In the latest episode of DrDoctor Will See You Now, we explore the intersection of technology and operational change in the NHS with Milica Stjepanovic, Deputy Divisional Director of Operations at Imperial College Healthcare Trust NHS. With Tom, she discusses giving patients options, control and agency through "Choice Booking" of appointments, as well as the impact this has had on operations and the patient experience.
The scale of the problem: Imperial was booking around 1.7 million appointments per year but only seeing 1.2 million attendances - a gap of 500,000. Every time a patient is rescheduled, their DNA rate increases by 2.5%, creating a self-perpetuating cycle of waste.
The letter burden: The traditional booking model generates a cascade of letters - booking, cancellation, rescheduling and rebooking - often arriving simultaneously, without any direct conversation with the patient.
The case for Choice Booking over Partial Booking: The term "Partial Booking" has negative connotations from failed previous attempts. "Choice Booking" reframes the model around patients actively selecting their appointment, with the booking treated as confirmed.
The 8-6-4-2 model: Inspired by the 6-4-2 model already used in theatres, the 8-6-4-2 model structures the booking window as follows: at eight weeks, patients are texted to validate whether they still need their appointment; at six weeks, capacity is locked and patients are invited to self-book; at four weeks, any remaining gaps are mopped up; and at two weeks, urgent and cancer patients fill remaining slots.
The importance of a centralised booking team: Centralised booking is presented as a prerequisite for this kind of change, enabling quicker implementation, phased rollouts, consistent training and proper management of the waiting list.
Managing the unbooked waiting list: Patients are placed on the waiting list the moment a referral is accepted but are traditionally left with no contact until letters arrive. Imperial introduced proactive 12-weekly text updates to reassure patients they have not been forgotten, significantly reducing anxiety and inbound phone calls.
The single waiting list principle: Having multiple waiting lists - DNA lists, cancellation lists, new and follow-up lists - makes it impossible to prioritise fairly or consistently, and impossible to digitise. A single new and a single follow-up waiting list is presented as a fundamental requirement for effective management.
The staff experience: Constantly shifting priorities and complex multi-list working creates stress and inconsistency for booking teams. Standardisation and clear processes allow staff to focus on meaningful patient-facing work rather than administrative churn.
Using technology to free up human capacity: Automating routine communications allows staff to focus their time on the roughly 20% of patients who genuinely need additional support rather than processing high volumes of rebooking.
Serving deprived communities: Imperial serves one of the most deprived populations in London. Technology-enabled automation frees up staff to provide more personalised support - including language and translation assistance - to those patients most likely to disengage from the system.
Tom: So welcome to another episode of DrDoctor Will See You Now, the podcast where we explore the intersection between technology and operational change in the NHS. And I'm really excited to say today we're going to be talking about a subject which is really close to my heart, which is giving patients choice, control and agency through partial or choice booking of appointments. Now, this is special to me because the very first thing I ever did when I worked in the NHS was help Coventry Hospital move from a full booking to a partial booking process many, many years ago. And so I'm absolutely delighted to have Milica, who's Deputy Director of Operations at Imperial Healthcare NHS Trust, I hope I've got that right, and an expert in patient booking models to come and talk to us about the work they've done. Milica, would you like to introduce yourself?
Milica: Yeah, hi, thank you so much. I'm Milica, everyone calls me Mel, that's how I'm known at Imperial. And I'm Deputy Visual Director of Operations, I work for one specific division. And I was responsible for rolling out the digital booking and changing our model of how do we book the patients from moving us way too far in advance into restricting to only six weeks in advance.
Tom: Amazing. And I can't wait to hear more about this, because from what I hear, the results have been really fantastic and has been incredibly positive on the operational side. Can we start right at the beginning? So before the project started, what was the problem? And why did you start the process?
Milica: Yeah, so I've worked for Imperial for a while now, just over 10 years. And I went through the ranks, did almost every single banding and job, moved around different departments. And I was always fascinated about the bookings and how we can improve that, because that's the start of the patient pathway. It's the first contact that patients usually have with our hospital or any hospital. And if you don't get that right, everything else falls down. Opportunity, patient experience, trust performance, income, it just snowballs. So I was always very interested at it. And then opportunity was presented to me to take out patients department at Imperial, which I took actually. I really wanted to do it. So it's not that they had to beg me. I probably begged my boss more and said, please, please let me do it. I want to give it a go. I spoke to quite a lot of managers around the hospital and a lot of our clinicians to kind of just do a little bit of a deep dive. What is it that's frustrating them and what is it that doesn't work for operations? And then we work very closely with Helix, who did quite a lot of user insights for us, specifically on the side of the patients. What is it that they're frustrating them? The number of letters that we are sending, which is just, you know, crazy amounts of paper, you know, in the 21st century was something that really bugs me and I wanted to resolve it. And also the impact on the finances is significant. It's not cheap to send that many letters. And then I looked into some of our data. For example, we looked into the number of booked appointments and then the number of patients actually attending the appointment and the contracts is quite significant. We booked around 1.7 million and attended around 1.2 million.
