InsightMed Research Hub

Systematic review search strategy builder

A systematic review search combines subject headings and free-text synonyms for each concept in your question. This builder helps you assemble those terms and produces draft search strings in the syntax of each major database, ready for you to test and refine.

Open the search strategy builder

How it works

  1. Start from your question as a review title or as PICO elements (population, intervention, comparator, outcome).
  2. Get suggested terms. With your permission the tool looks up MeSH headings in the US National Library of Medicine database. A heading is pre-selected only when it matches your whole concept; partial matches are shown unticked with a warning. When AI suggestions are available, every MeSH heading the AI proposes is checked against the NLM database and unverified headings are removed.
  3. Review each concept, untick terms you do not want and add your own synonyms.
  4. Copy the drafts for PubMed, Embase.com, Scopus, Web of Science, Cochrane CENTRAL, MEDLINE (Ovid), Embase (Ovid) and trial-registry prompts. Options add UK and US spellings, hyphen variants, truncation (off by default) and the Cochrane RCT filter for PubMed.
  5. Record what you ran in the search log (database, date, exact strategy, results) and export it as CSV for your methods section.
Drafts, not validated searches. Emtree headings are suggestions and are not verified. Test every string in the live platform, check that it retrieves known key papers, adapt the syntax where needed, and have the strategy peer reviewed by an information specialist, for example using the PRESS checklist.

Privacy

Search terms are sent to the NLM MeSH service only after you allow online lookup for the session. If AI suggestions are on, your topic or PICO text is sent to InsightMed’s AI service, which runs on Cloudflare Workers AI. Do not enter patient-identifiable or confidential information. When the AI service is busy or its daily limit is reached, the tool falls back to the MeSH lookup.

Sources

Lefebvre C, et al. Chapter 4: Searching for and selecting studies. In: Cochrane Handbook for Systematic Reviews of Interventions. McGowan J, Sampson M, Salzwedel DM, et al. PRESS Peer Review of Electronic Search Strategies: 2015 guideline statement. J Clin Epidemiol 2016;75:40–46. MeSH data from the US National Library of Medicine.

Frequently asked questions

Which databases does it support?

PubMed, Embase.com, Scopus, Web of Science, Cochrane CENTRAL, MEDLINE (Ovid) and Embase (Ovid), plus prompts for ClinicalTrials.gov and the WHO ICTRP.

Can I use the generated search directly in my review?

Treat it as a draft. Run it in each database, check that known relevant studies are retrieved, adjust it, and have it peer reviewed before you rely on it.

How are MeSH headings chosen?

The tool looks for headings whose name or entry terms exactly match your concept. Only a whole-phrase match is pre-selected; partial matches are shown unticked so you decide.

Does it include the Cochrane RCT filter?

Yes, optionally, for the PubMed string. CENTRAL drafts do not add a trial filter.

What happens to my research question?

It stays in your browser unless you allow online lookup. Then terms are sent to the NLM, and to InsightMed’s AI service if AI suggestions are on.

Related

PRISMA flow diagram · Effect size converter · All tools in one workspace · Guides

Free browser tool · No sign-up · Last reviewed: October 2026. Independent educational tool; not endorsed by PRISMA, Cochrane or the US National Library of Medicine.