Literature Search

Ten databases. One query. Fully deduplicated, PRISMA-ready results — in an afternoon.

The literature search is one of the most time-consuming and technically demanding steps in a systematic review. Different databases use different syntax, different controlled vocabularies, and different interfaces. Researchers typically spend two or more weeks crafting search strategies, running them separately, and manually deduplicating results. mapped does this in a single session.

---

Supported Databases

mapped searches across 10+ academic databases simultaneously from a single interface:

• PubMed / MEDLINE
• Cochrane Library (CENTRAL)
• Scopus
• Web of Science
• Embase
• PsycINFO
• CINAHL
• IEEE Xplore
• Google Scholar
• PROSPERO
• Additional databases added regularly

You don't need separate accounts or search interfaces. mapped connects to each database and translates your query into the appropriate syntax.

---

AI-Powered Query Generation

Building effective search strategies is part science, part art. mapped's AI assists by:

1. Generating optimized search strings for each database's specific syntax (PubMed uses MeSH; Embase uses Emtree; Scopus uses its own field codes)
2. Mapping MeSH terms — identifying the correct Medical Subject Headings and exploding them appropriately
3. Expanding synonyms — capturing variant spellings, abbreviations, and related terms
4. Balancing sensitivity and specificity — ensuring you capture relevant studies without drowning in noise

You review and modify every generated query before execution. The AI accelerates the process; you control the strategy.

---

How It Works

1. Define your search criteria — based on your PICOS framework from Step 1
2. Review AI-generated queries — one per database, each tailored to that database's syntax
3. Execute searches — mapped runs all queries simultaneously
4. Automatic deduplication — identical records from multiple databases are merged
5. Review results — browse, filter, and proceed to screening

---

Automatic Deduplication

When you search 10+ databases, duplicates are inevitable. The same study indexed in PubMed, Scopus, and Embase appears three times. mapped's deduplication engine:

• Matches records by DOI, title similarity, and author/year combination
• Handles formatting differences across databases (e.g., different title capitalization, abbreviated vs. full journal names)
• Flags near-duplicates for manual review when the match is uncertain
• Reports exact duplicate counts for your PRISMA flow diagram

---

PRISMA-Compliant Search Documentation

PRISMA 2020 requires detailed search strategy reporting. mapped handles this automatically:

• Full search string for each database, exactly as executed
• Date of search for each database
• Number of records retrieved per database (before and after deduplication)
• Limits applied (date range, language, study type)

This documentation exports directly into your manuscript's Methods section and PRISMA flow diagram. No manual logging needed.

---

Export Formats

Search results can be exported in standard reference formats:

• RIS — compatible with EndNote, Zotero, Mendeley
• BibTeX — for LaTeX-based workflows
• CSV / Excel — for custom analysis or backup

---

Why This Step Matters

A systematic review is only as good as its search strategy. Miss key databases or use poorly constructed queries, and your review will have gaps that peer reviewers will find. mapped ensures comprehensive, reproducible, and well-documented searches — the foundation everything else builds on.

---

Next step: Once your search is complete, move to Screening →