Stephen O'Brien MP: The latest NHS IT shambles

Stephen O’Brien, MP for Eddisbury and Shadow Minister for Health, questions the future of the supercomputer programme in light of Fujitsu terminating its contract.

In September 2003, Patricia Hewitt said:

“The principle, the civil liberties and the practicalities of great big IT projects and databases have a horrible habit of going wrong.”

Yesterday it transpired that the Government are terminating the contract with Fujitsu to deliver the NHS IT programme to the South East and South West of England. It seems that in the Government’s attempt to ‘reset’ the contract (code for ‘pouring more taxpayers’ money in’), Fujitsu have decided that they cannot deliver what is asked at the price given to them, and they intend to cut their losses, after what will no doubt be an ugly legal scuffle.

The question before us is, does this put in jeopardy the whole ‘supercomputer’ programme - the core of which is to create a national electronic patient record database, to enable hospital appointments to be booked electronically, and prescriptions to be created and administered electronically - and how much of a problem is that?

Despite New Labour embracing an internal market in the NHS, and welcoming the private sector ‘at a local level’ (i.e. imposing it from above) after the (more) disastrous early years, the NHS IT programme remains a monolithic, soviet-style mechanisation project (which is why, no doubt, it has survived Mr Brown’s purges). What is more, getting to the truth of any matter is near impossible, as tractor production figures remain the order of the day. In fact, so obfuscating is the quango that runs it, ‘Connecting for Health’, that it was recently criticised in a Government sponsored report for a “defensive and controlling approach to the release of information”. Even the Labour-dominated Health Select Committee had to deal with “vague and inconsistent” officials when they reported on the electronic patient record.

So, for example, Connecting for Health routinely tells us that “x” million bookings have been made to date, whereas against the target that matters (i.e. 90% of GP referrals to first consultant-led outpatient services should be made through Choose and Book by 31 March 2007), the Government are running at 44%. 

Or on electronic prescriptions – the target is that 100% of all sites should be capable of issuing Electronic prescriptions by 31 December 2007:  the Government are doing better against this target: 64.1% of sites are able to issue Electronic prescriptions. However, EPS is being used for only 1% of daily prescription messages, and only about 0.05% of prescriptions are actually dispensed electronically (assuming 773 million prescription items per annum). Furthermore it is submitted in parallel with a paper prescription, neutralising any anti-bureaucratic benefits.

The Care Records Service itself was supposed to be, albeit tentatively, ready by December 2007. Under this system everyone in the country will have their demographic data held on a central database accessible across the NHS, in a ‘summary care record’. Those of us who do not opt out will also have details of allergies and prescriptions added, and whenever an individual is treated, that information will be added to the record with their consent. This will complement ‘detailed care records’ held locally by the providers who treat us. Aside from the security concerns, progress is slow - the first pilots only went live in March 2007, and we are still awaiting a timetable for full roll out. At March 2008 153,188 Summary Care Records had been uploaded to the Spine in the pilot areas, circa 0.24% of UK medical records.

Last week the National Audit Office confirmed that:

“Current indications are that it is likely to take some four years more than planned – until 2014-15 – before every NHS Trust has fully deployed the care records systems.”

Today’s exit by Fujitsu follows Accenture’s departure in September 2006, after they had written down $450m losses against the programme. Serious questions loom – with only two of the original four suppliers left in the market, will either take on Fujitsu’s contract at the original price? Furthermore, on the one hand the duopoly increases the negotiating power of the suppliers, and on the other if these suppliers fail, the Government will be forced to sacrifice either the project, or swathes of taxpayers’ money.  But that two suppliers have now quit brings in to question whether the programme can ever be delivered.

Make no mistake, Information Technology could be bringing huge benefits to patient care, frontline healthcare professionals and informing patient choice. Tony Blair was right when he said in 1998 that ‘The challenge for the NHS is to harness the information revolution and use it to benefit patients’. The mistake he made was to try and meet that challenge from his sofa, rather than allowing the hardworking staff and knowledgeable users of the NHS to solve it themselves. Information Technology should be used to meet local need, and tailored to local geo- and demo-graphic situations – the State should step back, support local innovation, and enable diverse systems to talk to one another, not build a one-size-fits-all structure from Whitehall – but with a ‘control freak’ Prime Minister, is that likely to happen?

