Background: The prevalence of delirium among hospitalized patients ranges up to 56% and results in increased hospital mortality and duration of hospitalization. Symptoms of delirium may be subtle in early stages and may present clinically as hyperactive, hypoactive or mixed type. As such delirium often evades early detection. In addition, delirium in the hospital setting is often multifactorial and single interventions are less prospective in leading to resolution. We developed a comprehensive model for treating hospital-acquired delirium, by involving all member of the healthcare team, working in concert to evaluate and treat delirium as appropriate.
Methods: A multidisciplinary structure was established through the inclusion of primary service providers on the medical units (attending physicians, residents and midlevel practitioners), the nursing staff and nursing assistants, the pharmacists, physical therapists, and nutritionists. Appropriate teaching and instruction was provided to all levels. A “Delirium orderset”, a comprehensive set of linked interventions, was created to facilitate basic evaluation and treatment of delirium. The Confusion Assessment Method (CAM) screening tool was performed by the nursing staff twice a day on all admitted patients. Hospital acquired delirium was defined as patients with a negative CAM score at admission, who subsequently had at least two consecutive positive CAM scores. Inclusion criteria were all patients admitted to the general medical floors, who were at baseline mentation, had a negative CAM score at admission, and who subsequently had two consecutive positive scores. Providers were then notified for evaluation. CAM screenings were continued for these patients until discharge. A baseline period for data comparison was selected from 9/1/2016 to 9/25/2017. The project launch date was 9/26/2017, and data reviewed through 12/1/2017.
Results: From 9/26/2017 – 12/1/2017, there were 2395 inpatients of which 2277 (95.1%) had a CAM tool assessment. Of the 2277, 125 (5.5%) had a CAM tool assessment of positive with an average length of stay of 12.7 days, a 30 day readmission rate of 16.7%, and a non-hospice mortality rate of 7.2%. CAM tool use increased from 1.2% of inpatients to 95.1% of inpatients since inception of the order set. Of the 125 CAM tool positive assessments, only 2 had orders through the order set.
Conclusions: The education and reinforcement measures that were provided to staff, resulted in heightened awareness and increased use of the CAM assessment tool. An increased use of the tool allowed better identification of hospital-acquired delirium patients. Although order set use is low, orders may be entered by bypassing the order set. Provider education is needed on order set usage. As this initiative recently launched, more data are needed to determine the impact of outcomes.