Webinar Transcript: How Texas Oncology has leveraged machine learning to unlock and increase infusion capacity

With consistent pressure to do more with less, and large time and cost barriers to increasing physical infusion capacity, cancer centers must examine how to best utilize the resources they already have. Tammy Sayers, Chief of Operations at Texas Oncology, shares how 20+ Texas Oncology locations were able to increase volume of daily scheduled appointments up to 20% while using existing resources with the help of iQueue for Infusion Centers.

This webinar was originally hosted at ACCC. The transcript is below.

TS: We partnered with iQueue for Infusion Centers to leverage machine learning, and unlock and increase our infusion room capacity in our centers across the state of Texas.

My name is Tammy Sayers and I’m the Chief of Operations at Texas Oncology. I have been with the practice for a little over 25 years. My team and I across the state develop and implement policies and procedures, we roll out strategy objectives for the practices, and monitor metrics and our financials to evaluate the success of the practice and measure the quality of care we deliver to our patients. I also utilize my background in nursing which has been what I feel is an asset to my role in this organization.

The agenda today: I’ll give you a quick overview of who Texas Oncology is, and why we reached out to partner with iQueue for Infusion Centers, and what were our pain points. I’ll talk at a high level about the implementation and the expansion across the state. I will give some fairly detailed results on three of our cancer centers that have very positive outcomes. Then I’ll discuss the best practices we were able to achieve, any challenges and next steps, and then hopefully we’ll have a little time left at the end for some questions.

So who is Texas Oncology? We’re obviously an oncology practice in the state of Texas. But as you can see across the bottom of the slide, we also have surgical specialists, breast, urology. We do have a Texas center for proton therapy, and we have an interventional surgery center. But our main business is around the care of cancer patients.

You can see the stars on the map reflect the around 210 locations we have across the state, all the way up from Amarillo to El Paso down to what we call The Valley in Brownsville. Then the red dot in the middle there by Dallas is actually our Proton Center, which we’re extremely proud of. We have about 500 physicians around the state. As I said, the majority of those are medical oncologists. Partnering with US Oncology, we’ve played a significant role in more than 100 FDA approved cancer fighting drugs. We are very proud to say we are a community-based program, and we also believe in evidence-based care and value-based care models. We focus on the overall value and also the most important thing, our patient outcomes, and how we can improve patient cancer care.

Just a few little facts about Texas. There are about 1.9 million new cases of cancer, cancer that are expected across the United States. Of that 1.9 million, about 134,000 are expected to be diagnosed in Texas. 42,000, or a little over 42,000, are expected to die from cancer. Texas does rank third in the nation for the most new cancer cases, and also third in the nation for expected cancer deaths.

So why did we get to this point where we reached out to our resource and partner at iQueue for Infusion Centers? We really were experiencing growth across the state, more significantly in pockets. But we had already expanded our footprint over the last 15-20 years, to where we felt we had geographically covered the state, as we tried to be where the patients were so that they didn’t have to drive far for their cancer care.

But with our growth, we were finding we needed to increase capacity in our infusion centers. So we kind of took a step back and said “okay, we need to increase capacity in certain centers.” Is there any way we can increase that capacity without actually spending capital on expanding the footprint of the clinic and also on expanding the resources that are needed to support that footprint? We thought without having done any real analytics — we’ve done very minor analysis around it — but we felt we had an uneven utilization of infusion chairs. We knew just by visually watching the traffic at our infusion centers that we had that 10-to-two spike that occurred in the majority of our centers across the state, where it was just crazy chaos from 10 in the morning till two in the afternoon, and then it would drop off and be really almost too peaceful in the afternoon until five o’clock. So this 10-two phenomena was leading to longer patient wait times, unhappy patients, unhappy physicians. Nurses were exhausted, they were missing their breaks. They weren’t getting their lunches. So we knew we had to do a better job around scheduling in our infusion centers. 

Then as we all know, COVID hits and in Texas, it really hit hard into March 2020. We had the need to not only maximize our staffing resources, but also the existing space we had where we already were finding the need to increase our capacity. So we had a double whammy when COVID hit. 

