One Epic to rule them all?

In the first week of October, digitalhealth.net published an interesting op-ed. Long-standing editor Jon Hoeksma asked if NHS England was thinking of doing a national deal with Epic for a single electronic patient record.

“Rumours have been circulating over the summer that Tim Ferris, the new head of digital transformation is a big advocate of the system [which is used by Massachusetts General Hospital, where he used to work] and has told colleagues it should be used by more trusts,” he wrote. Before adding that there were hints at the HETT conference that “there have been high-level meetings with Judy Faulkner, the founder of Epic.”

Hoeksma himself acknowledged that the piece was speculation, and the NHSX press office gave an entirely non-committal response when asked about the idea. But if nothing else, it suggests that NHS England is thinking about the digitisation of hospitals. Which has to be a good thing, because the subject has been curiously absent from recent discussion of healthcare technology.

Unthinkable, surely…

Until recently, the idea of a big, central EPR procurement for the NHS would have seemed unthinkable. Hoeksma says in his article that “the idea of procuring a single, national EPR system for all hospital trusts in England was the central pillar of the failed National Programme for IT, that ran from 2003-11.”

In fact, even NPfIT didn’t propose a single EPR for the entire country. It split England into five regions whose local service providers placed contracts with three EPR providers – which, over time, were whittled down to two LSPs and two systems: CSC/Lorenzo for the north, midlands and east and BT/Cerner Millennium for London and some trusts in the south.

Still, the point holds. NPfIT achieved some of the things that it set out to do: a national network, some national products including a booking system, a prototype digital front door. But, with the partial exception of London, it signally failed to digitise hospitals; and nobody has wanted to try its model since.

Indeed, when former health secretary Jeremy Hunt wrangled £4.2 billion out of the Treasury for NHS IT in 2015, he asked US ‘digital doctor’ Robert Wachter to conduct a review of the lessons to be learned from NPfIT before deciding how to spend it.

Many of Wachter’s conclusions focused on trust readiness. He called for more clinical engagement with IT, a better trained informatics workforce, and money targeted to trusts most ready to use it. The result was the global digital exemplar programme, which aimed to take a dozen acute trusts with ‘good’ IT and turn them into an ‘ivy league’ of HIMSS level 6/7 trusts.

The idea was that others would then emulate them via a ‘fast follower’ and a ‘blueprinting’ programme; although, in the event, the arrival of a new health secretary, Matt Hancock, and a new digital agency, NHSX, meant only one round of the GDE programme was run.

Since, then nothing has replaced it. NHSX was badly distracted by the Covid-19 pandemic and the high-profile failure of its test and trace app. But it always came across as more interested in ministerial initiatives like axe the fax, the professionalism aspects of the Wachter review, and standards and data, than EPRs.

It did set up a ‘health systems support framework’ to encourage trusts to buy standards-based systems (interestingly, Epic wasn’t on it) and funded some projects at ‘digital aspirant’ trusts. It has just issued a What Good Looks Like document, which says there is a need to ‘level up’ on EPRs.

And it has put out a Who Pays for What document linked to a Unified Tech Fund. However, once that’s gone, both documents say that integrated care systems will become responsible for IT strategy in their patches and for funding this out of their own resources. Over and out. 

On the other hand…

Having said all that, it’s possible to find straws in the wind to suggest that the EPR climate is changing. Hancock has gone and been replaced by yet another health and social care secretary, Sajid Javid.

New secretaries of state like to make big announcements and it’s interesting that in his otherwise poorly received speech to the Conservative Party Conference, he chose to note that while “some parts of the health service have joined-up electronic patient records and surgery robots, others are literally using fax machines.”

NHSX is also being abolished or, formally, wrapped up into NHS England’s new transformation directorate, headed by Tim Ferris. Ferris told HETT that while it’s “hard” he wants to “drive transformation faster”; and he’s likely to find that particularly hard if some trusts lack the basic EPR functionality to tell them where patients are in the system and whether they are being efficiently and effectively treated.

The Health Service Journal reported that NHSX head Matthew Gould told the same conference that around 10% of trusts are still paper based; with many more only “semi-digitised.” Unfortunately, it didn’t say what, if anything, NHSE/X is proposing to do about this; but it might be another indication that the subject is on the table.

Four more straws: There is a comprehensive spending review coming up, and past history suggests the Treasury is more willing to put cash into big systems that hold out the prospect of a measurable return on investment than other kinds of IT programme.

Two: NHS England has a mechanism to run a big EPR procurement, if it wants to. WGLL and WPFW say they are handing IT over to ICSs, but they also say there will be central funding for national infrastructure, national products and systems that it makes sense to buy once. WPFW lists systems like Microsoft 365 as an example of the latter; but an EPR might qualify, if a good enough case could be made.