Tom: Right. Okay. A huge amount of waste there.
Milica: A huge amount of waste. And if you just think from the operational perspective, rebooking all of those patients, that takes not just staff, but it's a huge amount of work and the likelihood of getting it right reduces every time when you cancel the patient. And then we finish in a problem where patients are not booked in chronological order. The RTT is not improving. Patients are frustrated. I don't even want to talk about that because that's just a poor experience. And I use example of the airport and airplanes. And I said, can you imagine if we are booking our planes, tickets, and then we book 1.7 million, then we just cancel 700,000. That would not be viable business in any shape or form.
Tom: It wouldn't exist.
Milica: It wouldn't exist.
Tom: It wouldn't exist.
Milica: So why it's okay to run the public organisation like that? And I was adamant that we can improve it.
Tom: Yeah. And we see these sort of stats in every hospital we work in. So 30 to 40% of appointments booked never happen. Hospital initiated cancellation, patient initiated cancellation. And if you think about it in terms of resource, it's probably 30 or 40 people full time just managing that churn. And then the other really fascinating stat that the team have found is that every time you reschedule a patient, the DNA rate goes up by 2.5%. So as you say, we're creating work for ourselves and a worse patient experience.
Milica: And every time you do that, you send a letter. So there is a booking letter, cancellation letter, rescheduling letter, rebooking letter, and then it just goes around. And majority of times all of those letters arrive at the same time. And we do that without even speaking to the patient once. So as a patient myself, we're all patients. We all have attended hospital in some shape or form. For me, it was just, okay, can I just get one letter? I just want to know when is my appointment and I hope you won't cancel it. And I have started for myself and then worked through the problems with quite a lot of people. It wasn't just me. There was a full team of people at Imperial who was really on board on this journey and said, you know, we need to try. We just need to try and give it a go. And it worked.
Tom: Yeah. You should always try and give it a go. And of course you're really lucky at Imperial because you do have Helix and you do have all of that additional resource. And I presume that helped, but thinking about that problem statement, I assume this is the same for everyone. So if there was somebody listening to this who perhaps works in an organisation that doesn't have that additional resource, how do they get started?
Milica: So we do have a quite ambitious outpatient transformation programme. So yes, we are very, very lucky in that sense. But it is possible to be done just with operational management involved who wants to change. Now, the setup in each organisation is a little bit different. And I believe in order for this to work, there has to be a centralised booking team so that first of all, you can implement change quicker. You can do the phase way and roll it out across organisation in more structured, safe way. And also you can train your workforce to do exactly what needs to be done. So we were quite lucky that Imperial is majority centralised. We're in the process of finishing that and getting all services on board. But I think that played a quite big part because as somebody who was responsible for outpatients, I was responsible for all booking. So when I got the pilot services on board and clinicians wanted to try something new because nobody could have argued that it works. So I didn't have to make a case for change in that sense. I just had to make the case that this is different and that I believe that it's going to work now. I'm a very energetic person and I have big personality. So I'm happy to use that and just get people motivated, get them on board and say, you know, we need to try or we're just going to year on year complain that it doesn't work.
Tom: Exactly. I think you tell a great data story as well, because I think that's the other side of this is once you start looking at some of those numbers, you realise how endemic this problem is. And it's not a sensible way to run anything. And, you know, we know that there's a direct correlation between DNA rate and how far away you book an appointment. We know that for a fact. Your point about clinicians is interesting because I certainly remember when I first started looking at patient self-booking and choice, one of the arguments that landed really well with my clinical colleagues was giving them some flexibility, because obviously the challenge is at the moment, if you want to book leave, it means you have to do a whole load of rescheduling appointments. Were there any other conversations you had with clinicians early on in this sort of journey that either positive or negative?