What we did is partner with iQueue for Infusion Centers, then spend a lot of time educating them on what our needs were. I worked with my team and we developed a list of 20 cancer centers that were kind of our Hit List of where we wanted to start. So we worked with iQueue, spent many hours educating them on what the needs were in the cancer centers, how things flowed, how we scheduled all the ins and outs. They took a lot of time to hear how we practice in our world at that time.

The 20 clinics we chose to be on our initial list, they varied in size. We had a couple that were six, eight, 10, 20s was probably the average, but in the 20s to 30s 40s. Then we also had some really large centers with 100+ infusion chairs. These centers sat in urban areas, suburban, rural. We tried to choose various clinics across the state.

We started in November of 2020. Our first site went live in January of 2021, we went live in our largest center in the Dallas market with over 100 chairs. And now we are today in October of this year. We are currently at 20+ that are live. Today we’ll only talk about the 14 that had enough time to really measure, but we do have 20 live at this point in time. We have plans to continue our expansion across the state. We are in discussions with my team and the iQueue team to talk about the best sites, the characteristics of the sites that would most benefit from the iQueue scheduling platform.

The last thing I want to mention — I want to give a big shout out to the iQueue team. And in doing the planning for this iQueue rollout, we were also in the planning stage for the rollout of our new electronic medical records, which is going to be a year long rollout across the state of Texas. So we actually simultaneously planned our iQueue rollout to align with the rollout of our new EMR, so we would have the sites that we’re going to go live on iQueue go live prior to their rollout on the EMR. That presented an extra challenge, but the iQueue team rose to the challenge and worked very closely with us to understand how we needed to work through the schedule.

Early results — we continue to see these kinds of results improving. 12 out of our 14 locations had an increase in average scheduled appointments. Nine out of the 14, or 64%, had an increase in average daily scheduled hours in the infusion center. You can see that the ones that didn’t have this increase or improvement either stayed flat or decreased by a very small percentage, so it wasn’t anything that was alarming.

Remember, we were up against a lot of obstacles. We were up against COVID mainly, and trying to keep social distancing and limiting who was in our infusion rooms. So it’s quite miraculous to me that we had this increase in a time when we were having to go through all the COVID restrictions. 

11 out of our 14 had a more level loaded chair utilization through peak hours. These graphs show you how that looks. Eight out of the 14 had a higher median utilization during the afternoon hours, which was an outcome that we really wanted to see in all of our centers. In the graphs I’m going to show you in the next few slides you can see where we attained that outcome.

This is the first graph. We’ve got three cancer centers we want to show you the results for. This is Amarillo Cancer Center and before I get into the results, just a little story about Amarillo, it’s up there in the Panhandle and there’s really no other Texas Oncology Centers around it. They really have to support the cancer care in that community on its own without using partners in areas that are closer to them.

So they were experiencing some significant patient growth and the infusion room was busting at the seams. The nurses wanted more chairs, the doctors wanted more chairs, patients were not happy because they didn’t get the time that they wanted. We were talking about spending capital to expand the footprint so we could get more chairs in this infusion room.

But what nagged at us was that we knew we were already in the planning stages of building a bigger cancer center in the Amarillo markets. It was just a hard pill to swallow that we were planning this large Cancer Center that should be up in the next three years. But we have this immediate need. So with iQueue, knowing that, we set about trying to improve and get more people in the infusion room in a timely manner.

If you look on the left, I’ll try to explain these charts. The left hand side shows the number of chairs in the infusion room. This one had a max of 20 chairs, along the horizontal line you can see the times and the infusion room. It goes from 7:30 to 5:30 in the afternoon. You can see 26 weeks before, the average, before iQueue, we used the 26 week average. You can see they had a big slot, a big startup early between eight and 10 in the morning and then 10 o’clock, they would be going strong getting to their peak. This site did a nice job of taking a break, slowing down at around noon for their nurses. But then another little peak again hits real fast, and then they quickly shut down their day.

By utilizing the iQueue platform in the scheduling tool, we were able to get some better flattening out or ramping up in our day. So you can see the ramp up in the day evened out a little bit more, it was a little bit slower ramp up. Then they get to their peak level and they are able to extend — they still have their nurses getting lunch, but they were able to extend that peak later into the day. It looks like it’s been around three to 3:30, while they were at their peak previously at around 1:30 or two. 

