Guide · Databases
Which databases to search for a literature review
Each database indexes a different slice of research and exports a different set of fields. This guide compares the fourteen databases SLReviewFlow imports, so you can choose the right ones for your question and know what each file will bring into your review.
By Kushan Liyana Arachchige, Research MindUpdated
- Health and clinical
- PubMedEmbaseEBSCOhostCochrane CENTRAL
- Engineering and computing
- IEEE XploreACM Digital Library
- Theses and grey literature
- OpenAlexGoogle ScholarProQuest
01Every database at a glance
A literature review usually searches several bibliographic databases, because each one indexes a different part of the published research. This table lists the fourteen that SLReviewFlow imports, how you can get access, which file to export, and whether the file carries the reference list of each article. The name of each database links to its full export guide or to its notes further down this page.
| Database | Access | Export | Cited references |
|---|---|---|---|
| ScopusElsevier | Subscription | CSV, every field, with references | In the file |
| Web of ScienceClarivate | Subscription | BibTeX, Full Record and Cited References | In the file |
| PubMedNational Library of Medicine | Free | Save, Format: PubMed | No |
| OpenAlexOurResearch | Free | CSV (Excel), every column including Work ID | Fetched from OpenAlex |
| IEEE XploreIEEE | Subscription | CSV | No |
| EmbaseOvid | Subscription | Excel, Complete Reference | No |
| EBSCOhostCINAHL, APA PsycInfo, ERIC and others | Subscription | CSV | No |
| Cochrane CENTRALCochrane Library | Free | CSV | No |
| Lens.orgThe Lens | Free | CSV, every field | In the file |
| DimensionsDigital Science | Subscription | CSV, full record or bibliometric mapping | Bibliometric export only |
| Google Scholarthrough Publish or Perish | Free | Publish or Perish CSV | No |
| ACM Digital LibraryACM | Subscription | BibTeX | No |
| ProQuestDissertations & Theses, ProQuest Central and others | Subscription | Excel | No |
| ScienceDirectElsevier | Subscription | RIS | No |
| Anything elsecitation searching, registers, websites | Any | RIS | No |
Checked against SLReviewFlow's import templates on 26 Oct 2026.
02Which databases suit your field?
Choose databases for the literature your question needs, not for the ones you already know. Multidisciplinary databases such as Scopus, Web of Science, OpenAlex and Lens.org cover most fields, but each has gaps, and subject databases index journals, conference papers and reports that the large ones miss. Bramer and colleagues found that no single database retrieved all the studies included in a set of systematic reviews, and that combining several was needed to find nearly all of them.
- Health and clinical: PubMed, Embase and Cochrane CENTRAL for trials, with CINAHL (through EBSCOhost) for nursing and allied health.
- Psychology, education and social science: APA PsycInfo and ERIC through EBSCOhost, and ProQuest for theses and social science collections.
- Engineering and computing: IEEE Xplore and the ACM Digital Library for conference papers, which are central in these fields, plus ScienceDirect.
- Theses and grey literature: ProQuest Dissertations & Theses, OpenAlex, and Google Scholar through Publish or Perish.
If your institution does not subscribe to Scopus or Web of Science, OpenAlex and Lens.org are free, export in full, and both supply cited references.
03What to do before you export anything
Four habits keep your searches reportable. PRISMA-S, the reporting guideline for literature searches, asks you to report each database, the platform it was searched on, the full query and the date, so record these as you go rather than reconstructing them later.
- One search, one import file. Export the results of each search separately, and upload each file under the search it came from.
- Record the date, the query and the number of results. Copy the query exactly as the database shows it, including any filters you applied. The result count lets you check that the export is complete.
- Export every field the database offers. Abstracts and subject terms help screening; affiliations, funding and references feed later analyses. A field left out cannot be added later without exporting again.
- Keep every part of a split export. Most databases limit how many records one file can hold. When you export in batches, upload every batch under the same search.
04What does each export bring into your review?
Every export carries the title, authors, year and source of each record. Beyond that, databases differ a great deal. The matrix shows which fields arrive in SLReviewFlow from each database's recommended export.
| Database | Abstract | Keywords | Subject terms | Affiliations | Funding | Times cited | Cited references | Open access |
|---|---|---|---|---|---|---|---|---|
| Scopus | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| Web of Science | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| PubMed | Yes | Yes | Yes | Yes | Yes | No | No | No |
| OpenAlex | Yes | Yes | Yes | Yes | Yes | Yes | Fetched | Yes |
| IEEE Xplore | Yes | Yes | Yes | Yes | Yes | Yes | No | No |
| Embase | Yes | Yes | Yes | Yes | No | No | No | No |
| EBSCOhost | Yes | Yes | Yes | No | No | No | No | Yes |
| Cochrane CENTRAL | Yes | Yes | No | No | No | No | No | No |
| Lens.org | Yes | Yes | Yes | No | Yes | Yes | Yes | Yes |
| Dimensions | Yes | No | Yes | Partly | Partly | Yes | Partly | Partly |
| Google Scholar | Partly | No | No | No | No | Yes | No | No |
| ACM Digital Library | Yes | Yes | No | No | No | No | No | No |
| ProQuest | Yes | Yes | Yes | Yes | No | No | No | No |
| ScienceDirect | Yes | Yes | No | No | No | No | No | No |
Abstracts matter most, because title and abstract screening depends on them; a database that only gives snippets, such as Google Scholar, leaves more decisions for the full-text stage. Keywords and controlled subject terms, such as MeSH in PubMed and Emtree in Embase, help you check that the search found what you meant. Affiliations, funding and citation counts support descriptive and bibliometric analyses of the included studies.
