Development and Evaluation of a Standardized Nursing Language Clinical Decision Support System for Long-Term Care Nurses Using GPT-4.0–Generated Nursing Scenarios

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Purpose: To develop and evaluate the usability of clinical-decision support systems (CDSS) tailored for skilled nursing facilities (SNFs), incorporating standardized nursing languages to support optimal nursing care. Method: The current study used a user-centered design, with four participants from a single SNF. Large language modeling was used to train and test CDSS using a total of 130 Generative Pretrained Transformer 4.0–generated scenarios of simulated SNF residents. System usability was evaluated using the System Usability Scale (SUS); Usefulness, Satisfaction, and Ease-of-Use (USE) questionnaire; and heuristic evaluation. Results: Mean SUS score was 60 ( SD = 3.65). Mean USE score was 6.62 ( SD = 0.55), with the ease of learning subdomain rated highly. We identified usability issues through heuristic evaluation with nursing faculty experts, receiving overall positive feedback. Conclusion: CDSS has the potential to significantly enhance care quality, marking a pivotal advancement in the application of health care technology to support nursing professionals.