Article type
Abstract
Title
Following up on the evidence: creating an organisational repository of included studies in systematic reviews
Background
The primary units of systematic reviews are the included studies in their analyses. For all review projects at the Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU) this amounts to approximately 1,200 included references each year. The reviews are published on our website, including apprasials and lists of included and excluded studies. However, as of today SBU does not have a structured database of included studies.
Objectives
The aim of this project is to present and discuss areas of use for an organisational repository of included studies: to follow up on research integrity issues, such as retractions and predatory journals; bibliometric analysis; better internal and external transparency on which studies are included and excluded; how different projects and topics relates to each other and if studies appears (and have been appraised differently) in more than one review project; and if there is a correlation between the risk of bias level for each individual study and the quality of the journal in which it is published.
Methods
All studies included based on relevance in SBU reports for the last three years will be compiled in a structured database. Main data source will be the bibliographic database Scopus (Elsevier), which includes interoperable metadata such as Digital Object Identifier (DOI), International Standard Serial Number (ISSN) and PubMed reference number (PMID). All studies will be manually indexed with SBU related information such as report specific metadata and risk of bias assessments. The Journal Impact Factor and the Norwegian Register for Scientific Journals, Series and Publishers will be used as proxy for quality of journal and compared to the assigned level of risk of bias (low, moderate, serious or critical) based on established critical appraisal tools.
Results
Results will be presented from the bibliometric analysis, such as aggregated data on author affiliations, publication and study types, and correlation between critical appraisal and journal.
Following up on the evidence: creating an organisational repository of included studies in systematic reviews
Background
The primary units of systematic reviews are the included studies in their analyses. For all review projects at the Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU) this amounts to approximately 1,200 included references each year. The reviews are published on our website, including apprasials and lists of included and excluded studies. However, as of today SBU does not have a structured database of included studies.
Objectives
The aim of this project is to present and discuss areas of use for an organisational repository of included studies: to follow up on research integrity issues, such as retractions and predatory journals; bibliometric analysis; better internal and external transparency on which studies are included and excluded; how different projects and topics relates to each other and if studies appears (and have been appraised differently) in more than one review project; and if there is a correlation between the risk of bias level for each individual study and the quality of the journal in which it is published.
Methods
All studies included based on relevance in SBU reports for the last three years will be compiled in a structured database. Main data source will be the bibliographic database Scopus (Elsevier), which includes interoperable metadata such as Digital Object Identifier (DOI), International Standard Serial Number (ISSN) and PubMed reference number (PMID). All studies will be manually indexed with SBU related information such as report specific metadata and risk of bias assessments. The Journal Impact Factor and the Norwegian Register for Scientific Journals, Series and Publishers will be used as proxy for quality of journal and compared to the assigned level of risk of bias (low, moderate, serious or critical) based on established critical appraisal tools.
Results
Results will be presented from the bibliometric analysis, such as aggregated data on author affiliations, publication and study types, and correlation between critical appraisal and journal.