Speaker1: [00:00:00] Hello, I'm Suhail Mirza and welcome to Voices of Care, the healthcare podcast. I'm delighted today to welcome my guest, Pritesh Mistry, Fellow at the King's Fund. Pritesh, welcome. Thank you for giving us your time. 

Speaker2: [00:00:11] Hi. Thank you so much for inviting me.

Speaker3: [00:00:16] This is Voices of Care, the healthcare podcast.

Speaker1: [00:00:20] It's a pleasure. It's an exciting year for King's Fund, a new Chief Executive. A busy year for health and social care. 

Speaker2: [00:00:28] Yeah, it's always busy. And as you mentioned, we've now got Sarah at the helm. So some exciting changes ahead, I'm sure.

Speaker1: [00:00:34] And your area of speciality. I know you've worked within the NHS and the Royal College of General Practitioners in your career. The theme running through it for me is technology, research, innovation. So your area of focus at the King's Fund is very much around the technology piece.

Speaker2: [00:00:48] Yes. So my work tends to encompass technology, data and innovation within health and social care. So my background, as you rightly say, is broadly technical. So engineering and physics, but a lot of the work that I now do is focusing on the enabling environment. So culture, leadership, infrastructure. So how do we get the best from technology.

Speaker1: [00:01:10] So it's that sort of intersectionality between technology and the human.

Speaker2: [00:01:13] Absolutely. Yeah. So the human factors, the technology, the culture, and the environment are so important. So we see a lot of amazing things with technology, but that tends to happen in isolation and in pilots. And when it hits the real world, you don't often see that huge potential realised. So what is it that we need to do to create the environment for that technology to have the benefits for patients and staff?

Speaker1: [00:01:36] I think by the end of the podcast, we're hoping with your wisdom, we're going to shed a lot of light on that. I'm going to start with a four-letter word, if I may, which is not often usual, but the four-letter word I'm thinking about is data, because that's really a central theme running through all of these technologies, whether it's AI, etc. and I want to specifically focus and get your view on the initiative of the Federated Data Platform. It's a centrepiece for the NHS. It's a transformative technology, potentially. Can you give us a little bit more context around that, its rollout and actually where it sits in the policy spectrum?

Speaker2: [00:02:10] Absolutely. So the Federated Data Platform was procured at the end of 2023. And I suppose it's trying to solve a historic problem with the NHS in that the digitalisation and the approach to digitalisation of the NHS is happening in silos. So you'll have individual organisations or even departments digitalising and as they digitalise, the technology tends to be procured or purchased independently without consideration of how those systems may work together. And so that's created systems. So technologies like software with the data captured in that which is in different formats and in different structures. And they don't link up well together. And so you might have information about a person in a hospital record, in a GP record, in a shared care record, and in various other places as well. And so the caregiver can't have a full view of what's happening to that individual, where they've been. And so what we end up having is patients having to tell the caregivers time and time again about what's happened to them. And that can be, as you might imagine, quite difficult sometimes if it's a serious condition. Also, it relies on the memory of the patient. So the technology, the Federated Data Platform, is intended to pull all that information for an individual together in one place, so the user, the healthcare staff, can access the information that's relevant for that person at that moment in time to be able to provide the best care possible.

Speaker1: [00:03:45] And it's been broadly welcomed. Matthew Taylor, NHS Confed welcomed it. Doctor Nicola Byrne, who's obviously the National Data Guardian for Health and Social Care, says it will have a transformative effect. And as you say, it's there to drive efficiency, save money as well and make care delivery more seamless. It's based on a couple of predicates, though, and one of them is to have sufficient level of trust in terms of it's one's personal data, what's being done about that? And that is a legitimate point that people have raised.

Speaker2: [00:04:18] Yeah. So I mean, there are a few dependencies, let's call them. There are both technical, cultural and trust factors that come into this. It's focusing on the kind of the trust element. The NHS as a system has not been very forthright in engaging with patients. And so we've seen a number of missteps when it comes to data. Care.data is the big one. That was quite a few years ago. More recently, there was a GP data for planning and research purposes. Exactly. And what we saw with those examples is an erosion of trust in the public, in how data is used in the NHS, in particular the GP data planning, research purposes that showed an uptake. So an increase in the number of people opting out of their data being used in the NHS. And what that means is that people are less likely to have their data and representative data, which has implications for unbiased systems. So, systems that are equally effective for all being used in the NHS. And now, what we're seeing is that there has been a move to try and increase trust in the NHS. So the Federated Data Platform is for direct care purposes, so people can't opt out of their data being used for, for that application. However, it's not just a federated data platform. There are additional contracts that have been provided around privacy-enhancing technologies, so that should improve the privacy, the security and enable tracking and understanding which staff members are accessing data and for what reason.

