So when you talk about coordinated visits or length deployments we need an infusion of appointments where the patient must be seen by their oncologist. Earlier on the same day coordinating visits have some pretty clear benefits to the patient. For example, they reduced the number of trips the patient needs to make. The cancer center per their treatment cycle. But they also can complicate infusion center operations. Some of the negative impacts of linked appointments are forcing the schedulers to make a trade in trying to select optimal time for coordinated appointments. So the trade there is the infusion time is too early. The patient may arrive late for their infusion because there was a delay from the clinic and if the infusion time is too late.
The patient may need to wait a while after their clinic appointments to get their treatment. Additionally delays from the clinic can lead to unpredictable arrival for individual patients. Even if you’re typically getting the booking window light. There are always the patients who arrive much later, much earlier than expected. And another impact is that resources or staff may sit underutilized in the early mornings or late afternoon due to the way the infusion centers a schedule in the clinic schedules link up. So one example of that would be if the infusion center and the clinic schedules both start at 8:00 AM patients may not be making it through their clinics into the infusion center until something like 9:30. So that leaves a whole hour and a half or the infusion centers operating and staffed.
But can’t take back that many patients because we’re all doing their planning appointments and then on the other end of the day. Sometimes the clinic federal start to taper down in the early afternoon say around 2:00 or 3:00. And that means that those patients wouldn’t want to start an infusion in the late afternoon hours say five or six. If the infusion center is open. Then that time is also under utilized at the highest level. There are three steps needed to run an infusion center. The first step here is optimizing templates. So having optimized templates and that means that you’ve made scheduling corporate templates that have maximize the chance that the resources you need to start a patient will be available on time.
These templates should also rigorously track when nurses and chairs are being used by what patients in order to protect resources for planned utilization especially in the middle of the day. The second step here is looking at planning. So why the first step optimizing the templates takes place months before or saves the day of the appointment. The booking and planning step takes baby weight takes place weeks and days before the appointment. In this process the schedulers are booking patients to the thoughts on the optimized templates and then the clinic staff is also reviewing the plan for the day to identify bottlenecks and make sure there is adequate staffing.
The third step here is the delivery of care. So getting the treatment on the day of the misstep is sort of very immediate taking place right before, after, and during the treatment and their operational efforts to accommodate variability or other issues. In each specific patients flow through the center. So this is when a patient comes late where I put them Oh, I need an add on or I put them. What I do is if a patient treatment is running longer than expected et cetera only steps need to sort of happen in order to have a truly optimized flow of patients through the treatment center.
So now I’d like to introduce the concept of a booking window. The booking window is the idea that instead of leaving some static amount of time between the clinic and the infusion appointment let’s say, one hour you would be able to leave any range of times between the clinic and the infusion appointment. So a booking window would be something like 45 minutes to two hours, and that window is great because it gives us the flexibility to put patients and the correct template slots or the correct time of day. That will work for the infusion center without negatively impacting the flow of the patients. But the booking window has two components minimum time and maximum time the minimum time must actually be achievable for the patient based on your clinic’s historical performance.
So if you’re at a center for say third default just leaving 15 minutes between the clinic and the infusion appointment you might look at your data and see whoa. It actually takes for most of my patients a whole hour to get from their clinic to their infusion appointment. So in that case, you would want to stop using that 15 minute estimate or sort of school time and switch to using a more realistic minimum time lapse that has them derived from your past performance. The other component of the booking window is maximum time, which is typically determined by institutional standards and patient preference. One example is if a center typically doesn’t have wait times for the clinic or their infusion center. They may want to have a maximum window or something like two hours because they know that people can float pretty quickly through the clinic and the infusion and they want to be able to minimize the visit length for those patients.
But for a different center saved or having two hour wait time in clinic and two hour wait an infusion they would be maybe more comfortable with a longer maximum time on their booking window because say three or four hours first because you have patients already waiting that long anyway. So what’s to happen is if you have that bigger booking window. It’s going to let your infusion center stick to the slots better, which will reduce the overall wait times in the infusion center and you slowly close that sort of reduce the total visit time for those patients. So here’s an example that illustrates the power of the booking window. Well, there are two scenarios here. One on the left hand side and one on the right hand side, the one on the left hand side is the fusion schedule that we made by using just a static time between the clinic and the infusion appointment. So here it.
