Article type
Year
Abstract
Objectives: there is no published tool to support assessment of qualitative evidence synthesis (QES). The objective of this paper is to present such a tool and results from a feasibility test. The target audience is primarily guideline developers and authors of overviews of systematic reviews.
Methods: we developed a tool (Figure 1) based on the Tong and colleague's 2012 (1) framework for reporting the findings from QES. The tool focuses on risks influencing the confidence in the findings rather than the reporting per se. It is based on domains and supporting questions in analogy with tools such as ROB2 and QUADAS2. The items were discussed and refined with an expert in qualitative research and several methodologists at SBU (Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU), Stockholm, Sweden), with experience in conducting QES. We pilot tested the tool in an overview on experiences and expectations of care in eating disorders.
Results: in the pilot project, 29 QES fulfilled the inclusion criteria, which we assessed with the new tool. The tool was easy to understand and to use for the external expert that had not been involved in the development of the tool. Two signalling questions however were perceived as not supporting a risk assessment. These two questions: 1) whether the synthesis went beyond a summary of results of included studies and 2) about the use of CERQual in an appropriate manner, are currently being handled separately.
Conclusions: this tool worked well and supported a rapid assessment of limitations. However, if confidence in QES is to be formally assessed with CERQual, an in-depth assessment including contacts with the authors of the QES is necessary.
Patient or healthcare consumer involvement: the target audience is primarily guideline developers and authors of overviews of systematic reviews, but in the end it is also the goal that the research will be available to the patients themselves.
Methods: we developed a tool (Figure 1) based on the Tong and colleague's 2012 (1) framework for reporting the findings from QES. The tool focuses on risks influencing the confidence in the findings rather than the reporting per se. It is based on domains and supporting questions in analogy with tools such as ROB2 and QUADAS2. The items were discussed and refined with an expert in qualitative research and several methodologists at SBU (Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU), Stockholm, Sweden), with experience in conducting QES. We pilot tested the tool in an overview on experiences and expectations of care in eating disorders.
Results: in the pilot project, 29 QES fulfilled the inclusion criteria, which we assessed with the new tool. The tool was easy to understand and to use for the external expert that had not been involved in the development of the tool. Two signalling questions however were perceived as not supporting a risk assessment. These two questions: 1) whether the synthesis went beyond a summary of results of included studies and 2) about the use of CERQual in an appropriate manner, are currently being handled separately.
Conclusions: this tool worked well and supported a rapid assessment of limitations. However, if confidence in QES is to be formally assessed with CERQual, an in-depth assessment including contacts with the authors of the QES is necessary.
Patient or healthcare consumer involvement: the target audience is primarily guideline developers and authors of overviews of systematic reviews, but in the end it is also the goal that the research will be available to the patients themselves.
PDF