Fig. 3From: How are missing data in covariates handled in observational time-to-event studies in oncology? A systematic reviewBreakdown of multiple imputation (MI) usage. 1 2 did not specify the type of multivariate MI model used, similarly 1 for univariate. 2 1 study ensured the sample size stayed the same for different models. 3 3 studies did not clearly state that they were using complete-caseBack to article page