Risk Factors for Delirium Transition During an Icu Sleep Quality Improvement Project

1Johns Hopkins University, Baltimore, MD
2Johns Hopkins Hospital, Baltimore, MD
3Emory, Atlanta, GA

Meeting: Hospital Medicine 2014, March 24-27, Las Vegas, Nev.

Abstract number: 704


Delirium is common in the intensive care unit (ICU) and associated with increased length of stay, cost, and long‐term mortality and cognitive impairment. Disrupted sleep is common in critically ill patients and a potentially modifiable risk factor for ICU delirium.


As part of a multi‐faceted quality improvement (QI) project to promote sleep in a medical ICU (MICU), we examined the association of daily perceived sleep quality ratings, as measured by the validated Richards Campbell Sleep Questionnaire (RCSQ), with subsequent transition to delirium, measured twice daily using the reliable and valid Confusion Assessment Method for the ICU (CAM‐ICU). Additional potentially relevant demographic and ICU variables evaluated for bivariable associations with delirium and potential subsequent inclusion in a multivariable logistic regression model using generalized estimating equations (if bivariable p<0.10) included: age, sex, race, ICU admission diagnosis, nighttime mechanical ventilation status, and daily sedative (i.e. benzodiazepines and/or opiates) administration via bolus and via infusion routes.


During this 201‐day project, 362 unique, consecutive patients were admitted to the MICU, accounting for 1705 patient‐days eligible for delirium assessment (i.e., days without coma). In this analysis of transition to delirium in the ICU, there were 131 patients accounting for 590 eligible patient‐nights for evaluation (i.e. nights for patients who were delirium‐free at MICU admission and had at least 1 night of sleep in between two consecutive days of delirium assessment). Among these 131 patients, 31 (24%) transitioned to delirium in the MICU. Patients transitioning to delirium in the ICU versus those maintaining normal cognition demonstrated no significant difference in median [inter‐quartile range] ratings of perceived overnight sleep quality (60 [29‐77] vs. 57 [32‐78], p = 0.64). In multivariable regression analysis, there was no significant association of sleep quality with delirium transition (OR 1.00, 95% CI 0.99‐1.01); however, continuous infusion of sedatives was significantly associated with delirium transition (OR 4.74, 95% CI 1.70‐13.24). Also, patients reporting use of sleep aids at home were less likely to transition to delirium (OR 0.21, 95% CI 0.06‐0.74) (Table 1).


During our QI project to promote sleep in a single MICU, we found no association between daily perceived sleep quality ratings and transition to delirium, but infusion of sedatives (i.e., benzodiazepines and/or opioids) was strongly associated with transition to delirium in the ICU. The use of sedative infusions is an important, modifiable risk factor for delirium in ICU patients

To cite this abstract:

Niessen T, Kamdar B, Colantuoni E, King L, Neufeld K, Bienvenu O, Rowden A, Collop N, Needham D. Risk Factors for Delirium Transition During an Icu Sleep Quality Improvement Project. Abstract published at Hospital Medicine 2014, March 24-27, Las Vegas, Nev. Abstract 704. Journal of Hospital Medicine. 2014; 9 (suppl 2). https://www.shmabstracts.com/abstract/risk-factors-for-delirium-transition-during-an-icu-sleep-quality-improvement-project/. Accessed May 23, 2019.

« Back to Hospital Medicine 2014, March 24-27, Las Vegas, Nev.