Digital transformation: How PCNs are making the most of new technology

Beth: Can you give an outline of how your PCN is getting on with digital transformation?

Shanika: West One PCN has a population of around 57,000 patients. We’re one of six PCNs working very closely with our GP federation.

We’ve appointed a digital transformation manager at our PCN, which we’ve found extremely valuable in helping us support practices with digital.

Some of the work we’ve done includes a virtual chronic kidney disease (CKD) pathway, which enables consultants to go into EMIS to look at the referral and the patient’s record and then input directly and provide a summary care record.

We’ve also embraced AI in recognition that it’s a big part of digital transformation. Some of our practices are using AI tools to help with triage. There are also transcribing tools used by clinicians in day-to-day practice to reduce workload. I use that myself and I have to say it does significantly reduce workload; I can pretty much see two patients in the time that I would normally see one.

We are also piloting a skin analytics tool, which is an AI tool, to identify skin lesions. We’re working very closely with our trust colleagues to have a look at whether this can support them in their two-week referrals because they’ve mentioned that up to 40% of the people being referred don’t need to be seen in the suspected skin cancer clinic.

Neil: SMASH PCN in Cheshire covers 70,000 patients across seven very mixed practices, ranging from my surgery of 28,500 patients to a very small one of 3,500.

We’ve got quite a lot of issues around geography because we’re spread out. We’ve got Crewe town, rural areas, deprived areas, commuter-belt developments with lots of new housing and long-standing farming communities. It’s probably three towns in one PCN, which gives us a lot of issues.

We’ve done a lot of stuff around digital and have lots of plans and it’s something I’m very interested in.

John: As well as being clinical director at North West Ipswich PCN, which has 40,000 patients, I’m also chief clinical information officer and clinical safety officer for Suffolk and North East Essex ICB and non-executive director at Suffolk GP Federation. I find these roles do overlap.

As a PCN, we’ve centralised quite a few of our services, which has helped with digital transformation, and we’ve appointed a digital transformation lead. We’ve centralised coding, set up a pharmacy hub, and launched Welby, which is a long-term-condition management system. It stops the duplication of people coming in for a blood test when they already had one done a couple of days before. So, for example, if you put someone on an ACE inhibitor, it will automatically text the patient to book a follow-up appointment for the renal function in a couple of weeks’ time. It’s also useful for CQC because it’s based on NICE guidance.

‘The problem with AI is that it’s a very unequal playing field – that’s challenging for individual practices’

Dr John Lynch

The other thing we’re doing, which I’m sure others are doing too, is to use messaging to manage long-term conditions, by sending out protocols and templates for your patient to fill in for medication reviews or asthma reviews or reviewing their blood pressure etc. If the patient’s bloods have been done by the healthcare assistant and the results are ok, then some of those things can be done at a distance rather than the patient having to come in. So that’s improved efficiencies.

And we’re looking at how we automate some of our coding because that’s becoming an issue. We do it centrally and have a protocol for all the practices, but we’re finding that the number of letters is going up, and it takes a long time to train coders, so we’re working with a company to deal with processing about 70% of our letters. We’re just going through the governance side of that at the moment.

We’ve also been doing a lot of work in the federation around ambient voice technology, and we’ve held practices back because the Information Commissioner wasn’t happy with the consent process. We’ve now got to the point where we’re happy to sign off this one particular product, and we sent packs out to practices to help them implement that.

But I do think that the problem with AI is that it’s a very unequal playing field, and it’s quite challenging for individual practices. Not all practices are aware that they need to complete a DCB 0160, quite often. And that the manufacturer has to have a DCB 0129, and that you’ve got to go through all the regulations, and quite often, the algorithms may be difficult to understand.

But that’s where the ICBs and digital teams can help – the assurance processes to make sure that the product is appropriate and safe for your patients.

Phil: I was clinical director of Foundry PCN in Lewes for five or six years before I passed on my mantle earlier this year. I now retain a kind of a financial director role, as well as data analytics and looking at patient segmentation and workforce optimisation.

As a PCN, we’ve been lucky in terms of the simplicity of our organisation because we’re a single practice PCN with 28,500 patients across five sites. We’ve got two DTLs in post because they do slightly different things. One’s more project management and systems, and the other is data analytics and coding.