05Which databases include cited references?
Cited references are the reference list of each record: the works it cites. They let you search backwards from your included studies, which is a recommended supplementary search method, and they make citation and co-citation analyses possible.
Only some sources supply them. Scopus, Web of Science and Lens.org include them in the export when you choose the right option; Dimensions includes them in its bibliometric mapping export only. For OpenAlex, SLReviewFlow fetches each record's references from OpenAlex after you import the file, which is why the export must include the Work ID column.
06Notes on each database
Scopus · Elsevier
Export: CSV, every field, with references. Watch for: Turn off "Truncate to optimize for Excel", or long abstracts and reference lists are cut.
Identifiers for duplicate matching: DOI, EID, PMID · Read the Scopus export guide
Web of Science · Clarivate
Export: BibTeX, Full Record and Cited References. Watch for: "Full Record" alone leaves out cited references; exports come in batches of 500.
Identifiers for duplicate matching: DOI, UT, PMID · Read the Web of Science export guide
PubMed · National Library of Medicine
Export: Save, Format: PubMed. Watch for: Choose the PubMed format, not CSV; up to 10,000 records per file.
Identifiers for duplicate matching: PMID, PMCID, DOI · Read the PubMed export guide
OpenAlex · OurResearch
Export: CSV (Excel), every column including Work ID. Watch for: Work ID is not in the default columns; add it, or the file is rejected.
Identifiers for duplicate matching: DOI, Work ID, PMID · Read the OpenAlex export guide
IEEE Xplore · IEEE
Export: CSV. Watch for: The DOI is the only identifier, so records without one are matched on title, authors and year.
Identifiers for duplicate matching: DOI
Embase · Ovid
Export: Excel, Complete Reference. Watch for: If Ovid splits the export into several files, upload each one under the same search.
Identifiers for duplicate matching: Embase ID, PMID, DOI
EBSCOhost · CINAHL, APA PsycInfo, ERIC and others
Export: CSV. Watch for: A search across several EBSCOhost databases is one search; each database is counted from the file.
Identifiers for duplicate matching: Accession number, PMID, DOI
Cochrane CENTRAL · Cochrane Library
Export: CSV. Watch for: Many records are trial registry entries rather than articles.
Identifiers for duplicate matching: CENTRAL ID, PMID, trial registry IDs
Lens.org · The Lens
Export: CSV, every field. Watch for: Keep the References column, or the cited references are lost.
Identifiers for duplicate matching: Lens ID, DOI, PMID
Dimensions · Digital Science
Export: CSV, full record or bibliometric mapping. Watch for: The full record has no cited references; the bibliometric file has no document type.
Identifiers for duplicate matching: Dimensions ID, DOI, PMID
Google Scholar · through Publish or Perish
Export: Publish or Perish CSV. Watch for: Abstracts are short snippets and one work can appear several times as different versions.
Identifiers for duplicate matching: DOI, when present
ACM Digital Library · ACM
Export: BibTeX. Watch for: The ACM Guide to Computing Literature also lists other publishers' work.
Identifiers for duplicate matching: DOI
ProQuest · Dissertations & Theses, ProQuest Central and others
Export: Excel. Watch for: Theses have no DOI, and author names come in mixed orders.
Identifiers for duplicate matching: ProQuest ID, DOI
ScienceDirect · Elsevier
Export: RIS. Watch for: Every record arrives typed as a journal article, so document type can't be used to limit.
Identifiers for duplicate matching: DOI
07A source that is not on the list
Citation searching, trial registers, organisation websites and hand searching are reported separately from databases in a PRISMA 2020 flow diagram. Collect those records in a reference manager such as Zotero or EndNote, export them as an RIS file, and add them in SLReviewFlow under Other methods, with the date and a description of how you found them.
08Doing it in SLReviewFlow
Bring every database into one pool
- Add each database as a source, then record each search with its date, query and result count.
- Upload the export under its search. SLReviewFlow reads the file and tells you if anything is missing, such as cited references.
- Build the pool. Duplicates across databases are matched on their identifiers, and the flow diagram counts each database.
09Questions
How many databases should a literature review search?
There is no fixed number. Most systematic reviews search at least two databases that index different literature, chosen for the topic; a single database misses studies that only other indexes hold. Scoping and other reviews follow the same logic, scaled to their aims. Write the choice and the reason into your protocol.
Is Google Scholar enough on its own?
No. Google Scholar cannot export more than a limited set of results per query, shows only snippets of abstracts, and its ranking changes over time, so a search cannot be repeated exactly. It is useful as an extra source for grey literature, alongside at least one database you can export in full.
Which free databases can replace Scopus or Web of Science?
OpenAlex and Lens.org cover a similar range of fields and both export in full. Their coverage is not identical to the subscription databases, so say in your methods which ones you used and why.
Can I combine exports from different databases?
Yes. Upload each export under the search it came from. The pool matches the same work across databases on its identifiers, then on title, authors and year, and your flow diagram counts each database separately.
10References
- Rethlefsen ML, Kirtley S, Waffenschmidt S, 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
- Bramer WM, Rethlefsen ML, Kleijnen J, Franco OH. Optimal database combinations for literature searches in systematic reviews: a prospective exploratory study. Syst Rev. 2017;6:245. https://doi.org/10.1186/s13643-017-0644-y
- Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. https://doi.org/10.1136/bmj.n71