We were joined for a discussion on the future of digital in supporting elective care and waitlist management by a panel including Sally Mole, senior digital programme manager – digital portfolio delivery team at The Dudley Group; Susannah Cleary, assistant director of operations and delivery for elective care at NHS England; and Barry Mulholland, CEO and founder of MBI Health.
The session focused on the role of digital in supporting elective care and waitlist management, taking point of recent plans, and looking ahead at the potential for data and digital to further enhance elective care and waitlist management in the future.
Susannah introduced herself and her role overseeing operations and delivery within the Elective Care programme at NHSE. “I have a wonderful team that works on patient choice and implementing tactics to improve awareness,” she said, “and also who look at what we can do to improve people’s experience of care whilst they are on a waiting list.”
“I’m the CEO of MBI Health,” said Barry. “We’re a healthcare technology and advisory business with a simple mission to restore confidence in elective care in the NHS. We work with over 90 NHS organisations to improve access times, reduce wait times, and make sure that the data driving decisions is as accurate and trustworthy as possible.”
Sally shared some details about her role as digital programme manager and head of the PMO at The Dudley Group. “I manage a portfolio of about 70 items, 36 of which are active, and a multitude of which are in relation to elective care,” she told us.
Elective care – the current position
Barry talked about how elective care and waitlist management is “becoming more mainstream”, noting his 25 years’ worth of experience in this space. “For me, the key dial that we have to think about is public satisfaction,” he continued, “and we have the lowest levels of patient satisfaction with the NHS in history, so that’s a huge problem.” Losing the public means starting to lose political consensus, particularly around funding, he considered, “and ultimately we run the risk of losing the NHS”.
Elective care represents 75 percent of all contact with the health service, and how we fix that involves looking at what Barry calls the “front log” problem, representing the front of the pathway. “If you think about the patient, the biggest area of angst is access to primary care, and then if I get referred to secondary care, getting seen and diagnosed quickly.” Although the shift to neighbourhood health has been talked about from a policy perspective, there is a need to consider whether patients will be seen and diagnosed faster, he highlighted, “as that’s the only thing that will improve satisfaction in elective care, and from a digital perspective, that opens up questions about what digital tools will help us do that”.
Susannah agreed with the importance of treating and diagnosing people quicker, adding: “It’s necessary, but not sufficient to improve people’s experience of care – we know from the British Social Attitudes survey that 65 percent of respondents said they were dissatisfied with the length of wait to get to hospital, and from data I’ve been collecting, shorter waits are very much linked to experience.”
A major factor in that, according to Susannah, is a dissatisfaction with communication around waiting times. “We’re used to buying things online and immediately getting updates on expected delivery,” she went on, “and that gulf between our experience with retail and the NHS experience is widening.” Susannah also spoke of hopes that her team’s work will, over time, become “setting what we would call some minimum standards that are clear about the offer people can expect while they are waiting”.
Improving the NHS App as the front door to the NHS and the underlying architecture behind it for booking and managing appointments as well as waitlist validation would be a step in the right direction, Susannah said. “This is all happening, but it’s not happening consistently, and we need to get more trusts and service areas on board with it, make it a consistent offer to users, and ensure digital inclusion so people know how to use these functions.” She also noted the importance of communicating proxy access and ensuring reasonable adjustment flags follow patients throughout their journey in improving public satisfaction.
The key thing Dudley has been looking at is understanding the workflows and processes of clinical teams, Sally shared. “It’s important to fully understand our services and the challenges they have, because that’s where the opportunities are. We’re trying to hone in on AI in areas like clinical decision support and getting quicker diagnostics, things around risk stratification and getting the patients in that are urgently required to come in.” AI is capable of stratifying patients “much quicker” than a human, she considered.
Elsewhere, her team has been looking at patient empowerment, Sally said. “We’ve got lots of disparate systems, and from the patient perspective, it’s how we make that seem seamless, so we’re looking at existing solutions and how we can optimise them.” There will be a big push this year around automating, she went on, “to help the administrative team be able to do the things they should be doing, and not the constant chasing of waiting lists and making sure data validation is correct”. Improving clinical efficiency and processes, as well as having better operational oversight with the help of AI and automation, will create a better experience for patients in the long run, she added.