A key role of DTLs is bridging IT and clinical practice. A lot of time goes into standardising your own systems – whether that’s the PCN or a federation or practices – because unless you do that and have a shared understanding of what’s happening then it falls apart when you try to bring in other IT systems. It’s as much about project management, people, and relational aspects as it is about the IT itself.

We’ve spent the last six months bringing in long-term-condition management system Welby and we’ve got Anima as our digital front door –  our clinicians still do the triage, but it picks up information at the front door. And we’ve been using AI listening tools with data agreements in place.

Beth: It sounds like you all have a digital transformation lead. Would you say they’re having an impact at this stage?

John: The impact can be so variable. The most successful PCNs are those with a dedicated team for the whole PCN, not bringing one in from another practice and trying to do a little bit here and there. You need that core team that does the change management.

I organise yearly digital days for DTLs and others who are interested in digital and it seems that, in their first year, digital leads struggle to feel empowered within their systems to get on and do things, to write a strategy and so on.

So, the effectiveness of DTLs will come down to 1) have they got the time? 2) have they got the permission? And 3) are they able to collaborate and lead that process? It’s quite a difficult skill set to find in one person.

But when you form a PCN, you need to agree at the outset what practices want to achieve and that sets the direction of travel. For example, at the very beginning, we agreed that we would all be on the same system – that’s absolutely essential if you want to get real change within the PCN.

What I’ve found is that you don’t always get everybody on board. So, you start off with, say, two of the four practices, and once you’ve proved value, the others will join in because they see the benefit.

Shanika: Another limitation is how practices want to work. I’ve got seven practices that work very differently and want to continue working very differently because it works for them and their patients. When you have seven different models of care, it’s very difficult for one person to go in and say, ‘Let’s transform and all go onto one model’. That mindset and culture shift is difficult for one person to lead. It becomes difficult for PCNs with practices using a range of different systems or processes.

Neil: I’d second that. It’s exactly the same story with us – seven practices with different cultures. They’re all delivering good healthcare, but very differently.

We all get on and there are no personality clashes, but we’ve struggled to get them moving in the same direction. And getting them to choose the same product is almost impossible.

Like Phil, we have two DTLs to support practices, but creating a cohesive answer is quite difficult.

‘There is huge power in FOMO.’

Dr Neil Paul

Phil: I’ve done consultancy in different PCNs, looking at the RAG (red, amber, green) rating and workforce. And I’ve realised how lucky we are at Foundry. Though, at Foundry, we put in the hard work five years ago when we were getting three quite different practices to agree.

I think the learning from that is that you can’t force people.

You must create something people want. Unless you can produce something that shows there is a benefit, you’re never going to convince them, no matter how good it is.

Evidence that systems save time is vital. AI transcription was easy to roll out because the first couple of people who did it got immediate benefit and everyone else followed. And literally, within a couple of weeks, everybody’s doing it because they see the benefit immediately.

The problem is that some of the more difficult transitional changes require a lot more systematic change within the organisations. And a lot of this goes back to what Neil was saying about the different ways people work and practices feeling their model of care is right for them, and it works well for their geography.

Getting them to change is nigh on impossible, so it’s nudging – if you go too quickly, you lose the trust because you’re trying to force it too much. We’ve found that using DTLs and using transformational change is about creating a vision of something in the future. If you can sell that vision, then actually you can get people moving along bit by bit. It’s a slow process.

Neil: FOMO [fear of missing out] is a huge power. One of my DTLs is slowly implementing RPA (robotic processing automation) across practices and we go to the innovator practice first, show results, then others follow.

It’s got to be more than stories and emotion. It’s got to be results, such as how much time has been freed up, and practices tend to be very bad at coding this sort of thing well. Our EPRs aim to record clinical content of consultations, not managerial ones.

Our MSK first-contact practitioner project took almost a year to tease out data. We had this worry that we’ve just moved the cost pressure of running physios from community to our own budget. It does look like they have had an effect, but understanding that is so difficult.

DTLs need so many skills – data collection, understanding the coding, and analysis of the data. And I’m talking deep analysis of data and multiple data sources. Then, you’ve got to think about how you might present that data.

And then there’s understanding the money. We found that some of the people who come from outside core general practice don’t really understand how practices work – how the money works – and that can be a real issue.

The right DTL is absolutely transformative, but their backgrounds and skills sets are variable. I do wonder if there needs to be some thought around competency frameworks and accreditation for DTLs. It’s a valuable role that perhaps needs a bit more regulation. Sometimes it’s just the practice manager who’s taken that title and a bit of the ARRS funding.

