03.11.10
The case for NPfIT 2.0
Experts agree that electronic health records are the key to developing safe, affordable 21st century health care. IT has delivered gains in industries such as finance, commerce and the media. There is widespread dismay and frustration at the slow rate of dissemination of IT in health care. NPfIT was the grand programme that would correct this, says Dr DerekMeyer
Massive IT programmes are notorious for going wrong. Public sector IT programmes are notorious for going wrong. A massive public sector IT programme? When the press is full of articles highlighting NPfIT failures and politicians call for the system to be scrapped it is all too easy to accept this stereotype.
NPfIT may have put in the foundation needed for effective, affordable 21 century health care.
The UK has traditionally adopted a 'close follower' IT strategy and quickly adopts US developments in new technology. While the US has undisputed world leadership in health care and in information technology, it still has a health system that relies mainly on pen-and-paper. Why is health care so behind? Why is health care informatics so much more difficult to implement than financial services or commerce?
In the US various factors have been identified. Their system is too fragmented and an encouragement of competition prevents the necessary co-operation needed for system integration. There are disincentives to IT investment because the costs are borne by practitioners but the benefits flow to insurance companies. The medico-legal barriers are too high.
In England, we have avoided all of these. We started with a unitary, single payer health system. The government provided strong support and leadership from Tony Blair down. The programme had generous funding, strong and dedicated management and was not sidetracked by special interests or mired in inter-departmental strife. Contracts were awarded to the world's leading IT companies, all of whom had a strong track record of successful project delivery.
So what happened?
Infrastructure, such as N3 and NHSMail was successfully deployed.
The strategy of centrally-procured local systems makes sense, as it’s cheaper to buy in bulk and standardised systems are cheaper to maintain. The strategy was successful for PACS but other local hospital information system programmes have been less successful.
PACS allows radiology departments to get reports out quicker and this makes them more effective and more efficient. Cautious departments noticed the improvement achieved by early adopters and followed suite, even though this required modifications to the way they worked.
In contrast, early adopters of hospital information systems experienced pain and simply did not see the improvement in efficiency and effectiveness needed. As a result, cautious hospitals were not prepared to modify they way they worked to fit in with the IT systems and instead insisting that the systems be modified to accommodate their working practices. This diminished the benefit of central procurement.
Healthspace is used by a fraction of the patients who could use it and the process of uploading summary care records has been partially halted. Choose and Book is unloved by GPs and has not been embraced by patients and there is no evidence that patient choice has in practice been significantly enhanced or that indirect measures of efficiency, such as DNA rates, have improved. GP system integration such as GP to GP record transfer and software to uploaded summary care records has been successfully deployed and summary care records made available.
However, the core product, the NHS Care Records Service, has just not delivered.
So if everything was done by the book, why did the National Programme fail to deliver the improvements in efficiency and effective that health informaticians confidently promised? Thomas Edison allegedly said: “I have not failed 700 times. I have not failed once. I have succeeded in proving that those 700 ways will not work. When I have eliminated the ways that will not work, I will find the way that will work.”
While it is also possible to find areas that could be improved, NPfIT did many things right. Management was reasonable. There was substantial political backing and sufficient financial resource, procurement was honest and robust and contractors in the main diligent.
One failing that has been identified is lack of clinical consultation.
Clinicians as a group are not shy and inarticulate and the views of clinicians are well known. Clinicians do not storm into hospital managers’ offices to complain about antiquated paper-based information systems and to angrily demand more IT investment. When summary care records were being uploaded, the BMA was concerned about issues of patient consent and recommended the upload of records be suspended. They did not urge that outstanding issues be resolved rapidly because lives would be at risk if summary care records were not available. US clinicians use pen-and-paper because high quality electronic patient records are not that important for clinical care.
It’s not that health informaticians do not consult clinicians. Clinicians' views are ignored because they are not what informaticians like to hear.
The truth is that clinicians are not punters. They do not study a patient's 'form' in order to determine how a condition will progress. Clinicians are more like airline pilots, dealing with the the here-and-now. It is important to know the aircraft's current height and airspeed. Its height this time last week is of little interest.
There may be good economic reasons why retail banking has computerised and health care has not. Health care generates a huge amount of information but the value of the information decreases rapidly. It is important now to know how a patient slept last night or their current haemoglobin. In six months’ time this information will be worthless.
Electronic health records require a huge investment in data capture and this is taxing for frontline staff. They are expensive to keep secure and accurate and for the most part contain largely useless information. When this is taken into account the overall value of electronic health records may well be negative and putting in more IT investment may actually decrease the efficiency and flexibility of the health care system.
Is the focus of health care IT investment simply wrong? Instead of investing in collecting and storing old information, perhaps IT investment should be directed at getting current information to the person who needs it quickly.
How would such a system look?
Twitter demonstrate the value of information to the recipient depends not on the profound nature of the information but on its provenance and timeliness. While profound information retains value and trivial information perishes rapidly, Twitter suggests that the value of trivial information may be very high initially. For example, knowing if a patient passed urine last night may be very valuable to their doctor now. Knowing that a hospitalized child asked for (and got) ice-cream for breakfast would be very valuable to their parents on that morning. There is no need to store this information indefinitely.
Social network sites like Facebook and LinkedIn show the importance of describing social networks - having a map of how people connect, so they can send messages along these relationships to each other.
No one would seriously suggest that sensitive health information is put on Twitter or Facebook. Secure networks and trusted identities are an absolute requirement.
This infrastructure is exactly what NPfIT has developed. N3, NHSMail, Healthspace, ePrescribing, Choose and Book and Summary Care Records provide exactly the building blocks that are required if a general purpose NHS wide social network were to be developed to allow messages - whether concerned with appointments, prescriptions, investigations, dietary preferences or other matters - to reach the people who need the information rapidly.
Forget health records. The new NPfIT 2.0 should concentrate on the here-and-now.
The NHS can be made fit for the 21 st century. Once the false starts and badly allocated resources are forgotten, NPfIT may be remembered as the programme that made this possible.
Dr Derek Meyer is lecturer in biomedical informatics at St George's, University of London.
Tell us what you think – have your say below, or email us directly at [email protected]