Speaker2: [00:06:00] So that should improve trust in how data is used. And there's also approximately £2 million worth of funding that's been allocated at the end of 23. So 2023. All around patient engagement and public trust. So that's going to deliver on the three commitments from the Data Saves Lives strategy. So that includes things like transparency statement, a data pact, and standards for public engagement which is fantastic to see. I think that's absolutely necessary. It's the right thing to be doing. Unfortunately, those commitments were due to be delivered in March last year. So in March 23rd, we're now in March 24. So they should have been delivered 12 months ago. It's great to see that it's actually moving now and something is happening. But alongside that 2 million, what's going to be covered has been expanded as well. So my challenge with that would be yes, absolutely necessary. It shouldn't be a one-thing or one-time thing. It needs to be ongoing and we need to make it meaningful. So not just ticking off things, but meaningful, ongoing conversation with publics and the patient, as we've seen in various different parts of the country like London, like Manchester, Frimley as well. So amazing stuff happening. But it needs to be supported.

Speaker1: [00:07:18] No, absolutely. I think that's it's something that we'll keep a close watch on I'm sure. And I think it's part of this broader piece where technology has really been a central element. We've had the budget, of course. March budget in 2024, significant increase in investment, planned over 2025/26, over three years. But I want to take up the trust point that you've talked about, and broaden that because the technological transformation is vital. The government has put a lot of money behind it. It's been trailed in many policy areas. But we also need to understand the impact on the leadership within the NHS, etc. And you've written before very eloquently around the disruptive impact of technology on leadership. Can you explain about that? Because I loved your analogies. Very rare I see commentary that will quote Elvis. And we need to, we can't go on with Suspicious Minds. And the role of leadership and trust, both within the stakeholders that work within the NHS and the public. Can you just expand? That's a really important point.

Speaker2: [00:08:27] Yeah. So the trust is super important. And when we're engaging with the public, it should obviously include the staff and the leaders. Then the leaders know how to support their staff, and the staff know what patients and the public expect. And going back to what I was saying, that the NHS has been digitalising in silos. So you get often what happens is that you then have technology that is developed and delivered in a particular part of the system. So for example, if you have GP practices that are highly digitalized but hospitals aren't, then you find that there are ripples in those impact that causes burden on staff, and it causes more pressures in the system. That has knock-on implications for staff and for patients. And so your hospital sends letters to your general practice. Your general practice has to scan them in. And that takes time. It's inefficient. It doesn't work well. And this is because parts of the system are not working collaboratively and not working well together. So going back to what I was saying about digitalising in silos, what we're seeing is that when leaders step into that space and work collaboratively, your systems that you're buying are purchased with the intention to work together.

Speaker1: [00:09:43] The funds are distributed across many different aspects of the system. 

Speaker2: [00:09:48] The funds are distributed, but also the requirements for the technology are distributed. So rather than saying this technology is absolutely what we need, it's also thinking what's the trade-offs, if it's absolutely what we need, but it's not going to join to the other parts of the system then actually it's going to cause other pressures. So if we think about things like virtual wards, for example, we're now moving into an era where the technology, it pushes against the boundaries. So virtual wards, you'll find that the money comes into hospitals. Generally that means that your resources lie there. They can increase their number of staff, they can deliver the care, but then the patients will be in the home, in the community. That means there's potentially more work for community staff. That money doesn't go to them, but they're holding more risk. And so where we've seen it working really well is actually those parts of the system communicating together with the leaders, helping to bring parts of the system together, moving towards a shared goal. And in doing that, it works out better for staff and for patients.

Speaker1: [00:10:52] And the subtheme within that in making sure systems speak to each other. Interoperability. You've talked about, a bit of an arcane subject. However, that too is not just a technology issue. It is also around relationships.

