HTN was joined by a panel of healthcare leaders for a discussion on digital in supporting pathway redesign, covering challenges, approaches, and examples of success. We considered what good looks like in this space, along with national guidance, what’s working and what isn’t, with an aim to share practical insight and learnings.
Making up our panel, we welcomed Lee Rickles, CIO at Humber Teaching NHS Foundation Trust; David East, Rego director at NEC Software Solutions; and Lydia Salice, divisional director for digital at Moorfields Eye Hospital NHS Foundation Trust.
“My role at Moorfields Eye Hospital is lead for digital clinical services,” Lydia told us, “which covers all of the digital services our patients access directly, rather than the more back-end digital services, and working closely with transformation teams, IT teams, and so on, to do what we can to transform eye care.”
As well as being Humber Teaching’s CIO, Lee shared that he also has a role as a director of the Interweave digital programme, an NHS social care-owned information interoperability platform supporting things like the Yorkshire and Humber Care Record. “When talking about supporting pathway redesign, we have to look beyond individual organisational boundaries,” he considered. “At Humber Teaching, we’re a complex multi-specialist trust, and for us, the clinical pathway is rarely confined to a single building or service – it’s around the patient and probably multiple systems. From a regional point of view, it’s looking at how we connect that personalised care or personalised experience, and use those platforms to redesign pathways so they work better for staff and patients alike.”
David introduced NEC Rego, a referral and pathway optimisation platform, and his role engaging with ICB customers. “We’ve had over ten years of experience in pathway redesign, largely within our dental community, where we have just over 50 percent UK coverage. We’ve now moved into wider medical conditions and specialties, and my role is to ensure we are continually evolving the platform to meet the needs of our user communities and stakeholders, focused entirely on patient outcomes, and how we improve productivity and efficiency across the system.”
Journeys to date
Lydia took us through some of the latest developments in pathway redesign at Moorfields, highlighting work on referral management and triage. “In ophthalmology, it’s so important that we accept referrals directly from high street optometrists, rather than having to ask patients to go to their GP, and placing that extra burden on the GP,” she considered. “We wanted a way to be able to manage those well from the outset, so we started with one patient, then worked up to 100 patients, and then a pilot of a region. In London, we’ve now been commissioned for this service for all of North Central London and all of South West London, and we have a pilot in North East London, as well.”
An independent review showed a saving of £800,000 per year in North Central London alone, Lydia shared, and 95 percent of referrals are currently triaged within 24 hours. Eight percent of referrals are avoidable and are either sent back with advice and guidance to be managed in the community, or sent to community eyecare pathways, she continued, “and of our advice and guidance requests, we’ve managed to do 46 percent without an appointment”. A lot of referrals were incomplete, affecting the ability to make decisions on getting patients to the right place, and that led to an education programme with referrers, talking about what kind of information should be included and what testing should be completed, and that brought complete referrals up from 49 percent to 81 percent.
Other impacts of this work have included a 48 percent reduction in discharge at first appointment, according to Lydia, meaning those coming in for appointments are those who actually need care and treatment. “We’ve also had lots of urgent referrals coming in, and 41 percent of those are de-escalated to routine,” she noted. “That’s really important, because we have so much pressure in our urgent clinics, and we can make sure those urgent slots go to those who really need them. On the other hand, three percent of our routine referrals are actually upgraded to urgent, and one percent are escalated to emergency care – often something that could be a stroke, for example, is presented as an eye condition, and we need to get patients to A&E straight away.”
David shared some of his team’s work, reflecting back on their very first digital referral optimisation programme ten years ago. “We were taking an ICB from a paper-based traditional way of working; there were high rejection rates, lost emails or letters, missed information,” he said. “Effectively what we did with the region, particularly around their dental pathways, was build a suite of algorithms incorporating NICE guidelines and GIRFT, as well as feedback from engagement with user communities and clinical teams to understand requirements.”
