Geisinger Health System, located in Danville, PA, is an integrated health services organization serving more than 2.6 million residents throughout 44 counties in central and northeast Pennsylvania. Its physician-led system is comprised of more than 21,000 employees, including a 1,100-member multi-specialty group practice, eight hospital campuses, two research centers, and a 467,000-member health plan.
Waiting for the “next big thing” before proceeding with an IT implementation can be a lesson in futility. Geisinger Health System faced such a conundrum when choosing an RTLS (real-time location system) solution, and here’s what they did to expedite the process.
![]()
Geisinger Health System, located in Danville, PA, is an integrated health services organization serving more than 2.6 million residents throughout 44 counties in central and northeast Pennsylvania. Its physician-led system is comprised of more than 21,000 employees, including a 1,100-member multi-specialty group practice, eight hospital campuses, two research centers, and a 467,000-member health plan.
Geisinger’s Director Supply Chain Technology and Process Engineering Kevin Capatch started at Geisinger in 2009 and one of the first projects assigned to him — the selection of an RTLS system — had been under way for almost four years. It was while working on this project that Capatch first met Sharon Kemberling, director of Geisinger’s Health System Transfer Center.
“She said, ‘You need to take this one over the goal line,’” Capatch said, laughing. “When you look at a program that’s been going for four years, and you start to say, ‘Well, what took four years? What’s going to keep it from taking another four years?’ I think what happened was new technologies were coming out at an alarming rate. We were in an almost endless cycle of looking for the next big thing.”
Capatch is right, of course. Finding the right time to install new technology can be tricky. But Geisinger eventually selected its RTLS system and Capatch shares his insights on the process and the impact the solution has had on the health system to date.
Q: What were the first steps you took when tasked with selecting Geisinger’s RTLS system?
![]()
Capatch: I first asked what work had already been done and was told the system under consideration was pretty much equipment-specific as we were looking for visibility into our physical assets. Somebody asked if it would track staff and patients, and that’s where I joined the party. We began the vetting process again with the vendors that had been chosen for the equipment-tracking solution to see if their product could track staff and patients.
It took a year to cycle back through those vendors and have them present their human tracking approaches and really boiled down to either what we called Wi-Fi-based location technology or non-Wi-Fi-based. Those were the two predominant options, and that’s when I get a call from Sharon who said she’d like me to look at TeleTracking’s RTLS solution.
Kemberling: In 2011 we switched from a thick client version of TeleTracking’s bed-boarding software to a Web-based version to better coordinate care for a number of different locations. This system has afforded us the opportunity to provide a level of transparency and visibility across multiple hospitals and care sites without having to physically walk the sites. It also left me with a good impression of TeleTracking, and I felt they deserved a chance to bid on the RTLS project.
Q: You were five years into the selection process when Sharon called. What were you thinking?
Capatch: I remember thinking opening it up would not be a smart thing to do. But TeleTracking deserved at least some consideration, and I also knew they had a legacy technology, which is what they were going to try to share. In my mind, it would be an easy open-and-shut case. It turned out they had their newest technology, which we didn’t know about. Again, that was that cycle. Do you let that new technology come in?
Q: What was it about TeleTracking that resonated most with you?
Capatch: The one significant difference TeleTracking had at the time was the ability to create a virtual light curtain. Most of the location technology is assuming, with triangulation, you’re somewhere. With a virtual light curtain, I can set up a room, give everyone in it a badge, and not have a physical wall using their co-location infrared technology. This technology was not on the market and wasn’t that difficult to deploy because the devices were battery operated.
What really drove this point home was the reaction of one member of the selection committee, our network engineer. In the year and a half I’d been working with him, I had never seen him lean forward and pay attention to any vendor presentations. When TeleTracking was demonstrating the virtual light curtain, he started asking questions and got very animated. After everybody left the room, I said, “Bob, what do you think?” He said, “I’m thinking, if this is really true, this is the way we need to go.”
I now had to tell people we needed more time, and we challenged TeleTracking to show us the technology in action. We made arrangements for them to come in on a Sunday, take over one of our conference rooms, and re-create what would be a physical space you would see in a hospital, and they literally demonstrated live. We could see the tag come into a room, leave the room, and interact with a hand-washing dispenser.
This was an off-network solution with a level of granularity. We weren’t planning on being able to have it, and the question became, “Can we afford it?” We went back and entered pricing, and it fell kind of in between the two previous options. With Wi-Fi you’ve got most of your infrastructure and you’re buying tags, but if you’re going to put a sub-network system in, you have to put in a little bit more infrastructure. We didn’t expect it to be cheaper than Wi-Fi, but when it fell in between, it became quick internal marketing to make people confident this was the right direction to go.
Q: What did you need RTLS to do at Geisinger? In particular, what physical assets were you most interested in tracking?
Capatch: I have to say, we might have been a little narrow at the beginning because the ROI was done on a sampling of intravenous infusion poles and pumps. Actually, it became one of the leverage points I used to sway the decision maker as we were going to do a conversion of infusion pumps. I said, “Well, this could be interesting, because from a ‘hunting, courting, and hiding’ standpoint, you’re going to have to create a threshold event to bring new pumps into the system.”
![]()
What we found, now that the system is up, is we identify a new class of assets we want to tag. We’ll find 70 percent of the assets pretty quickly, then struggle over the next two clinical engineering cycles trying to find the other 30 percent. That was what our ROI study actually proved when they went out and tried to do the validation on infusion pumps. They started the week with 200 pumps identified and every day went back to try to find those same exact 200 pumps. By the end of the week, there was 30 percent less than the original count. They just disappeared. Now you would eventually, if you really worked hard at it, have found them. But that’s not what you’re trying to do with an RTLS system. The RTLS keeps these items from being hidden or lost in the first place.