Speaker2: [00:11:08] Absolutely. So if we think about interoperability is a technology issue, then you can tick off your computer systems and your data and your data formats, and you soon get lost in technical jargon. But what you find is it still doesn't mean the data or the information flows. What we find is that people who want to share data will share data and will share information. The technology needs to help, but really it comes down to people trusting each other, different professions from different departments and different organizations, trusting each other, having that agreement of information governance in place. The technology helps that. But if you know how other people are going to be using that information and it's going to be for the benefit of the patient, it's going to be shared.

Speaker1: [00:11:59] And that also has a big impact. We've had the Messenger Review now for some time and the importance of leadership. So this plays into this. And there's going to be some very important training and development that's going to be required in terms of this intersectionality of leadership and technology.

Speaker2: [00:12:13] Absolutely. So I think there's going to be substantial changes to leadership because of technology. Part of it is how people work together, but also it's how do you share responsibilities, how do you share the burden? How do you share the risk when it's going to be distributed unequally across different organizations and different departments. And then it's also how do you start creating more of a holistic culture. Department level or organization level. But when you've got staff that are potentially on site and off site and different professions and new professions coming in, it's all changing. You've got underpinnings of digital. So leaders need to be able to use digital and be familiar and confident with that, to be able to engage with staff in a two way conversation to create a good, cohesive culture.

Speaker1: [00:13:01] Watch this space. I think it will be interesting. I want to just expand the conversation slightly. We've talked about the public and the importance of trust, how data, how technological interfaces are now going to impact community primary care, tertiary care, acute care. There has been a large increase in registrations on the NHS app. Big increases in visits to the NHS website. But there is of course, an issue around digital inclusion. The NHS have issued a framework on this point. There are significant numbers of people in our society that don't have or are either offline and find the challenge of interacting with the NHS quite significant. And there are those people who are not yet sufficiently digitally literate. Can we talk about how important as a golden thread, in my view, inclusion plays in this whole digital transformation? 

Speaker2: [00:13:56] Absolutely. So I think it's important to remember that healthcare, traditional healthcare before digital was not perfect. And now with the incorporation of digital, it's improving, but it's not one or the other. It's both. It's making healthcare more accessible, more impactful, better outcomes, a better experience for staff and for patients. So digital inclusion is a key thread, I completely agree. We published some work in March 2023 looking at talking to people, talking to members of staff, and trying to understand what's working, what's not working. And, I think there's a lot of misconceptions about who is excluded. And to be honest, I think we need to take away some of those myths. We need to stop thinking that there's particular groups and particular demographics, we need to stop thinking that once everyone's online, that will be solved. Because in reality, anyone and everyone can be digitally excluded. And by that I mean it can be down to your personal circumstances. So cost of living crisis, the increasing number of people unable to afford devices and data, they're more at risk of being digitally excluded. People who don't have skills and confidence will be digitally excluded. But also we're talking about healthcare. As your healthcare changes, you may have less physical dexterity, for example, less able to use devices. 

Speaker2: [00:15:17] And so as your health changes, there's likelihood that you may become more digitally excluded. And so what we need to ensure is that we've got services that are digitally enabled and traditional services that meet the needs of the public. And so what our work showed is that there's three fundamentals. There's your devices, there's your internet data, and then there's your skills and confidence. And these are absolutely necessary to just be able to be considered to start being digitally included. But then it's not just those things. It doesn't stop there. It's also about how your services and your digital tools are designed. And this is where we need to start engaging with people. So a lot of what we've heard is there are people who are very digitally capable. They do their online banking. One lady described her setup as something from Star Trek with multiple monitors and all sorts, but she finds herself digitally excluded because the tools and the services are not set up around how she needs them to be set up. They're not set up, co-developed and co-produced with her. So what's absolutely necessary is those fundamentals. But services that are co-developed and co-produced with people in community settings. And this means that the NHS can't do it alone.

Speaker1: [00:16:37] This includes a very important role for the voluntary sector that you talked about I think.

Speaker2: [00:16:41] Absolutely. So the NHS can't do it alone, nor should it. There's a lot of good stuff happening out there. There's a lot of communities that are around voluntary organisations and charities. So how can the NHS and we're seeing this in parts and they're doing such good great work. They're working in partnership, in collaboration with charities and voluntary organizations to talk to people, understand what their challenges are, how they want to access services and then reconfiguring services to have your continued traditional services mid-level, where you've got volunteers and tech in the public area, and then you can use your tech at home and that's your more high-level tech service. And by having that, optionality means that people can choose where they go. They can help people to move up the digital ladder, and we can continue to use services in the way that people expect and need. 

