ICU sedation, delirium & recovery
Clustering of critically ill patients using an individualized learning approach enables dose optimization of mobilization in the ICU
Authors' conclusion
An artificial intelligence-based learning approach was able to divide a heterogeneous critical care cohort into four clusters, which differed significantly in their clinical characteristics and in their mobilization parameters. Depending on the cluster, different mobilization strategies supported the likelihood of being discharged home enabling an individualized and resource-optimized mobilization approach.
Study context
- Population
- Critically ill patients in two interdisciplinary ICUs in Munich, Germany, April 2017–May 2019
- Participants
- 948 patients
- Comparison
- Mobilization dosage within AI-derived patient clusters
- Primary endpoint
- Discharge home
Methodological detailscluster analysis, k-means cluster analysis, artificial intelligence-based, two ICUs
- Cluster analysis
- K-means cluster analysis
- Artificial intelligence-based
- Two ICUs
Primary result
As reported in the abstract
Likelihood of discharge home by cluster and mobilization parameter
Adjusted odds ratio (CI)P value
Early mobilization (< 72 h)"Young Trauma" cluster
10.0(2.8–44.0)
P < 0.001Ambulating"Middle-aged" cluster
24.0(7.4–86.1)
P < 0.001Standingolder, non-frail cluster
4.7(1.2–23.2)
P = 0.035Ambulatingolder, non-frail cluster
8.1(1.8–45.8)
P = 0.010Original publication
Clustering of critically ill patients using an individualized learning approach enables dose optimization of mobilization in the ICU. Crit Care. Published 3 January 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →