Skip to main content
Fig. 3 | BMC Medical Informatics and Decision Making

Fig. 3

From: 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 studies

Fig. 3

Mean square error of effect estimations under different adjustment strategies. Mean square error was presented according to different a ORAY, b ORUA, c ORUY, and d rUU in the target population; e number of cases and f matching approach in the original case–control studies; and g number of studies and h pooling method in the meta-analyses. The y-axis limits differ between plots. OR, odds ratio; A, exposure; Y, outcome; U, covariate. For all scenarios, pooled full-adjusted ORs showed the least mean square error. With insufficient or improper adjustment of covariates in original studies, the pooled effect estimations were away from the true value

Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us