Kemberling: Every day, in our daily bed huddles, the need for infusion pumps was brought up. We’d search for pumps for patients and, once we got them tagged, we went from multiple incidences of needing to find one to not being able to remember the last time we had to have a conversation about it. I think there were some thoughts that the zone-level just wasn’t going to get us where we wanted to be. But I have to say, it was very successful in the way the zones were put together, to be able to be almost entirely transparent about location of the specific devices.
Q: How are you leveraging RTLS in terms of managing the staff and patient population?
Capatch: When we went live, we wired the EDs and the ORs for patient and staff. Before, if somebody moved from post-anesthesia care unit or pre-op to OR or OR to post-op, it was tracked manually. In the future, it will be done with at a minimum as a combination. At some point in time, it may get to the point where you might accept the RTLS system without having that secondary confirmation, which will provide a different level of visibility to that workflow, and that was the second reason we did it. We wanted to start to have workflow information so we could look at our response times, find out how long a patient is in the room before a clinician presents — those type of things.
![]()
I think we’re starting small enough that we can be specific to the key reasons people wanted tracking. They wanted to know what response times were, so that they could start to have conversations. It wasn’t somebody’s opinion; it was actually data that you could look at and have a clarifying conversation about the data.
Kemberling: We’ve done a lot of work on predictive modeling as well with our analytics team. Having the ability to overlay the actual numbers with projections has been instrumental in leveraging decision making around those kinds of situations. But we’re looking beyond that to the point of, say, from an infection control perspective, or exposures, who’s been in contact with which patient for which staff member, and helping to navigate the logistics of that as well.
Capatch: I think hand washing among nurses will be our next pilot. There are a couple areas of the hospital where we have enough granularity that we could turn that on, and I expect we’re going to see two things: a change in compliance, and then a sustainment of that compliance. What we won’t see are what’s happening with staff members not wearing tags. That’s going deeper, but somebody has to go first. I’m sure the nurses will ask their other care team members to join them. I think it will give people confidence in practice.
Q: Is there a “big brother” feel when people start wearing tags?
Kemberling: Initially there were some questions, and there were certain individuals who expressed that concern. But, as a whole, when presented with the reasons and potential gains they were all professionals and understood the benefits. Now, it’s quite the opposite. They’re thinking, “This is really cool.”
Q: How are you taking action on your findings?
Capatch: If you’re going to ask people to do things, you’ve got to give them time. In the Chief of Care Support Services group, we did a manual time study of nursing time on a floor. Most people realize nurses don’t get to spend a whole day nursing. A significant amount of a nurse’s time is spent doing what we classified as logistics activities.
![]()
To do a time study manually takes an engineer or somebody trained to hold a handheld device, determine who’s doing what and where, and then enter a code in the device. We did this study manually, and we’ve redone it using RTLS technology to validate. The RTLS technology will give us the same information, but our ideal goal is to give nurses back time by designing a different logistics role on the floor, called a care support associate. There’s somebody up there who supplants nursing at a level great enough to give them time to do the things that they need to. We are actively working on a give-back-time program and are using the RTLS system to help us measure that time to give back.
Today, we know we’re not very good at doing that, but we’re doing something about it. That’s one of the game changers at Geisinger: We don’t want to just know we have a problem; we want to know we’re solving it. RTLS will definitely tell you that you have a problem so it becomes an asset issue. We’re going to be able to move assets, because we don’t want to waste the care support associate’s time.
We don’t want to simply replace the nurse running around with the logistics person on their arm; we want to replace that with execution. The secret is visibility and maintaining that visibility to be trusted.
Geisinger Health System, located in Danville, PA, is an integrated health services organization serving more than 2.6 million residents throughout 44 counties in central and northeast Pennsylvania. Its physician-led system is comprised of more than 21,000 employees, including a 1,100-member multi-specialty group practice, eight hospital campuses, two research centers, and a 467,000-member health plan.
Capatch: I first asked what work had already been done and was told the system under consideration was pretty much equipment-specific as we were looking for visibility into our physical assets. Somebody asked if it would track staff and patients, and that’s where I joined the party. We began the vetting process again with the vendors that had been chosen for the equipment-tracking solution to see if their product could track staff and patients.
What we found, now that the system is up, is we identify a new class of assets we want to tag. We’ll find 70 percent of the assets pretty quickly, then struggle over the next two clinical engineering cycles trying to find the other 30 percent. That was what our ROI study actually proved when they went out and tried to do the validation on infusion pumps. They started the week with 200 pumps identified and every day went back to try to find those same exact 200 pumps. By the end of the week, there was 30 percent less than the original count. They just disappeared. Now you would eventually, if you really worked hard at it, have found them. But that’s not what you’re trying to do with an RTLS system. The RTLS keeps these items from being hidden or lost in the first place.
I think we’re starting small enough that we can be specific to the key reasons people wanted tracking. They wanted to know what response times were, so that they could start to have conversations. It wasn’t somebody’s opinion; it was actually data that you could look at and have a clarifying conversation about the data.
To do a time study manually takes an engineer or somebody trained to hold a handheld device, determine who’s doing what and where, and then enter a code in the device. We did this study manually, and we’ve redone it using RTLS technology to validate. The RTLS technology will give us the same information, but our ideal goal is to give nurses back time by designing a different logistics role on the floor, called a care support associate. There’s somebody up there who supplants nursing at a level great enough to give them time to do the things that they need to. We are actively working on a give-back-time program and are using the RTLS system to help us measure that time to give back.