Each patient is booked one hour after their clinic. So patient one has a clinic at 7:00 and then their patient at and. Patient 2 has a clinic at 7:15 and an infusion at 15. And then two, three has a clinic at some 30 and then their infusion at 8:30. But if you look at the schedule say you only had two nurses you know for these three patients. It would actually be too many starts to reasonably achieve it. Once you see thinking that each. If you assume that each patient needs a full hour to get set up for their infusion this one will take this one hour infusion here. Nurse two will take the three on Fusion but then will be no one to take this five hour infusion down here at 8:30 because nurse 1 is busy until 9 and the nurse who is busy Toma in 15.
So this schedule isn’t good because these starts aren’t achievable. It’s already planning to either a require another nurse or B to force this patient to wait 30 minutes. And so it’s a less than ideal schedule because it’s already got wait time built into it. And doesn’t even consider that the wait time is going to be added on during the day of operations. And then over here on the right hand side scenario, we used a flexible booking window of 2 sorry 45 minutes to two hours to book each appointment into a more optimal space on the template. So you can see here that patient one has the same appointment the clinic at 7 and then infusion at 8 but patient 2 has the clinic at some 30. But doesn’t have their infusion until 9:30 or 915 all the way down here. And then patient 3 has the clinic at 7:30.
But then has their infusion just 45 minutes after their clinic. And what we’ve done by using the booking window is that we’ve been able to space out these infusion appointments that have achievable spacing that starts are doable for just two nurses. So you can see nurse one would take this one hour rotation nurse 2 would be available to take this 5 hour infusion here. And then by the time this patient 2 comes along. But either nurse one or two would have already set up their patients and be ready for the next person. So I just want to take this opportunity to point out that a booking window shouldn’t be the end all be all in scheduling. It certainly is possible to book outside on either end of a booking window, but you would want to encourage your schedulers to do it as the exception rather than the norm because you really want to have about 80% of your patients using a proper booking window to really feel the benefit of this practice. So how do you determine the right booking window for your schedule.
Well, the most difficult to determine part is the minimum time. That’s going to be the bottom of the window the time that you know you must leave between a clinic and an infusion appointment every three steps to sort of figuring out what that time should be for you. So the first step is to get for each patient to visit the clinic appointment time to the scheduled appointment time and the infusion arrival time or that the check in at the infusion center. So you can see up here examples of each. The clinic appointment should be a nice round number because we typically tend to book on the Tens or the 15th of the hour. And then the infusion check in or infusion rival stamp should be an actual time stamp and ideally, that represents when the patient has actually shown up in the infusion center and would be ready to receive their infusion. Everything we received that we use this time is because the typical flow through an infusion center is something like patients go to their labs they go to our clinic and then they usually go more or less right to the end treatment center.
Once they’re done with their clinic. So it’s a pretty good estimator of the minimum time that that person was truly available to start their insurance. So then the next step is you calculate the difference between those two times. And here I have them in minutes. So for example, for this patient. I had their clinic appointment at 9 AM there feed and check in at 10:33 am. The difference here is 93 minutes or an hour and 33 minutes and then you want to go ahead and calculate that difference for all the visits you have in your data set. Typically we recommend that you use a data set that represents at least three months of days three months of visits in your infusion center because you do want to have quite a bit of data to make sure the values you calculate are representative of what you guys can truly do. So to have all that data and the difference calculated you can go ahead and use a percentile function. And what that will do is it will transform your values here into a summary table that summarizes the distribution of those values.
So for example, here’s the percentile for this set of numbers. You could say, for example, that it’s four minutes is the 25th percentile. That means that 25% of the values here are less than four. And then conversely it means that a whole 45 percent of values are greater than that. So you can calculate that for a number of percentiles here at the 50th percentile you guys also know is the median in here is the 75 percentile, which means 75% of your patients arrive before 73 minutes after their clinic appointment. And what we recommend with this table is that you use the value either at the 16th to the 75th percentile as the minimum time and your booking window. And we simply recommend something like something a little higher, like 70 somebody fits instead of the same median as because the median or the average truly represents what about half your patients can make what is achievable for half your patients.
