Fig. 1From: Dealing with missing delirium assessments in prospective clinical studies of the critically ill: a simulation study and reanalysis of two delirium studiesBias in the association between delirium duration and the cognitive outcome for each imputation strategy stratified by the proportion of missingness, missingness mechanism, and association of missingness with the auxiliary variable. Results are derived from the analysis of 1000 simulated datasetsBack to article page