Speaker1: [00:17:35] I think the message for me there is that digital exclusion and inclusion needs to be defined very carefully, and we've got to keep a watching brief on it. Now, you've talked about the voluntary sector. We've talked about the NHS can't do this on its own. We're very focused on the NHS here. Quite rightly. We've had the budget for technological transformation etc.. And there is another segment of the healthcare landscape of course, called social care. And I just want to touch upon that briefly. We just don't have the time to cover it entirely. But I know that you and I think your colleague, Simon Bottery, have done some really fascinating work looking at the potential and already the promise of technology within a challenging social care environment. Tell us a little bit about that. I think you talked about there being a very wide potential for technology from whether it's going to be virtually assisted technology or helping with the hospital discharge. 

Speaker2: [00:18:23] Yeah. So I'd say there's huge potential for technology in social care. Unfortunately, it doesn't have the attention and support that it needs. So the current focus is around digitalizing social care. So digital social care records and falls and prevention. So preventing falls and prevention. So there's a lot more work that needs to happen to support social care. It is a system, as you rightly say. So it's not just about the NHS, it is about social care and how those parts of the system work well together. There's huge benefits that could be derived from technology and social care, and people expect it as well. So it's about getting some of the digital basics right. Unfortunately, I would say that there's similarities and challenges across the NHS and across social care. So there's insufficient workforce. The workforce doesn't have the support and skills and education needed to be able to use the digital tools to the best ability. There's a lack of digital infrastructure as well, and you'll see this as a commonality across the NHS.

Speaker1: [00:19:28] This came out also in the Hewitt review when they were looking at systems and the importance of having people who are technically qualified in terms of data and technology. 

Speaker2: [00:19:34] Yeah, absolutely. And then some of the evaluation I feel is incomplete and doesn't often look at the implications on social care. So virtual wards we mentioned already, when they're used in some instances there's no evaluation on the impact on carers, for example. And there's a real risk that technology like virtual wards asks more of carers and that causes carer burnout, more burden on carers, which then has knock-on implications for the patients, for the NHS and for social care as well. And so we need to consider more holistically who's going to be impacted by the technology and how we make sure the technology works for all of those people. Something really interesting that came out of that roundtable was around the potential for technology to do some of the stuff that we often overlook. So technology is great. It can do a lot, but we often focus on the number of widgets. But what we heard from one individual was the opportunity for technology to give her more control, more independence. We don't have metrics for control of their environment, 

Speaker1: [00:20:45] But they're so important to the humanity. 

Speaker2: [00:20:48] Absolutely. So there's a huge potential. So she was speaking about how something very simple, like a voice assistant, can give her more control of her lighting and her blinds and all these other things that enables her to feel like she's much more in control of what she has and what she can do. And there's much more that technology can do, but we just haven't really had the opportunity to do that yet.

Speaker1: [00:21:11] But plenty of promise within social care.

Speaker2: [00:21:13] So much promise that we should be pushing more towards.

Speaker1: [00:21:17] Absolutely. One to end now, we've referenced Star Trek earlier. Having grown up watching it, space is the final frontier. So here in technology, we're now going to touch upon AI and large language models as the new frontier, maybe the final frontier. This has been trailed a lot in mainstream media all over the place. But in healthcare, I mean, the government have made a bet in the March budget of this year in terms of funding, the Institute of Public Policy Research produced a report looking at all of the public sector and talked about tens of billions. I think it's 24 billion of savings that could happen across all government if AI rolled out. Bringing it back down to earth, which I think you have a brilliant, uncanny habit of doing, looking at these things with helping us who are not experts. Talk a little to end today about the promise of AI, but also some of the demands it's going to make in terms of how do we define that and regulation.

Speaker2: [00:22:13] So we could do a whole series of podcasts on this.

Speaker1: [00:22:17] I'm sure we will. 

Speaker2: [00:22:19] So I think, okay, to start with, what do we mean by AI? And it's very difficult to define what we mean by AI. It's an umbrella term. It's been around from the 70s. And what it means has changed substantially in that duration. I would say it's like talking about vehicles. What do you mean by vehicles? Do you mean bikes? Do you mean cars? Do you mean airplanes? They're all vehicles. But they do very different things. And AI has a parallel. So I suppose the way I would frame AI is thinking about human cognition and human capabilities. What is it that humans can do that technology hasn't been able to do and is now starting to be able to do? And so what we're seeing is there have been leaps and bounds in AI, in the capabilities you can do. And that that gets us very excited about the potential for AI. And so I think I'd categorize it in four main buckets. So there's AI that can partially do a task. And the implications around that would be that it reduces workload for a member of staff.

