Parliamentary IT adviser Tom Brooks argues that NAO report overlooks crucial issues
Enough evidence now exists to make everyone aware that there are serious doubts as to whether the National Programme for IT will ever be able to provide the necessary IT capability to underpin the modernisation of the NHS.
Everyone, that is, except for ministers, who seem in a state of denial, and the National Audit Office, which appears overwhelmed by the task of peering into the deepest recesses of NHS Connecting for Health, examining the evidence and interpreting the facts.
The NAO's report on the National Programme for IT in the NHS, said, "NHS Connecting for Health has taken positive action to ensure the contractors are managing their tasks well."
How can that possibly be true when all the main contract deliveries are running so late?
The NAO was taken in by the Department of Health's propaganda, congratulating it for the procurement of contracts being "completed commendably quickly". If that was so, how then does the NAO explain why the deliverables are appearing so slowly?
The NAO report virtually salivated with excitement about the speed of procurement, blithely recommending that other organisations learn lessons from the experience of Connecting for Health, in particular the advantages supposedly gained through the swift procurement exercise.
Yet the same NAO report pointed out that the delivery details in the national data spine contract had to be reorganised and replaced the following year. The core care records element of the Accenture contract had been revised into four releases, the last of which was 13 months later than the original target date. Customers of CSC fared even worse, with the contract rescheduled into five releases, the last of which is now due 22 months later than originally planned.
Despite being the main contractor for the data spine, BT was 14 months later than contracted in deploying a London system to operate with it. Fujitsu, having been pushed into a "common solution project" with BT by Connecting for Health, faced the embarrassment of having to change its principal subcontractor just over a year into the contract.
The reality of Connecting for Health's "swift procurements" was a record of delays and timescale renegotiation before the programme really got started. It is hardly convincing evidence of success.
So what confused the NAO into believing the procurement was sound? Should it not have spent more time exploring the root cause of these contractual failings? Did the NAO go back to the original swiftly drawn-up contracts to examine their quality? If it had done so, it might have noticed that the contracts postponed agreeing a great deal of the detail until after the contracts were signed.
According to supplier sources, documents in various contracts had to be produced after the contracts were signed. The missing details included service level specifications, helpdesk interworking procedures, annual implementation plans, component system descriptions, quality plans, disaster recovery plans, module testing plans and specifications - the list is interminable.
The NAO does not appear to have uncovered the extent of the contractual holes that existed at contract signature time, nor examined how much the absence in early 2004 of such important detail from the contracts led to the subsequent rescheduling and delays.
In May 2004, Computer Weekly revealed that only five months after the deal had been signed it had run into contractual issues. A leaked BT document revealed that the issues the supplier was facing arose from "detailed definition of requirements and practical deployment not envisaged at the effective date of the agreement". Was it a case of sign in haste, correct at the contractor's pleasure?
The NAO has not revealed how these differences were resolved, nor at what price to the taxpayer. Subsequently, nearly all the main contracts were redrawn to adopt a much slower product release plan than the one originally envisaged. Whatever time was supposedly gained in the procurement phase has certainly been lost in the implementation. How on earth did the NAO miss this crucial point?
The NAO claims that although there have been delays in delivering the NHS care record service, suppliers - not the taxpayer - have borne the cost of overcoming the difficulties in delivering the software.
But elsewhere in its report the NAO recorded that the dissolution of the contractual commitment for the common solution project between BT and Fujitsu had added £62m to the original £934m cost of the Fujitsu contract. Who is paying this £62m if not the taxpayer?
Before Richard Granger's arrival as chief executive of Connecting for Health and director general of NHS IT, a well-constructed consultation exercise took place to define what the NHS needed from the care records service. The NAO report said that the final output-based specification had been issued to prospective suppliers early in the contract negotiations.
But soon after contracts were awarded, the requirements were redefined into "bundles". Connecting for Health broke the links with the output-based specification requirements when it defined different bundles for the northern iSoft and southern IDX solutions. Bundles were then allocated to releases and phases, which have been redefined several times.
Confused? So, apparently, was the NAO. It recommended, "While some adjustment of suppliers' milestones for the delivery of functionality may be a necessary pragmatic response to suppliers' difficulties in delivering, it should not allow this to compromise the eventual achievement of the vision of the fully integrated care record service that was the objective of the programme at its inception."
Does this mean that timescale slippages are OK as long as there is no admission that the end date will be missed?
There is no longer a published specification of what suppliers are contracted to deliver. Connecting for Health has never published content specifications for each software release and phase that is remotely equivalent in detail to that in the output-based specification. Yet if, as the NAO report recorded, "The plan remains for the entire implementation to be completed by 2010 in accordance with originally contracted timescales," a detailed specification of each phase must surely exist.
How likely is it that this NHS IT implementation will succeed and the 2010 target be met? The NAO seems too confused to express an opinion. On the one hand, its report commended "the notable progress and tight control of the central aspects of the programme". On the other it nervously noted that "successful implementation of the programme nevertheless continues to present significant challenges".
There is no guidance in this report for MPs concerned that billions of pounds of NHS money could be wasted. The need for parliament, independent of the Department of Health, to commission an open, transparent study into the state of the National Programme for IT is greater than ever.
Whitehall remains firmly in denial that a problem exists. Ministers are badly misguided. The NHS is an institution revered by most of the population. It deserves all the support it can get to deliver the 21st century healthcare that patients require. The NAO report does not help it achieve that goal.
● Tom Brooks is a long-standing member of the Parliamentary IT Committee, who has for several years helped MPs with their scrutiny of NHS IT development
Vote for your IT greats
Who have been the most influential people in IT in the past 40 years? The greatest organisations? The best hardware and software technologies? As part of Computer Weekly’s 40th anniversary celebrations, we are asking our readers who and what has really made a difference?
Vote now at: www.computerweekly.com/ITgreats