2.5. Statistical Analyses

The proportion of adequately reported items was considered as the number of adequately reported items divided by the total number of applicable items. A proportional meta-analysis was performed using OpenMetaAnalyst software to estimate the overall weighted proportion of each outcome of interest. The data are presented as the overall proportion with a 95% confidence interval (CI) using the random-effects DerSimonian-Laird model. Assessment of the consistency of effects in studies is an essential part of a meta-analysis.

We assessed the heterogeneity using I² statistic. The heterogeneity test examines the null hypothesis that all studies are evaluating the same effect. I squared (I²) represents the percentage of total variation in studies due to heterogeneity instead of chance. I² can be calculated and compared across meta-analyses with diverse types of studies, sizes, and type of outcomes. Higgins et al. [24] elaborated an approach to quantify the effect of heterogeneity and gave a measure of the degree of inconsistency in the results of these studies.

We calculated the agreement between raters using Fleiss’ kappa (k) statistics for multiple raters. The classification of agreement was as follows [25]: <0: no agreement; between 0 and 0.20: slight; between 0.21 and 0.40: fair; between 0.41 and 0.60: moderate; between 0.61 and 0.80: substantial; and finally, 0.81–1: almost perfect. This statistical analysis was done using various packages in R software.