Study appraisal and data extraction
We critically appraised the quality of analytical studies on the basis of a modified quality criteria rating scale [25]. Studies were rated on a scale of excellent, satisfactory, or unsatisfactory using five major criteria: (1) study design appropriate for analysis; (2) representativeness of the study population; (3) use of appropriate survey instruments for the questions being asked; (4) control for confounders; and (5) whether there was comprehensive explora- tion of the research question. Comprehensive exploration distin- guished an excellent study from a satisfactory study, by indicating that all quality criteria were thoroughly addressed in addition to methodological factors that could not be uniformly assessed, such as explicit theoretical grounding. Qualitative studies were ap- praised on the basis of the following quality criteria: (1) use of a qualitative sampling methodology such as purposive sampling; (2) appropriate sample size to reach data saturation; and (3) contain a clear recruitment strategy. Those for whom there were questions as to whether they qualified for inclusion, both authors reviewed them. Only those publications that rated satisfactory or better were included in the review (n 30).