The role of data and digital in improving elective care
“I think the most urgent issue we face is the quality of data,” said Barry. “I don’t think enough people are aware of how poor it actually is, and I think digital has a huge opportunity to do something about that. We run a national data quality monitoring tool, for example, and our average removal rate of RTT pathways is running at about 30 to 40 percent when we engage with organisations. So if you’re in a situation where 40 percent of your waiting list is not correct, that’s a big problem.” The amount of money and time spent trying to fix that manually is “crazy”, he continued, “and we can do a lot through automation – we’ve done a lot with the NHS App, but I think there’s a lot further to go”.
The most promising use of AI in this space is around speeding up diagnosis, Barry went on, particularly in the diagnostic stage. “I see a huge change coming there,” he said, “using AI rather than having to get humans involved. In terms of specific things to reduce the numbers, we’ve already got examples on the market, like a product that uses AI to read clinical documentation and compare that to what has been recorded in the EPR, and we find that we can basically have everybody validated at all times.”
Susannah pointed to the challenges around “dummy slots”, or appointments that automatically go out to patients or carers for timeslots like 2am, when they actually shouldn’t be for them to see. “By some accounts, there are about 200,000 of these dummy slots pushing through every month, and it’s confusing,” she said. That comes back to surfacing the right information for users in a way they understand, “and while there’s a lot of this whizzy technology that has a place and should be rolled out; often there isn’t the thought leadership or drive to improve NHS culture around communications”.
In the last 12 months, Dudley has been working on a number of projects, including one looking at inefficiencies in theatre management processes. “It wasn’t so much that the tools we had in place weren’t the right tools; it was more about data flow across those tools,” she explained. “We made a standardised, streamlined process for all proformas, so when a patient is ready for surgery, that is fed into the EPR, which in turn feeds into multiple different systems. So, we have the pre-op database, we have the theatres database where we do emergency and elective care, and we’ve got the EPR where the entire trust looks at the data for the patient.”
Describing getting that compliance from clinicians around the standardised process as “quite a challenge”, Sally told us how having the data electronically meant it was easier to monitor. “We’re worlds ahead when we start these projects now, because we’ve done all of the hard work and built that background knowledge and engagement with our team,” she noted, “so we have more room to concentrate on delivery and improvements”. She also shared details around a project which went live in December 2024, providing a centralised calendar for theatre schedules, to provide a clear view of any gaps or cancellations, and to have all of that information in one place. “It really has improved workflow efficiencies, and the approach that we took was to look for a minimum viable product that could help fix the challenges right now, and how could we then optimise moving forward, which enabled us to deploy that at pace.”
Giving an example of a project MBI Health has been working on in this space, Barry told us about four months ago his team was approached by a trust to come in and risk stratify 130,000 patients that had been “lost to follow-up”. To trawl through those patients was going to cost an estimated £600,000 in manual effort, he said, and would take about eight months. “They presented us with the challenge of finding a faster way of doing this, identifying whether patients still need to be seen, and how much clinical risk we are carrying,” he went on. “We ran our AI solution against the 130,000 patients, and it took us three weeks. We did identify groups of patients that really did need to be seen urgently, but we also identified a huge amount of that 130,000 that had just not been updated. That’s an example of how you can do this at scale quite quickly.”
Overcoming barriers and choosing the right digital tools
Sally’s team has a multitude of requests coming in for solutions, she shared, “and sometimes people know what they want, and sometimes they just have a challenge they want us to solve”. Part of that assessment is considering whether the trust has existing systems that can facilitate the request, or whether they can be onboarded onto something in the portfolio about to go live. “If it’s not something we’ve considered, we’ll do some in-depth scoping and business analysis,” she said. If it’s something being procured from a third party, making sure the solution fits the service’s needs is key, “but it’s also important to build those relationships with stakeholders and ensure expectations are clear – understanding what problem you’re trying to fix, and looking at prototypes or demonstrations to clarify it’s going to work in clinical practice”.
Susannah highlighted challenges around integration and adoption. “My sense is that around leadership and culture, current uncertainty can filter down to all of the levels in terms of how decisions are made, how policies are set, how they’re deployed, and how they’re implemented from the top down. Not having that national strategy or that roadmap has made it quite tricky to take work forward.” The publication of the 10-Year Plan and the Dash review has been helpful, she went on, in allowing her team to understand the direction of travel and to get things moving. “And like Sally alluded to, that commercial process and getting projects signed off and approved, is just as tricky at the top as it is perhaps on the front line.”
“We’ve got to break the monopoly of the EPR system,” said Barry. “I think a lot of the technology companies can do interoperability, but we’re blocked by the big EPRs, who don’t want us to do it, so we’re working with one hand tied behind our back.” He also mentioned the procurement process, how confusing that is, and how “every organisation has their own process”. Information governance is another source of frustration, he went on, “because we’ve got solutions deployed in 28 organisations, we’ve done the DCB0129 and the various mandatory things, but then we go to a new organisation and they tell us their IG process is completely different, which takes months”.