Three: There is a roll-out mechanism. The independent evaluation of the GDE programme has just been published and it concluded that it basically worked. Handily, the latest round of NHS reforms, will not only create integrated care systems, but provider groups and integrated care provider organisations. They will be big enough to get the value out of a big EPR and lend themselves to buddying arrangements.

Four: NPfIT casts a long shadow, but central bodies are feeling pretty chipper about tech at the moment. The Covid-19 app didn’t really work out, but the deal with Palantir did, and so did initiatives like Covid Oximetry @Home. Why not keep going?

What are the options?

Epic for everyone:

Hoeksma discussed the pros and cons of a national procurement for Epic on the basis that it really would be used across the NHS, by all trusts, or at least all acute trusts.

He listed out a number of pros: Epic is a big, well-integrated system that is already used by a lot of tertiary centres; the ‘Epic way’ drives consistent deployments and working practices; and a national roll-out would bring additional benefits, such as familiarity for clinicians who move between hospitals and easier information sharing.

He also identified some obvious cons: it would be staggeringly expensive and it would only address the IT needs of hospitals when integrated care systems are being told to integrate them with other providers in new, digitally enabled pathways of care.

In fact, one of the challenges of a national Epic deployment would be defining its scope. NHS England has been setting up clinical networks and regional networks for pathology and imaging. Increasingly, they’re choosing their own IT, and they’re not choosing Epic, they’re choosing established imaging and pathology suppliers.

At the same time, acute trusts are looking for remote outpatient and ‘hospital without walls’ solutions, and while Epic has a patient portal, it doesn’t really do what the more advanced trusts are looking to do. Still, all systems need to be integrated with others at some point, so while structuring a deal would be complex, it shouldn’t be impossible.

Impact on the market: Hoeksma’s scenario would require NHS England to not just run a national procurement for Epic, but to require trusts to use it. The commissioning board might be able to do this, if it could make using Epic a condition of a trust’s licence. But the competition issues and push-back from foundation trusts would be huge.

Clearly, if it happened, the EPR market in England would close. Other EPR vendors, and the larger number of suppliers working with ‘best of breed’ or ‘open platform’ trusts would cease to win business and would, presumably, see their technology removed as Epic was rolled out, presumably in phases, over a number of years.

Likelihood: Still very unlikely. The option of literally buying Epic for the entire NHS has the allure of being a big thing to do. But it would not only be staggeringly expensive; it would wipe out the sunk costs of EPRs in trusts that have digitised; including those that have achieved HIMSS 6/7 using other systems.

Those trusts would see progress stall as they waited for Epic. And it’s unclear where innovation would come from. The additional benefits to the NHS of using just one EPR system would have to be enormous and well-accepted to overcome these problems.

Epic for some, other systems for others:

Hoeksma has evidence that NHS England has been talking to Epic. The NHSX statement says it regularly talks to vendors and has been discussing WGLL and WPFW with them.

So another scenario, if a national procurement is in the offing, is that it would be for Epic for trusts that want it and find the current local, regional, national and Treasury governance difficult to navigate. Alongside this, NHS England might run similar procurements for a handful of EPR systems for trusts that wanted them.

If this was going to address the problem of a significant minority of trusts still lacking EPR functionality, then it would also have to be backed up by a requirement for trusts to deploy one of these national systems; perhaps as a condition of their licence, or a high ranking from the Care Quality Commission.

This might sound NPfIT-ish, but it would be closer in spirit to the aims of the GDE programme as interpreted by Tim Ferris’ predecessor, Matthew Swindells, who wanted to use it to identify “four or five” EPRs with solid credentials, put them on a framework contract, and allow suppliers to promote them with roll-out support from the GDE teams.

Impact on the market: The EPR market in England would close once all trusts had picked a provider. In all likelihood, some small vendors, with just one or two sites, or under-developed offers, would be pushed out of the market, as their customers came under pressure to pick one of the national systems.

However, there would be more than one player in the market, which would help to address competition concerns, trusts would be able to retain their sunk investments, as long as they had gone with one of the national providers, and there would be some scope for innovation, as long as trusts were able to switch EPR in the future.

What about best of breed? An interesting question in this set-up would be what would happen to trusts that didn’t want to buy from a single supplier, but wanted to stick with or pursue a ‘best of breed’ or ‘open platform’ approach.

As the Highland Marketing advisory board pointed out recently, there is a tendency for providers like Epic (and Cerner, and to some extent System C, Allscripts, and Meditech) to talk about EPR as a ‘thing’ when really it is a suite of functionality that supports organisational and clinical operations.

