Electronic versions

  • Heather Menzies Munthe-Kaas
  • Claire Glenton
    Knowledge Centre for the Health Services at the Norwegian Institute of Public Health, Oslo
  • Andrew Booth
    University of Sheffield
  • Jane Noyes
  • Simon Lewin
    Knowledge Centre for the Health Services at the Norwegian Institute of Public Health, Oslo
Background
Qualitative evidence synthesis is increasingly used alongside reviews of effectiveness to inform guidelines and other decisions. To support this use, the GRADE-CERQual approach was developed to assess and communicate the confidence we have in findings from reviews of qualitative research. One component of this approach requires an appraisal of the methodological limitations of studies contributing data to a review finding. Diverse critical appraisal tools for qualitative research are currently being used. However, it is unclear which tool is most appropriate for informing a GRADE-CERQual assessment of confidence.
Methodology
We searched for tools that were explicitly intended for critically appraising the methodological quality of qualitative research. We searched the reference lists of existing methodological reviews for critical appraisal tools, and also conducted a systematic search in June 2016 for tools published in health science and social science databases. Two reviewers screened identified titles and abstracts, and then screened the full text of potentially relevant articles. One reviewer extracted data from each article and a second reviewer checked the extraction. We used a best-fit framework synthesis approach to code checklist criteria from each identified tool and to organise these into themes.
Results
We identified 102 critical appraisal tools: 71 tools had previously been included in methodological reviews, and 31 tools were identified from our systematic search. Almost half of the tools were published after 2010. Few authors described how their tool was developed, or why a new tool was needed. After coding all criteria, we developed a framework that included 22 themes. None of the tools included all 22 themes. Some themes were included in up to 95 of the tools.
Conclusion
It is problematic that researchers continue to develop new tools without adequately examining the many tools that already exist. Furthermore, the plethora of tools, old and new, indicates a lack of consensus regarding the best tool to use, and an absence of empirical evidence about the most important criteria for assessing the methodological limitations of qualitative research, including in the context of use with GRADE-CERQual.
Original languageEnglish
Article number113
JournalBMC Medical Research Methodology
Volume19
DOIs
Publication statusPublished - 4 Jun 2019
View graph of relations