Evidence Synthesis: Systematic & Scoping Reviews for the University Park Campus: Data Extraction & Quality Assurance

Services for Systematic Reviews Outside of Medicine

Data Extraction

 

The data extraction process for Non-Medical Science Systematic Reviews (NMSSR) is in flux and currently requires researchers to be creative while remaining compliant with PRISMA and JBI standards. However, the current software Covidence does not do a good job of data extraction and quality assurance for NMSSRs because it was developed for Randomized Controlled Trials (RCT). The data extraction and quality assurance templates for NMSSRs must be able to answer the study's research questions and comply with traditional SR information.

Currently, multidisciplinary and NMSSRs have been pulling the final datasets from Covidence after screening and utilizing other tools to extract the data and perform quality assurance. Qualtrics, Survey Monkey, and Excel have all been used for published NMSSRs at USC.

As of 2023, USC has been involved with beta-testing a new software called Cognetto. This partnership is now operating under a large National Institute of Health grant and is an option for evidence synthesis teams that are partnered with a USC Faculty Librarian. Currently, we are using it for data extraction and quality assurance. However, the long-term goal is that Cognetto will replace the need to use multiple software (Zotero for downloading, Covidence for de-duplication and screening, and the third software selected for data extraction and quality assurance. In its current stage, it allows researchers to customize the data extraction and quality assurance templates. In its completed stage researchers will be able to upload searches directly into Cognetto, which will de-duplicate, perform abstract and full-text screening, and complete the process with advanced data extraction and quality assurance. It will also have an Artificial intelligence (AI) option to expedite screening and extraction and quality assurance if the researchers want to use AI. This will greatly expedite the whole process of completing NMSSRs at USC. Note: The use of Cognetto requires a PI to sign an NDA and convey to the team that they will not develop a competing product. The NIH grant is awarded in Phases, USC is Phase 2 funding, and USC libraries are receiving NIH funding to beta-test and help offer UX advice to the Cognettto team.

Quality Assurance