Development of a Simple Scoring System for Predicting Discharge Safety from the Medical ICU to Low-Acuity Wards: The Role of the Sequential Organ Failure Assessment Score, Albumin, and Red Blood Cell Distribution Width

Abstract

Despite advancements in artificial intelligence-based decision-making, transitioning patients from intensive care units (ICUs) to low-acuity wards is challenging, especially in resource-limited settings. This study aimed to develop a simple scoring system to predict ICU discharge safety. We retrospectively analyzed patients admitted to a tertiary hospital's medical ICU (MICU) between July 2016 and December 2021. This period was divided into two phases for model development and validation. We identified risk factors associated with unexpected death within 14 days of MICU discharge and developed a predictive scoring system that incorporated these factors. We verified the system's performance using validation data. In the development cohort, 522 patients were discharged from the MICU, and 42 (8.04%) died unexpectedly. In multivariate analysis, the Sequential Organ Failure Assessment (SOFA) score (odds ratio [OR] 1.26, 95% confidence interval [CI] 1.13-1.41), red blood cell distribution width (RDW) (OR 1.20, 95% CI 1.07-1.36), and albumin (OR 0.37, 95% CI 0.16-0.84) were predictors of unexpected death. Each variable was assigned a weighted point in the scoring system, and the area under the curve (AUC) was 0.788 (95% CI 0.714-0.855). The scoring system was performed using an AUC of 0.738 (95% CI 0.653-0.822) in the validation cohort of 343 patients with 9.62% of unexpected deaths. When a cut-off of 0.032 was applied, a sensitivity and a specificity of 81.8% and 55.2%, respectively, were achieved. This simple bedside predictive score for ICU discharge uses the SOFA score, albumin level, and RDW to aid in timely decision-making and optimize critical care facility allocation in resource-limited settings.

Keywords: SOFA; albumin; discharge; intensive care unit; prediction; red cell distribution width.

Figure 1

A flow chart showing the inclusion and exclusion of the patients in the study. In the development cohort, we developed discharge scores using cohort data containing 522 patients > 19 years who had been discharged alive from the medical intensive care unit (MICU). We validated the discharge score performance in different periods of the MICU cohort dataset, which was composed of 343 patients > 19 years who were discharged alive from the MICU. MICU, medical intensive care unit; ICU, intensive care unit.

Figure 2

Receiver operating characteristic (ROC) curves for the discharge scoring system in the development cohort. (a) The area under the curve (AUC) and 1000 bootstrap samples are presented in the discrimination. (b) The calibration plot; the Hosmer–Lemeshow goodness-of-fit (H-L) test was performed in the calibration. A calibration plot graph compares the model’s predicted probabilities with the actual predicted probabilities. This graph should be close to the 45° cutoff and have a p value greater than 0.05, indicating that the discharge scoring system predictions are appropriate. ROC, receiver operating characteristic; AUC, area under the curve.

Figure 3

Receiver operating characteristic (ROC) curves for the discharge scoring system in the validation cohort. The discharge score predicted short-term mortality after intensive care unit (ICU) discharge more precisely (area under the curve [AUC], 0.738; 95% confidence interval [CI], 0.653–0.822) than the SWIFT score (AUC, 0.607; 95% CI, 0.516–0.699). When the cut-off value of the discharge score was set to 0.032, sensitivity was 81.8%, and specificity was 55.2%. ROC, receiver operating characteristic; SWIFT, Stability and Workload Index for Transfer; ICU, intensive care unit; AUC, area under the curve; CI, confidence interval.