Speaker1: [00:23:21] So productivity is a potential.

Speaker2: [00:23:25] And then there's the extension to that, which is entirely doing a task. So then that starts changing people's roles and responsibilities. And then you've got AI that can absolutely make things easier. So it changes your risk and it makes things more intelligible, more easier to do. So that means that that task could move from an expert to another healthcare professional, or potentially even a patient themselves. So things start moving around, I suppose, as something that is similar but isn't using AI is online triaging. So whereas you used to call and then you used to have to tell someone your information, you now do that online so that task or that activity is moved to the patient doing it through a digital tool. And the last thing is that AI can do things that we've never seen before. And that's where we talk about the more transformative AI. And by that, you can now get AI that will analyse an ECG so your heart rate measures or the electrical pulses of your heart. And from that, it can start analysing the risk of you having a stroke or a heart attack. And so there's information that we have been looking at visually. But I can now look at the data within that and has a whole other level of interpretability within that. And so there's huge potential for AI, but it does come down to the data which we've spoken about and the quality of the data and the representation of that data. So the data needs to be representative of the people that it's going to be used for. If it's not, then it might be biased. You end up with potential for worsening care outcomes, widening health inequalities. And so yes, there is huge potential, but I think we're still quite early on that route.

Speaker1: [00:25:22] And one of the key points that I've drawn from what you said there is that there is a potential there for more enhanced co-production with the patient, but there will also be the gap of ensuring that the patient population has sufficient health information literacy. So it goes back to the data. It goes back, I guess, ultimately to that quintessential humanity and the trust that people are going to be required to have in what AI can do.

Speaker2: [00:25:49] Yeah. So staff need to be able to trust the AI. People need to be able to trust the AI, and there's a lot of AI that's being developed and proof of concepts, which are the amazing things you can do. But once you start looking at more in depth, you see that it works well for some people, less well for others. People won't trust that, and that's going to be a negative. So we need to do this properly. We need to make sure that we can evaluate it effectively. And that means changing how we evaluate tools. It's not just about evaluating it now but having a longitudinal monitoring approach. So how is it continually helping, benefiting or not? How do we go back and reevaluate our tools as an ongoing basis? And going back to what you're saying, there's a lot of focus on AI for this healthcare system, which is great, but we need to also move towards more of a patient-centred approach to AI. So I potentially, as I mentioned, can move tasks. And one of the big benefits that large language models have is it can change information, not just text-based information, but let's focus on that for a minute.

Speaker2: [00:27:01] It can change text-based information from very specific, jargon-heavy, medicalized language to something that was much more approachable and understandable for a layperson. That empowers people, it gets people more involved in their health, more understanding in their health, their well-being, and empowers them to be able to do more to look after themselves. But it can also, you're now seeing a move on from large language models to multimodality. So that's using different formats of information. So voice animation, video imagery, a mixture of that with text as well. And so now you can move away from relying on text to people who have no literacy or very low literacy levels, are unable to read but can look at an image and notice that something is going up, which means that that's not necessarily a good thing and that can help them again on their health journey. We can start moving away from these traditional barriers of literacy, numeracy, health literacy that you need to keep building up and use technology to circumvent some of those to really empower people. 

Speaker1: [00:28:07] So actually, it's potentially is a fabulous tool done properly to democratize not only diagnosis but delivery of care.

Speaker2: [00:28:16] Absolutely. 

Speaker1: [00:28:17] On that very, very positive and upbeat note, I'm sure we'll be inviting you back to unpack that more as more and more developments. I'm sure it'll be high on your agenda in the work that you and the King's Fund are doing. But on that note, Pritesh Mistry, thank you so much for your time and your wisdom today. 

Speaker2: [00:28:32] Thank you. 

Speaker1: [00:28:33] Pleasure. If you've enjoyed this episode of Voices of Care, please like, follow, or subscribe wherever you receive your podcasts. And if you want to find out more about how we are transforming health and social care, please visit newcrosshealthcare.com/voicesofcare. In the meantime, I'm Suhail Mirza. Thank you very much and look forward to seeing you on the next episode. 

Speaker2: [00:28:49] Voices of Care, the healthcare Podcast.