In the last year, Rego has generated around 1.3 million referrals across the board, David told us, noting: “We’ve got a 93 to 95 percent acceptance rate, so it’s right place, right time, at scale. That means we can help to protect the front door of secondary care. Across the suite of specialties we run, I think on average we’re looking at only 75 percent actually being retained in primary care, and a much smaller number will make its way into secondary care.” There’s a lot of genuine system collaboration in the programmes his team is involved in, he suggested, “and patient choice is King; whether that’s geography based, distance from home, or waiting times – for us it’s all about the impact we’re making, whether we’re meeting the needs of ICBs, and whether we are satisfying the patients being seen.”
At Humber Teaching, the shared care record is being used to message, alert, and share information within EPRs and its own portal. “Digital maturity can be variable, particularly when it comes to care homes, nursing homes, and people that predominantly look after those on end-of-life pathways,” Lee observed. “We’ve done end-of-life planning with ambulance services, community health, Macmillan, local authorities, care homes, and so on; without that buy-in across the whole pathway, it’s not going to work.”
2,000 conveyances have been prevented in the last two years, Lee shared, “and we achieved that by getting everybody together, looking at user-centred design, considering who is involved and where, and how we could do things with the least possible friction”. For ambulance services, that meant ensuring they had access to the information they needed to be aware of as soon as a call came through, he went on. “With the care homes, it wasn’t necessarily technology-oriented, it was more how you could create visual indicators and processes on top of technology, and providing simple information appropriate to what that person should see.”
A similar thing has also been done with neuro in children’s and adult services, where waiting lists were a challenge, according to Lee, putting a number of technologies in to help move that along. The Do-IT profiler is being used as a tool to assess people’s needs when they initially come into the service, he indicated, meaning the potential to redirect or offer ongoing support outside of the hospital setting. “That’s starting to have an impact, and we’re seeing waiting lists drop quite significantly, but we’re still quite early in the evaluation stage.”
What works, what doesn’t, and overcoming challenges
Lydia shared how her team has been running webinars as education sessions, open to anyone in the region, to go through common referrals and common challenges, getting feedback from referrers on what would work for them, and what information is required to help support decision making at different points in the pathway. Advice and guidance is at 46 percent at the moment, she told us, “and this is quite niche to ophthalmology, but where we get a scan image sent through, we can go straight from a referral to list for surgery, which can be essential for certain conditions.”
This kind of thing is happening across the NHS, Lydia noted, but is being repeated again in primary care, secondary care, etcetera, because information can’t be shared. “If we could all have better ways of sharing information, like in the NHS App or single patient record, that would make a difference. In a few weeks we’re going live with photos from our patient engagement portal, which links to the NHS App, and we’ll be able to request photos directly from patients for review.”
The main challenges for Rego have historically been around early engagement, said David, due to having quite a complex stakeholder group. “As well as making sure the services that we design are fit for purpose and tailored correctly, we also need to ensure we have the right adoption and onboarding processes, technical integrations, and understanding – it’s certainly not a one-size-fits-all.” The earlier in the process that communications are ironed out and engagement is carried out correctly, the better, he noted, “and we have to have that continuous engagement, pathway review groups, so we can see that the system is performing well against changing guidance, services, and so on”.
Digital is “one of the main culprits” in making the neuro pathway challenging for his team, Lee shared, since greater awareness of services driven by social media has not only helped with the education piece, but also led to a “significant number of inappropriate referrals” to mental health services which might be better handled in other places. “The other thing is that referrals coming in were often quite poor on the level of data they provided, which can sometimes be because the NHS number isn’t always used as the key identifier.”
Solving that meant going back and looking at the experiences of young people and their families, finding out what would give them confidence, and how they saw a service operating, Lee said. “Fundamentally, the care we provide is not going to massively change; it’s understanding how we can help – whether to send emails asking people if they still want to be on the waiting list, rather than actually what is needed in the early days, which is some additional support or some really honest understanding. That’s why we went the route of, if there’s something when a referral comes through, we ask whether they can make use of this tool to assess their current requirements, so we can prioritise and see whether this is somebody we need to help quickly.”