Tracking benefits and outcomes
Sally shared that her team had a business case investment for additional staff in order to bring their portfolio in from six years to three and how, as part of that, they needed to demonstrate the benefits and impact of the recruitment. “A lot of the work that we’ve done to date has been very standardised,” she said, “and we’ve got a benefit realisation SOP that goes through a number of steps to track benefits throughout projects, which feeds into our project portfolio checklist. So for every project we’ll do a high-level baseline of benefits and what we’re expecting out of the solution, before moving in to key metrics as we move toward delivery.” That includes projected savings and cost reductions, which are agreed prior to go-live.
“Predominantly we work against the green book standards, which is an evidence-based framework looking at cost savings, income generation, risk avoidance, and so on,” Sally continued. “We also do time-in-motion studies, looking at the before and after, so we can actually see how efficient teams are being and whether we need to move people around.” The trust has forums to get feedback from service users, and a benefits realisation check is performed at the six-month mark “because when you first implement, activity usually drops by about 30 percent”. Areas such as patient flow and clinical risk are also measured, along with patient feedback. “Last year, we had a remote health coaching project for elective care, giving health coaching to patients ready for surgery to help them recover faster, and we’ve seen reduced length of stay and better outcomes from that.”
Barry agreed with Sally’s point about the importance of agreeing and baselining metrics upfront. “I could say that one of our solutions does a lot around reducing clinical risk, but when I go into an individual organisation, what does that actually mean, and how am I going to track it? We do a lot of work on reducing waiting lists and ROI relative to what the trust would have paid to do it themselves or how long it would have taken, versus getting us to do it.” There are a lot of claims being made about different projects on the market that are difficult to link to reality, he went on, “so as a CEO I”m very conscious about working closely with trusts, asking what metrics they want to move, and agreeing upfront what the baseline is and how we’re going to track it over time”.
“That’s an important point,” Sally noted, “as it’s very easy to make these claims, and it’s easy to do a pilot in one area and then apply that at a larger scale across the whole organisation, and that’s not really reflective of reality.” Dudley tends to pilot solutions in multiple different areas, which often brings about “very different” outcomes for the same solution. “We’ve just done an AI ambient voice pilot in our SDEC area and had good feedback from clinicians, but we’ve then piloted the same thing in rheumatology for the letter backlog, and we’re seeing that it’s working to reduce that backlog, but it’s two different use cases and the output is very different.”
Looking ahead
Susannah talked about the potential for AI to make an impact in this space in the future. “The short answer is that yes, I see it having a big impact. In my sense, the public are up for it, but they’re a bit nervous sometimes, and we have to be mindful of that and the risks involved.”
Touching subjects already mentioned around data quality and faster diagnosis, Barry said: “I also want to float a concern I have at the moment, which is that we’re not doing enough from an elective care point of view about clinical variation. That’s where the predictive analytics starts to come into play, where we start looking at differences between population groups and outcomes. We’ve got the datasets and the processing power to really drill into why two separate patients were on that waiting list for the exact same procedure, at the exact same time, but went through a very different clinical pathway and achieved different outcomes.” Having those conversations about variation is important, he considered, “and that’s from the front end too, because I think we also have big variations in referral patterns and demand”.
Focusing on early diagnostics is key, Sally told us, talking about an AI solution Dudley has deployed that identifies tumours and when patients have had strokes “much quicker”, as well as a risk stratification tool for patients preparing for surgery that tells clinicians which patients need to be seen face-to-face. “At the moment in the NHS we’re rightly very risk averse,” she said, “and there’s a lot we have to go through in terms of the clinical safety and the guidance, but there’s still lots of learning to do as we go through, which means there is lots of caution. Whilst it’s good to scale things up at speed, we have to make sure things are done correctly and they’re working as expected.”
Sally also mentioned triaging as an area where AI has the potential to help with ensuring patients are on the right pathway and prioritising those who are most urgent. “I also agree with Barry about the data because we piloted an automation tool to look at the population and where patients were coming in, in terms of their diagnosis, and what we were finding was that when the patient arrives for the first time, they’re not diagnosed, so you have to ratify them with other symptoms, which could look like a thousand different outcomes. So I think we need to do better with our data to help inform our care pathways and workflows.”
We’d like to thank our panellists for joining us, and for sharing their insights in this space.
