Fig. 5From: Can statistical adjustment guided by causal inference improve the accuracy of effect estimation? A simulation and empirical research based on meta-analyses of case–control studiesForest plot of an empirical meta-analysis on passive smoking and breast cancer [18]. SE, standard error; IV, inverse variance; CI, confidence interval. A weak but significant association between passive smoking and breast cancer was detected in primary analysis. Underestimated results slightly shrank the effect and gave a false negative estimation, while overestimated results substantially amplified the effectBack to article page