NAO logoJon Hoeksma

Rarely can the news that a project is two years late and will cost double the £6.2bn price originally quoted provide such a welcome read as Friday’s National Audit Office (NAO) report into NHS National Programme for IT was for NHS Connecting for Health (CfH).

The tone is almost relentlessly upbeat, apart from the escalation of the total programme costs and delays to the national summary record — but this is blamed on recalcitrant clinicians yet to see the folly of their ways.

Instead of being the warts and all account many had expected, this report, two years in the making, offers an upbeat assessment of the programme in its early years.

It could barely have made better reading for CfH had it authored the document itself. At times it reads almost as though someone at the NAO accidentally hit “accept all track changes” on a draft back from CfH.

Sources suggest that the NAO was ground down in a war of attrition with CfH, who fought a dogged rearguard action to keep out criticisms it found unpalatable or unacceptable.

At the DH press conference Richard Granger said he had only been personally involved for the past two weeks. But his office and organisation have been working with NAO far longer.

Though there are caveats and provisos made by the NAO, they are the exception rather than the rule: "We have examined progress to date in these areas and may return to carry out a further examination at a later date should this prove necessary," says the report in a rare flash of steel.

The NAO acknowledges that the birth of the report was gruelling, with the document going through "an exhaustive process". A spokesperson told EHI: "It’s not our customary post-project review".

Out of date?

Interestingly, the extent of delays to the summary record have grown even since the publication of the report. Speaking the day before the publication Lord Warner told the BMA’s Local Medical Committees conference that trials would now start in 2007 with full implementation unlikely to begin until 2008.

However, at last Friday’s press conference Lord Warner offered an assurance that the full NHS CRS, consisting of both a local detailed and national summary record, would be fully implemented by 2010.

Even the news that the cost of the project has mushroomed to £12.4bn wasn’t as bad as the figure of £20bn Lord Warner had volunteered in interviews just weeks earlier. Similarly the news of a two-year delay to the national summary record was also bad, but not nearly as bad as the delays to the local electronic record systems the NAO declined to mention.

The NAO report also fully details the extent to which the National Programme for IT (NPfIT) has underspent. Against an expected spend against contracts of £1,448m by the end of March 2006, the programme had actually spent £654m. The report makes clear that the taxpayer has not had to pay software suppliers for systems that don’t work.

The report says this “reflects the slower than planned delivery of some systems and the successful operation of contractual provisions”. So while contractors are quite rightly not getting paid for systems they have yet to deliver, NHS trusts have also not been getting the systems they need to deliver better care to patients.

However, the implementation numbers quoted in the report only paint a partial picture. The figure of 9,600 is undeniably impressive, but not all implementations are equal. Strip out the figures for broadband connections, software upgrades for Choose and Book and electronic prescriptions or even websites like Map of Medicine and the numbers look much thinner.

Even the claim of 36 patient administration system (PAS) implementations quoted doesn’t tell the full story. A PAS implementation in an acute trust, migrating up to 20 years of data, is a huge undertaking, while a PCT getting a PAS for the first time should be a smaller task. Also, would NHS trusts left to their own devices or working in the kind of consortia formed before the national programme have made more progress on electronic records by now? The NAO doesn’t speculate.

Last Friday, Richard Granger said he was surprised a programme to deliver systems to improve quality of care and save lives should prove so controversial. This seems rather disingenuous as it equates criticism of NPfIT’s delivery with opposition to modernising NHS IT.

And as the NAO report makes clear, once upgrades and infrastructure are removed from the totals, relatively few clinical systems have yet been delivered.

Strategy undelivered

The central strategic vision, against which contracts were awarded, of fully integrated clinical solutions replacing all existing hospital, community and GP systems remains unproven and undelivered at this stage. The NAO is reticent on this point, largely leaving CfH to set out its position.

Instead the NAO report more narrowly focuses on the early years of the programme and its procurement phase. It largely avoids asking why delays have occurred and whether the programme is unlikely to be delivered in its original form.

Only cursory attention is given to why the local and national components of the NHS Care Records Service (CRS) are over two years late, and whether there now exists a credible delivery schedule.

