LONGITUDINAL TRENDS IN INCIDENCE AND PREVALENCE OF DELIRIUM IN THE INTENSIVE CARE UNIT

Abstract

Introduction Delirium in the intensive care unit (ICU) is associated with increased risk of dementia, institutionalization following hospital discharge, and increased risk of death in patients older than 65. While recent efforts have been made to prevent delirium, any change in delirium incidence has not been described. The aim of this study was to determine the longitudinal incidence and prevalence of ICU delirium. Methods Data was obtained from participants enrolled in a randomized clinical trial testing the effectiveness of a multicomponent delirium reduction strategy (PMD trial). They were admitted to the medical-surgical ICU at an urban, academic-affiliated hospital between May 2009 and January 2014. Twice daily sedation and delirium assessments were performed using the Richmond Agitation Scale Score (RASS), and Confusion Assessment Method for the Intensive Care Unit (CAM-ICU), respectively. Demographic and clinical data was obtained from the study and the Regenstrief Medical Record System (RMRS). A mixed effects logistic regression model was used to determine longitudinal trends in incident and prevalent delirium, and odds of delirium were reported with 2014 as reference year. Results In this study, 3381 patients were included. The mean age of the study sample was 57.7 years (SD 16.0), with 33% (n=1114) over the age of 65. The overall prevalence of delirium was 19.6% (n=663), and overall incidence was 28.1% (n=209). Delirium prevalence and incidence decreased yearly. The odds ratio for delirium prevalence was 3.32 in 2009 (95% CI 2.13-5.16, p Conclusions We found a longitudinal decrease in delirium incidence and prevalence observed in this study. Further research is needed to analyze whether this decrease may be associated with implementation of bundled care programs to prevent delirium, especially in elderly adults who are at greater risk of complications following delirium. This research was funded by MB is supported by NIA R01 AG040220-05, AHRQ P30 HS024384-02, CMS 1 L1 CMS331444-02-00 and NIA R01 AG030618-05A1. BK is supported by NIA K23-AG043476.