NPfIT ‘needs re-evaluating to be a success’

Connecting for Health must look at three key things – scalability, complexity and availability if the National Programme for Information Technology is to be a success, according to a panel of industry professionals.

Speaking at the Healthcare Computing 2007 conference in Harrogate, the panel said that the size and complexity of the NPfIT programme had been under-estimated, and counselled Connecting for Health (CfH), the agency responsible, to re-evaluate its methods.

The panel members provided a long view of the current NPfIT project, saying that it appeared to have been on a ten year cycle from the days of Information for Health in the late 90’s to the current day, citing the re-emergence of local information strategies under the NHS Local Ownership Programme.

They stressed though that the core aims of NHS IT policy has remained the same, to deliver a common nationwide record, which offers information at the point of care and at the point of need.

Summarising progress so far, industry veteran Phil Sissons of Magic Consulting said: “Not much has been delivered, not much cost to the government and none of the smart procurement that Bill Gates [Microsoft’s co-founder] proposed and Richard Granger [director of CfH] wanted.”

Among the many high risks identified, the panel said that the programme has suffered from providing strategic fixes to yesterday’s problems, ‘gold-plated’ solutions to trivial issues, creating unnecessary technical challenges and suffered from unintended political interference and moving goal posts.

Murray Bywater, director of Silicon Bridge Research highlighted the difficulties entailed in the unprecedented scale of the systems and data required: “Things can work well, but scaling up makes issues more difficult. No country has gone beyond creating patient records for four to five million people before; NPfIT will be creating records for 50 million people.

Bywater contrasted this with the scale of the US Veterans Administration system electronic record system, often cited as an exemplar: “The VA system in America was set up for seven to eight million people with 128 systems implemented, but when pressed, even they admit that they are implemented differently. As it stands NPfIT is a pretty tall order.”

Opening the panel session, a call was made by the chair of the panel, Bud Abbott of the British Computer Society and conference committee urged CfH to take note of an old saying – ‘”the existence of facts or statements about any situation is of no value if no-one perceives their significance.”

Nicholas Brown director of medical imaging consultancy MIMIC suggested that it was not viable to replace existing systems with new systems. Instead, interoperability with existing legacy systems should be looked at in the bid to create a single patient record.

“What is being proposed is a very excellent system for health informatics in the future, to replace existing systems with new systems, but it is now being seen that it is not viable to do this. Good use of old and new systems, and interoperability should be the main focus.”

Professor Colin Tully of Middlesex University and a member of the group of 23 academics who proposed a technical review of NPfIT, advocated an approach based on the creation of a "system of systems".

He explained: In a nutshell, it sets out to deal with the situation where the kinds of systems that we traditionally think we know how to develop and deploy need to be grouped into a "system of systems" (or supersystem); and where those individual systems, and the super-system, and the ecosystem within which they exist, have long life cycles during which they all co-evolve.

"They also all display emergent properties, which are important in influencing their evolutionary paths. The approach applies to large-scale inter-disciplinary government projects, developing multiple, heterogeneous, distributed systems, bringing together independent, disparate organisations and entities to achieve overall coherence.

"This approach is likely to be required when the complexity being faced breaks down the assumptions on which traditional systems engineering is based, such as requirements that are relatively stable and well-understood, a system configuration that can be controlled, and a small, easily discernible set of stakeholders."

The panel agreed that reflecting on past experiences and knowing what needs altering is something CfH should not back away from doing.

Professor Tully added: “How can observing yourself and your problems and seeing what works and what doesn’t be bad?”

Another observation made by the panel was that NPfIT has at times struggled to adapt to an evolving policy priorities, for example incorporating initiatives such as telehealth into its agenda.

Sissons warned: “Requirements have changed… but the national programme has failed abysmally to focus on areas like telehealth. The use of proper NPfIT devices seems to be like a phantom.

He continued: “It’s almost as if you create a blind man’s buff or pass the parcel, who is left holding the parcel? The LSPs face a very expensive and big challenge for the national programme.”

One audience member went as far as to say that there is no purpose to the national programme, the project seemed to be bigger than man going to the moon.

Concluding the session, the panellists felt that the key three issues raised by the current review of the programme by NHS chief executive, David Nicholson must be addressed if the national programme is to be a success: the need for NHS management and users to be properly re-engaged; the need to reduce the scope of NPfIT and there should be a change in CfH’s mentality.