A standardized coding form and manual (available from the principal author) were created using an adaptation of Zaza et al.'s Data Abstraction Form for Systematic Reviews[56]. The coding form also included Downs & Black's Study Quality Appraisal Checklist[57], which consists of 27 items and five subscales that assess methodological quality. A higher score indicates better methodological quality. Because of ambiguity in the ‘power’ item, the checklist was modified to assess whether the study authors reported power calculations that indicated an appropriate sample size for detecting clinically important effects (0 = no, 1 = yes). Although the checklist does not have a pre-specified cut-off for acceptable studies, the mid-point score of 14 was used as a guideline to distinguish between low- and high-quality studies.
