The Uses of Artificial Intelligence in Nursing & Healthcare

The Current Uses of AI in Healthcare

AI has already been integrated into many different aspects of healthcare: “We are still at the beginning. It has just started,” says Dr. Shaw. “Things are happening fast. Currently, in some electronic health records, there are algorithms that may or may not be considered AI that process data on patients. These algorithms can have clinical decision support tools that may identify health trends, interactions between medications, or might identify when a patient has high or low limits. It can act as a bit of a guide. It’s not that sophisticated yet, though.” This clinical decision support can help providers make decisions more quickly and potentially identify concerns that the provider may not have noticed. 

Another integration of AI in healthcare is electronic health records (EHRs): “Epic, one of the nation’s largest EHR providers, announced a few weeks ago that they are integrating ChatGPT-4 into their platform. When that happens, that will be transformational because you can now ask intelligent questions about the data in a health record. Data that isn’t structured just as numbers. Now, we could ask questions of the written notes a lot easier,” explains Dr. Shaw. 

Remote monitoring is also coming into play, allowing providers to gather patient data when they are not in the hospital or clinic. “AI is being used along with remote monitoring of patients. We can give them a device to monitor blood pressure or blood sugar. Things we would monitor in person can now be monitored at home. When the data comes into their clinical record, we can apply algorithms that notice trends and alert care providers,” explains Dr. Shaw.

The Future Uses for AI

It is impossible to guess exactly where AI will take healthcare delivery and nursing. However, Dr. Shaw has some predictions based on current trends and developments: “Algorithms in AI will someday be able to notice a change based on patient data and associate that change with a potential outcome,” he explains. 

“For example, if a patient has had consistently high blood pressure, plus their weight is going up, it can trigger the system to reach out to the client to come in to be seen. We don’t currently have the capacity to reach out to patients. Usually, patients come to us. Hopefully, this will allow us to move from a reactive system to more proactive care.”

Overall, the more data physicians can have on their clients, the better they can make decisions. “Due to people now having smartphones and the internet, there’s the capacity to collect health information beyond just traditional places like clinics or hospitals. That gives providers better insight into people’s everyday health,” says Dr. Shaw. 

However, there are concerns about the quality of the data received. “There is the promise that we could put smartwatches on people or give them devices to monitor their health remotely,” shares Dr. Shaw. “But how can you be sure that data is true? Is it the actual patient or someone else? Did they take the measurements correctly? The devil is in the details here, and there are still many things to work out.”

“I think one of the biggest promises that we all hope for is that AI will help with the documentation burden for care providers,” says Dr. Shaw. “Documentation takes so much time and isn’t direct patient care, but we have to do it, and it’s causing burnout. People spend hours on this per day. It’s so expensive. If AI can help with that, that would be huge.”

Ethical Considerations

While the integration of AI in nursing and healthcare delivery holds great promise for improving efficiency, clinical decision-making, and patient outcomes, ethical considerations must be taken into account: “It’s a bit scary because we don’t know how reliable it is. The AI is probably biased as there’s always bias in these kinds of algorithms. Also, the outputs are based upon the data that it has, and it can’t ask questions or gather more data,” posits Dr. Shaw.

Another concern with AI centers on equity and access. “Equity is a concern in multiple ways. One is, who has access to these tools? And not only do people have access, but do they have the capacity to use devices?” asks Dr. Shaw. “Who tends to go to hospitals and clinics? Generally speaking, they tend to be sick and might be older. Can they look at a phone? Do they have the dexterity to use it? Do they have the mental capacity while they are sick to interact with some of this new technology? We must consider different ways the technology fits into the care model.”

He adds, “This new technology might empower a nurse to visit a home and meet with a patient. It may be in some of these cases where we don’t give people these tools but rather use them to help them. But you must understand that different clinics and hospitals with different budgets and access will have these tools before places like a community health center might.”

There are still a lot of questions to be answered: “I don’t think we should move too fast. There are lots of unknowns right now. There’s a lot of caution in healthcare about integrating these tools into practice because we need to ensure they’re accurate. If you’ve used ChatGPT, you know it’s great, but also, it’s not always right,” warns Dr. Shaw. “Depending on the decisions that you’re making in healthcare, you need to be able to trust the source when you make your decisions. For example, if AI shows you an incorrect dosage, that can really harm someone.”

Advice For Nurses and Healthcare Providers

Undoubtedly, AI and more advanced technology will become a part of standard healthcare delivery. Nurses who want to stay up to date with the latest changes will need to be proactive. “Get involved with learning how to use technology that your patients may like wearable monitors. Become familiar with how to use apps as well. Check ChatGPT and have an understanding of how useful it might be, but also what its limitations are. Don’t be afraid to use tech and become literate in it within your workplace. Sit on committees at work where decisions about technology implementations are made because you can provide the perspective on how to tailor it to your work environment,” encourages Dr. Shaw. 

At Duke, Dr. Shaw and his colleagues keep the curriculum current to ensure students have the most relevant training. “We must teach students the material that meets the current market needs. We also need to teach them for the future, but you can’t cram everything in. As fast as we can and is appropriate, without overshooting, we are adding new things into the curriculum,” he says. “For example, we are increasing the integration of remote monitoring technologies and are beginning to pull those into clinical scenarios to match the experiences that students get when they go through their clinical rotations.”