ICU sedation, delirium & recovery
Natural language processing diagnosed behavioural disturbance phenotypes in the intensive care unit: characteristics, prevalence, trajectory, treatment, and outcomes
Authors' conclusion
NLP-Dx-BD screening enabled identification of three behavioural disturbance phenotypes with different characteristics, prevalence, trajectory, treatment, and outcome.
Study context
- Population
- Critically ill patients in three medical-surgical ICUs
- Participants
- 2,931 patients · 3 ICUs
- Exposure
- Agitated vs non-agitated vs combined behavioural disturbance phenotypes
- Agitated phenotypen = 225
- Non-agitated phenotypen = 544
- Combined phenotypen = 667
- Outcome
- Prevalence of behavioural disturbance phenotypes
Methodological detailsmulticentre, natural language processing screening, time-dependent multivariable analysis
- Multicentre
- Natural language processing screening
- Time-dependent multivariable analysis
Main finding
As reported in the abstract
Observational data: associations, not effects.
Further findings
| Outcome | OR (CI) | P value | |
|---|---|---|---|
| Receipt of antipsychotic medicationsAgitated phenotype vs Non-agitated phenotype | 1.84 (1.35–2.51) | P < 0.001 | |
| Receipt of antipsychotic medicationsAgitated phenotype vs Combined phenotype | 1.27 (0.86–1.89) | P = 0.229 | |
| DeathAgitated phenotype vs Non-agitated phenotype | 1.57 (1.10–2.25) | P = 0.012 | |
| DeathAgitated phenotype vs Combined phenotype | 4.61 (2.14–9.90) | P < 0.001 | |
| Death, patients receiving mechanical ventilation (subgroup)Agitated phenotype vs Combined phenotype | 7.03 (2.07–23.79) | P = 0.002 | |
| Death, patients receiving mechanical ventilation (subgroup)Non-agitated phenotype vs Combined phenotype | 6.10 (1.80–20.64) | P = 0.004 | |
| 0.050.21520Odds ratio (log scale) |
- Receipt of antipsychotic medications: time-dependent multivariable analysis, adjusted for key outcome predictors
- Death: time-dependent multivariable analysis, adjusted for key outcome predictors
Original publication
Natural language processing diagnosed behavioural disturbance phenotypes in the intensive care unit: characteristics, prevalence, trajectory, treatment, and outcomes. Crit Care. Published 4 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →