A comparison of different forward citation chasing tools for complementary searches for Cochrane systematic reviews

Article type
Bracchiglione J1, Requeijo C2, Santero M2, Savall O2, Selva A3, Samsó L2, Roqué-Figuls M2, Escobar Liquitay C4, Solà I1
1Institut d’Investigació Biomèdica Sant Pau (IIB SANT PAU), CIBERESP, Barcelona
2Institut d’Investigació Biomèdica Sant Pau (IIB SANT PAU), Barcelona
3Clinical Epidemiology and Cancer Screening, Parc Taulí Hospital Universitari. Institut d’Investigació i Innovació Parc Taulí (I3PT_CERCA), Sabadell
4Associate Cochrane Centre - Research Department, Instituto Universitario del Hospital Italiano de Buenos Aires, Buenos Aires
Background: Forward citation chasing, defined as the use of a citation index to retrieve references that cite a source, is currently recommended as a complement to find all possibly relevant research for systematic reviews (SRs). Recently, new tools have been developed to ease this task, but their performance has not been compared yet.

Objectives: To compare the performance of different forward citation indexes and tools for forward citation chase in a sample of Cochrane SRs.

Methods: We searched for Cochrane SRs with at least two published versions. Starting from an ‘index reference set’ (i.e., references of the included studies in the original version of the SR, plus the reference of the original SR itself), we conducted a forward citation search using the following resources: Web of Science (WoS), Scopus, Google Scholar, Citationchaser, and Paperfetcher. We assessed the performance of each tool regarding the identification of the ‘target reference set’ (i.e., primary study references included only in the updated version of the SR but not in the original version) in terms of sensitivity and precision.

Results: Preliminary results (10 pairs of SRs) show a median sensitivity/precision of: WoS: 58%/0.41%, Scopus: 64%/0.46%, Google Scholar: 66%/0.26%, Citationchaser: 59%/0.37%, Paperfetcher: 60%/0.48%. We found no significant differences among the tools

Conclusions: From our preliminary analysis, we could not identify a citation tool with a significantly better performance over the rest. Forward citation search should not be used as a standalone method to update a SR.

Patient, public and/or healthcare consumer involvement: None.