And you want that minimum window time to be only achievable for half your patients because in that case, only half your patients would show up on time. So you want it to be a little higher. So you can say something like most of my patients will show up on time. So that’s why we re max recommends being in sort of in higher values here. And then conversely we don’t recommend going anywhere above the 90th percentile because after that, you start to run into patients in cases who truly had some very unusual delay or extraordinary circumstance. And it’s OK if those patients show up late. You typically just want to have that window being so that most your patients can show by the bottom of the window. But not all of your patients need to show up by then.
There are a couple other factors that can affect the clinic to infusion delays. Some of the common ones that we have found our day of week time of day disease group, individual provider or the individual oncologist who is seeing a patient in clinic location. A lot of these factors are actually related to sort of the clinic sessions. So you know a provider might have a Tuesday morning clinic and then a Wednesday afternoon clinic et cetera. So they tend to just be related to how crowded those clinics are. This level of granularity is quite detailed and not all centers are going to need it. For example, you can see here that we’ve split it up by clinic appointment time morning or afternoon. And the department, the individual provider within the department. And then over here is the minimum time for each of the providers appointments in that frame.
So what you would want to do sort of a deep dive like this rather than using just one booking window is if you feel like you have a lot of variation in what the appropriate booking windows are. So sometimes you find that just a couple of providers will stick out if they are very popular or tend to run very crowded clinics. And in that case, you wouldn’t need to single out every provider. You could say something like for everyone except Dr. Tompkins afternoon clinic. You should use a minimum time of two hours. But for that specific g.I. clinic in the afternoon, you’d want to leave in the noon time a three hour because you know that’s truly how long you’re taking and having more than one booking window or splitting up the booking windows can be advantageous because it gives you an even greater chance of booking infusion appointments where the patient can make it. So again, not necessary for everyone. It’s typically helpful for people to see to figure out if they need it.
But it doesn’t necessarily need to be quite this complicated. I put aside one of the common questions we are asked is at the clinic schedules need to be changed in order to optimize the flow into infusion. So almost always the answer is no. And that’s because the infusion templates and the infusion scheduling practices should be adapted to the demands that they get from clinic and other sources of patients. So not patients as well. However, we can tell you what we look for in a good clinic schedule clinic several better, easier to work with than others. And so one of the things we look for is having an equal number of patients in the morning in the afternoon. And that’s because an infusion China typically have an equal number of staff in the morning and afternoon.
So there is equal space in both those pre noon and post room times and it’s better if the clinics are able to send you patients to fill that source spread out over the day rather than sending you all your patients in the morning or all your patients in the 10 to 12 hour. Another thing we look for in a clinic schedule is that they’re sending you roughly the same number of patients on each day of week. And again, it’s not like aha moment. You know you can accommodate it if they’re sending you only 30 patients on Monday and 60 patients on Tuesday. But what that does mean is that you’re probably pretty over utilized tend to stay in you’re probably pretty under utilized by Monday. So the more that they can spread out how many patients they’re sending me on each day. And with the better. And when I say that I don’t mean each provider has to send you the same number of patients each day, week.
Many providers don’t practice every day week. It’s more like the amalgamation of the clinic schedules that are working on my day. Ideally would be roughly the same. Again, because you have roughly the same number of staff and chairs many decisions are everyday weeks. Are you really trying to get the demands that clinic is sending you to match the supply. You have in the infusion center. And the last thing we look for in clinic schedules is to have a well coordinated start of both the clinic schedules and the infusion center. And I’ll go back to the example I mentioned a little bit in the intro is the case where if both infusion center and the clinic open at 8:00 probably patients are going to be available from the clinic to be treated in the infusion center until 9:00 or 9:30 or even 10.
So ideally don’t open at the same time. And even more ideally, if it is all possible for the clinics to open before the infusion sire that’s really the best case scenario because then the clinic can take some time to get those patients ready to go. And then the infusion center won’t be sitting there staffed. But with no patients. So having an early start in the day relative to the infusion center is like a good practice for clinic schedules. However, most of us live in the real world and have to live with clinic schedules that aren’t perfect. But that’s OK because there are some things that the center can do to accommodate that. So the first is having your fundamental infusion templates be a good match for the demand that you’re getting from clinics. So one example would be is if the clinic is sending you say more patients on some days of the week than the other. You want your clinic tablets to match that.
