EHI interview: Matthew Swindells

Matthew Swindells carried out a major review of NHS informatics during the year that he acted as its first chief information officer. So he’s well qualified to consider the future of the National Programme for IT in the NHS.

Swindells, who left the Department of Health for a position as head of the Tribal Group healthcare practice 18 months ago, says the project can still be achieved; but that “it is running out of time.”

His view was apparently confirmed soon after his interview with E-Health Insider, when Alistair Darling told the BBC’s Andrew Marr Show that the programme is “quite expensive” and “not essential for the front line.”

The Chancellor’s comments prompted a spirited defence of NPfIT by health secretary Andy Burnham, and in the end, the Pre-Budget Report only “trimmed” it by £600m. However, Swindells feels it is at risk of losing sight of the original vision of integrated shared records that support patient flows.

“There is a danger that problems in the national programme are leading to compromises with the central vision,” he warns. “If they compromise on the core vision of integrated systems to support integrated care we will have spent a lot of money and not delivered the infrastructure that the NHS needs to transform quality and cost of care.”

Vision on

Swindells was chief executive of Royal Surrey County Hospital NHS Trust and then a health policy advisor to the health secretary, before he became director general for informatics and programme management on the NHS Management Board and ultimately acting chief information officer at the start of 2008.

He was closely linked with the wave of NHS reform that saw primary care trusts reconfigured and a new interest in shifting acute care into the community get underway with the Our Health, Our Care, Our Say white paper.

He has no time for the idea that everything was better before the programme came along. “There are too many people looking back with rose-tinted glasses on where we were 20 years ago,” he says; adding that for all its delays and problems NPfIT is still preferable to the piecemeal lucky-dip that used to be NHS IT.

The health service, he adds, must step up and take responsibility for many of the current problems, and not just blame suppliers and contractors.

“In Cerner, we have the world’s leading hospital IT system, installed in hundreds of hospitals around the world. There is no question that the suppliers under-estimated how hard it would be to meet the NPfIT specification, but we should also look to ourselves and ask why we are making it so hard.”

With regard to the national programme’s other ‘strategic’ electronic patient record product, he says: “The vision of Lorenzo is fantastic, but it is a vision that requires joined up care and that requires interoperability with the NHS Spine and other products.

“Lorenzo is probably a next generation product. We should treat it as a product that we want to succeed. But we need to be honest about how long it will take to get to a stable product; if this is going to be another 18 months we should say ‘that’s OK’ and have a plan for the interim.”

Swindells rejects the suggestion that the NHS could have found a simple solution by buying proven products, gaining the benefits and then building on these foundations. “The NHS had spent two decade buy big HISS systems or trying to tie together ‘best of breed’ and the position was pitiful,” he says.

“To meet the NHS’ aspirations, it had to be prepared to be at the bleeding edge. The mistake wasn’t in deciding not to buy off the shelf. You could argue that its mistake was in being too traditional in the systems it bought, not too radical.”

Misunderstandings

Instead, Swindells feels that some of the problems the national programme has had to grapple with are the result of mistakes and “issues of psychology.”

“First, there was a specification of systems that assumed the NHS was a single organisation managed by the Department of Health, when, in fact, it should be viewed as a complex system with many organisations within it,” he says.

“For the duration of the national programme, the NHS has been behaving less and less like a single entity and encouraging individual units to behave autonomously; chief executive David Nicholson has been encouraging independence of thought, telling the NHS organisations that they should ‘look out not up’.”

Then, there was the general difficulty of trying to implement systems in a standardised fashion across very different organisations within this changing system – a mistake he says many private sector organisations have made as well.

“The very process of implementing an IT system changes people’s expectations of what they want,” he argues. “Look at the work of academics like Nandish Patel at Brunel University. He talks about the concept of ‘deferred design decisions’ – allowing IT system specifications to evolve with the organisational systems that they aim to support.”

Swindells says the problem was not as simple as a lack of clinical engagement, but that there was a failure to set and manage expectations. “It was a mistake to think you can capture all the requirements of the NHS, put them in a document, go and buy against it and deliver within five years,” he says.

Another major contributing factor was the shocking state of NHS IT. “I don’t think there was a comprehension of how poor the state of NHS IT was. The absence of useful software, resilient networks and secure working practices meant that it was a long journey to the start line.”

The view from inside the NHS

Nicholson commissioned Swindells to carry out a review of NHS informatics with the long-drawn out departure of director general of NHS IT Richard Granger.

He says the purpose of the review was to determine whether the programme could deliver; and by when. As part of the work, he drew up a new structure and governance arrangements for NPfIT. “I drew up the new NHS chief information officer role and was then asked to step into it,” he says.

However, Swindells left for Tribal before the review reported. He says he left when he did to avoid any conflict of interest: “Because I’d taken on the role with Tribal, I was no longer able to deal with contractors for the DH.” But he acknowledges: “My strong preference would have been to stay longer and publish the informatics review before I left.”

In July 2008, a Health Informatics Review was published and positioned as supporting Lord Darzi’s Next Stage Review of the NHS. This argued that while the ‘overall direction of travel’ for the national programme was still to implement ‘strategic’ electronic patient record systems, ‘interim solutions’ would be needed.

In particular, it put a new focus on getting the Clinical 5 – a patient administration system with ‘sophisticated’ reporting, order communications and diagnostics reporting, letters with coding, scheduling for beds, tests and theatres, and e-prescribing – into hospitals.

Swindells says that what was published was a shortened version of his review, but still his work. He feels that the emphasis on the Clinical 5 was particularly important because there were “too many clinical areas that had been waiting for new systems for too long.”

And the view from outside

He also argues that when he left for the private sector in spring 2008 things were looking good for the national programme.

“I thought the programme was on a roll when Barts went live [with Cerner Millennium] and the Royal Free was next on the stocks, when we had the first release of Lorenzo Version 1, and a pipeline of implementations in place,” he says. “But the problems at the Royal Free in particular set the programme back really badly.”

He now worries that compromises made during the constant renegotiation of NPfIT contracts “are resulting in the demands of local service providers being downgraded.”

“We now have cliff edge between LSPs, how does that fit with the original vision?” he asks, adding there needs to be a renewed focus on integration. “Patient flows should not be constrained by the LSP contracts. Enabling patient flows is absolutely vital to achieving the vision of a 21st Century healthcare system.”

He also says it is vital for the programme to deliver to hospitals, saying he is not sure that the changes that have been made “get the programme into a mature roll out”. “Can we do Imperial and North Bristol?” he muses. “Only when we have shown we can do a really major implementation will we back on track.”

Looking to the future, he says the agenda will eventually shift away from the technology to how it is used. “I think the informatics agenda will come to the fore and we’ll see a big opening up of aggregated organisational data. Information saves lives, not the computer system.”

As the financial climate for the NHS gets ever colder, he says NHS informatics as a whole must “step up to the plate” and help NHS trusts and commissioners achieve the transformation of services they need to deliver.