Survival of Frail Elderly with Delirium

Figure 1. Flow diagram of the recruitment process.

Figure 2. Kaplan–Meier survival curves for patients diagnosed with delirium at admittance (prevalent cohort) versus patients without delirium diagnosis. Data were censored after 1 month in plot (A), 6 months in plot (B), 1 year in plot (C), and 2 years in plot (D). Note: In the plots, delirium = 1 for patients with no delirium diagnosis (blue curve in the plots). Time was expressed in days. The p-value shown in the plots corresponds to the log-rank test comparison of no delirium versus delirium survival probability curves.

Figure 3. Kaplan–Meier survival curves for patients diagnosed with delirium at admittance or during the stay (incident cohort) versus patients without delirium diagnosis. Data were censored after 1 month in plot (A), 6 months in plot (B), 1 year in plot (C), and 2 years in plot (D). Note: In the plots, ConfusionalSyndrome = 1 for patients with no delirium diagnosis (blue curve in the plots). Time was expressed in days. The p-value shown in the plots corresponds to the log-rank test comparison of no delirium versus delirium survival probability curves.

Figure 4. Hazard ratios of demographic variables and frailty scores for the prevalent cohort, data censored at 2 years. Note: left to right columns are the variable name, population size, HR (95% confidence interval), graphical representation of HR (95%CI), and-p-value Pr>z. AIC—Akaike Information Criterion; signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1; concordance is the probability of agreement between two random observations; global score log-rank test was significant at 0.05.

Figure 5. Hazard ratios of demographic variables and frailty scores for the incident cohort, data censored at 2 years. Note: left to right columns are the variable name, population size, HR (95% confidence interval), graphical representation of HR (95%CI), and p-value Pr>z. AIC—Akaike Information Criterion; signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1; concordance is the probability of agreement between two random observations; global score log-rank test was significant at 0.05.

Figure 6. Hazard ratios of comorbidities for the prevalent cohort, data censored at 2 years. Note: left to right columns are the variable name, population size, HR (95% confidence interval), graphical representation of HR (95%CI), and p-value Pr>z. AIC—Akaike Information Criterion; signif. codes: 0 ‘***’ 0.001‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1; concordance is the probability of agreement between two random observations; global score log-rank test was significant at 0.05.

Figure 7. Hazard ratios of comorbidities for the incident cohort, data censored at 2 years. Note: left to right columns are the variable name, population size, HR (95% confidence interval), graphical representation of HR (95%CI), and p-value Pr>z. AIC—Akaike Information Criterion; signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1; concordance is the probability of agreement between two random observations; global score log-rank test was significant at 0.05.

Figure 8. Hazard ratios of pharmacological variables for the prevalent cohort, data censored at 2 years. Note: left to right columns are the variable name, population size, HR (95% confidence interval), graphical representation of HR (95%CI), and p-value Pr>z. AIC—Akaike Information Criterion; signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1; concordance is the probability of agreement between two random observations; global score log-rank test was significant at 0.05.

Figure 9. Hazard ratios of pharmacological variables for the incident cohort, data censored at 2 years. Note: left to right columns are the variable name, population size, HR (95% confidence interval), graphical representation of HR (95%CI), and p-value Pr>z. AIC—Akaike Information Criterion; signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1; concordance is the probability of agreement between two random observations; global score log-rank test was significant at 0.05.