Improving Delirium Screening and Detection in the Emergency Department (ED): The Implementation and Evaluation of the ED Delirium Detection Program.

BACKGROUND

Delirium is missed in over 75% of older adults presenting to the Emergency Department (ED). The study aimed to determine the efficacy of the ED Delirium Detection Program (ED-DDP) to improve delirium detection while evaluating implementation outcomes.

METHODS

The ED-DDP consisted of a train-the-trainer model where delirium champions (DCs) trained ED nurses to perform delirium screening using an electronic health record (EHR)-embedded Brief Confusion Assessment Method (bCAM). The ED-DDP was implemented across 3 diverse EDs using a stepped-wedge cluster randomized trial, consisting of control, implementation, and intervention periods. The RE-AIM (Reach, Efficacy, Adoption, Implementation, Maintenance) framework guided outcome assessments. Efficacy was defined as delirium detection between control and intervention periods. Implementation outcomes were assessed via quantitative (surveys, training logs, EHR) and qualitative (semi-structured interviews) methods.

RESULTS

Across the 3 ED sites, 94.4% (n = 17/18) of DCs completed delirium training and 94.4% scored ≥ 80% on the post-workshop assessment (Implementation/Fidelity). Over 90% (91.1%, n = 195/214) of nurses agreed to receive training by DCs (Adoption); the average score during nurse bCAM spot checks was 73.3% (Implementation-Fidelity). After ED-DDP implementation, the odds of delirium screening were 11.5 times higher (95% CI: [6.0, 22.3], p < 0.001), when adjusting for time and site clustering (Penetration); screening varied from 6% to 80% across the 3 sites (Reach). The proportion of older adult encounters with a positive delirium screen increased from 0% to 2.2% (p = 0.002; Efficacy). Qualitative data revealed that although DCs and nurses thought delirium screening was a priority for patient care quality and safety, competing priorities were a barrier to consistent and accurate screening (Maintenance).

CONCLUSIONS

Although a comprehensive ED delirium training program successfully increased delirium screening, detection remained low. These findings suggest that sustainable delirium detection in the ED requires not only robust training but also deeply embedded workflow solutions and clear post-detection action pathways.