Background/Objectives: Nursing diagnosis is a complex process that requires clinical judgment, time, and resources and whose implementation is hindered by factors such as workload, lack of time, and resistance to computerized systems. This study aimed to compare the quality and efficiency of care plans generated by nursing professionals versus those produced by an artificial intelligence (AI) model, using the NANDA, NOC, and NIC taxonomies as criteria. Methods: An observational study was carried out with three simulated clinical cases. Thirty experts, fifty-four nursing professionals, and the ChatGPT model (GPT-4) were included. The experts established the referral plans using the Delphi technique. Responses were evaluated with a validated rubric (EADE-2) and analyzed using nonparametric tests. Professionals’ perceptions on the use of computer systems were also collected. Results: ChatGPT scored significantly higher on several dimensions (p < 0.001) and resolved all three cases in 35 s, compared to an average of 30 min for practitioners. Professionals expressed dissatisfaction with current diagnostic documentation systems. Conclusions: AI demonstrates high potential in optimizing the diagnostic process in nursing, although for its implementation human supervision, ethical aspects and improvements in current systems must be considered to achieve effective integration.
Loading....