Luisa: In my capacity as a DTL, my primary role involves collecting and analysing relevant data, which I then review with my clinical director to assess the position of practices within the PCN, or with the practices themselves.

Given the abundance of available data, I recognise that not all of it is presented with sufficient context. But I am committed to understanding each practice’s position on various projects to provide a narrative that highlights the great work being done, ensuring it is not merely seen as a numerical figure.

Within SMASH, we have divided responsibilities: one DTL focuses on digital transformation, bringing a background in pharmacy and business management, while the other specialises in project management and transformation, with experience working with our community services provider.

In addition to data management, we continuously evaluate new digital systems and provide feedback to our practices. Currently, we are implementing digital tools PACO GP [patient and care optimiser that automates processes] and PACO Connect [for cross organisation appointment booking] from Blinx across both PCNs through the ICB pilot project. We are gradually transitioning shared services to this new platform to improve collaborative working.

Beth: What challenges have you faced over the past year in digital transformation?

Phil: Plenty. Thinking about the financial and the management side of things, we come from a background of general practice – small corner shop organisations that were small enough to manage themselves. We’re now in a situation where the organisations are dealing with £5-6m turnover and 100-plus staff and you can’t run them in the same way.

I think part of what holds people back from working together is fear. It’s the fear of what happens when you come together. If you haven’t got the systems in place to track changes to see whether where you’re going will be better than where you were, then you don’t do anything. You need to take data at the beginning, understand it and say, ‘This is my hypothesis, this is my proposal and we’re going to check afterwards and only if it works will we keep doing it’.

I’ve been involved in helping practices with rostering to help them understand the actual costs and what people are doing – in terms of, am I seeing the patients at the top of my level and am I doing the right things with them? And am I being supported by systems, by AI, to make sure that all the right stuff is happening?  If you can measure and plan accurately, then you can see whether the changes are a more efficient way of working.

The challenge is that we haven’t been collecting this information because we haven’t been looking at things in that way.

Shanika: One challenge is that the support from the system has got less and less as the years have gone on. When PCNs started, there was quite a lot of digital investment.

For example, we were given quite a few laptops to help run the Covid vaccination programmes. Now, trying to get a laptop from the ICB for an ARRS role is impossible. Another example is that during Covid a lot of us switched to AccuRx as a tool for text messages, chronic condition reviews, and patients have got used to it for self-booking. Then the cost of text messages came about and this whole thing about who’s going to continue funding it as a tool for practices. With the ICB funding cuts, there is a risk – a huge risk – that the digital funding and support is going to be reduced even further for practices.

Any digital innovation or transformation needs to come with support, including financial support for practices in terms of tools. And if one practice goes for a tool, it’s going to be much more expensive than if you go for it as a PCN. We need our system to support us with the digital transformation agenda and I worry that the support is diminishing day by day.

‘Fear is part of what holds people back from working together.’

Dr Phil Wallek

Neil: I’d echo a lot of that. As GPs, we’re in this strange situation where we see ourselves as independent contractors and independent businesses, but then we want the system to pay for everything for us. I know that’s difficult, but there are times when we do just have to pay for things ourselves. A lot of practices don’t want to do that, despite the objective evidence that it may make them more profitable in the end.

Data quality is interesting. Increasingly, there seems to be huge teams of people wanting to extract data from everybody and stick it in a big analysis pot. They come up with all sorts of allegedly useful information – often, that’s naming and shaming what they perceive to be poorly performing surgeries. But nobody’s talking about making sure that the data quality is any good.

Back in the primary care trust days, we did quite a lot of analysis of the billing data produced by hospitals. We found the data quality was absolutely abysmal. They were getting the wrong patient, wrong practice, wrong area, and wrong operation. We need to be talking about how we improve the data we’re recording, especially in this world of population health, big data and insights.

And finally, our local secondary care organisations appear to be computerising and putting in electronic patient records. They’ve got massive infrastructures and IT teams and we’re in real danger of being swamped by the way that the hospitals want to do it.

John: One of the challenges for practices is that it’s a bit like the wild west out there in terms of apparent solutions. How do practices decide what is best? Where are they going to get that information from? Where is the leadership telling them?

