For a recent HTN Now panel discussion, we were joined by David Hammond, deputy chief strategy officer at Wirral Community Health and Care NHS Foundation Trust, and Julia Lake, associate director for digital (data) at Leeds Teaching Hospitals NHS Trust, to talk about how their organisations are currently using data, how data can support the move from reactive to proactive care, and how data can support operational service delivery.
We started out by asking our panellists to offer a brief introduction to themselves and their role. Julia went first, sharing that she has been the associate director for digital (data) at Leeds Teaching Hospitals for just over two years, after working her way through the ranks within the NHS: “In my role, I look after the clinical coding team, the data quality team, and the information and insights team. I have a technical data warehouse team, and a business intelligence team, and we’re working to bring all that together. I’ve got a massive project underway at the moment, which is to move us to an Azure data platform.”
Next up with the introductions was David, who told us how in his role as deputy chief strategy officer at Wirral Community Health and Care, he has a strategic view, and supports the board in the development of strategy, as well as managing the trust’s programme management office.
“I’m very interested in how we turn that strategy into reality, and one of the strategic projects I’m leading at the moment is around population health management, and particularly how we work with primary care networks, in terms of better coordinated, more person-centred care for those who will most benefit.”
Using data to drive transformation and insights
Moving on to discuss some examples of how each organisation is using data to drive transformation and deliver insights, Julia told us how at Leeds Teaching Hospitals, data is used in nearly everything the trust does, including direct patient care and research.
“We’ve got a big agenda around health inequalities, and we use data for a lot of quality improvement initiatives within the trust, operational planning, and so on. My team sit on different boards, including the unplanned care board, and the diagnostic board, helping them to understand the uses of the data and how we can help them transform services.”
A recent project, Julia continued, was a proof of concept on the trust’s new data platform, and involved streaming near real-time data from three different systems to provide a live bed state, looking at things like how many patients were waiting, how many patients needed a bed, and what kind of specialties were required.
“That offers a helicopter view of what’s happening in the bed base at the moment, and you can filter by specialty, different areas like intensive care, and different bed types. Then it also looks at what’s happening at the other end of the spectrum – how many patients are ready for discharge, any delayed discharges, and you can drill down to that patient-level detail.”
As far as outcomes, Julia said that “it’s really helped us to understand the kind of flow and where we need to concentrate our efforts at the discharge end”.
“Another example is that one of my team has just built something for our local ICB around paediatric asthma. It’s a dashboard with live data of re-attenders, so children who keep turning up to the hospital. It’s looking at how we could help the child’s parents, and looking into why they are turning up at the hospital, so we’re using it for that proactive care and to try and change the way that those people interact with our services.”
Switching to look at some of the work with data being done at Wirral, David offered us an insight into some of the recent projects he has been involved in, and how his organisation is using that data toward delivering more proactive care. Like Leeds, he told us, “we use data in the day-to-day operational management of the organisation and governance processes”, and “it’s built into the organisational governance and performance management structure; we’ve got our trust information gateway which pools near real-time information from a whole range of different sources; and it’s understanding performance, understanding progress against KPIs, looking at waiting lists, and that kind of thing”.
Wirral is also working with some local primary care networks on a population health management programme which David says the team hopes to “roll out universally”. He shares that work is “all about understanding which people will benefit the most from proactive contact and holistic care planning, and making sure that a care plan is developed with that person and their family, to avoid the loop of multiple contacts”.
He also shared some of the work going on at an ICB level with the development of “a whole suite of tools for the data interaction programme, “looking at different population groups and levels of population need, and looking for opportunities for intervention”.
“Some of that is around waiting list management and combining information to understand at place or organisation level who sits on which waiting lists, and whether there are people we need to pay particular attention to. Alongside that is a whole set of different dashboards, looking at different population groups, frailty being one of those, and fuel poverty, and so on. There’s a lot of that kind of developmental activity going on, which is all around directing service delivery.”
Tackling data literacy
On the topic of data literacy, Julia and David talked a little bit about ways of making sure that the importance and impact of data is something that is understood across their organisations.
David told us that making data “part of day-to-day work” is essential, “whether that be finance data, performance data, quality data, or risk management”.