Milica: Yeah, I mean, there was both, you know, delivering change of this scale in an organisation as big as Imperial is never straightforward. And not all of us are on the same page, but I believe in a journey and taking us through the pain and through the success. There was a couple of things. Our consultants, probably every consultant across the country, they like to see their clinics booked. They like to see them booked in advance. They like to know and we cannot blame them for it. I like to come to work knowing what meetings I have in my diary and then you can plan your day. So I have understood that. But I think when I presented them the case why I want to try this and why I believe in it, it was difficult to argue that it's okay to cancel patients three times in average. I think that's difficult to argue for what reason. We should not be cancelling. So we should not be booking patients into capacity that we don't have, that is not confirmed. If we do want to do that, then we have to change how we do it. There is no point of promising appointments that we know 40% of the chance it won't happen. And I think when we went into those conversations, there were consultants and services, it was like, okay, let's try. Let's give it a go. We had to develop quite a lot of other things, the dashboards to support them to see the lack of capacity that we have to understand the backlog. I wanted to trust to understand the backlog, not just managers. I think it's all of our problem. It's not just one person's problem. And I think when we started going through that, then it became a little bit easier. Now we have quite a lot of consultants at Imperial. So everyone manages clinic in a different way. And the point that I have made is we can't manage organisation like that. It has to be standardised or as close to it as possible so that we can then manage administration in a way to deliver. Productivity is quite a big thing in the last two years, more than ever. It's almost the only thing that we're talking about. It's the productivity. If you're cancelling 700,000 appointments a year, just the pure administration that I need to run that, to run it properly so that we don't lose the patients, the patients are not missing their appointments. I think it's impossible. And I'm yet to see any organisation who managed to do it properly.
Tom: Your point of standardisation actually does create flexibility ultimately.
Milica: Correct.
Tom: And when everything's done differently, you cannot run a productive organisation. And I love, and I would like to spend a bit more time on later, the backlog is everyone's problem. It's not the management's problem alone. It's as an organisation, how are we going to solve this problem? And that's exactly the right attitude. So Milica, can you just dive into a bit about how you design the 8642 model?
Milica: So the 10-year plan from NHS England came out and it was that every patient needs to be validated before it's booked. And we all believe in that. Every manager does it. But the pure volume, validating 1.7 million appointments, it takes a bit of time. It's not straightforward. Now with technology, it becomes a little bit easier. So we have used doctor-to-doctor technology, but there is also the different providers there. And we were like, okay, so we want to validate all the patients. How we can do that? I can't put the workforce on it. I don't have it. So, okay, let's start texting them. Do you still need this appointment? It's actually as straightforward as that. And then when should we do that? Okay. Now you can go private, your circumstances changes, you move out, loads of things, loads of reasons why the patients don't attend. So we decided for that to be eight. Now we run 642 in theatres. It's a well-established methodology and it works very well. So for me, it was like, well, if it works in theatre, why can't we work it out? Why do we need to have so many different methodologies? Consistency again. Correct. Much easier to train people, much easier for managers to understand it. It's just easier, isn't it? So I was like, okay, how we can do the 642 now in outpatient settings? And then I was thinking about consultant's annual leave. Now in our organisation, the policy is six weeks. I'm not so sure if that's everywhere. I think that's quite consistent. But I do think it's quite consistent. So, okay, in six weeks we will lock capacity. We know what we have. We know what we don't have. Of course, all these cancellations, hospitals are big, we have a lot of pressure. But majority, we know that on six weeks we will have capacity locked in. And then we can start texting patients, book yourselves in. So we decided to restrict capacity to six weeks. Four weeks is when we, okay, let's mop it up. What is not booked, short-term cancellations, additional clinics, lots of things can happen in the operational world. So we have locked it on the four. And then on the two is just, let's book the urgent that we have not done, cancer patients that we have not done. Let's populate those clinics. And that's how we came up with 8642.
Tom: That's nice.
Milica: Now it was too long.
Tom: Got the eight.
Milica: Not catchy. So our lovely communication team says, well, actually Milica, what you're doing is providing the choice to the patients. And then that's how the choice was born.
Tom: And that's how you have choice booking. So this is one of my bones of contention here, Milica. So talk to me, because this is partial booking to me. But you don't like partial booking. You like choice. Tell me, what have I been wrong about for the last 12 years?
Milica: I don't think you're wrong. I think the word is wrong. So partial booking means that's how at least Imperial did it a long time ago. You book the patient partially. Okay. And then somebody needs to call the patient and confirm the appointment.
