Do you know what you call an advanced mobile medical cart with a wireless thin-client desktop, a 24-inch monitor, a barcode printer, a Bluetooth handheld scanner, and a kitchen sink? Answer: a very popular, but expensive request from nursing. Some folks call these, computers on wheels. But we call them workstations on wheels, or “WOWs”, at MD Anderson.
When our IT Engineering team, led by Emil Patel, designed and built the standard workstation on wheels for clinical use, Chris Belmont, our chief information officer, asked me to make sure we tagged every deployed WOW using our RTLS solution. The WOWs were deployed shortly before and throughout our OneConnect (Epic and other integrated clinical information systems) go-live last March 2016. An active RFID tag was affixed to each WOW frame, as directed by Belmont.
..Being able to “track” WOWs required more knowledge about where WOWs belong, and to whom they belonged, than just their current location.
One day in late March Chris challenged me with a simple question, via text message, that enhanced my perspective on RTLS and asset tracking forever. The exchange went like this:
Chris: Are you tracking WOWs like we talked about?
Toby: Yup, we sure are. We’ve got around 400 in the system and we have about 125 more coming up to be deployed, and we’ll be tagging them once they’re ready to go. Why what’s up?
Chris: Can you define tracking for me?
It hit me right then when Chris asked me this question. I had decided that we were “tracking” WOWs, but my definition of tracking was really only “locating.” I could locate workstations on wheels within 10 to 20 feet of a spot on a map. But being able to “track” WOWs required more knowledge about where WOWs belong, and to whom they belonged, than just their current location. Also and just as importantly, being able to know when WOWs were “out of place.”
So I challenged my RTLS team to begin thinking differently about our use of the RTLS technology, beginning with workstations on wheels. We made a few changes.
The first thing we did was create “Home Zones” in our RTLS tool based on where each workstation on wheels were assigned. For example, our inpatient ICU clinic needed seven (7) WOWs assigned to them. So we created a group of WOWs called G11 – Intensive Care Unit NW, and setup a “virtual zone” in the RTLS tool initializing a Wi-Fi “boundary” around the clinic where these WOWs belong. When a WOW drifts outside of its “Home Zone”, an alert can be sent to a nurse manager (or not) indicating that a WOW assigned to them is missing. A simple webpage can also be opened indicating where the WOW is at the current time, and which Home Zone it actually belongs to.
That’s all fine and good, but arguably its still not, “tracking.” Well I guess it is if you were satisfied and wanted to stop right there, but we weren’t and didn’t. We wanted to know a couple of more things about these WOWs. One of them was, “how often do they move?”
We wanted to know a couple of more things about these WOWs. One of them was, “how often do they move?”
To detect movement, we needed to dig deeper than the standard reports provided by the vendor. We needed to get into the raw data to track every (x , y) coordinate movement on the map. We went back to trigonometry class and inserted a formula to calculate a 4-foot circle, then built a SQL query to isolate WOWs that hardly moved at all over a period of time. In fact, we found more than 75 WOWs that hadn’t moved more than 4 feet in 60 days! Other WOWs of course, traveled much more often and in larger patterns. Those that rarely moved were labeled as “low mobility”, and those that moved a lot, were labeled as “high mobility.”
Its one thing to be mobile, but it is another thing to actually be utilized. Full disclosure here, but we didn’t get utilization data from the RTLS. Who cares though, right? Rarely does one system give anyone the full picture. Many times, the best decisions are made based on several inputs rather than just one.
In this case, we joined the mobility data captured through the RTLS, with login information from our thin-client VM machines and active directory logins. We didn’t care “who” logged into each machine. We just cared how often the machines were actually logged into and for how long. With this information, we were able to measure utilization. The machines that were logged into the least were labeled as “low utilization” and those machines that were logged into most, we labeled as “high utilization.”
Figure 1 illustrates these concepts.
These workstations on wheels are not cheap. They are around $8500 a piece with all devices mentioned earlier included. So with 600 WOWs having been deployed, and more requests come in, a hospital can’t just order more machines based on a perceived need. A little fiscal responsibility has to come into play here, right?
So once the new request for another WOW is validated by business need, instead of putting the order in for a new one right away, we are now able to run reports to decide if there are WOWs that can be reclaimed.
Referencing Figure 1 above, the “Low/Low” charted WOWs are under-utilized and immobile. These WOWs are not being used by the clinics for which they were assigned and should be reclaimed for use in other areas needing more WOWs. For each WOW we identify as “low utilization/low mobility”, we save $8500 through reclamation and redistribution of the WOW to another clinic. That’s big.
The WOWs that fall in the “high utilization/low mobility” category can be acted upon as well. The IT leader should recommend that a stationary computer be permanently placed in the area where the WOW is being used. At an average cost of $1000 or less, a stationary computer is much cheaper than a workstation on wheels. After stationary computers are installed, the WOWs that fall into this category should be reclaimed and redistributed to other areas where mobility is more important. This will save the hospital $7500 per redistributed WOW.
For each WOW we identify as “low utilization/low mobility”, we save $8500 through reclamation and redistribution of the WOW to another clinic. That’s big.
So by using RTLS technology and data in a more innovative way, we are now able to avoid new equipment costs with ‘real’ information in hand, rather than someone’s opinion. This is a far more transparent and well-informed way to make decisions, and much more bulletproof I might add.
Please feel free to share similar ideas from your experiences, and how you save money as a result!