Solving Operational Bottlenecks in Health Care With Artificial Intelligence

Michael Coen, Chief Product and Technology Officer, TeleTracking Technologies

Michael Coen explains how artificial intelligence is transforming hospital operations by converting vast, complex data into real-time, actionable insights that empower clinical teams, streamline patient flow, and reduce inefficiencies during even the most demanding periods of care.

Please introduce yourself by stating your name, title, and any relevant experience you’d like to share.

Michael Coen: My name is Mike Coen. I'm the chief product and technology officer at TeleTracking Technologies. I've been with TeleTracking for 9 years managing health care operations.

Many hospitals are rich in data but struggle to act on it in real time. What makes artificial intelligence (AI) uniquely capable of bridging that gap between insight and execution?

Coen: Hospitals have lots of data. They have too much data, and they can't sift through that data in an efficient way to make anything out of it.

AI is perfect for that because it can take massive amounts of data—complex data—sift through it as fast as real time, and provide context to that data and next step actions.

Humans can't always see the patterns in the data. Clinicians are extremely busy. Using AI to do all of the complex thinking allows humans to do what they do best, which is care for people. AI is built perfectly to do this because of its ability to do things that humans can't do at speed.

How can AI-powered operational platforms help health systems better manage patient throughput during times of peak demand, such as flu season or unexpected surges?

Coen: It is pretty straightforward because clinicians and nurses need to know the next best action for them. What should they be doing right now? AI can, as I said earlier, sift through that data very fast and is able to provide insights on what needs to be done and when. AI can make the recommendations to ensure that the nurse is doing things that are enhancing activity during those peak domain times.

Optimizing flow is essential to meet peak demands, such as during flu season or COVID surges. When nurses are struggling with the care that they need to provide, they should only be thinking about what they should be doing next. AI can reduce the bottlenecks of health care operations by making the decisions—or at least making recommendations very prescriptively—for nurses to say, "This is what you need to work on next."

For example, in the world of discharge optimization and discharge planning, how do you figure out the pattern of who to work on next? Nurses aren't doing that. They're not thinking in those ways. They're thinking about the care they need to give a patient. AI can tell them that in order to have a positive impact on the demand side of the equation for the beds, they need to be working on patient X first and then patient Y immediately following that.

There is a lot that can be done using AI that can make it much easier and better for the nurses during those peak demand times, and in general.

What are some of the most common operational bottlenecks hospitals face today that AI can directly alleviate, and what does that look like in practice?

Coen: One of the main things that I think of is putting the right patient in the right bed at the right time. Quite often, inefficiencies occur when a patient is placed into a bed, then has to be placed into another bed, and then again into another bed.

If they're in the right bed at the right time at the beginning of their care journey within the acute setting, they have a better likelihood of having a shorter length of stay and having better clinical and financial outcomes for the patient and health system respectively. That's one area.

Another one is inefficient transport. Transporters are being tasked to go off and move a patient from their room to the discharge location or from their room down to an MRI setting or to an operating room (OR) procedure. In that process, quite often that transport routing is very inefficient. The transporter that takes the next task is there because it's the next task to do. Is that transporter the right transporter? If you know geospatially where that transporter is, you can make decisions better in real time on what's the next best job for that transporter.

AI can help with simplifying things for people who have to go do work and can tell them, "Here's what you need to do," especially when no direct clinical outcomes are involved. Better or more efficient transport routing is a huge area where inefficiencies exist that AI can improve upon.

As hospitals adopt AI for operational decision-making, how should leaders think about change management and staff adoption to ensure success?

Coen: That is a great question. The way I look at it is you have problems, and you need solutions. Start with a real problem and try to solve that real problem using AI. That's first and foremost.

Secondly, focus on being extremely transparent about the purpose and the outcomes of what that AI is doing for you. It's not there to replace you. It's there to make you more efficient and make you better.

I think hospitals need to actually measure and share the impact of AI with the people who are using it. Show them how leveraging AI has made their job faster, made them more efficient, and helped them to make better decisions.

Finally, I would empower the users to want to do this. This is not something that is, as I said earlier, meant for replacing what they do. This is purposely meant to empower them to do better in their job or role. For that transporter, what we're trying to do is make them more efficient so they don't have to keep walking back and forth. Instead of 15 000 steps a day, if we change it to 10 000 steps a day, that's good because you are more efficient. If you've transported 15 people instead of 10 people, you've increased efficiency within the health system.

So, by leveraging AI, showing them the results, and training them on how to use the technology, they'll feel more empowered to get more done. They’ll also be able to see the results of their work through better outcomes. Those are probably the most important items I would focus on.