De-duplicating for systematic reviews

From HLWIKI Canada
Jump to: navigation, search
Hierarchical view of "evidence"
Creative Commons source:
" we go down the pyramid from the top, the strength of the evidence becomes less as we consider other sources of information such as published reviews, textbooks, CME education, clinical experience, practice guidelines, patient preferences; and what is in the general domain, such as social media, Internet, monographs, and promotional materials. All of this influences clinical care, but what would be optimal is using the strongest data to make clinical decisions..."
Are you interested in contributing to HLWIKI International? contact:

To browse other articles on a range of HSL topics, see the A-Z index.

Last Update

  • Updated.jpg 16 February 2018


See also Expert searching | Hand-searching | Network meta-analysis | Rédaction de revues systématiques: recherche documentaire | Software for systematic reviewing | Systematic review librarian

De-duplicating for systematic reviews must be accomplished with the same attention to detail as systematic review searching, and yet, few tools that accomplish deduping have been validated for their accuracy (e.g., RefWorks).

One of the major challenges in searching many bibliographic databases for the systematic review is retrieval and deduplication of citation duplicates (and even triplicates). Duplicate and triplicate citations must be removed from aggregated sets in EndNote Web or RefWorks prior to the research team reviewing the papers one at a time. Removal of duplicates ensures researchers do not waste valuable time screening the same (or very similar) papers twice. Many bibliographic and reference management software tools provide some form of deduplication (close or exact duplicates) and may even employ algorithms to match records that appear to be exact duplicates. As many librarians have shown, such a process is rarely successful on its own, and in fact requires further curation, editing and manual removal (one at a time). This is time-consuming and leads to wasted time and other resources. For more information and assistance, ask your medical librarian in your hospital or health centre. There is a freely-available deduplication tool developed by the CREBP - Centre for Research in Evidence-Based Practice. There are also a few papers in the literature (see below) that explain how medical librarians and researchers are going about the deduping process. Some excellent methods are even emerging such as the Bramer method (2015) which was tested by a group of health librarians and whose steps were published in the JMLA in July 2016 (see Bramer WM, Giustini D, de Jonge GB, Holland L, Bekhuis T. De-duplication of database search results for systematic reviews in EndNote. J Med Libr Assoc. 2016 Jul;104(3):240-3.)


Personal tools