Effect measures and Synthesis methods
We used the extracted data to calculate the percentage of children with DCD in each sample, and this estimate’s 95% confidence interval (CI). Random effects models were used due to the presumed variance in effect sizes extracted from each study. Data analysis for the prevalence of DCD in preterm, the prevalence considering categorization of prematurity, and considering assessment tools and different cut-off criteria for DCD, were performed on Excel, using a spreadsheet developed by Neyeloff, Funchs & Moreira (2012).16 Categorization of prematurity was considered according to World Health Organization (WHO);17 preterm children are those who were born alive before 37 weeks of gestation. The sub-categories based on gestational age: extremely preterm (< 28 weeks), very preterm (28 to 32 weeks), and moderate to late preterm (32 to 37 weeks).17
Data analysis for comparison between preterm and full-term groups was conducted using Review Manager Software 5.4. Random effects models and risk ratio were used. Data analysis for comparison between preterm and full-term groups was also conducted considering categorization of prematurity, and considering assessment tools and different cut-off criteria for DCD. The full-term group was recruited from those studies that presented results for this population. Heterogeneity among studies was evaluated using the I2 statistic with low, moderate, and high I2 values of 25%, 50%, and 75% respectively.18
We analyzed the most restricted cut-off criteria from studies that considered more than one cut-off criteria. When there was more than one time point assessment, we considered the one with larger sample size.