Why Are We Missing Delirium in the ICU? Quality Improvement Implications of a Nursing Survey on Delirium Screening


Luisa Sperry, MS41; Michael Mintz, MS41; Aaron Pinkhasov, MD2, 1. Stony Brook University School of Medicine, Stony Brook, NY, 2. Chairman, Department of Behavioral Health, NYU Winthrop Hospital, Mineola, NY


Despite its high prevalence in critically ill patients, delirium frequently goes unrecognized. The Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) is a validated delirium screening tool with three possible ratings: positive (delirium present), negative (delirium absent), and unable to assess (UTA). The purpose of this study is to evaluate baseline CAM-ICU adherence and delirium recognition in the ICU. We also sought to assess nurses' views on delirium screening and perceived barriers to using the CAM-ICU in order to determine areas for improvement.


CAM-ICU Assessment Comparison: Two trained investigators performed CAM-ICU assessments twice daily for all patients aged 65 and over admitted to the medical ICU at NYU Winthrop Hospital during a two-week period in December 2016. Investigators' assessments were compared to corresponding nursing assessments. Survey: ICU nurses completed an anonymous paper survey distributed by the unit nurse manager. The 13-question survey consisted of 10 five-point Likert scale questions, two yes/no questions, and one free text question.


CAM-ICU Assessment Comparison: 58 unique patients were included for a total of 156 CAM-ICU assessments. Investigators identified 43 (28%) positive delirium assessments. However, nurses missed delirium in 32 (74%) instances, instead assigning an inappropriate negative (34%, n = 11) or UTA rating (41%, n = 13) or not performing the CAM-ICU at all (25%, n = 8). Survey: The response rate was 77% (33 of 43). Nurses feel it is important to screen for delirium (88%) and routinely use the CAM-ICU (79%). Furthermore, they believe the CAM-ICU is a valid and reliable way to screen for delirium (73%) and feel they have adequate time to use this tool (73%). However, a minority of nurses believes their screening will ultimately improve patient care (45%), and few feel that physicians consider screening results when making treatment decisions (6%). Additionally, nurses do not believe that the CAM-ICU is validated for use in intubated and sedated patients (66%) or those with preexisting dementia (81%).


ICU nurses believe it is important to screen patients for delirium and are willing to do so. However, they do not believe that the CAM-ICU is validated for use in intubated and sedated patients or those with preexisting dementia. This finding may account for many UTA ratings inappropriately assigned to delirious patients. Additionally, nurses do not feel that their assessments influence physicians' management decisions or patient outcomes. These findings suggest that interventions to improve delirium screening should focus on providing education on assessing patients who are intubated and sedated or have preexisting dementia. Furthermore, these findings highlight the need to develop a system for physicians to be alerted to positive delirium screening assessments made by nurses, so that this information can be used to improve patient care.


1.Oxenboll-Collet M, Egerod I, Christensen V, Jensen J, Thomsen T (2016). Nurses' and physicians' perceptions of Confusion Assessment Method for the intensive care unit for delirium detection: focus group study. Intensive Crit Care Nurs 2016 Sep 5. doi: 10.1111/nicc.12254

2.Swan JT (2014). Decreasing inappropriate unable-to-assess ratings for the Confusion Assessment Method for the Intensive Care Unit. Am J Crit Care, 23(1), 60-8. doi: http://dx.doi.org/10.4037/ajcc2014567.

3.Terry KJ, Anger KE, Szumita PM (2015). Prospective evaluation of inappropriate unable-to-assess CAM-ICU documentations of critically ill adult patients. J Intensive Care, 3(52). doi: 10.1186/s40560-015-0119-y.

Back to the October 2018 issue of ACP IMpact