Comparative effectiveness of delirium recognition with and without a clinical decision assessment system on outcomes of hospitalized older adults: Cluster randomized controlled trial.

Background: Early recognition of delirium is essential for effective management, but it often goes unrecognized, resulting in adverse outcomes. Clinical decision support systems can enhance adherence to guidelines and improve patient outcomes. We developed a mobile-based clinical decision assessment tool (3D-DST) based on the 3-minute diagnostic interview for confusion assessment method-defined delirium (3D-CAM). Implementing the 3D-DST may enhance delirium recognition and adherence to interventions among healthcare professionals, potentially improving outcomes in older adults.

Objective: To test whether improved recognition of delirium could lead to better clinical outcomes in older adults.

Design: A cluster randomized controlled trial with pair-matching.

Setting: A tertiary geriatric hospital.

Participants: Patients aged ≥65 years.

Methods: Four general wards were paired and randomly assigned to the intervention group (two wards) or the control group (two wards). The intervention included routine delirium assessments by nurses using either the 3D-DST or the 3D-CAM, along with delirium prevention and intervention measures carried out by a multidisciplinary team. Outcomes measured included delirium incidence, duration, severity, length of stay, and adherence to the delirium assessment, prevention, and treatment protocol. A trained nursing researcher collected data on demographics, clinical characteristics, and primary and secondary outcomes.

Results: 211 eligible patients participated (106 in the intervention group and 105 in the control group), with 21 identified as delirium-positive using the 3D-DST. The median Charlson comorbidity index score among older adults in the intervention group was 1 (1-2), compared to 2 (1-3) in the control group (P = 0.032). Nurses' adherence to delirium assessment was significantly higher in the intervention group than in the control group (73 % vs. 31 %). The recognition rate of delirium among nurses was 89 % in the intervention group and 42 % in the control group. There were no statistically significant differences in delirium duration (6 [3-9] vs. 7 [2-14], P = 0.967), incidence (8.5 % vs. 11.4 %, P = 0.500), severity (2 [1-3] vs. 2 [1-4], P = 0.891) or length of hospital stay (15 [14-18] vs. 18 [13-22], P = 0.568) between the intervention and control groups.

Conclusions: The 3D-DST enhanced adherence to routine delirium recognition by nurses. However, effective strategies are urgently needed to strengthen multidisciplinary collaboration and enhance adherence to delirium management among healthcare professionals.

Registration: Chinese Clinical Trial Registry, Identifier: ChiCTR1900028402.