Development and usability evaluation of a nurse-led clinical decision support system (AI-AntiDelirium) for management of intensive care unit delirium: A mixed methods study

Abstract

Background

Clinical decision support systems (CDSS) have been identified to aid clinical decision-making, but few studies focus on the application of CDSS in intensive care unit (ICU) delirium, and particularly usability testing is not employed. We aimed to develop and conduct usability testing of Artificial Intelligence Assisted Prevention and Management for Delirium (AI-AntiDelirium), a CDSS designed to identify delirium and modifiable risk factors and prevent and manage delirium in the ICU.

Methods

Between January and April 2021, a cross-sectional study including 117 ICU nurses recruited for usability testing from four ICUs in two university-affiliated hospitals was conducted. The development of AI-AntiDelirium included needs assessment, function design, iterative design, agile development, and usability testing using the Delirium System Usability Evaluation Scale (Delirium-SUES).

Results

Based on the needs assessment, AI-AntiDelirium was developed to contain four main modules—delirium assessment tools, risk-factor assessment, nursing care plan, and care activity list—and was designed to provide individualized interventions based on patient risk factors. The mean Delirium-SUES score was 184.64 (full score: 210), indicating that AI-AntiDelirium was acceptable in terms of usefulness, ease of use, attitude, use tendency, and long-term effects.

Conclusions

Our study developed AI-AntiDelirium, a CDSS perceived as useful and easy to use. Incorporating usability evaluation when designing AI-AntiDelirium may be effective in and enhancing clinical staff use.