Tools for Deduplication
When gathering citations from a systematic search all of the search results are downloaded and will go through deduplication then screening.
Videos in the Systematic Reviews YouTube Playlist will be useful for learning how to download citations from commonly searched databases, whether uploading to EndNote first or directly into Covidence.
As a best practice, all citation files downloaded from databases should be saved and archived similar to any raw data used in a research project. A standard naming convention for files should be used. Recommended naming convention:
projectID_database_date.xxx
xxx will be the file type. Typically citation files are in the following formats: .ris, .nbib, .xml, txt
Covidence and EndNote desktop software are the most commonly used resources to conduct citation de-duplication. Although EndNote has the added benefit of providing citation management software support, it takes more effort and has a steep/er learning curve when it comes to de-duplicating search results.
If the team prefers to use Covidence, please see the Data Extraction tab for links to sign up for an account. Click here for instructions on importing references into Covidence. Covidence currently supports three formats for file imports:
Covidence is the recommended method for deduplication. Covidence automatically deduplicates uploaded citation files using unique identifier such as DOIs (digital object identifiers), PubMed IDs, etc.