Comprehensive literature reviews typically require searching multiple databases, since no single database indexes all journals. Each database applies different indexing practices, subject headings, and search syntax such as field tags, quotation marks, and Boolean operators. It is important to understand the distinction between subject headings and keywords and how databases organize and retrieve content.

Indexing is the process by which databases assign standardized terms to represent an article’s main topics. Because each database has been developed independently and uses its own indexing policies and practices, these terms often vary among databases.  

Subject headings are standardized terms from a database’s controlled vocabulary that are assigned by indexers to describe the main topics of an article. Here are some examples of databases and their respective controlled vocabulary:  

• PubMed: MeSH (Medical Subject Headings)
• Embase: Emtree
• CINAHL: CINAHL Headings
• PsycINFO: APA Thesaurus of Psychological Index Terms

Keywords are natural-language terms that are either supplied by authors or appear in searchable text fields such as the title, abstract, and keyword fields of an article. Keywords capture author language and terminology that may not yet be indexed or standardized.

Field tags are used to limit a search to specific parts of a record, such as the title, author, abstract, or journal name. They help improve precision by focusing the search on where terms appear, which can reduce irrelevant results. Field tags often use abbreviations that vary across databases, along with database-specific search syntax, as shown in these examples: 

Translating a search

When translating a search strategy, both subject headings and keywords must be adapted to match each database’s syntax and indexing system.

This table shows how subject headings and keyword fields differ across databases: 

*Note that some databases, such as Web of Science and Scopus, do not use controlled vocabulary, so searches there rely primarily on keywords.

The examples below demonstrate how keywords and subject headings can be combined into a search strategy in each database:

PubMed

("Continuity of Patient Care"[MeSH] OR "care continuity"[tiab] OR "care coordination"[tiab] OR "continuity of care"[tiab])

Embase

('patient care'/exp OR 'care continuity':ti,ab,kw OR 'care coordination':ti,ab,kw OR 'continuity of care':ti,ab,kw)

CINAHL  

MH ("Continuity of Patient Care") OR XB ("care coordination" OR "care coordination" OR "continuity of care")

Web of Science

TS= (“Continuity of Care” OR "care continuity" OR "care coordination")  

Scopus

TITLE-ABS-KEY ("care continuity" OR "care coordination" OR "continuity of care")  

PsycINFO

DE ("Continuum of Care") OR XB ("care continuity" OR "care coordination" OR "continuity of care")

Tools for database translation

Tools exist to support the process of translating search strategies across databases, but they should be used as an aid rather than a replacement for careful manual review.

When you use Embase, the Query translator can be used on PubMed searches to convert elements such as MeSH terms and PubMed field tags into their Embase equivalents. Researchers can paste their PubMed search into the tool and review the translated strategy for accuracy before running their search in Embase.  

Another useful tool is the Polyglot Search Translator, which helps convert search strategies among databases such as PubMed, Embase, CINAHL, Web of Science, and Scopus. The tool automatically adjusts field tags and syntax across databases. However, it cannot reliably translate proprietary subject headings or database-specific indexing terms, which must be reviewed and adapted manually.

As useful as automated translation tools can be, the initial results require careful manual review and refinement. In particular, database-specific indexing nuances must be verified by the researcher. These tools should support — not replace — expert judgment.