The transition from Spine to Spine 2 - a success story or yet another NHS IT failure?

Sooraj Shah reports on how the implementation of the Spine 2 - a platform that connects clinicians, patients and applications - is helping to restore the NHS's reputation for project management

Since the winding down of what had been described by MPs as the “worst and most expensive contracting fiasco in the history of the public sector”, the National Programme of IT (NPfIT), little has been left for the NHS to savour.

That is aside from NHS’s Spine, the part of the national infrastructure that stores patient information and enables electronic messaging. The Spine is one of the only parts of the ill-fated NPfIT to have emerged from the disaster with any credibility.

“The Spine was one of the successes around the framework. It was the glue that allowed the NHS organisations to make the right choices,” says James Berry, CTO at healthcare solutions provider Advanced Computer Software Group.

The Spine handled about 400 million electronic messages a month and more than 2.6 billion information requests and amendments each year. The Health and Social Care Information Centre (HSCIC) claims that Spine services and systems were available 99.97 per cent of the time.

So if Spine was such a success, why did the HSCIC feel the need to replace it with NHS Spine 2? The decision revolved around changes to the supply chain as well as preparing the infrastructure for future demand.

The Spine went live more than 10 years ago, and the NHS’s contracts with the suppliers who delivered it were about to end, so some kind of revamp was inevitable. The HSCIC wanted the new Spine to be capable of proportioning significant increases in data volume, enabling messaging and sharing between NHS organisations to be faster and more effective than before. It therefore abandoned the original Spine’s Oracle relational database and moved to open source platforms.

“The NHS is looking for more flexibility going forward than a proprietary system can provide because more interoperability is needed and more endpoint devices need to be connected,” Ovum analyst Charlotte Davies explains.

“It has to be done in a more iterative way, it has to be done in an agile way,” she adds.

So out went Oracle, and in came the Riak NoSQL database from US developer Basho. According to Ovum analyst Laurent Lachal, this is a more flexible solution that has enabled the NHS to respond more easily to changing demands while at the same time cutting costs.

Advanced’s Berry adds: “Ten years ago Oracle was a sensible choice for the NHS, and now they’ve chosen another system that meets the reliability and performance requirement.”

Much of the criticism of the Department of Health’s handling of NPfIT centred on the fact that it signed long-term multi-million pound contracts with big-name technology vendors and was then held to ransom by some of them, having failed to check the small print with sufficient rigour.

“I believe in the old contract there were particular clauses when suppliers reached certain capabilities, now I think it’s a different type of contract and that could be significant,” Berry explains.

According to Stuart McCaul, managing director of EMEA at Basho, who was present in the commercial negotiations between the US company and the NHS, HSCIC has completely avoided vendor lock-in by using the government’s G-Cloud framework.

“It’s a big departure from the government contracts in the past, including Spine 1, which was a 10-year contract. So now the maximum time that a contract can be signed is for two years,” he tells Computing.

Crucially, unlike many of the services derived from NPfIT, HSCIC will be in control of Spine 2.

“They are taking assistance from vendors rather than outsourcing the project, they have a lot of control and therefore a lot of flexibility,” McCaul says.

Modifications to the previous Spine were held up not by the technology, but by the contracts, Berry points out.

Basho is not the only supplier on board, with IT consultancy BJSS also collaborating with HSCIC, taking over a role previously held by BT.

“It’s interesting that it isn’t managed by BT any more [as was was the case in the era of NPfIT]. BJSS, which is a smaller British vendor, is the consultancy that has been contracted,” says Ovum’s Davies, adding that the sidelining of BT is surprising given that it has many big contracts within NHS trusts and local authorities.

The telecoms giant had a £889m contract with the NHS for the central NHS Spine, and various other contracts for implementing patient record systems, which accounted for an additional £1.5bn. But problems with the NHS deal were thought to have contributed to a £134m loss that BT announced in 2009. The NHS may have wanted to move away from one of the traditional vendors to a smaller supplier in line with the government’s strategy to give more business to SMEs.

According to Davies, going for smaller projects is the best way forward in healthcare.

“People look at things as too big to fail; I think of it as too big to succeed,” she says, adding that the NHS seems to have learned from its mistakes with the NPfIT and would be more careful with the SLAs in the future.

Along with speed and flexibility, suppliers and users are looking forward to new features available with Spine 2, too.

“We’re looking forward to being able to deploy is child protection-related enhancements, which mean that notifications from a local authority that a child has a protection order will be pinned to a chart so that at an A&E centre a clinician can be notified of this,” says Jon Payne, an engineer at software provider InterSystems who has been directly involved with the Spine 2 integration.

Payne says Spine 2 will also make it much easier for clinicians to access Summary Care Records.

“The best thing about Spine 2 is that it’s the gateway to making things easier and more accessible to the broader NHS community and will enable tangible benefits to be delivered in a cost-effective and timely way,” he says.

There seems to be a multitude of reasons for the NHS to switch to Spine 2, but was the transition as “successful” as HSCIC has claimed?

According to Payne, the transition went smoothly and none of InterSystems’ customers has complained of any issues.

Advanced’s Berry believes the NHS should be “congratulated” for the way it handled the transition.

“Our employees went home on Friday and came in on a Tuesday morning and the transition to Spine 2 had been completed. As some of our apps need to be turned on for 24 hours a day, our users were actually logged in to our system between the transition and weren’t affected,” he says.

So far, he says, Spine 2 seems faster than its predecessor, though it is still early days.

While users will not have to learn any significant new skills, Basho’s McCaul says that the main difficulties lie in getting them accustomed to the new technology. NHS end users may find themselves interacting with Spine 2 through an unfamiliar web interface, and may require some training.

As for the NHS itself, McCaul believes that it has prepared effectively for having more control over the Spine by up-skilling its own IT staff and recruiting further specialists.

“The NHS asked me to speak at a recruitment drive, where it was trying to entice the latest batch of software engineers to come and join HSCIC and to see it as an innovative and attractive place for them to work,” says McCaul.

“They had a few key architects, and we’ve provided extensive training to the development team of the NHS so they would be able to reliably run the infrastructure by themselves. We had to make it user-friendly and it does require a few days training. We’ve spent several days on site at NHS to do this,” he adds.

Overall, McCaul says the project “proves that the NHS can take on a large complex project”.

“The NHS has vertically integrated this project instead of outsourcing it - it’s an amazing success,” he concludes.