Skip to main content
Fig. 2 | Journal of Intensive Care

Fig. 2

From: Prognostic significance of delirium subtypes in critically ill medical and surgical patients: a secondary analysis of a prospective multicenter study

Fig. 2

Association of delirium subtypes versus no delirium with secondary outcomes. OR odds ratio. A Shows the adjusted odds ratios with 95% confidence intervals for both delirium subtypes as compared to non-delirious patients, as analyzed with mixed-effect logistic regression models. These models were adjusted for age, admission diagnosis and APACHE IV score, and a random effect for hospital. An interaction term for delirium subtype with APACHE IV score, centered on its mean, was added to the model if significant. Since the odds ratio for outcome Antipsychotics, ever was larger than the other secondary categorical outcomes, this outcome was not shown graphically in the figure to improve visualization. The unadjusted and adjusted odds ratios are shown in Additional file 3. B Shows the adjusted differences with 95% confidence intervals of both delirium subtypes as compared to non-delirious patients, as analyzed with mixed-effect linear regression models. These models were adjusted for age, admission diagnosis and APACHE IV score, and a random effect for hospital. An interaction term for delirium subtype with APACHE IV score, centered on its mean, was added to the model if significant. The unadjusted and adjusted differences are shown in Additional file 3

Back to article page