The NHS is to be wired on an epic scale - though experience shows the larger the project and tighter the deadline the more likely an IT project will fail. But it's not all gloom for the government's ambitions, reports Michael Cross
The government's latest plan for health service reform is based on an extraordinary gamble - that the NHS can computerise its activities on an unprecedented scale.
On April 1, 2003, work will begin on a parallel set of national projects that will remodel healthcare around the needs of patients, not institutions.
Two years later, if all goes well, hospital appointments will be booked by computer, as airline reservations are today. Each patient's health records will be available on screen wherever they are needed and GPs will order prescriptions electronically from pharmacies rather than writing out illegible scrips.
Increasingly, patients will seek medical advice via the internet or digital TV channels, rather than going to their GPs in person. Similar "telemedicine" technology will allow GPs and nurses to seek advice from specialists anywhere in the country - or the world.
For the first time, researchers, auditors and budget-holders will have immediate and accurate information about which treatments (and doctors) are harming patients. Last but not least, the system will for the first time give the government and citizens accurate information about how the NHS is performing. "For the first time," according to the top-level strategy document Delivering the NHS Plan, "citizens will have independently validated information of how their money has been spent on healthcare in their own area and what progress has been made."
The snag is that these developments, together with other crucial elements in the NHS Plan, depend on a huge investment in information technology. Their estimated bill is £5 billion - on top of the £1 billion that the NHS already spends each year on IT. Deadlines are fearsome: most of the new systems are due to be in operation by December 2005, which assumes that development, procurement and installation will all happen in record time. If history is any guide, the whole project is a set-piece for disaster.
Repeated IT failures in the public sector have shown that the larger a project and the tighter the deadline, the more likely it is to go wrong. History also suggests that healthcare is a particularly difficult service to computerise - not just in the UK but around the world.
The NHS has had two previous IT master plans over the past 10 years: one in 1992, the other in 1998. Although neither was anything like as ambitious as the present scheme, called the National Strategic Programme, both were notorious for missed targets. Projects such as a new NHS identity number for patients, and electronic patient records in hospitals ran years behind schedule.
Anyone who has spent time in a GP waiting room or a casualty department knows the consequences of the poor state of NHS IT. At best, this means being asked over and over again for details like name and address, and being kept waiting for hours for a bed to be allocated or a discharge letter to be produced.
At worst, poor information-handling can kill. In late 2001 a kidney transplant operation at a Glasgow hospital failed because the recipient's body rejected an organ that should have matched. Although the organ itself was healthy, it had to be urgently removed and thrown away. The mistake was traced to a simple error made when a secretary typed data from one NHS computer system into another incompatible system. Every day, the NHS makes mistakes because its information systems are not properly integrated.
Meanwhile, the Audit Commission reported earlier this year that nearly half of hospitals in England and Wales are unable to collect enough data to know if they are meeting government standards on waiting times.
Medical records are a particular scandal. Today, most patient records are kept on paper, usually stuffed into "Lloyd George envelopes" (originally introduced to serve the Liberal government's National Insurance reforms).
Paper medical records have three basic problems.
They are incomplete: each NHS organisation records only its dealings with the patient. They are inaccurate: records are usually completed after the patient has been treated, often in a hurry, by low-paid and overworked staff who have other priorities. They are usually in the wrong place. Every day, hospitals cancel appointments and delay the discharge of patients because records can't be found. At best the folder is sitting on a trolley in the wrong department. At worst, it's in the boot of a car that's been stolen from the consultants' car park.
In an age when the internet, text messaging and digital photography are everyday consumer items, why has healthcare proved so resistant to IT?
Don't necessarily blame the NHS. Despite a run of disasters in the early 1990s, such as the waste of £43 million by the former Wessex Regional Health Authority and the failure of London Ambulance Service's computerised despatch system, the NHS has done reasonably well by international standards. A GP in the UK is much more likely than his counterpart in the US and many continental European countries to work with a computer.
There are several good reasons why healthcare everywhere has proved more difficult to computerise than, for example, banking.
Firstly the subtle language of medical professionals is more difficult to capture digitally than the language of finance or statistics, which is mainly numbers. Medical records contain much "free text" - usually handwritten today - as well as images of X-rays and ultrasound scans.
Secondly most of the crucial processes of healthcare do not take place at a desk, but in a casualty cubicle, operating theatre or an accident scene. These are not convenient places for computer terminals, which means that clinicians update their patients' records as an afterthought and there is little incentive for accuracy.
Finally there are cultural issues. Easily accessible electronic medical records raise issues of confidentiality. Some doctors are uneasy about their patients having access to online medical information through services such as Medline, once the preserve of professionals. Most GPs are already familiar with patients walking into the surgery armed with the latest journal articles about their condition, downloaded from the internet.
Inter-professional relationships must also be considered. Booking a hospital appointment electronically, for example, is a more subtle process than booking an airline seat. Technically, when GPs refer a patient to a hospital specialist they don't order an appointment, they "request" one.
Of course the NHS, as the world's largest integrated healthcare system, adds a few complexities of its own. Firstly its very size: at the latest count, the NHS had 1.2 million employees. Secondly, a legacy of fragmentation. Over the past 10 years, most decision-making on IT has been left to individual hospital trusts and even GP surgeries, creating vast differences in progress. At least two dozen different hospital information systems are in use across the NHS; probably no two are identical. When a hospital decides to upgrade its IT system, the procurement can take two years or more to get through the necessary bureaucracy.
There is a legacy of under-funding - although the NHS's total IT bill is considerable, the amount spent per employee or as a percentage of total budget is much lower than in private industry.
Nonetheless the government seems confident that the National Strategic Programme, published in June, will confound the sceptics.
Ministers - and the IT industry - say there are several reasons for optimism. These include:
Top-level commitment. For years the IT portfolio was a booby prize to be passed to the most junior health minister, whose main hope was to avoid disaster. Now it has the Prime Minister's personal blessing, the whole atmosphere has changed. Top level careers - perhaps the next general election - depend on it being made to work. Real money is available. Today the NHS in England spends about £1.1 billion on ICT (information and communications technology). The Wanless report into the NHS's future funding suggests that this doubles next year (2003-4) and peaks at £2.7 billion in 2007-8. More importantly the money will be ring-fenced for IT - and not diverted into more pressing priorities such as waiting-list initiatives. Technologies have advanced. A large number of the NHS's IT problems arise because its computer systems cannot exchange information. The technology that made the internet possible - so-called "markup language" - removes this problem. Other developments include handheld computers that can be used at the "point of care", communicating with the central system through secure coded wireless links.
Of course other challenges remain. Ministers admit that there is a conflict between the centralisation implied by the IT programme and the "front-line freedom" being offered elsewhere. NHS organisations will lose their freedom to pick their own IT systems, instead choosing from a limited number of products delivered by a handful companies called "prime service providers" (PSPs). IT companies will fight bitterly to become PSPs. It is not yet clear what, or how, they will deliver. But the real challenge is cultural change, by doctors, nurses, health managers - and patients. The NHS must win the public's trust in its ability to handle sensitive personal data securely. Everyone must learn to cope with the issues raised when medical knowledge comes under the glare of information transparency, rather than being buried in racks of dusty and illegible files.