So if you’re only getting 30 patients on Monday you only really need 30 or 35 slots on Monday. And then if you’re getting way more patients on two days a 60. You need all 60 slots there on Tuesday because if you sort of like underestimate or underprovide the number of slots for what clinic is going to send you what’s going to happen is those patients are going to go to some other day. What’s going to happen is those patients are just going to get added on top of your tablets and they could be added on top of the temples and spaces that don’t work well for your infusion center. You want to make sure you have enough slots pre decided on pre optimized to take all the patients you’re getting from the clinic and then another thing you can do to sort of deal with maybe an uneven clinic schedule especially one that’s sending you a lot of patients between 10 and 2 rather than spread all across the day is still the early morning and late afternoon slots that are currently under your life with patients that don’t require a linked visit.
So we can talk a little bit about how to do that. I just want to go through this example here real quick. And you can see here that this is the chair utilization image for a center that has 17 chairs. So that’s chairs on the y-axis. And then time of day over here on the x-axis. So they’re operating from eight to 7:30. And you can see here in the middle of the day. And this one to four hours that they’re actually planning their schedule has already had them planned to go over their max capacity for planning to run out of chapters with this tool because they haven’t they haven’t utilized the and in the beginning of the day, really well. They have this peak here. That’s going to cause them some is when they go to operate the day. And then you can see down here. And there’s really only 10 patient hours here above this chart line.
There’s really only these 10 patient hours that are having the issue. So what they’ve done here is they moved some of those patients in their patient hours in their sort of treatment duration over down here to the afternoon slot. So it was just a couple of patients. So maybe three or four patients moved to this afternoon has level loaded their day in a way where they’re not planning to run out of chairs and they have a chance of operating without doing that. So we have this rule of thumb, which says you just really need to move 10% of patients from this peak of the day to the edges of the day. And that will free up a lot of head room.
So you can see they have maybe they have no buffer chairs throughout the day here. And here they have buffer chairs from most of the day except this one little picture. So it really can make a big difference just to move a couple patients. And you can find a good match for patients who would be amenable to moving to the early morning and late afternoon time are looking for a non ecology patients who often don’t require providers of it on the day of using look for treatments where the patient wants to get it before or after work. You could look for patients who would like to avoid rush hour trafficked patients who have planned supportive care only so someone who’s only coming in for a disconnect and doesn’t need to see their oncologist or someone who is coming in for hydration the day after the chemo.
And also, it I need to see there call this those would be good candidates for early morning or late afternoon slots and then this last item here at not all centers allow this, but some organizations do let their oncology patients see their provider up to 24 to 48 hours before their chemotherapy. And if you are, you’re living in a world where that’s possible. This can actually be a great tool for moving patients away to the peak and offering these off peak times and actually a little bit of a patient satisfy her, especially for people who again fall into these categories of wanting to work or avoiding rush hour and things like that. So that’s something to keep in mind as well. As far as getting patients to accept these off peak times that can be a little difficult, especially if people really want to come in after the morning rush hour and then leave before that and that sort of peak time.
There’s a lot of demand here, even not even not considering the need for late visits in an unlinked appointment patients may want to come at that midday time. And so a couple of centers have had a few strategies that worked well for getting patients to be accepted early morning and late afternoon times. And so one strategy is to actually have a strict apology a policy that stands non length appointments from the tended to our from those peak times and some something surely do need to do that because they need to reserve every chair and every nurse from tended to for those like patients. But other centers have tried something a little bit less harsh something a little more flexible and we actually recommend having the schedulers do something like offer off peak times.
First to any patient with whom they are speaking. So for example, if a schedule is trying to work with the patient to book a two hour infusion they may tell the patient. OK We have a slot available eight or 8:30. And in that case, the patient is likely to accept either either of the early morning slots that is offered to them. However, if the scheduler would have offered her the sort of 8:30 or 12:30. The patient probably would have taken the midday option. So it’s just a little bit about how you present the options to the patient and how much you need them to push back until they’re allowed to have a peak time that each individual scheduler can have on their own. So there is one other issue.
