President Obama vows to “address the crushing cost of health care” by, among other things, “invest[ing] in electronic health records and new technology that will reduce errors, bring down costs, ensure privacy, and save lives.” Sounds great. Let’s take a quick trip across the Atlantic to see how that’s worked out for the British.

In two words, not well. In 2002, Tony Blair ordered Britain’s National Health Service to participate in the National Programme for IT, and thus launched an ambitious attempt to computerize the entire Service. The initial projected cost: 2.3 billion pounds. It was widely described as the largest IT project in the history of the world.

Fast forward to late 2008. Costs had risen to a minimum of 12.7 billion pounds, and since the system was 4 years late – and counting – its final cost, actual full in-service date, and even whether it would ever work at all remained up in the air. The Financial Times described it as exhibiting an unhealthy level of ‘Great Leap Forward Syndrome.’

So what went wrong? Where to start. There was the crass malignancy of the government’s attitude to its suppliers. The official responsible for delivering the system compared the project to a sled team and failing suppliers to dogs who would be shot and fed to the survivors to motivate them. A year later, he abruptly quit the project after one key firm was accused of accounting irregularities and another major contractor dropped out and threatened to sue the government to escape from the affair.

Then there was the incompetence of the Department of Health. The 2007 ‘competence review’ of the Department by the Cabinet Office is richly amusing. Of the 12 assessed areas, the Department got a green light in zero of them. The “Base Choices on Evidence” area was described, for instance, as an “Urgent Development Area.”

In other words, the Department was pathetically bad at the elementary job of making decisions based on facts, and equally bad at learning from past errors. The House of Commons’ Public Accounts Committee produced a 180 page report on the system in April 2007 listing a similar avalanche of misjudgments: no delivery timetable, no cost controls, and no cost-benefit analysis.

Above all, there was the faith in the top down model, and the resulting failure to realize that creating a system linking all doctors, hospitals, and administrators was going to be immensely expensive and complicated, and that building the system had no necessary connection to solving substantive problems. The Public Accounts Committee’s report summed it up by noting that the entire project was afflicted with a worshipful attitude to IT, which led to the delusion that it would be easy to fix complex social problems and simultaneously save money by applying large amounts of centralized technology.

All of this led a senior executive at Fujitsu, which at the time had an 896 million pound contract for the system, to publicly compare it to a camel, a tightrope walker, and a sinking ship, and to sum up by saying:

It isn’t working, and it isn’t going to work. There is a belief that the national programme is somehow going to propel transformation in the NHS simply by delivering an IT system. Nothing could be further from the truth. A vacuum, a chasm, is opening up.

When Gordon Brown appeared before the Commons Liaison Committee two weeks ago, he was still defending the indefensible. Meanwhile, the chief executive of London’s Royal Free Hospital described the system’s trial deployment as “incredibly disappointing” and said it would cost the hospital an addition 10 million pounds a year to cope with it. In 2007, 91 percent of all doctors opposed it. A later survey found that most NHS hospitals were simply refusing to implement parts of the system. The Department of Health’s response was to impose massive fines on any hospital that refused to cooperate in full.

And then there’s the question of privacy, which Obama raised. Many individual doctors are refusing to use the system: as one put it, “Any lock that can be opened by 50,000 keys is not secure.” That’s a wise attitude: in late January, the government came forward with a proposal, hidden in an obscure corner of the Coroners and Justice Bill, to allow ministers to “draw up new information-sharing orders that would allow them to release private data – such as tax returns, personal details or medical records – to any public or private body.

Naturally, there will be safeguards. Naturally. But the only defense the government could make of the proposal was that it would “reduce the number of people who need to be notified of a death, thereby helping to relieve distress.” That’s right: a vast, comprehensive medical database is being compiled in Britain, with widespread information sharing by ministerial order, just so Britons will be spared the pain of notifying the gas company that their aunt has died. If you believe that, you probably believe Obama’s proposal will work too.