ICU sedation, delirium & recovery
Identifying subgroups of intensive care unit (ICU) patients with high mortality rates using machine learning: A nationwide, population-based study
Authors' conclusion
Interpretable machine learning can identify ICU patient subgroups with ≥80% 6-month mortality using routinely collected data. These groups are predominantly marked by impaired consciousness and reduced urine output.
Study context
- Population
- Dutch ICU patients admitted 2013–2023, National Intensive Care Evaluation (NICE) registry, 84 hospitalsICU admissions, 84 hospitals
- Participants
- 807,727 patients
- Outcome
- Subgroups of ICU patients with high (≥80%) mortality
- Follow-up
- 6 months post-ICU admission
Methodological detailsnationwide, population-based, registry-based, machine learning, temporal validation, Model trained on 70% of hospitals, validated on the other 30% (2013-2022); temporal validation on 2023 data
- Nationwide
- Population-based
- Registry-based
- Machine learning
- Temporal validation
- Model trained on 70% of hospitals, validated on the other 30% (2013-2022); temporal validation on 2023 data
Main finding
As reported in the abstract
High-mortality (≥80%) subgroups · 6 months
High-mortality subgroups identified
10
Observational data: associations, not effects.
Further findings
External validation of subgroup mortality
- Median absolute deviations in mortality
- −1% and −2%
- Subgroups falling below the 80% mortality threshold
- 1
Original publication
Identifying subgroups of ICU patients with high mortality rates using machine learning: A nationwide, population-based study. J Crit Care. Published 27 June 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →