How to Use the Systematic Review Search Translator Prompt to Move a PubMed Search Into Embase and CINAHL

A systematic review search is only as good as its weakest database. Many teams build a careful PubMed strategy and then copy it into Embase and CINAHL with a few edits, which misses records because each database has its own thesaurus, field codes, and proximity syntax. Translating concept by concept is slower but far more reliable. The Systematic Review Search Translator: PubMed Strategy to Ovid Embase and EBSCO CINAHL with MeSH to Emtree and CINAHL Headings Mapping, Field Codes, Proximity, and PRISMA-S Reporting prompt does that translation the way a health sciences librarian would: concept table first, then platform specific strategies, a syntax table, limit warnings, benchmark checks, and a PRISMA-S reporting block.

What the prompt produces

1. A concept table that splits the PubMed search into concepts and lists MeSH terms, free text terms, and Emtree and CINAHL Headings candidates marked for verification.
2. A numbered Ovid Embase strategy with exploded headings, title, abstract, and keyword fields, proximity, truncation, and combined lines.
3. An EBSCO CINAHL strategy with exploded headings, TI and AB fields, N3 proximity, and numbered sets.
4. A syntax translation table that shows what changed and where meaning shifts.
5. Limit translation with warnings about filters that are not equivalent across databases.
6. A validation step and PRISMA-S reporting block for each database.

How to fill the inputs

ReviewQuestion states the question and its PICO or PEO elements. Each element usually becomes one concept block.

PubMedStrategy is your final PubMed search, pasted exactly. The prompt keeps every free text term unless it explains why one was dropped.

TargetDatabases names each database and its platform. Embase on Ovid and Embase on Elsevier's own interface use different syntax, so the platform matters.

Limits lists what you applied in PubMed, such as dates, languages, or a humans filter.

BenchmarkArticles are known relevant studies the search must retrieve. They are your quality check.

SearchDates records when each search will be run, which you need for reporting.

Reading the example output

The example translates a search on home based telehealth exercise for older adults with type 2 diabetes:

• Headings are candidates, not assumptions. The concept table suggests Emtree and CINAHL terms for each MeSH heading and marks every one VERIFY IN THESAURUS, because heading names and trees differ.
• Phrases become proximity searches on purpose. The PubMed phrase home based exercise becomes home, remote, or video within three words of exercise. The syntax table notes this was widened deliberately so it can be reported.
• Automatic term mapping is called out. PubMed quietly adds mapped terms, but Ovid and EBSCO only search what you type, so free text has to carry more weight.
• The human limit is handled carefully. Embase uses the standard animal NOT human exclusion instead of a checkbox, and CINAHL relies on the population concept.
• Benchmarks are tied to lines. Each known study is linked to the lines that should retrieve it, with advice to inspect its indexing if it is missed.
• Counts are left blank. The PRISMA-S block includes platform, coverage, date, and limits, with record counts left for you to fill after running the search.

Tips for better results

• Open each thesaurus while you work and check every suggested heading, including its scope note and narrower terms.
• Run the benchmark check before you finalize. A missed known study is the clearest sign a concept is too narrow.
• Save each strategy exactly as run, with the date, for your appendix.
• Ask a librarian to peer review the translated strategies using a structured checklist if your review team has access to one.

Mistakes to avoid

• Do not copy PubMed field tags into Ovid or EBSCO. They will fail or behave differently.
• Do not assume a MeSH heading has the same name or tree in Emtree or CINAHL.
• Do not rely on database limit checkboxes without understanding what they exclude.
• Do not report hit counts you did not record at the time you ran the search.

Who it is for

Health sciences librarians, graduate students writing systematic or scoping reviews, nursing and allied health researchers, guideline developers, and evidence synthesis teams who need reproducible searches across several databases.

Related PromptDig links

Start with the Systematic Review Search Translator: PubMed Strategy to Ovid Embase and EBSCO CINAHL with MeSH to Emtree and CINAHL Headings Mapping, Field Codes, Proximity, and PRISMA-S Reporting prompt and paste your final PubMed strategy. For more prompts for researchers, Browse more prompts. If you have a literature search or evidence synthesis prompt that works well, Share a prompt.