Resources / Tool
Reference Deduplication Tool
Paste your exported search results (one citation per line) to flag exact and likely near-duplicate records before you start title/abstract screening. Nothing is uploaded — it all runs in your browser.
How this works
Exact duplicates are found by normalizing each line (lowercasing, stripping punctuation, collapsing whitespace) and comparing them directly. Near-duplicates are flagged when two lines share 75% or more of their words — common when the same record is exported from two databases with slightly different formatting (e.g. author initials, punctuation, or journal abbreviation).
This is a lightweight first pass, not a replacement for dedicated screening software. It works on plain citation text, not on structured metadata (DOI, PMID) the way tools like EndNote, Rayyan, or Covidence do — always spot-check flagged pairs before removing anything, and use a dedicated reference manager for your final, citable dedup pass.
Need help with your full search and screening process?
Our team designs and documents reproducible search strategies across multiple databases, with full deduplication and screening records for your PRISMA flow diagram.
See Search Strategy Support