Tom: Right. Okay.
Milica: Okay. I don't believe in that because you need to call 1.7 million patients. It's a lot of patients. There is no way you will be doing that. And I mean, you will try maybe for six months and then everything will collapse, which then means the DNAs will go up. The patients will start being lost in the system. They didn't get the letters. So then you rush with the letters and it just snowballs down, isn't it? But I do like the concept, but not the word.
Tom: Okay. Okay. So we're going to agree from now on, it's choice booking.
Milica: I believe it's choice booking.
Tom: We're going to go with choice booking. We're going to put partial booking in Room 101.
Milica: We're going to put it in the bin.
Tom: It's in the bin. I would put it in my path box.
Milica: Fair enough. Because to be honest, what I would historically have called partial booking is what you called choice booking.
Tom: Exactly. It's the same principle. I think loads of organisations had a bad experience with it. It's not just Imperial went through a little bit of a problem. I think that we all experienced it and it's because of that. It was just the way of how I believe if you book the patient, it's confirmed. It has to be confirmed. There can't be, oh, let's now maybe call them, play out, let's move them. It has to be, you lock it, patient is locked. You get rid of all that churn and all of those cancellations.
Tom: Correct.
Milica: Okay. So we've accepted that we're doing choice booking. That's great. So just tell me a little bit about how you manage your unbooked waiting list.
Milica: Unbooked waiting list is a capacity. And I'm trying to separate those two together. How you book the patient and the capacity you have, it's close together, but they are independent. The patient gets referred to the hospital. The moment the referral is accepted, the patient is on the waiting list. That's how we run it at Imperial. Now those patients sit on those waiting lists. We don't contact them. There is radio silence until you receive the letter and then you receive the cancellation letter a couple of times.
Tom: Great experience.
Milica: So yeah, it's fantastic. So what we have done is we start with your product, we designed the welcome message. And we are texting them every 12 weeks to tell them that they are on our waiting list. We have not forgotten about them. We have a bit of a problem with news because it's a different system. We are on ERS there.
Tom: And ERS is a whole conversation that I feel that we need to come back to. Both positive, but also really challenging for some of the operational change. So I'd love to explore ERS, but I'm going to put a pin in that. Keep telling me.
Milica: But we still can text the patients. So we have found a way around it by using the Dr. Dr. product because you have your waiting list. If you have your waiting list, you can upload and you can text. You can always find a way to do it. It's not maybe the streamlined, but there is always a way to do something. So that's why we implemented the eight weeks text messages. Do you still need appointment and welcome message? We know you're on our waiting list. Of course, it's our communication team who did the text messages, not me, because mine would be very simple, probably wrongly spelled.
Tom: To the point and people would do what they said.
Milica: I just believe straightforward, without the fluff, but they're very, very lovely messages that my colleagues came up with. So we want to keep them engaged so that they understand that they are not lost. Because one of the biggest fears that our patients have and the feedback we got is that they think they're always lost. And then that happens, but it doesn't happen as often, I think, as we think it happens. But it's one of those things that just becomes a fact.
Tom: And it holds people back from moving to choice booking, this fear of lost patients, doesn't it?
Milica: Correct. So what we did to solve that problem, I just presented, okay, those are the waiting lists. They're not lost. They're there. We can see them. We're just not booking them. And there is multiple reasons why patients are not being booked properly. We have not one waiting list, but there is multiple waiting lists. But you only have one slot. So how do you decide which patient goes into that slot? So all of those are real problems that operational management has and that we need to resolve. And with this process, we have managed to improve it. We have not resolved it because for that, we need a single waiting list. I think it's a must that you have one new, one follow-up waiting list. That's it. Nothing else. There is no DNA waiting list. There is no cancellation waiting list. Everything is one.
Tom: One source of truth.
Milica: Correct. So that you can then, because if you have a structured data, you can digitalise that data. You cannot digitalise something that you have five different waiting lists, one slot.
Tom: Yeah, it's impossible.
Milica: How that goes.
Tom: And so often the rules around this are just in someone's head, right? And this is a big challenge, I think, with scaling anything in the NHS is really talented people who are doing the booking, who know the rules, but they're not well documented. You can't digitalise it and it's really hard to scale.
Milica: But also it's a priority. So how do you go about it? What is more important? And we are leaving those decisions to somebody who potentially should not be making those decisions. And that's not fair on our staff.
Tom: It's inconsistent again.
