Simple Architectures for Complex Enterprises: A Case Study in Complexity

Current Status of NPfIT

NPfIT has been in crisis almost from the first day. By mid-2004 (barely a year into the contract), both Fujitsu and BT, two of the five primary regional vendors, were having trouble with their IDX (regional CIS vendor) relationships, and this trouble never let up. According to a confidential draft audit by the National Audit Office (NAO, a British government audit office),

• By mid-2004 NHS Connecting for Health was concerned about the effectiveness of supplier management of both BT and Fujitsu, and the performance of IDX... However, by April 2005, even though NHS Connecting for Health [the British government bureau responsible for NPfIT] had been applying increasing pressure, working with the prime contractors, to encourage IDX to match its planned deliveries, insufficient progress had been demonstrated and Fujitsu lost confidence in IDX’s ability to deliver the Common Solution project.

The CSC/iSOFT partnership was faring little better. According to that same confidential NAO audit,

• CSC, the Local Service Provider for the North West Cluster, agreed to a remediation plan with NHS Connecting for Health for the delivery of Phase 1 Release 1 as it was having problems meeting the original target dates... Further delays led to a second remediation plan which pushed the deployment dates for two elements of Phase 1 Release 1 further back into 2006, some 19 to 22 months later than originally planned.

But of all the partnerships, the one that probably fared the worst was the Accenture/iSOFT partnership. By September 2006, Accenture had decided that the pain associated with this project was not worth it, and abandoned the project altogether. According to a baseline case study³⁶ in so doing it walked away from almost $4 billion in revenue writing off $500 million it had already spent, and agreeing to pay $100 million “to settle its legal obligations.”

iSOFT was involved in three of the four partnerships, and the strain on that company might bankrupt it or, at the very least, force its sale. According to its financial results released in December 2006, the company took an almost $800-million loss for the fiscal year ending in April 2006—a huge loss for a company that had total revenues of the year of only $340 million.³⁷

As you can see, every major company involved in the regional clusters has taken a severe financial hit from NPfIT. It seemed that everybody underestimated the complexity of this project. The costs of this underestimation will likely be measured in the tens of billions of dollars.

In the area of user confidence, NPfIT is in serious trouble. There are three critical constituencies that have been alienated by NHS’s approach to NPfIT: health care workers, patients, and IT professionals.

A good indication of how the health care professionals feel about NPfIT is found in a recent editorial of the British Journal of General Practice (May 2005). It says,

• The impact on patients and professionals has yet to be seriously addressed. A very different approach is needed to nurture culture change... The £30 billion question is not just whether NPfIT will get the technology right but whether it can also win the hearts and minds of the people on whom the NHS depends every day.

Patients are also unhappy about NPfIT, even at this early stage of the project. Most of the patient concerns are directed at the ability of NPfIT to protect records. This distrust is illustrated by a Web site, http://www.TheBigOptOut.org, which states,

• This system is designed to be a huge national database of patient medical records and personal information (sometimes referred to as the NHS ‘spine’) with no opt-out mechanism for patients at all. It is being rolled out during 2007, and is objectionable for many of the same reasons as the government’s proposed ID database... You will no longer be able to attend any Sexual Health or GUM (Genito-Urinary Medicine) Clinic anonymously as all these details will also be held on this national database, alongside your medical records. For the first time everyone’s most up-to-date and confidential details are to be held on one massive database.

But of all the constituent groups that have expressed unhappiness with NPfIT, the most vocal by far has been the IT community. In January 2005, The British Computer Society (BCS) sent a position paper to the NAO describing a number of concerns with the NPfIT approach, including the following:

• Failure to communicate with health care users

• Monolithic approach

• Stifling of innovation among the health informatics market

• Lack of record confidentiality

• Quality of the shared data

In April 2006, 23 highly respected academicians sent an open letter to the Health Select Committee. In this letter, they made some harsh statements:

• Concrete, objective information about NPfIT’s progress is not available to external observers. Reliable sources within NPfIT have raised concerns about the technology itself. The National Audit Office report about NPfIT is delayed until this summer, at earliest; the report is not expected to address major technical issues. As computer scientists, engineers and informaticians, we question the wisdom of continuing NPfIT without an independent assessment of its basic technical viability.

In October 2006, this same group sent another open letter to the same committee:

• Since then [April] a steady stream of reports have increased our alarm about NPfIT. We support Connecting for Health in their commitment to ensure that the NHS has cost-effective, modern IT systems, and we strongly believe that an independent and constructive technical review in the form that we proposed is an essential step in helping the project to succeed... we believe that there is a compelling case for your committee to conduct an immediate Inquiry: to establish the scale of the risks facing NPfIT; to initiate the technical review; and to identify appropriate shorter-term measures to protect the program’s objectives.

The BCS offered to help the NHS with a review of the NPfIT architecture. What did NHS think of this generous offer? Not much. Lord Warner, head of the NHS responded forcefully:

• “I do not support the call by 23 academics to the House of Commons Health Select Committee to commission a review of the NPfIT’s technical architecture. I want the Program’s management and suppliers to concentrate on implementation, and not be diverted by attending to another review.”³⁸

Soon after, Lord Warner apparently had had enough. Like Accenture, he was bailing out. In December 2006, he announced his retirement from the NHS. He was followed in July 2007 by Richard Granger, Director General of IT for NHS, the man who was widely blamed for most, if not all, of NPfIT’s problems.

At this point, nobody knows what the eventual cost for NPfIT will be. Estimates range from $48 billion to $100 billion. It seems likely that the project will go down in history as the world’s most expensive IT failure.