“Unless you build it in and make it normal, you don’t necessarily see the benefit,” he went on, “and that’s absolutely what we’ve done”. Making sure the data is good quality is another important element of this, he shared, and “our business intelligence teams do an awful lot of work around data validation.”
From a population health management perspective, David told us that Wirral has been “focused on people with moderate and severe frailty, because that’s a group we believe we could achieve the greatest benefit for, both for individuals in terms of joining up care; but also because we know those are people who are massively overrepresented within unplanned care; so if we get it right at the level of the individual, we should be getting it right for the wider system”.
Julia talked us through her findings from Leeds, saying that when she first started in her role, she wanted to “change the way data was used in the organisation and put some real structure around data management”.
“We’ve set up a data literacy forum, and my team is in that, and they’ve got guest speakers who come and talk about their projects and the data needed for that, but that needs to broaden out, because at the moment it’s quite a small group. We want to use the data platform as a trust-wide asset, so we want people to be able to use it, but we understand that there are limitations to peoples understanding of the data.”
Having strong leadership and digital advocates has helped, Julia continues, and the team have started to get a more joined-up view as a result. Part of it has been ensuring that when reports are sent out, “we’ve got information on the front cover about where the data comes from, because we’ve started using Power BI a little bit more; what it can be used for, the limitations, or the assumptions we’ve had to make in doing that, and that helps with people understanding it”.
The ideal, Julia told us, is that “we get to a point where people can self-serve”, but to do that, “we want to give them curated datasets so they can’t go far wrong with it, and it’s using those digital advocates to make sure that people are understanding those datasets when we’re putting them out”.
Using data to support the move from reactive to proactive care
Taking a closer look at how Wirral is using data in designing care models and moving from reactive to proactive care, David shared some insight into his team’s approach, saying that the first thing to consider is “who for?”, because “there are some groups where this is time well spent, but we shouldn’t, and couldn’t, do it for everyone”.
“There’s the intensivist approach, looking at proactive identification, and then holistic and person-centred assessment and care planning; and then there’s the extensivist view, which would look at more public health driven approaches, whereby we’d say people with certain characteristics might benefit from certain messages or prompts and things that we might do that are maybe less resource-intensive.”
A project on the intensivist end of the spectrum, David said, would be with people who have been clinically identified as having moderate to severe frailty, and who are “typically the highest users of health and care services”.
“In conversations I’ve had in the past, there’s an underlying anxiety that we’re going to identify more work, but the data shows that these are people who are already in daily contact with our services, having GP contacts, having home visits, having unplanned care episodes. The higher levels of frailty that people experience, the more likely their care needs are to be unplanned, and that’s not good for them or the system.”
One thing that data has already shown, says David, is that “frailty will affect people younger in more deprived areas”, and “the highest proportion of unplanned versus planned care happens in the most deprived quintiles of the population; so really this is also a health inequalities model that we’re talking about”.
The work doesn’t stop at identifying these groups of people, however, as it’s also about “pulling together existing information about them”, which presents a data and information sharing challenge with having access to the right systems and “making sure that people who are working together in that space are able to draw on that information, so they’re not repeating things that have already been done”.
“It’s important that this happens within the primary care space, on a primary care network footprint, because that’s where continuity of care needs to live, and because primary care is the one part of the system that never discharges anybody. We need to maintain continuity and communication, and within a 30,000 population primary care network there’s probably 1,000 people who are going to need that ongoing continuity of care, not only in an assessment and care plan, but also involving other services, and identifying when things change which may warrant a review.”
The data challenge around proactive care in this intensivist model, David concludes, is identifying those people and then drawing the information together in one place, as well as “looking at how we can share information effectively and how we maintain some level of contact with that individual and their family, so we understand what’s happening and whether we can do something different to respond to that”.
Using data to respond to challenges
Julia and David moved on to talk about the ways that their organisations are using data to change models of care, and to solve challenges.
Julia talked about waiting lists, saying that the current model is not sustainable, and sharing some of the work in Leeds to make the change to proactive care, including scanning at the point of care, and some predictive elements around risk stratification, predictive analysis, waiting list management and so on.
“We’re doing more with genetics around the genome project and seeing if we can identify patients with family history of long term conditions, and whether we can start to look at early screening for those patients to start them on that journey earlier.”