Milica: Correct. They need to come to work. This is the list, crack on until we digitalise that. You know, there shouldn't be, oh, check this list. Now check that list. Now validate this list. It just creates this machine.
Tom: It's so interesting because we think about it a lot from the patient side and that, you know, feeling like you're in a bit of a black hole. I know from our stats that 40% of the phone calls to most booking centres are people chasing their referral, which just shows. And we know that when you send those messages to let people know what's going on, it reduces the stress, it gets rid of the phone calls. But I hadn't thought about the staff side. I hadn't thought about someone coming into work and being given this really stressful situation of those five or six lists.
Milica: Correct. But also our priority changes all the time. So one day the priority is let's clean all 65 week breaches because that's the right thing to do. And we shouldn't have patients waiting 65 weeks and off. But then what about the rest of them? And then, okay, let's go for 52. No, now it's a cancer. And you're consistently moving your people and the administration. Now we're a big team. We're talking about, you know, 50 people.
Tom: Yeah. Yeah. You can't just move the goalposts all the time.
Milica: And people need to know what we need to do with as little changes on the day as possible. And I'm the first one who drives my team mad because one day this will be priority. Then I will be told something else is priority. And it just raises this anxiety, the stress and the work just becomes really unpleasant. I find operational management probably one of the most exciting professions out there. And I want people to enjoy it because, you know, we serve a population.
Tom: You serve people and it's a fun job, right? Because you get stuff done and you make a real difference to people on a daily basis.
Milica: You can feel that. Patients are happy. It's, you know, it puts a smile on a lot of faces. At least it puts a smile. I still call our patients. So I would still do them if the DNAs went up or something. I would say, come on, give me 10. Let me have a chat to see. Yeah. Why? What doesn't work? I think that's very important to keep that.
Tom: Yeah. Stay close to the problem, right?
Milica: There are patients, you know, we serve 2.4 million. Sorry. Yeah. So for me, it was I wanted to make administration a little bit more happier. I was a little bit tired of listening how administration is not successful. It doesn't work. It's always a booking problem. It's always, oh, the booking team can do something better. Yeah, absolutely. We can. No question about it. But we also need tools. You know, you put both of my hands behind my back and ask me to do better. It's a little bit difficult.
Tom: I couldn't agree more. I feel like admin teams get beaten down on a lot by people and I think it's really unfair. Yeah. You know, I've spent my whole career working in hospital admin outpatients and the people that I meet care about that job so much. Yeah. They care so much about every single patient. They really, really do. And they're so important because without them, we would not have a hospital. We would not. Okay. Yeah. It's actually that simple. Yeah. And I think that they're getting quite a lot of bad press. But what I said to the team that I manage and I said to everybody that I know, without you, our patients will not have any chance. No, they would not. They would have no chance to get around all of these processes, paperwork, governance that we have put in place. What a reason to come to work. You know, there's real meaning in that. I think so. This is my sort of vision and dream, I guess, for Dr. Doctor is I want to be every administrator's assistant. I want to be there helping them do their job, getting rid of the drudgery so that they can be helping people every day.
Milica: Exactly. That was my vision. My vision is let's automatize everything that we can in order for the workforce that we have is actually focused on the patients that need support, not all of the patients need support. It's maybe 20 percent. Yeah. And then you can spend time with them. Correct. And do it properly and support them. You know, if you look into the DNAs and we looked in our organization, we are very focused on trying, you know, we serve one of the most deprived population areas in London. So we're very focused on trying to reduce the DNAs in that specific cohort. And yeah, I want administration to do that. Yeah. And we've certainly seen from sites that those less well served populations, if you can spend the time with them, work with them on, you know, find people that speak their native language from their community. The engagement levels are really high. And that's you know, this is what the NHS is amazing at, is when you're in the system, it cares so well for people. And the job of, I think, you know, the administration teams is to extend that into the community as much as they possibly can.
Tom: Yes, absolutely. It's spending time a little bit with our communities. It's caring. Yeah, exactly. It's caring. It's not just transaction. You know, this is the regulation, it's actually spending time, have a chat a little bit. Listen, we're noticing your DNA in quite a lot of our appointments. Is there anything that we can do to support rescheduling transport, support organising the translation? There is quite a lot of work that we can do, but we can't do that if we need to reschedule 700,000 appointments a year. And I'm really excited by how technology can help with this as well, because things like the translation and language requirements as we bring in more AI sort of automatically translating, I think is going to be really exciting.