The NHS journey to interoperability is hampered by cultural resistance and its traditional capacity for building large legacy data silos. But while for many hospitals and trusts it remains a fraught work in progress, some projects are now cracking the interoperability puzzle, reaping the benefits and seeing the impact as more data sources are brought in from the cold.

While interoperability is frequently cited as a key enabler for the advancement of integrated and connected healthcare, the comprehensive delivery of truly interoperable systems remains an elusive challenge across much of the NHS.

The will is undoubtedly there. According to the 2018 NHS IT Leadership Survey, 82% of respondents said that interoperability – enabling systems and staff to share patient information – was their highest priority. But reality is falling short of the goal.

Phil Colbourne, Manager, EMEA healthcare sales at information management specialist Hyland estimates that more than 80% of hospitals cannot adequately achieve this goal today.

“Interoperability is reliant upon the use of industry standards like XDS and FHIR,” he says. “Wide-scale adoption of these standards would allow disparate systems to integrate onto a common data sharing platform and thereby deliver the interoperability that the NHS needs.”

The main stress points, suggests Colbourne, are common access to systems (sign-ons), an unwillingness to utilise the existing technology that could be brought into play, and insufficient focus on cloud – and how it can be leveraged to deliver a centralised data sharing model.

“Faster adoption of standards and cloud would accelerate the process and provide a more efficient interoperability solution,” he says. “Our focus is on a centralised, multi-tenanted architecture and the use of industry standard to store, manage, protect and provide access to data. A simple user interface (OnBase Patient Window) provides enterprise wide access to all patient data via a single pane of glass.”

For Portsmouth Hospitals NHS Trust IT technical architect Nick Sargeant, the benefits of interoperability have emerged steadily throughout his experience as a consultant to the Salisbury, Wight and South Hampshire (SWASH) consortium’s project to implement an XDS-based vendor neutral archive (VNA), designed to enable the sharing of radiology imaging data across four trusts.

“I come from an IBM history so I know what it’s like working with a supplier that’s trying to hold on to its customer base,” he says. “I’ve seen so many times when the NHS has got itself into a proprietary corner.

“For me, interoperability between systems, data types and the way data is shared means I can get the best of both worlds. It encourages competition in the market place – I know suppliers will work together, and I can encourage them to speed up development because they know they are not the only game in town.

“As an architect on this project, I was keen from the start on open standards and interoperability. The initial proposition was that a VNA would help with imaging migration ahead of a new system contract: the whole imaging history would then be available from day one of the new service. As the procurement phases progressed, the value of designing it around open standards quickly became a common feature of all requirements.”

The flexibility of participating suppliers was essential to the project’s success and it soon became clear that Hyland stood out when it came to adopting open standards – and would be a suitable platform for the data hosting and sharing aspect of the new radiology system.

Inevitably, the benefits of interoperability bring new challenges for innovation – not least in the exponential increase of data volumes. But there is also a ripple effect which begins to bring other data sources into the interoperability infrastructure.

“We started with radiology but other important sources of images are being archived into our VNA as part of the wider patient record. We are picking different ‘ologies off and they can take a long time to integrate,” says Sargeant. “It starts to go beyond just being an archive. It provides a federation for all those imaging sources. For example, if we know a patient is attending an appointment in two days’ time, the VNA can push the last 10 images from wherever they sit down to the local radiology solution.”

He estimates that just one CT scanner has quadrupled its data output since the VNA was implemented in 2013. By 2020, the system will hold 400 terabytes – a prospect that is driving the consortium’s cloud strategy.

“The large volume of data presents its own implications for migration to interoperable systems, and it’s something we need to keep a close eye on,” says Sargeant. “At the start of the present contract, we forecast the amount of data we’d be talking about – and we missed it by 10-15%! Despite all the practicalities of our forecast it has been difficult to flex system capacity. Moving to the cloud, we’ll be able to do that more readily. Price trends in storage are falling, and the cloud allows you to buy capacity through the project, as required, at the price of the tin.”

The ability to take on board new data types more quickly is central to the consortium’s interoperability strategy, he explains. There is no appetite to cull data in radiology.

“We have always envisaged that the VNA would outlive the PACS,” he says. “It’s a long-term solution, but it requires you to have all the components you need as the project develops. A VNA is essentially software; interfaces need to be standards that you can move forward with in the future. We might want to keep the software but change the way the data is stored. Interoperability operates at all levels of the technology stack.”

Sargeant suggests that desktop convergence is the trend that is lagging behind in the quest for interoperability. Quite simply, there are too many applications. Greater focus on standards for a common user interface would mean that clinicians, who all have their own view of the world, could adapt more readily to new applications. This is already the case for the consortium’s radiology system, which provides standard access via Hyland’s OnBase Patient Window.

Beyond that, archiving is likely to be the next big interoperability challenge, particularly as data volumes grow – converging, for example, from digital pathology and micro-biology sources.

When a specimen is prepared, says Sargeant, rather than providing a 1,000- word description of the tumour, they will simply take a picture of it.

“These slides are big compared with traditional radiology images,” he points out. “We are talking about two, three or four times the size. And pathology carries out four times as much work in a hospital as radiology. The volumes will get big really quickly.”

The long-term challenge for interoperability is that if slides are to be kept for 20 or 30 years, they are essentially being preserved for access via an unknown medium of the future. At least if data is transcribed to open standards today, changes in archiving technology will not ultimately preclude access.

For Hyland’s Phil Colbourne, it all comes down to standards. He says the underlying storage architecture is only part of the solution.

“We believe in disconnecting the applications from the storage via the use of an abstraction layer,” he says. “In that way the application can talk to the storage ‘natively’ but use industry standard protocols to provide access to and sharing of the data.”

It sounds so straightforward. But bearing in mind the diversity and range of databases across the NHS, the scale of the interoperability challenge remains huge. Hospitals need to take their lead from innovative projects as the SWASH radiology consortium, and set their own ripple effects in motion.