literature) includes information or materials that are not published or indexed in the traditional databases, indexes, peer-reviewed journals, and books. These materials are referred to as grey or fugitive because they can be difficult to identify and obtain. Examples include dissertations, theses, government reports/publications, academic reports/publications, internal company reports, policy documents, and poster and paper presentations. (Adorno, Garbee, & Marix, 2016; Booth, 2012; Health Sciences Library McMaster University, 2019)
To minimize publication bias, it is recommended to search the grey literature as noted in the following select guidelines and standards:
There are numerous sources of grey literature. The Grey Literature Network Service (GreyNET) has a comprehensive list of the different grey literature document types. The Canadian Agency for Drugs and Technologies in Health (CADTH) has developed and made available a grey literature checklist, Grey Matters: A Practical Tool for Searching Health-Related Grey Literature.
While the Librarian typically searches 1-2 grey literature resources, the ES/SR Team may need to conduct their own grey literature searching. It is important to identify which grey literature types and sources are appropriate to search based on the research question and discipline.
Some suggested resources are included in the following sub-sections.
Relevant professional associations may also be searched for additional information.
To search in ProQuest™ Dissertations & Theses Citation Index, select the database from the “Search In” dropdown on the Documents tab on the Web of Science home page.
Preprints are drafts of research manuscripts that are intended to be shared publicly. There are numerous preprint servers which can be searched using a combination of keywords. As preprint servers maybe subject/discipline focused, it may be necessary to search multiple preprint servers. Additionally, it is important to remember that preprints have not been peer reviewed. (Bonato 2022)
Statistical data can be found on:
To minimize bias and perform a comprehensive, exhaustive search, it is usually recommended not to apply limits to the search. For example, database limits for language, publication date or publication format may introduce bias, reduce search sensitivity and unintentionally exclude relevant citations.
Another approach in lieu of applying filters or limits to the search strategy, the SR/ES Team may apply the limits or filters when screening the retrieved results as part of the inclusion/exclusion criteria.
If filters or limits are used, they should be applied at the end of the search strategy. Their application and reason for use must be documented in the methods section.
PRISMA for searching – Published in 2021, the checklist includes 16 reporting items, each of which is detailed with exemplar reporting and Rationale. The intent of PRISMA-S is to complement the PRISMA Statement and its extensions by providing a checklist that could be used by interdisciplinary authors, editors, and peer reviewers to verify that each component of a search is completely reported and therefore reproducible.
The PRESS (Peer Review of Electronic Search Strategies) Checklist provides a list of questions organized in six categories to consider when reviewing the SR search strategy. The categories encompass:
The full guideline statement and checklist document may be accessed via the open access article from the Journal of Clinical Epidemiology (McGowan et al., 2016b).
Cochrane Handbook for Systematic Reviews of Interventions, 2nd ed. (Wiley Online Library)
by
Finding What Works in Health Care: Standards for Systematic Reviews (NCBI Bookshelf)
by
JBI Manual for Evidence Synthesis
by
Piecing Together Systematic Reviews and Other Evidence Syntheses (ProQuest Ebook Central)
by
The Technical Supplement to Chapter 4: Searching for and selecting studies to the Cochrane Handbook for Systematic Reviews of Interventions Version 6.3, includes additional information on bibliographic databases other than MEDLINE, Embase and Cochrane CENTRAL.