Any national EPR procurement would probably be for a ‘big box’ system or systems. So what would happen for trusts that have a patient administration system from one supplier, but order communications and results reporting from another, and an e-prescribing or e-observations set-up from yet another… and so on. Two ideas:

One: We don’t really know if NHS England is planning a national EPR procurement, but if it is one of the reasons that it might want to run one is to choke out ‘best of breed’. Tim Ferris’ reported remarks suggest that he agrees with the Epic (and Cerner) line that an extensive, well-integrated system, with one database, one user interface, and well-defined workflows is what trusts need to run efficiently and effectively.

No ‘best of breed’ trust has ever proved otherwise by getting to the top of the HIMSS EMRAM model. And one reason that so many trusts are only “semi-digitised” is that the sheer effort of finding and integrating the right systems has been beyond them.

If NHS England ran a national EPR procurement, and if it was for just a handful of EPR systems, and if it found a mechanism to require trusts to deploy them, then a whole slew of functionality providers might be pushed out of the market.

Life would become a lot more difficult for new entrants with open data platforms addressing the data and integration challenges of the traditional ‘best of breed’ approach. Innovators might find their market was big providers who could take them into trusts, rather than trusts directly.

Two: NHS England might allow a coalition of suppliers onto the national EPR contract to keep a ‘best of breed’ option open. However, this would be complex to put together, and might take the commissioning board back into HSSF territory, with many suppliers on a framework contract that trusts fail to use.

Suppliers would have to make a very good case for being able to get trusts up to speed quickly. If NHS England specified a roll-out model along the lines developed during the GDE programme, they’d also need to find trusts willing and able to support deployments.

Likelihood: More likely. The big item missing from a set of national EPR contracts would be the efficiencies of scale that would be delivered by having one system, on which professionals could be trained from the outset of their careers, to do things the ‘NHS way’.

But it would tick a lot of other boxes. It would give trusts that want Epic a faster route to procurement than the current set-up, while creating a mechanism for “less digitised” and laggard trusts to obtain the IT that it looks as though NHS England wants them to have.

It could be spun into the government’s ‘levelling up’ agenda, might appeal to the Treasury, and would allow Sajid Javid to make a ‘big’ announcement on reform and technology. It wouldn’t be popular with all trusts, but if large, digitised organisations were allowed to keep and build on their existing systems, the NHS might wear it.

Indeed, some parts of the NHS would welcome it. At Highland Marketing’s advisory board meeting, it was noticeable that the two long-standing chief information officers who gave presentations were pursuing ‘best of breed’ and ‘open platform’ options and thought the way forward was to evolve them, connecting into their health and care communities in the same way.

But a new CIO, who had arrived in healthcare via the fashion industry, had not only deployed a Cerner EPR in a ‘big bang’ but was baffled as to how the NHS had “missed a golden opportunity” to develop a single EPR years ago. Not everybody remembers NPfIT.

One more heave on the strategy lever

The NHS is overdue a digital strategy. Personalised Health and Care 2020 wasn’t well focused and is now out of date. NHSX has been promising a replacement and, when it emerges, it will hopefully fill in some of the detail that is missing from the data strategy, WGLL and WPFW.

NHS England, its transformation directorate, and NHSX, could use a new strategy to try to achieve some of the benefits that they would get from a single national procurement for Epic or a suite of EPRs in other ways.

For example, they might define the kind of digital maturity that they want trusts to have and think they will need to feed the data platforms and apps outlined in the data strategy: either by defining functionality or setting a target for all trusts to reach a defined level on the HIMSS EMRAM or a similar maturity model.

They could back this up with a national ‘offer’, ranging from a full-on exemplar programme to blueprints and guidance. They could add carrots and sticks for boards from the usual mix of licence requirements, digital maturity targets, new funding pots, target spend, and regulatory activity.

Impact on the market: An NHS digital strategy with a strong steer to complete the digitisation of hospitals would be welcomed by the health tech market, which would see it as a “boost” to NHS IT. It wouldn’t change the mix of approaches available to trusts or the range of suppliers they might want to work with.

In principle, it ought to create new business opportunities, as suppliers would be able to pitch for business on the basis of a better defined requirement from the centre. However, this wouldn’t be guaranteed, since it would be ‘more of the same’ of what has been on offer to non-GDE trusts and all trusts since the GDE programme ended.

Likelihood: Very likely. In fact, it’s a certainty that there will be a new NHS digital strategy at some point. The big question: will a strategy do the trick, or will a national EPR procurement or procurements be required to give it effect? That may be the question that NHS England is wrestling with at the moment; with Epic and, hopefully, others.

Natasha Phillips: Health tech vendors and nurses must work more closely
Standing up for health tech and SMEs: Shane Tickell’s vision
The power of video marketing for health tech
Stabilising the NHS – it’s going to take a lot more than AI
Get it right, but get it done