What is the connection sort of more of an operational issue. And that’s what happens when the clinic a you a patient much earlier. And much later than expected. So even if you have your booking window is about right. This can still happen. And in fact, will still happen as you operate. So our rule of thumb for early patients is that those patients should wait for their assigned time and not. So they don’t tie up resources intended for other patients. For example, if a patient had an appointment at 9 AM but they came at 8 AM you wouldn’t want to automatically take them away. Take them back early because there may be a patient booked for 8:00 AM and if that any patient shows up on time, then they may not have a nurse available to take them back. And so that on time patients would have had to wait would have to wait, even though the early patient didn’t.
So there is an exception to this rule of thumb and that you can take a patient back early if there is both an available chair and an available nurse, but you have to check for this thing first. You have to have some visibility into the way your center operates. So to find a free chair we recommend looking at the your plan chair utilization for the day. You can see this down here. This is the same sort of chair your physician we saw on the last slide where chairs is here on the y-axis, and then time of day is here on the x-axis. And this is just the planned utilization for the day. So it doesn’t have the actual durations just what you expect for the expected durations for each treatment. And so if you’re looking for a free chair at eight what you could do is go down here to 8:00 and see that. I only have four first patients patients scheduled to be in chair at 8:00 and I have 25 chairs at 8:00. So it looks like I’ll probably have a chair to take that patient back early.
But then you also want to make sure you have a nurse and visibility into that is a little more complicated for people who are using an infusion template. Typically they can just look at their bookings for the day and see if there’s a free slot because we know anytime that there’s a unblock slot on the infusion schedule probably means that they’ll be under three at that time because that’s the rule making the template is that you must put a slot only when the nurse is expected to be free. So if you look at your chair utilization and see if retiring, you look at your schedule and you see a free nurse in that case, you can take a patient back earlier than they were scheduled for.
One quick tip to make this sort of make it a little easier to find where you have room if it’s possible to coordinate with Linux and central scheduling to make sure that you aren’t patient waiting for patients who aren’t going to show up. That’s really helpful. So that would include patients who have canceled say their clinic appointments often don’t intend to come to mean choosing centers that day for their infusions so staying abreast of that will actually and keeping your schedule up to date will actually make it much easier to identify where you have space for patients and where you don’t. And for late arrivals we recommend a pretty similar strategy as early patients, which is yes, you should work to them.
But you don’t know if you can work them right away until you identify that free chair and free nurse for them. So use the same process of looking at the chair utilization in the schedule to do that. The one exception for these late arrivals is that if their treatment duration would cause you to run past closed unless you take them back right away you might want to consider letting them skip the line. And that’s because yes, it’s important to have on time patients start on time. But it’s also important for all your staff to go home pretty close to the closing hours. So it’s typically it’s typically more helpful at that point to prioritize that late patients and get them all started on their inpatient treatment, and then to have those on time patients come back. So there’s lots thing I also wanted to cover in this webinar and that’s the new analytics. We’ve rolled out within the iQueue app.
This is the patient arrival time we just we wasted a couple days ago and can really help you work on this issue of are people able to arrive on time to their scheduled appointment time. And so you can look here and see in this first graph up here there is for each day in your data. And so other 13, 14, 15, et cetera. There is just a summary of how many what percent of patients on that day arrives late early ends on time. So you can see here on this day that a bunch of patients arrived in yellow on early and only a few patients right on time or late. But then you can also see the summary statistics for that time range. And this can be really helpful if you’re just trying to get an idea of when patients are arriving and then down here is a different analytic that looks instead at time of day for the time range.
So for each day in the time range. It sums up how many patients arrived at that time. And what their arrival status was. So again, early arrivals here in yellow on time or arrivals here in blue and late arrivals here in orange. And you can sort of see that over the day the morning most people are in time. But then as we get a little later, people start to arrive more and more early so that could be in this case, for booking window can actually be a little too long. I’m going to make one might want to adjust that. But it does sort of let you see trends over the time of day. And it can help you measure success. If you’re trying to move patients from being to mostly non-white mostly a arriving late status to arriving on time. All right. So that concludes the webinar.
Thank you so much, Daniel. We have a couple of questions before we head out. We have one about calculating the minimum time. How would you calculate this if you already have appointments for infusion visits. Patients would check in related to the infusion of blame time versus right after their clinic visit. So in that case, I’m wondering why the patients are. I’m wondering if the implication is that patients are doing are doing other things. And not going right to the infusion center after their clinic point. Clinic time. So say, for example, maybe you’re encouraging patients to go to lunch or something. And I think that could cause something like that. And in that case, it’s a little bit harder.
