Sonja Mensch

27 A systematic review of the literature (such as sports or school) or on the level of body functions (such as muscle tone or range of motion) (ICF-CY); (d) have been developed as classification instruments only; (e) measure quality of life only or (f ) normative measurements. Inclusion or exclusion of articles according to these criteria was done following predetermined steps. First, one author (SM) performed a selection based on titles. Second, two reviewers (SM, ER) independently screened the abstracts of the remaining articles and third, a consensus was reached about the in- or exclusion of the articles by both raters. In case of disagreement, further judgment was based on the full text independently by both raters, after which consensus was achieved. If the information in the full text articles proved inconclusive, additional information on the instruments was collected from their authors or manuals. Quality assessment and psychometric properties Methodological quality of the studies was evaluated using the Consensus Based Standards for the Selection of Health Status Measurement Instruments (COSMIN) protocol, providing standardised criteria for stepwise rating of methodological quality in a range of excellent to poor (Mokkink et al., 2010; Table 1). An example of a COSMIN box in appendix A (Terwee et al., 2007) was added. Table 1. Methodological quality criteria of the COSMIN boxes (Terwee et al., 2012). Rating Quality criteria Excellent All relevant COSMIN items are scored adequate Good Some things are not reported, but one can assume these issues are adequate, f.e. if it can be assumed that patients were not changed in a test-retest study, but this was not explicitly investigated Fair The value of the psychometric property could have been under estimated or estimated in a moderate sample or when there were other minor flaws in the design or analyses, f.e. reliability could have been underestimated due t the unstable patients or a long time interval in test-retest design Poor The results are not to be trusted because of major flaws in the design or statistical analyses, f.e. small sample size or inappropriate statistical methods Statistical outcomes were rated according to the criteria, shown in Table 2, as good (+), negative (-), and indeterminate (?) (Terwee, et al., 2007).