And we’ve got to be realistic. We’re not going to get any money for this – not in the short term, at least. So, one of the criteria that we look at when we’re in our digital group is whether it’s cost effective. If we put in this system to help us with our coding, will it pay for itself?  And will it release staff to do other things?  If it doesn’t tick that box, it’s just not worth it because practices won’t do it.

Then, governance is a real challenge, and no one really knows the solution. We haven’t got a road map for how we do that in primary care and the support isn’t there.

And with trusts putting in EPRs, there can be unintended consequences. They can affect our systems because of the way we interact. You can suddenly find that a change has been made by the trust and it has repercussions and then you need to make some changes to adapt to that.

The other thing driving change is the increasing demand. You might have a practice that is fine now, but in five years’ time, they’ll have double the number of people over 80. We say to them: ‘So, how are you going to prepare for that? Because it’s coming really quickly’. That’s an argument that I use, and my DTLs use, to get people to transform because they might be all right now, but will they be all right in five years’ time?

Beth: Looking at opportunities presented by digital, what are your hopes for the next two years in terms of digital transformation?

Phil: The opportunities are huge. The ageing population means that, if we don’t change the way we’re doing things, we are sunk. And there are lots of ways that digital can support that change.

We need to be supporting people to manage their own health care – giving patients their information and giving them ownership. Digital is really key to that because it can bring together that information and present it to people in a way that enables them to manage their healthcare better.

There’s huge opportunity with AI to support both patient management and our internal systems. I suspect that AI will help us with some of the challenges we have at the moment, such as EPR interactions and data analytics.

I think it’s probably a good thing that we may be shifting towards giving providers more autonomy in terms of giving them a budget. We need to give them control so that they can make a decision about what’s the best thing for them to use and provide the evidence that supports them in making that decision. We need national and regional teams who share best practice, what’s around that’s good at the moment, and who disseminate that down to the teams who are actually going to be using it.

‘It’s difficult to overestimate the impact AI is going to have.’

Dr Neil Paul

Neil: It’s difficult to overestimate the impact that AI is going to have, with people like Bill Gates saying you might not need doctors in five to 10 years. And yet it’s easy to underestimate the effect AI will have locally.

In Cheshire East, they did an exercise looking at where they thought the health economy would be in 2030. So, okay, that’s only five years away, but they did that piece of work without even mentioning AI. Yet it’s going to change absolutely everything and we need to be really thinking about how it will affect things.

And I’m revisiting the idea that hospitals are getting more data-driven too. We’ve just done a piece of work on outpatient referrals and the local hospital can’t even tell you how many they’ve got. They can’t tell you how many they’re getting, from whom, or how many they haven’t seen yet – it’s quite embarrassing. They’re so data poor. Yet I can buy something on Etsy or eBay and have it delivered from Vietnam, tracking it the entire way and getting a text when the driver is three stops away.

Once we’ve got end-to-end data, we can actually grasp that so-called left shift and understand it, not have work dumped on us. We can genuinely take on the pieces of work that we can deliver profitably with productivity that will ease the system.

But we’re going to have to really think across the boundaries, not just internally in primary care. We have to understand how much it costs us to deliver things. We’ve got the potential of a new GP contract coming so we really need to define primary care’s role and what we can do in the system to move the money and the work around to help patients and save money overall.

Shanika: If done correctly, there’s a lot of opportunity – reducing workload, increasing efficiency and standardisation in terms of the way we work. And using a data-driven insights approach to ask, ‘Is the work we’re doing cost effective? Is it working or not working?’

We need to bear in mind that while digital will work for the majority of patients, it might not be suitable for all. So, do we have a counteroffer for people who are not digitally savvy, or who need additional support with that?

But there is a lot of opportunity and we do need to embrace digital transformation.

John: I feel positive about the future. If you look at the investment in health tech, it’s the biggest single area of investment – it’s bigger than FinTech now. And what’s going on internationally is really exciting. You can get carried away by these things, of course, but the potential is very large.

It will make the job of a GP quite different in five- or 10-years’ time with a lot more support in managing complex long-term conditions. From a practice and a PCN point of view, we need to be looking at five to 10 years rather than two years. If you work using these enablers – around population health management, for example – it allows you to be proactive rather than reactive. And that’s encouraging because that has been shown to reduce demand, which will then help us cope with an ageing population.

I really encourage practices to get their patients using their app as much as possible because it allows them to look at their blood results and manage it. The doctor/patient relationship is changing and becoming more collaborative. That can only be for the good.