Delirium Severity Trajectories in Critically Ill Adults Using the Intensive Care Delirium Screening Checklist: A...

The delirium course of critically ill adults can be classified into trajectories based on the severity and duration of delirium as shown by a recent study. It is unknown whether these trajectories and associated outcomes are reproducible. We aimed to define delirium severity trajectories using the Intensive Care Delirium Screening Checklist (ICDSC) and delirium duration and evaluate the association of trajectory membership with clinical characteristics and 30-day post-discharge mortality.

Population-based retrospective cohort.

Fourteen medical-surgical ICUs in Alberta, Canada from January 1, 2014, to December 21, 2019.

We included adult patients (≥ 18 yr old) with an ICU length of stay of greater than or equal to 24 hours, an ICDSC score indicating delirium (≥ 4), and 30-day follow-up data were included. Group-based trajectory modeling identified trajectories over a 7-day period with SAS v9.4 (SAS Institute, Cary, NC).

None.

Demographic (age, sex) and clinical data (2 × d ICDSC score, comorbidities, illness severity, admission reason, procedures, length of stay, in-hospital, and 30-d post-discharge mortality) were captured from electronic medical records. In total, 21,071 patients were included, with a median age of 59 years (interquartile range, 46-70 yr), 59% male ( n = 12,547), and 3% died at 30 days ( n = 541). The five-trajectory model was selected. These trajectories followed previously defined patterns: 1) Mild-Brief (19.4%); 2); Severe-Rapid Recovers (18.5%); 3) Severe-Slow Recovers (31%); 4) Mild-Accelerating (14.1%); and 5) Severe-Nonrecovers (16.9%). Trajectory membership was not significantly associated with 30-day post-discharge mortality; however, clinically relevant trends were observed.

The current study substantiates the proof-of-concept model of five delirium severity trajectories. Trajectory membership did not predict 30-day post-discharge mortality. Further research is needed to understand the associations between trajectory membership, biological-based biomarkers, and patient-relevant outcomes.