Routine nurse delirium charting only moderately matched expert ICU assessments
AI-summarized from the linked source. Educational brief, not medical advice.
Brief summary
Routine bedside-nurse delirium documentation agreed only moderately with structured expert assessments in the ICU, and adding clinical context improved accuracy.
What NurseJet pulled from the source
Across 1535 paired assessments in 279 critically ill adults, agreement between bedside nurses' routine ICDSC screening (score of 4 or higher) and trained researchers' CAM-ICU assessments was only moderate (Cohen's kappa 0.42). A model that combined the individual ICDSC components with mechanical ventilation status and admission SOFA score predicted expert-confirmed delirium well (AUC 0.87; F1 0.72 on cross-validation), and the plain pass/fail cutoff performed worst of the models tested.
Why this matters for nurses
Delirium is common and harmful in the ICU, and routine charted screens are what downstream teams and registries rely on. This may matter because a single pass/fail nurse score may not capture delirium as well as a structured expert assessment, so a borderline screen that drives sedation, mobility, or escalation decisions is best read alongside the whole clinical picture rather than as a standalone verdict.
Bedside takeaway
Be aware that routine pass/fail ICDSC charting agreed only moderately with expert delirium assessments, so a borderline screen reads best alongside the whole clinical picture, not as a standalone verdict.
Key takeaways
- Agreement between routine nurse ICDSC documentation and researcher CAM-ICU was moderate (Cohen's kappa 0.42) across 1535 assessments in 279 patients.
- A model combining ICDSC components with ventilation status and admission SOFA score discriminated delirium well (AUC 0.87; F1 0.72).
- The simple validated cutoff alone performed worst; richer component-level detail improved accuracy.
- Patients were critically ill adults with acute respiratory failure or sepsis in academic-hospital ICUs.
Practice implications
- Score every ICDSC component deliberately instead of rushing to a single pass/fail number, and when a screen sits near the threshold, corroborate it with your own bedside picture of attention, fluctuation, and level of consciousness. Use complete, accurate delirium documentation to support ABCDEF/ICU Liberation discussions about lightening sedation and early mobility.
Limitations & cautions
- This was a single health-system, academic-ICU sample focused on respiratory failure and sepsis, so it may not generalize to all units. The optimized model was only internally validated by cross-validation and has not been tested prospectively, so real-world bedside performance is unproven.
- AI-summarized from the linked source. Review the original article before applying to practice.
Citations
Exact source links
Public citations are filtered to exact credible source pages. Homepage-only or invalid links stay in admin review and are not shown here.
Critical Care Medicine (PubMed)
Critical Care Medicine (PubMed). Optimizing Agreement Between Bedside Nurse-Documented and Trained Researcher Delirium Assessments in the ICU.
https://pubmed.ncbi.nlm.nih.gov/40970763/
Professional education only