The key part is understanding the worries and concerns of customers, and balancing those against the risks of the organisation, its professionals, resources, next steps, and the processes you can put in to improve that service provision, Lee shared. “That’s where the youth forum has developed from, because it’s effectively our non-exec directors on that service to say whether it is working as intended, and we’ve trained up some of that forum to provide materials and support information that then gets clinically signed off.”
From a Rego perspective, being used at the first point of contact with a patient and trying to make their initial consultation and triage as seamless as possible, is important, David shared. “The system pulls through patient medical history, any pre-existing conditions, any medications, and so on, effectively building that picture automatically for the referrer to make a decision,” he detailed. “What the algorithms will then do is incorporate all of that localised guidance into a simple and straightforward dynamic questionnaire, guiding them through based on all of the criteria and patient-modifying factors, to the suggested outcome. Clinical autonomy is key, so GPs and referrers can override the decision, but it’s presenting all of that information so they don’t have to search for it.”
Based on the algorithm and decision support tool, referrers are next presented with a list of services that suit patient’s needs, David continued, automating the process and helping to reduce rejection rates from 20 or 30 percent down to two percent in some instances. “We’re also making sure that mandatory fields, documentation, images, test results, and whatever is needed to meet that secondary care or acceptance criteria, is in place – we’re now at a point with certain services where they’re being auto-triaged, giving back 20 to 30 hours of consultant time per week, which can be redistributed to tackle other more problematic waiting lists.”
Lydia talked about linking systems to EPR to manage referrals and triage, noting work on a regional level. “I think where referral management is most effective is when all referrals go through a single point of access for a specific sector, and then are distributed with patient choice on the same criteria, especially where you have high-volume, low-complexity specialties,” she said. “We are using a system that integrates with ERS, and through that, integrates with all of the different EPRs. When we’re looking at sector-wide triage, and all of those different areas are integrating with ERS, that allows for information to flow, rather than us having to work with all of the different EPR vendors.”
Having two-way communication with referrers is integral to allowing referral quality to improve over time, particularly when feedback is given quickly, Lydia reflected. “If they’re referring a patient on a Tuesday, and by the Wednesday getting feedback on it and remembering that patient, they’re going to change their referral practice in future; whereas if that feedback is delayed for six months or more, it’s probably not going to have the same impact,” she added.
Looking ahead
Looking to the future of pathway redesign and what ambitions should be in this space, David spoke about the need for vendors to work together end-to-end to overcome siloes in technical integration and interoperability, stating: “Willingness from some of the dominant providers to work in partnerships and as an ecosystem would open up so much more opportunity. We could have have all of the open APIs and all of the technical capability in the world, but without a willing vendor at the other end, I think that just creates barriers. That would be where I would start.”
Lee also emphasised the need for greater collaboration, for “sweating that NHS asset”, and for moving away from working like separate franchises, with pilots or programmes being replicated across the country and not always being shared. “The number of pilots of AVT is a great example,” he went on. “Everybody’s going out to do the same stuff, whereas we should have been a lot more forensic and worked together to try and get that evidence. I’d like to see us be more connected, which should give us more resilience and allow us to be more responsive, reducing a lot of the resources we all use individually to create a far bigger capability for things like pathway redesign.”
“I agree with everything Lee and David said, and if we’re all working collaboratively, and our vendors are working collaboratively, if we have what we need for the NHS App, the single patient record, shared care records; all of those things will support us in what we’re trying to do,” Lydia agreed. “We know that in our pathways we’re doing many basic admin tasks that could be safely changed for AI, and that would help things move a lot faster. I think there’s a real opportunity there with clinical admin, and I’d love to see a bigger focus on that – it would be the safest place to implement AI, rather than starting somewhere with high risk to patients.”
We’d like to thank our panel for taking the time to share these insights with us.
