Fig. 4From: An informatics consult approach for generating clinical evidence for treatment decisionsNew synthesis of prior observational evidence. Meta-analysis of the association between warfarin use and the risk of ischaemic stroke in observational studies including approaches for automation. A Characteristics of observational studies included in the meta-analysis. B Forest plot depicting the hazard ratios calculated with the DerSimonian and Laird random-effects models. HR = hazard ratio; CI = confidence interval; SE = standard errorBack to article page