Following the second battle of El Alamein in 1942 Britain’s wartime Prime Minister Winston Churchill said: "Now this is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning."

Could the same now be said of the NHS IT programme? With the news that Richard Granger is to quit as head of Connecting for Health (CfH) there is a case for arguing that, five years after it was launched, we have reached the end of the beginning of the £12.4bn programme to modernise NHS computer systems.

Despite the undoubted problems the programme faces, there are significant successes that have been achieved that can now be built on. And with the National Programme for IT (NPfIT) local ownership programme (NLOP), the steady shift to a best-of -breed systems rather than single solutions, together with increased freedom of foundation trusts there is an alternative, more flexible approach, beginning to take shape which potentially offers a better prospect of success.

Any appraisal of NPfIT must take the long view of what it has and hasn’t been achieved. At the macro level its undoubted success has been to get very significant investment committed from the government; to negotiate contracts that mitigate some of the risks of non-delivery; to get IT on the agenda of politicians and NHS executives as an item critical to the future delivery of healthcare; and to successfully deliver a world-class national IT infrastructure for the NHS.

As an infrastructure project – refreshing and delivering the network and network services – NPfIT has been a success, though one that you sometimes have to leave the country to appreciate. At an April European e-health conference in Berlin a panel of executives leading health IT programmes in a dozen European countries were asked to identify their top priorities to achieve interoperable e-health systems, almost every one put a developing a master patient index as top of their list of priorities. Thanks to NPfIT the English NHS isn’t planning how it will eventually hook systems up to a national MPI, it is doing it now.

Similarly many other European countries are devising or piloting plans for an emergency care record. Although implementation has only just begun, NPfIT has started to roll-out of a national Summary Care Record system. Another big success attracting international attention is the QMAS system that links payments to GPs to the quality of care they provide.

Other notable achievements chalked up include plugging away at the delivery of Choose and Book; making headway on Electronic Transfers of Prescriptions – after first scrapping existing pilots; and getting the first stages of GP-to-GP record transfers working. Of course many of these developments were underway before the advent of NPfIT but they have been scaled up to programmes that can eventually be delivered nationally rather than stuck in pilot limbo.

In terms of clinical systems delivered the headline grabber has been PACS (Picture Archiving and Communications Systems), which will be 100% delivered by 2008. But there have also been notable successes in the delivery of community-based systems, including almost 600 implementations of SystmOne from TPP, software to support the single assessment process from Liquidlogic and RiO community and mental health systems from CSE Servelec. Over 50 PCTs and mental health trusts have also been provided the iPM PAS system by Computer Sciences Corporation (CSC).

Had the NPfIT programme set out to just deliver the IT described so far it could be judged a resounding success. The reality, though, is that its central purpose, and the focus of contracts awarded to local service providers (LSPs), was to deliver standardised, integrated shared clinical record systems by 2010 – and against this core requirement the programme has spectacularly failed to deliver.

Granger has stuck steadfastly to the doctrine that that the only way to deliver Integrated Care Record Systems – now termed Detailed Care Records – was to award billion pound contracts to international IT services and consultancy firms and to ‘drain the swamp’, replacing the mosaic of existing suppliers, replacing them with a handful of suppliers. This ‘route one’ approach of ruthless standardisation was to be imposed on the highly federated NHS for its own good, with suppliers paid only on delivery.

However, in the four years since contracts were awarded the LSP prime contractors and the software suppliers they were matched with have so far been unable to deliver anything approaching the integrated care records their contracts required them to provide. Instead there has been a litany of delay, confusion, waste and a steady scaling back of clinical functionality. Rather than deliver electronic patient records the programme has yet to get beyond delivering replacement patient administration systems (PAS) which for many trusts represented a backwards step from what they already had in place. Unless they have a pressing need many trusts have decided to sit tight.

Key suppliers have also departed: GE decided it wasn’t worth the candle after buying IDX, while LSP Accenture exited to stem its mounting losses. And neither of the two main software suppliers – Cerner and iSoft – has yet been able to deliver the promised next generation patient record systems tailored to meet fast changing NHS requirements. Barely 20% of acute NHS trusts have yet received even an ‘interim’ PAS system, and some of these have found they must introduce a variety of work arounds, paper systems and kludges.

In April the year House of Commons Public Accounts Committee concluded: "The Department [of Health] is unlikely to complete the programme anywhere near its original schedule". Four years in they said there was uncertainty "about the costs of the programme for the local NHS and the value of the benefits it would achieve".

The report said that suppliers were "struggling to deliver and noted: "At the current rate of progress it is unlikely that significant clinical benefit will be delivered by the end of the contract period." Industry analyst, Murray Bywater, director of Silicon Bridge Research says that for acute trusts things have barely moved on since Information for Health in 1997, and that in terms of delivering electronic patient record and order communications systems it has been "a lost decade".


Against this backdrop there has been simmering discontent from some NHS trusts, variously frustrated at NPfIT’s delays or facing urgent business and clinical requirements. With Granger at the helm of CfH the Department of Health has largely been able been able to keep the lid clamped down. Bradford was persuaded to rejoin the national programme after initially deciding it wasn’t prepared to take the iSoft software offered, though Northumbria and North Staffordshire both rejected the NPfIT systems offered by CSC. However, recently there has been far more public discontent including a letter by 79 staff at Milton Keynes deploring the Cerner Millennium system implemented by Fujitsu.

Recognition that the dogmatic ‘route one’ approach hasn’t worked contributed to NHS chief executive David Nicholson’s initiative to introduce the NLOP. Although at an early stage, there are indications this may prove to be letting the genie out of the bottle – a nationally determined and specified programme will be extremely difficult to square with allowing local priorities to be set. Last week Claire Mitchell, NPfIT programme director characterised it as a "healthy tension".

Much still depends on the extent to which the Additional Services and Suppliers Catalogue and GP Systems of Choice procurements actually allow local choice to become part of the programme. The big problem is that these two ‘choice’ initiatives remain extremely difficult to reconcile with the ruthless standardisation of the LSP contracts. Unless this can be resolved there is a danger that the LSP contracts awarded by NPfIT may be prove a straightjacket hindering NHS modernisation.

At least one of the LSPs, BT in London, is now openly pursuing a ‘best of breed’ approach to systems – using INPS for primary care, Cerner for acute and CSE Servelec for community and mental health. Fujitsu, the LSP in the South, now appears to be moving in a similar direction.

Although the UK supplier industry outside NPfIT has taken a severe pounding, those left standing have much to offer. Some have even emerged stronger. The NHS IT programme under Granger acted like a meteorite landing in a village pond. For better, for worse things will never be the same again. For ‘existing suppliers’ it has acted as a catalyst for them to innovate. In many cases they appear to respond to changing NHS requirements much more rapidly than the big players. In areas such as 18-week waits or patient access to records, for example, it is the ‘existing suppliers’ that have led the way.

Above all, if the NHS IT programme is to have a future that will deliver benefit to the health service it must extend its ‘market share’ and become relevant for the majority of NHS trusts and PCTs. To date it has delivered core systems to less than 20% of acute trusts and 10% of GP practices.

Adopting a big tent approach, based on stringent standards and use of the national infrastructure now in place, offers a real opportunity to cut the current Gordian contractual knot. To achieve this, however, continued drive and new pragmatic leadership will be required from someone not ideologically wedded to the approach pursued the past five years.