In terms of the technical side of things, Julia shared that Leeds is currently using the Symphony system in the ED department, a patient administration system and an electronic health record, which the team pulls data from.
“Our electronic health record contains all the clinical data around patients, and that’s where we look at their pathways, nursing observations are on there, and we’re lucky that we’ve got an in-house built electronic health record which gives us an advantage in the fact that we can build and develop very quickly, and it’s very clinically led, so we can bring that data in.”
Looking to the future, Julia says that Leeds has “only just got to the tip of the iceberg” in terms of what it has in place at the moment, adding that now the team is focused on bringing data assets into one place and “having that big repository where we can then start to harmonise the data and bring it into that common data model, so that we can query across all of the data”.
To overcome challenges such as augmenting data across lots of different systems, Julia says that the team wants to “move away from that”. She also shared that her head of technical information has recently started to talk to suppliers “to see whether we can get more real-time data, that we don’t need to re-import every morning”.
David also shared his perspective from Wirral, talking about his work with PCNs and how important access to different systems is to promote data sharing across teams.
“With frailty, referrals are received from GPs or community teams, and if its the GP, it can be prompted from within EMIS, so if they are entering information about somebody which suggests that is somebody with moderate to severe frailty, then there’s a pop-up that will appear, which asks whether they want to refer that person to the ageing well and frailty team.”
The team at Wirral has also made use of the Cheshire and Merseyside Graphnet tool, David says, which “allows a huge amount of different variables to be searched, such as unplanned care utilisation”.
“Once that has been done, essentially all of the activity happens within EMIS, because it’s the principle of continuity of care, within primary care, and so the customised assessment templates, the personalised care plan, it’s all happening there. The team has a shared digital folder containing all the information, but we’re looking at automating that within EMIS so that we can pull searches based on the activity of the team, and so that it’s all contained within one place.”
Overcoming data challenges
We asked our panellists to reflect on some of the challenges in using data for proactive care, and how they responded to these.
Julia placed data quality at the top of her list, as well as having real-time data around things like discharges and transfers, surfacing that data, and ensuring consistency in data across systems, adding, “we do have quite a lot of work to do around that and completeness of data”.
Working with suppliers can also be a challenge, Julia went on, since there can be “a lack of documentation, issues with getting the data required, and standardisation”.
“I think the absence of an NHS common data model isn’t great, and the NHS data dictionary is quite outdated, and those are things that would really help build that continuity nationally around what everyone is doing with data. Our data quality team have just written a new policy, and we’re trying to do lots of education with people that links into the data literacy programme. It’s about giving people the tools to be able to self-solve.”
The team at Leeds is also working on a suite of reports, visualisations and dashboards to send out to clinical service units, Julia told us, which provide an update on the quality of their data.
“That’s good practice, because it’s not all about the negative, so let’s look at the places where we’re doing really good stuff and try to raise everyone’s game. We’ve got a data architect working with us at the moment, because we’ve moved to Azure, and we’ve got really robust processes in place, and we want to take that learning and build on it.”
There is also a data catalogue, Julia says, “which will help us with things like data lineage, understanding the linkage, and when you do transformations it documents them, and you can just scan over those things and it will give you that visual representation of what’s happening”.
When it comes to the education and skills required to keep all of this in the air, Julia shared: “Our aim is to upskill, and hats off to the team because it has been a struggle to balance that with the day-to-day, but we need to push forward because we know we need to build for the future. The technology is only as good as the people and the processes, so we spend a lot of time on those. We’ve made mistakes along the way, we’ve used providers that have tried to do it for us and not with us, and I’ve done a couple of webinars where I’ve talked about the experiences and the pitfalls that you can fall into when you start from scratch.”
David shared that in his opinion, “identifying the people who will benefit most from proactive care and bringing together a set of data is not really the difficult bit”, but the challenge lies in “making this part of the day-to-day and bringing primary and community teams together”.
“It’s the nuts and bolts of what people are going to do, the workflow, and coordinating those people, as well as satisfying information governance and sharing requirements, and all that kind of stuff. We’ve only got a certain level of capacity, and the question is how we target that capacity. I don’t believe this is massively more activity, I think it’s about activity that we’re doing in a different way.”
We’d like to thank David and Julia for their time, and for sharing their insights with us.