Comments

[Eleanor McHugh]

These big government IT projects always seem to fall foul of the same basic problem: they envisage a single overarching system capable of meeting the needs of multiple levels of management whilst providing efficient service to millions of users.

Anyone who's ever designed software or IT infrastructure for a large corporate already knows some of the pain involved in such an undertaking, and a significant proportion of such commercial developments fail. But to amplify this process up to the scale envisaged in Whitehall departments - many of whom are insulated from both the real world and the world of technology - is a sure and certain recipe for disaster.

Instead of looking to the large corporates and trying to build everything in from the start, government should instead draw inspiration from the open-source community which has consistently created better solutions for large-scale problems than the commercial sector by allowing a free market in ideas.

Unfortunately the current state of affairs is typical of the dogmatic left-wing belief in central planning and monoculture which we all thought had been thoroughly discredited during the Thatcher years.

I design and implement software for a living and have worked on large infrastructure projects for a number European broadcasters as well as with UK internet and communications companies. My experience has been that the customer always knows what they want but very rarely knows what they need: not because they're stupid, but because they're not the intended user of the system.

I hope that the next Tory administration will work hard to encourage better use of information technology throughout government, the civil service and the public sector, but that it will do so with a far more laissez faire attitude.

Define clear standards for the interchange of information and let the free market satisfy those needs, paid for from local budgets by the people who will actually be using the software and hardware required to do the job.

Levels of acceptance would improve, the heterodox nature of the systems would make them more robust, and a thriving market for SMEs developing these solutions would ensure that costs were kept under control and quality continuously improved.

[Pisaboy]

Solid piece. Keep a close eye on them Stephen!

[Michelle Tempest]

It's another New Labour white elephant. They squandered the money on everything but patient care.

[David Parker]

It is difficult to find a single example of a NuLab IT initiative which has not been beset with major and continuing technical and budgetary problems.

Obviously, in any nationwide service there has to be some degree of centralisation, if only to ensure technical compatibility between the various installations of the system and the organisations using it. However,difficulties and disorganisation are are almost inevitable when the functions and applications of the system are specified not by those who will use it in practice, but by remote bureaucrats working to a political agenda.

Thus, with Choose & Book, to give just one example, call centres are often staffed with inexperienced or insufficiently trained employees, consequently patients are sometimes initially wrongly referred, with orthopaedic patients being sent to orthodontic, or even obstetric consultants!
But, as Stephen O'Brien points out, it is the long suffering clinical staff who are left to sort out these messes, and they are then blamed if this causes them to miss some totally artificial statistical target.

[snegchui]

Having worked with Consultants from Deloittes and Andersens, I can only say that whilst intellectually rigorous, they were like the Economist on the island with a tin, "Assume a tinopener". Wrong starting place, never mind the quality of your argument, you are not going to the right place. There were also masters at providing lots of documentation and drowning people in information and obscuring issues. Watching somebody costing £9000 pw to write manuals in the most obscurantist English was an interesting experience. I am sure there are places where Consultants are valuable, but this current obsession with their infallibility has to change, not only in the public sector who seem to think big salary = big brain and the private sector.
Cfh a quote "it also led to accusations of a project being imposed on the NHS with little input from the doctors, nurses and patients who would be using it." Well, ho ho , it is not fit for purpose, funny that. If Consultants were payed on long-term results and not short-term consultations things might change. "It was working when I left so my successors screwed it" vs "I had a right mess to clean up when I came in" do not constitute value for money for me. Like HP, like Dell, you need some miserable old bastards in who know trouble when they see it, know how incentives can perversely work and know how to say No to flashing lights and whistles that add no value. And if some gritty user champion had been employed to counteract the FUD of consultants masquerading as a way forward, you might have has something less ambitious, but which cost less and worked.
I agree with Elanor as well, that though the heterodoxy of systems sounds counter-intuitive, if you make your messaging infrastructures solid, it works, because people can do what they like internally on their departmental systems to suit themselves, and the cores systems only imposition is structured data at the point of export from one system to another, which is a lot easier than being made to alter your working practices to suit some interface designed for somebody doing a job different to yours.

[Teck Khong]

There are far more effective and patient-centred solutions than this ghastly, expensive and depersonalising system for storage and retrieval of medical information.

Intrusiveness of the nanny-state mentality is ultimately and ironically incompassionate and demeaning of individual rights.

As a patient, a doctor and a politician, I would call for a moratorium of the present fiasco and re-examine the case for a simpler and more cost-efficient yet ethical system.