The Hidden Cost of Hospital Inefficiency

In this episode, Sam Yeruva, Founder and CEO of Pycube, Inc., discuses the hidden cost of hospital inefficiency. Highlights of this episode include: What operational intelligence is How it changes the way hospitals function day-to-day How AI can be applied in hospitals Examples where operational improvements directly impacted patient care Kelly Wisness: Hi, this is Kelly Wisness. Welcome back to the award-winning Hospital Finance Podcast.  We’re pleased to welcome Sam Yeruva. Sam is the founder and CEO of Pycube, Inc., a company transforming the way hospitals operate behind the scenes. With a background in electrical and computer engineering and training from Harvard Business School, his decade working inside hospitals revealed a systemic problem. While clinical care is world-class, operations are often unpredictable. Motivated by a personal experience where critical biopsy samples were lost for 10 days, Sam launched Pycube to bring true operational intelligence to healthcare. Today, Pycube helps hospitals track assets and supplies in real time, saving caregivers hours and unlocking millions in recovered efficiency. In this episode, we’re discussing the hidden cost of hospital inefficiency. Welcome, and thank you for joining us, Sam. Sam Yeruva: Well, thank you, Kelly. Thanks for having me. Kelly: All right. Well, let’s go ahead and jump in. So, what is operational intelligence, and how does it change the way hospitals function day-to-day? Sam: Well, operational efficiency is a day-to-day operations that hospitals have. They’re like a well-oiled machine that actually runs millions of people who come in and get into the hospital. I call hospitals as nothing but– some people call them mechanic shops, which they actually– if you look at a repair shop where you take your cars, they actually make them better and send them back. The same way we all get sick and we go there and get taken care of, and then they fix us and they send us back into the productivity mode. I was talking to some doctors and they call it expensive hotel rooms. They’re providing a specific service. It’s just they’re full all the time, but they’re very complicated, run by very smart people, and they save our lives. So, it’s a well-oiled machine. It has a lot of components to it. There are very complicated things that they do to save our things. While doing that, they have to work with different disciplines to make sure that a particular patient is taken care of. While doing that, they have a lot of inefficiencies that pop up. It’s a process thing, right? People with good intentions come together who are well-trained in certain things. They do the job as well as they can. But you and I both know that recently the new technologies have come in. Now we are actually writing– when you go in patient registration, you have an iPad where you’re actually putting the information in. But when you go inside the hospital, there are a lot of places where there are still manual processes, they’re writing it on paper. There are good people, good nurses, and good people working in the health systems. They are doing the best they can to make sure they cater the patients. But what happens is things get lost. Things don’t appear. [laughter] It’s chaos inside that machine, and that is what we’re trying to fix to make sure it’s clean, it’s neat, the process flows are known so that the patient is taken care of properly in that area. So that affects us, you, me, and everyone who’s going into hospitals, and that’s what we are trying to fix. Kelly: That’s a great goal. I most definitely agree with all that. So how can AI be applied in hospitals in a way that is practical, safe, and measurable, not just hype? Sam: Well, [laughter] yeah, I’m very bullish about AI, and that’s a very good question that you asked. How do we do that? That’s a quick question. I was talking to one of the CFOs of a big health system and he would call me, and said, “Sam, is it true that I will have to change all my processes to make sure AI works in my environment?” I’m like, “No, no, no. AI should be used as a layer on top of what we’re doing. It should help us do things better. It should not change the way we are doing things drastically, but they should definitely help us do things better.” So, it’s a challenge to actually put AI in everywhere because it’s a common folklore right now that people think that, “Hey, this is not working. Maybe I’ll throw AI at it. Let’s see if it works.” No, it doesn’t work that way. I always go with a statement saying, “AI without PI is not going to work out.” Artificial intelligence is not working out– it will not work out without your practical intelligence. If you can’t fix it, if you don’t know how to fix it, then you can’t tell artificial intelligence to fix it for you, because it might give you something that you don’t like. So, there are definitely different ways. So, I think, first, there are different ways of implementing AI. The first, you have to understand, what is happening in the environment? What is going to happen in a particular workflow? A patient who’s giving you the sample, the sample is taken to the lab, which might be in that hospital or the hospital next door, or a couple of miles away. It’s been diagnosed, and then the report is given back to the provider, or in this case, the doctor. The doctor reviews it and gives you right diagnosis. In this entire process, there are many parameters that can go wrong, and you could be misdiagnosed or mistreated. So how do you ensure these things? If this is tightly coupled and if it’s tightly maintained, the data, if you’re collecting it, then you’re able to apply AI to make it better. But if anything in this entire thing is not properly working, then applying AI might give you wrong information. Garbage in, garbage out. So, it’s very important for you to have a digitized workflow which is properly maintained so that you can apply AI in a proper way and you can have a measurable outcome significantly improving the entire workflow efficiency and helping patients and helping providers to take care of their patients. Kelly: No, I love that. And I actually took down– when you said PI, the practical intelligence, I really loved when you said that. So, Sam, can you share a real example where operational improvements directly impacted patient care or reduced burnout? Sam: Oh, yes. So, I’ll give you an example in one of the hospitals, that healthcare we were actually working on. So, I’ll give you anecdotal– I shouldn’t say anecdotal, but without naming names here. One of the health systems where we are working, we showed them our tools where– it’s a big health system. It’s a big hospital that we’re working at. About a 700-bed hospital when they’re coming in. What happens is they collected the sample from one of the patients in the OR room, operating room, when they collected some samples. Millions of samples, thousands of samples that are collected every day, and one of the hospitals was actually doing the same thing. So, whenever you collect a urine sample or a blood sample, they’re supposed to go– based on their diagnosis or what they’re supposed to do, they’re supposed to go to A-lab, any lab, cyto lab, or a molecular lab, etc. It’s a clinical sample. And whenever you have a cancerous or a tumorous sample that they actually have a procedure on you, by giving you anesthesia or collecting them, they actually have to go to certain labs. We were just standing there and the nurse comes in– not a nurse, but a person who actually picks it up. He put them in the wrong spot. He picked up a sample which was supposed to go to A-lab, he put it in a B-lab, and the B-lab, he put it in A-lab and wrote it down and signed off and left. We were just watching there [inaudible] and I could figure that out. I’m like, “Oh, God, this might be in the wrong space.” It happened. And next day when we went back, and I was curious, and I asked the nurse, “Did you notice that?” And she was like, “Wow, would that happen? I didn’t know that.” She went back and started looking at it, and she caught it. And she tells me that these kind of things do happen. It goes to the wrong place. They don’t know where it is. Well-intentioned, but they’re all stressed out. They’re in a hurry. They write it down in the wrong space. That was one of the things that intrigued me because the intention of the person was not to go wrong, but when they’re writing it down, putting in the wrong bin, it just got routed to a different place. I’ll give you an example. We were standing in an ER room as well, and there was a patient coming in, and the nurse was actually looking for the tools. There are different kinds of tools that are required to take care of a patient. They couldn’t find it. They were running around. They were actually calling people to see– “Go find me some tools. I don’t have it.” They’re called PAMs. They’re different kinds of tools that they use for this ER. They were not able to find it. So, we showed them how to actually go and look for using the right tools with our technology, and they found it in five minutes. And they were very thankful to us because normally it takes– in a chaotic environment, imagine it takes about three months. Imagine you’re losing keys in your house, if it’s a big house or in your office, and you’re not able to find it. That’s the kind of chaos they have. So, we provide some technologies around it, which immediately they can go back and look for them. Having these tools, having this right technology to improve the workflow in the hands of the clinician...