The continuous outcomes were analyzed using an inverse variance method for pooling weighted mean differences and where the studies had different scales, standardized mean differences were used. Statistical heterogeneity was assessed by considering the chi-squared test for significance at p <0.1 or an I-squared inconsistency statistic of >50% to indicate significant heterogeneity. Where there was heterogeneity and enough studies, sensitivity analyses were conducted based on risk of bias and pre-specified subgroup analyses were carried out as defined in the protocol. Assessments of potential differences in effect between subgroups were based on the chi-squared tests for heterogeneity statistics between subgroups. If no sensitivity analysis was found to completely resolve statistical heterogeneity, then a random effects (DerSimonian and Laird) model was employed to provide a more conservative estimate of the effect. The means and standard deviations of continuous outcomes were required for meta-analysis. However, in cases where standard deviations were not reported, the standard error was calculated if the p-values or 95% confidence intervals were reported and meta-analysis was undertaken with the mean difference and standard error using the generic inverse variance method in Cochrane Review Manager (RevMan5) software. Where p values were reported as “less than”, a conservative approach was undertaken. For example, if p value was reported as “p <0.001”, the calculations for standard deviations were based on a p value of 0.001. If these statistical measures were not available, then the methods described in section 16.1.3 of the Cochrane Handbook ‘Missing standard deviations’ were applied as the last resort. For binary outcomes, absolute differences in event rates were also calculated using the GRADE pro software using total event rate in the control arm of the pooled results and presented in the “Clinical Summary of Findings Table” in the full version of the original guideline document. Pre-specified subgroup analyses were conducted for populations of interest. These are groups where it had been identified that the interventions were likely to have different effect (effect modifiers), rather than prognostic factors. Although prognostic factors are usually not good candidates for subgrouping in meta-analysis, it is often impossible to completely predict whether a potential difference in effect is due to a difference in how the intervention may work in a group, or in how it will affect all outcomes; for example active cancer is a prognostic factor, but can also possibly affect how anticoagulants work. When such subgroups are identified, studies were sub-grouped to observe whether there might be differences in effects between different groups of patients. Appraising the Quality of Evidence by Outcomes The evidence for outcomes from the included RCT and observational studies were evaluated and presented using an adaptation of the ‘Grading of Recommendations Assessment, Development and Evaluation (GRADE) toolbox’ developed by the international GRADE working group (https://www.gradeworkinggroup.org/ External Web Site Policy). The software (GRADEpro) developed by the GRADE working group was used to assess the quality of each outcome, considering individual study quality and the meta-analysis results.