So that allowed them to accommodate the growth that they were seeing in their infusion room. You can see they increase their average daily scheduled appointments by about 20%, going from 41 to 49. They also increased their daily scheduled hours by 19%, going from 98 hours a day to 117. Their total scheduled volumes went from about 5400 to 6000. This allowed us to really ramp up smoother in the morning. It allowed us to level load across the peak to support that large increase in volume and then we better utilized our afternoon hours.

This facility felt like we had really achieved the goals that they were looking for. Yes, they’re still in tight quarters. Yes, their cancer center is being built, so they see the light at the end of the tunnel. But continuing to use this optimization of the scheduling tool with iQueue for Infusion Centers, it allows them to continue with their volume and secure volume growth.

The next one is Odessa. Again, this is in West Texas. This is a smaller cancer center that doesn’t have the volume that Amarillo does, but they still were seeing the same ten-two peaks as most cancer centers. Their major goal was to increase that middle of the day utilization where you can see the boxes on the left and on the right. They wanted to get more patients in chairs on those days and then level out and plateau that peak.

So they also increased their volume. They increased their average daily schedule from 46 to 52 a day, which is about 13%. Their average daily scheduled patient hours  increased from 96 to 112. Showing by those boxes, they were able to increase the median utilization during their day. You can see the volume for them, their total volume went from 38 to 4500 during this period of time, but we did not have to hire any more nurses or expand the footprint in this facility either.

This last one was probably the biggest challenge because it is one of our largest cancer centers. It’s the one I’ve said had over 100 infusion chairs. It’s a crazy busy infusion center that is supported by, off the top of my head, around 80 physicians who feed into this cancer center or this infusion center. So there’s a lot of people to please, the physicians were trying to get their patients in when they wanted to get in, the patients were wanting not to have to wait for a chair, the nurses were frustrated because they couldn’t move fast enough to get people where they needed to be. They had some real explicit goals that they gave to the iQueue team around what they wanted to see with shifting to a more optimized schedule. 

You can see on the left their peak, it just kind of peaked and then boom, it dropped. We were like, “why aren’t you seeing more patients here?” and it was just such chaos. So we were able to optimize the schedule, use the AI tools that iQueue for Infusion Centers has, and really get a nice peak there, that was doable. It also kind of slowed down the day, optimizing those afternoon hours. They saw some significant increase in their volumes over this time period. They went from 108 average daily scheduled appointments to 125, about a 16% increase. They increased their average daily scheduled patient hours by about 25% and then across their optimization they level loaded significantly, much better, between their peak hours. Then they maintained that level such that they were able to get in more scheduled volumes. They rose from about 14,500 scheduled volumes to a little over 15,000. So they really accomplished their objectives. It was very much noticed by physicians, staff, and nurses, who were much happier with the outcomes.

So what did we get with our partnership? What were the best practices? We were definitely able to reduce appointment times, I mean appointments starting during lunch hours, we really were able to increase nurse satisfaction by giving them their lunch breaks. We have better scheduling of our patient arrivals. We knew when they were coming so we could accurately plan for nurses’ availability. Patients had less wait times, nurses weren’t running crazy for most of the day.

We used these advanced scheduling tools to more proactively plan. Nurse managers loved it because they could look at this tool in the future, or even at six o’clock this morning if they had a nurse that was sick, and quickly plan through their day using the scheduling tool. Obviously, you saw in our examples we were able to accommodate our increased patient volumes without increasing our fixed assets or our resources. We were able to better level load our schedules and see more patients.

The final bullet point, which needs to probably be at the top, is that the schedulers absolutely love this platform. They love the visuals in it and how they can see exactly where there’s a slot that’s available that meets the patients’ needs, the doctors’ needs, and the nursing staffing needs. The schedulers just think it’s the next best thing since sliced bread. They speak very highly of it.

Next steps — as I look at this, I don’t really consider these challenges, other than probably the middle bullet. We are going to continue to work with the iQueue for Infusion Centers team to further level load our day and optimize our afternoon hours. We continue to have COVID challenges, as everyone else does. This tool allows us to face those challenges head on and to continue not to strain our resources and better schedule our patients. So we do have a system wide expansion slated for the next few years and probably beyond. We look forward to our continued partnership and further developments in our facilities and further optimization.