An overall summary rating on quality for each type of quantitative evidence was given, using a four-point scale ranging from “Very Low” to “High”. This summary rating took all five quality dimensions into consideration.
Studies containing quantitative evidence rated as “Moderate” (moderate limitations) or “High” (no or minor limitations) in overall quality proceeded to data extraction. Studies containing evidence rated as “Low” (major limitations) or “Very Low” (serious limitations) were excluded from evidence synthesis after QA. In certain cases, where studies reported more than one type of evidence, some of the evidence may have proceeded to data extraction, while the remainder did not.
3.3.2 Quality appraisal of qualitative evidence Studies containing qualitative evidence on facilitators/barriers were assessed for methodological quality using an adaptation of a process developed by the National Centre for Social Research in the United Kingdom (Spencer et al., 2003; Appendix C.2). Two researchers independently conducted the quality assessment, and met for consensus.
To be consistent with accepted qualitative research methods, no fixed formula was used to determine inclusion of the qualitative studies. The major emphasis was, however, put on question 1, involving the overall credibility of the study. This judgment was carefully made on the basis of answers to the other 16 questions. Reviewers met to reach consensus on an overall rating of the study methodology as “High”, “Moderate”, “Low” or “Very Low”, with consideration of the “fit” of the study methods to the study’s purpose, whether the context of the study was considered where appropriate, the adequacy of the sample and/or analysis, and the explanatory value of the study. As with the quantitative evidence, those studies judged to be of Low or Very Low quality were not kept for final data extraction.
3.4 Data extraction (DE) Pairs of reviewers independently extracted data from the included studies, and then met to reach consensus. A copy of the data extraction (DE) form and the guidance given to reviewers can be found in Appendices D (Data Extraction (DE) Form) and E (Guide to Quality Assessment and Data Extraction). Data were extracted on the type of OHSMS (mandatory versus voluntary), study design, research question, study population characteristics, sampling strategy, participation rates, outcomes of interest to the review, statistical analyses, statistical power, and results. They also recorded the presence or absence of individual OHSMS elements, according to the framework presented by Redinger and Levine (1998).
3.5 Evidence synthesis Many systematic reviewers choose an explicit algorithm at the outset of the study for later translation of the findings into a summary statement about the level or strength of evidence they provide. (van Tulder et al., 2003; Franche
Institute for Work & Health