What you could do is look at maybe this late time analytic cure and that could be the place to start, for example, if you know that something like half of your patients are arriving late. It may be helpful to push that minimum time up, make it a little longer and you could experiment with that and see if that helps people arrive more on time or if those people still remain to arrive late. That could be a good test to do. Another way to do it is you know you can do for a week ask all your patients to come to the infusion center as soon as possible and get that data, that way. And the last thing you can do is if you have a clinic checkout time you could use that instead in an off set of 5 or 15 minutes to sort of like the patients simply go to your infusion center or use the restroom grab a drink et cetera.
And in retrospect, I think maybe getting the clinic checkout time would be the most ideal of those options. But those are a couple ways you could go about trying to estimate the time that the patients need after their clinic start. You ready. Right and we have another one for the booking window calculation. Is it something that you can provide after the data review. Yes, absolutely. The one caveat is in order to get this. We need to have a clinic staff time. And that can be a little more difficult than our standard data request. Not difficult to pull but just difficult to organize. Getting it added in. So that’s the one small thing. So if you would like this from us go ahead and let us know. And then we can either get it for you. We’re already getting clinic information or we can work with you to arrange to have that sent over. Are you OK. OK Another question.
My clinic appointments can have different expected duration. How can I account for this when determining my booking window. I can tell you. Let me pull up this graphic real quick. So you have a couple options here. The first option is if you only have one, 2, or three different sorts of expected durations for your clinic appointments say returning visit versus new if something we see a lot. You could go ahead and calculate the booking window differently for each one of those you would just have to segment the data set. So make sure to mark here. This is a return visit of 30 minutes or a new visit of an hour. And then you could do this analysis separately and just have the two booking windows.
The other thing you can do in this might be especially helpful if you have a couple different expected durations for those clinic appointments is to change the meaning of a booking window to mean the time you need to leave after the appointment would be expected to end too when the infusion center should start. So the way you would calculate that as you would take have your schedulers take the clinic appointment time. In this case, say it’s 9:00. AM if the expected duration is 15 minutes they take the clinic appointment time add the expected duration to get 9:15 and then add the booking window. So it’s a little bit of an extra step for the schedulers. But it would be a way to have a consistent booking window across all of those expected duration. So those are really your two options is to have two different booking windows or be manually add the expected duration. Right OK.
One final question. Why don’t we need to coordinate individual appointments on the clinic scheduling templates and then scheduling conflicts. That is a great question. I actually have a diagram that might help us answer it. So you can see here about the flow through the infusion visit is sort of a bunch of links to different operations right to the lab sort of has its own operation. The oncologist clinics have their own operations infusion center has their own way of managing their operations. And what happens to the patient when they flow through here is that they encounter delays at each one of these nodes as they’re completing their visit through the infusion center.
And so the thing about those delays is because they’re so unpredictable for any individual patients. It’s a little hard to have a tablet slot here and like then expect them to make a specific time just one hour later and I’ll call you and expect the same patient to make the their infusion appointment just one hour later in infusion. And when you really want to do is work them reducing those delays between each one of these visits. And then you’ll be able to coordinate their appointments. And so there’s an analogy that might be kind of helpful here. And that’s one of airplane connections. So airplanes live in a world where typically they their flight durations are very predictable right. You don’t get an airplane going from San Francisco to Dallas and have the pilot tell you.
Well, it’s going to be anywhere from like two to five hours until we’re in Dallas right. They typically know about when they’re going to land. And so that predictability is what makes having connections possible schedules connections possible within the airport. But symptom infusions higher tends to have more in the land of having many some hour variations in the duration of each treatment. So there could be an variation in how long you stay. And I have an hour or variation, how long you staying clinic et cetera. It’s actually better to schedule those appointments independently for now until you’re able to reduce the wait time. Each one of these operations so that you can then start performing like an airplane schedule.
Great looks like that’s the last of our questions if you have any more. Feel free to email me. And I can for those under Daniel but our time is up for today and a huge thanks to Daniel and of course, to all of you for participating. Keep an eye out on your inbox for the link to this recording of the session and laws announcements for future webinars. Also please complete the survey at the end of the session. We would really appreciate your feedback and thanks again.