Can I use the exact same search string in PubMed and Embase?

No. PubMed and Embase use completely different field tags, controlled vocabularies, and truncation rules. For example, title/abstract search in PubMed uses '[tiab]' whereas Embase on embase.com uses ':ti,ab,kw' or Ovid Embase uses '.ti,ab.'.

What is the difference between MeSH and Emtree?

MeSH (Medical Subject Headings) is the controlled vocabulary curated by the US National Library of Medicine for PubMed/MEDLINE. Emtree is Elsevier's controlled vocabulary for Embase. Emtree is substantially larger, particularly for pharmacology and medical devices, and terms often differ in phrasing and hierarchy.

How do wildcards and truncation work across databases?

In PubMed, an asterisk (*) truncates the root of a single word (e.g., 'random*'). Truncating multi-word phrases inside quotes is not supported and can cause silent query failure. In Embase and Scopus, truncation syntax differs slightly and requires caution when converting multi-word concepts.

What is PRESS review for systematic search strategies?

PRESS (Peer Review of Electronic Search Strategies) is an evidence-based guideline for peer reviewing search strategies. It assesses translation accuracy across databases, Boolean logic, controlled vocabulary explosion, and spelling.

SEARCH METHODOLOGY

How to Translate Search Strategies Between PubMed, Embase, and Cochrane

A systematic review search must be exhaustive and reproducible across multiple bibliographic databases. However, simply copy-pasting a PubMed search string into Embase or Cochrane will lead to severe syntax errors, missed studies, or false hits. This cheat sheet details field codes, controlled vocabularies, and conversion rules.

Published October 12, 2026

Automatic search strategy conversion in EvidenceFlow

Translate concept groups across PubMed, Embase, Scopus, Cochrane, and Web of Science instantly.

Build Search Strategy →

1. Syntax Comparison Across Major Databases

Each database uses unique field identifiers to specify where terms should be searched. Below is the standard syntax mapping across the five primary systematic review databases:

Database / Vendor	Field Codes	Example Query	Truncation Rules
PubMed (NLM)	[tiab] for title/abstract; [mesh] for MeSH headings	diabetes[tiab] AND metformin[mesh]	Trailing * on single-word terms
Embase (embase.com)	:ti,ab,kw appended to terms; 'term'/exp for Emtree	diabetes:ti,ab,kw AND 'metformin'/exp	Trailing * supported
Scopus (Elsevier)	TITLE-ABS-KEY(...) wraps whole concept	TITLE-ABS-KEY(diabetes AND metformin)	Trailing * supported
Cochrane CENTRAL	:ti,ab,kw field codes; [mh "term"] for MeSH	diabetes:ti,ab,kw AND [mh metformin]	Trailing * supported
Web of Science	TS=(...) topic tag covers title, abstract, keywords	TS=(diabetes AND metformin)	Trailing * supported

2. Translating Controlled Vocabularies: MeSH to Emtree

The most critical mistake in search translation is copying [mesh] tags directly into Embase. Embase uses Emtree, which requires separate identification:

Explosion: In PubMed, MeSH terms explode automatically unless specified as [mesh:noexp]. In Embase, you must explicitly write 'term'/exp to include narrower sub-terms.
Terminology differences: A MeSH term does not always match an Emtree term verbatim. For example, Heart Failure in MeSH is indexed as heart failure in Emtree, but other clinical concepts have different preferred descriptors.

3. Truncation and Phrase Quoting Pitfalls

Truncation symbols (asterisks) behave differently depending on whether you are querying single words or multi-word phrases:

Do not truncate inside phrases: In PubMed, writing "systematic review*" does not find "systematic reviews" or "systematic reviewing". Truncation only functions on unquoted single roots (e.g., review*).
Proximity operators: Embase supports NEAR/3 or NEXT/3 operators to find terms within three words of each other. PubMed does not support true proximity operators in standard title/abstract searches.

4. Checking Your Strategy Against PRISMA-S

The PRISMA-S extension requires reviewers to report the full, unabbreviated search strategy for every database searched, the exact search date, and whether search limits or filters were applied. After running your translated searches, import records directly into EvidenceFlow to deduplicate and screen with your team.

Ready to translate and run your searches?

Save hours of manual re-typing and ensure complete PRISMA-S compliance.

Start Free Project