The lack of focus on the NHS CRS is surprising as the opening paragraph of the report states: "The core of the programme will be the NHS Care Records Service, which will make relevant pats of a patient’s clinical record available to whoever needs to care for the patient.”

A further curiosity of the report, on which CfH might have been expected to set the NAO straight, is the apparent confusion between the detailed local electronic clinical records to be delivered by LSPs, and the thin summary national electronic record that forms part of the spine.

Throughout the report refers to just “the clinical record”, failing to specify which component of NHS CRS it means. There is a suspicion that this is a deliberate obfuscation fostered by CfH due to its continuing problems in delivering the central pillar of the programme.

But, in a firm tone rarely seen elsewhere in the report and despite skipping on the delivery it emphasises that NHS CRS remains the central objective of the programme and the deliverables should not be abandoned or watered down.

In its conclusions and recommendations the NAO says: “Whilst some adjustment of suppliers’ milestones for the delivery of functionality may be a necessary, pragmatic response to suppliers’ difficulties in delivering, it should not compromise the eventual achievement of the vision of the fully integrated care record service that was the objective of the programme at its inception.”

Coverage of delays

"We didn’t look at it. You’ve got to stop somewhere and that’s where we stopped"

— NAO spokesperson on the report’s lack of comment on LSP solution delivery

In the section of the report dealing with progress so far the NAO very clearly sets out the extent to which LSPs and their software suppliers have failed to meet contractual commitments on delivering electronic records.

Turning to the initial release strategy contained in contracts the NAO notes that Phase 1 was a contractual requirement in all LSP contracts to deliver a range of advanced functionality by June 2005.

“It includes full Choose and Book functionality; outpatient clinic letters; inpatient discharge summaries; report of the single assessment for elderly people; diagnostic imaging and pathology results; screening results; recording of care episode events; and routing of orders for some blood tests and diagnostic images.”

In the event, only Choose and Book has been partially delivered, with all the rich clinical functionality still being awaited. Where hospital trusts have received new systems they have either been replacement PAS or stand alone departmental systems.

Phase 2, ‘tentatively timetabled’ for completion by June 2005 and roll-out by December 2006, was meant to add “National Service Frameworks assessment and review record; secondary uses of Spine data; planning and recoding of the total care journey – integrated care pathways; full linking and electronic transfer or correspondence; pathology and image order and result; and integration of dental services”.

Again none of these have yet been delivered to the NHS. Phase 3, development of which was meant to be complete by June 2006, is also missing without trace.

CfH says the delays were caused by “difficulties by some suppliers in meeting agreed delivery dates, and requests from suppliers to pilot more extensively the clinical record before full scale implementation.”

Equivalent functionality will now be delivered in a different order, says CfH, and stresses that it is now delivering things not originally envisaged in areas such as PACS. Surprisingly, though, absolutely no revised dates or timetable are set out to replace the defunct release timetable against which LSP contracts were awarded.

Asked why the NAO didn’t comment on delays to the detailed clinical system that LSPs are meant to deliver a spokesperson said: “We didn’t look at it. You’ve got to stop somewhere and that’s where we stopped.”

Separate to the LSP-delivered part of NHS CRS BT was to have delivered the Personal Spine Information Service – better known as the summary spine record by December 2004; the project is now running two and a half years late. The NAO reports CfH as stating that the development of the clinical record is “now expected to commence in pilot form in late 2006, and in its full form from late 2007”.

No speculation

So is NPfIT then on track to deliver? The NAO offers no predictions, confining itself to considerations of value for money and procurement approach. “We don’t make any comment or speculate,” said NAO assistant auditor general Anna Simons. “The proof is in the pudding.”

Intriguingly though, at the NAO press conference last week Simons pointed out to EHI that last month’s NAO report into the termination of the PFI contract at the National Physical Laboratory showed that it was perfectly possible to have a successful, well run procurement and tight contract that ultimately failed to deliver. In the case of NPL, it took six years between signing and termination of the contract, resulting in huge costs and delays in building new laboratories.

Clinicians of course have long had a darker way of referring to the same phenomenon: “The operation was a complete success, unfortunately the patient died.”


National Audit Office

National Audit Office Report: The National Programme for IT in the NHS (June 2006) [PDF, 757K]