At Portneuf Regional Medical Center (PMC) in Pocatello, Idaho, nurses on the med-surg unit are receiving support through a new partnership that combines artificial intelligence (AI) with virtual nursing.
Amy Hemsley, MSN, RN, Assistant Chief Nursing Officer at PMC, says the hospital has partnered with care.ai to implement a virtual nursing solution enabled by AI inferencing and in-room sensors. Inferencing is the clinical decision-making process that caregivers use to devise patient care plans.
PMC uses virtual nurses who work on the same unit in a separate room and use the care.ai platform to continuously monitor patients for potential risks. These nurses can alert bedside nurses to problems before they arise, allowing care teams to be more proactive and less reactive.
Hemsley says virtual nurses can also check on patients via video calls. If the patient consents, a camera will be activated and the virtual nurse will appear on the in-room television. Patients can choose to opt out of using this feature.
“As we work through the implementation and fine-tuning of the virtual nurse process, our bedside nurses are seeing they have additional availability to perform hands-on caregiving, spend more time with patients, and focus on enhancing the quality of care,” Hemsley said.
While artificial intelligence in nursing isn’t new, the technology has grown in popularity with more hospitals using AI tools to improve patient care and support nursing staff. However, AI has also raised concerns among nurses and other healthcare professionals that the technology might compromise patient safety and ultimately make some nursing jobs obsolete.
Hemsley says virtual nursing and PMC’s AI-enabled solutions help the nursing staff provide quality patient care, but they will not impact current staffing levels.
“Nursing is a complex role and will always include tasks that require nurses to be in-person at the bedside,” she said. “As with all technology we implement, we aim to enhance the nurses’ role at the bedside.”
According to Hemsley, the AI platform has been well-received since it was first introduced in July. The hospital plans to slowly roll out the platform to other units.
AI in nursing personalizes patient care
At the University of California Irvine, Sue & Bill Gross School of Nursing, Jung In Park, PhD, BSN, assistant professor, says the university is leading the way in incorporating informatics and AI in both nursing research and education.
“Our faculty are increasingly incorporating AI and machine learning, which finds patterns in data and uses them to make decisions, into their research efforts,” she said. “By harnessing large and complex data sets, we uncover valuable insights that can elevate the quality of nursing care and improve the interpretation of medical records, including nursing documentation.”
Park says the impact of AI on health care is both wide-ranging and promising, particularly in its potential to enhance patient care.
She explains that AI algorithms can analyze data points from a patient’s medical records, lifestyle, and real-time health metrics. The data can then be used to forecast the patient’s risk of diseases like diabetes or heart failure, allowing providers to work with patients on disease prevention and personalized treatment plans.
While acknowledging the fear that AI might replace nurses or compromise patient safety, Park says it’s important to understand the limitations of AI in replicating human expertise.
“While AI can excel at sorting through large sets of data, making predictions, and even aiding in diagnostics, it lacks the understanding that nurses have of human emotions, ethical considerations, and real-time adaptability,” she said. “Nurses are often the first to notice subtle changes in a patient’s condition, due to shifts in behavior, appearance, or verbal expressions — elements AI is currently not equipped to fully understand.”
Park acknowledges that AI technologies come with their own challenges and risks.
“Like any tool that accesses and processes patient data, AI systems are vulnerable to data breaches if not adequately secured,” Park said. “Furthermore, algorithms can inherit biases present in the data on which they were trained, or from the people who created them.”
As an example, she says if an algorithm is trained predominantly on data from a specific ethnic or age group, its predictions and recommendations may not be as accurate or equitable for people outside that group.
“These concerns make it essential that the integration of AI into health care is managed with meticulous attention to data security and ethical oversight,” Park said.
Artificial intelligence in nursing education
Nursing students also are beginning to learn about AI technology in college — and for good reason. Park says, given that the future of nursing and health care will be shaped by AI, nursing students must have an understanding of how it works and how it can affect their paths as nurses.
“Our curriculum at UC Irvine is tailored to equip students with the skills they need to effectively navigate and utilize these tools in their future clinical practice,” Park said.
Beginning this fall, nursing students at the University of California, San Francisco (UCSF), will be learning about AI’s applications.
Yoshimi Fukuoka, PhD, RN, FAAN, a professor at UCSF’s School of Nursing, developed the school’s Artificial Intelligence course with a colleague and says fear of AI often stems from not fully understanding the technology’s potential applications and risk mitigation methods.
In addition to the AI course for nursing students, Fukuoka’s research focuses on using AI to prevent heart disease and type 2 diabetes.
“I’m developing a fully automated AI text-based chatbot to provide women with educational resources on heart disease,” she said. Chatbot uses AI and natural language processing and aims to mimic human conversation through text or voice interactions. Fukuoka cites Alexa and Siri as applications powered by natural language processing.
“We know heart disease is the leading cause of death in the U.S. and our chatbot will provide resources tailored to the unique health information submitted by each user, covering topics including risk factors, symptoms, and when to call 911,” she said.
In a pilot project that recently tested the algorithm, Fukuoka says that participants, two-thirds of whom were from diverse and ethnic backgrounds, reported improved knowledge about heart disease.
In response to the growing incidence of diabetes and obesity, Fukuoka and her colleagues are also working on an AI-based intervention to create a predictive model that can personalize a patient’s weight loss program. She says nurses can recommend both heart disease and diabetes/obesity AI-based interventions as preventive care measures to help at-risk patients achieve better health.
Moving forward, as technology continues to evolve, it’s safe to say nurses — and the public — will see more examples of artificial intelligence in nursing. However, as with any other tool being used in health care today, it will be necessary to closely monitor its implementation, safety, and usefulness over time.
Comment below and talk to your fellow nurses about this important topic by downloading the Nurse.com app.