Last month, the Commons’ public accounts committee published a pretty damning report on the NHS National Programme for IT in England, Connecting for Health. The committee’s analysis looked specifically at the development of a single care record. This is a set of electronic notes for each patient that can – with the patient’s permission – be accessed by clinicians working in different parts of the health service. MPs noted how progress on this central aim of the programme is running severely behind schedule and has failed to demonstrate value for money.
Single care records offer huge advantages over the alternatives, namely paper records or fragmented electronic records that can’t be accessed across the primary/secondary care divide. So has the time come for a radical rethink on how to deliver them for the NHS?
First a word about costs. A striking feature of Connecting for Health is the vast sums of money involved. The programme has an overall budget of £11.4 billion, with £2.7 billion spent so far on care records and another £4.3 billion in the pipeline. But I don't think we should necessarily be alarmed by these high numbers.
The NHS is a huge organisation, and around the world, top-performing health care systems typically spend above-average amounts on IT. For example, Kaiser Permanente recently bought a $5 billion IT system for its 8.7 million members (the NHS in England covers 50 million people). What's more, investment in health IT can sometimes lead to overall savings. For instance, an analysis published in Health Affairs last year found that the Veterans' Health Administration (VHA) had made a net saving of over $3 billion by investing in health care IT.
The new name of the game is "competition and collaboration" – personally, I can't see how either of these can be deployed effectively unless there is a single care record in place to facilitate them
Clearly, then, the point is what is achieved for the money rather than simply how much is spent. Like many of the world’s best health care systems, Kaiser and the VHA have fully-functioning, single electronic medical records. These records are accessible, with strict access controls, to clinicians working anywhere in the health care system – and they can typically be accessed by patients and their carers as well.
Currently in the NHS, we have first class IT within our GP practices. But the clinical record in hospitals is still almost always paper-based, supplemented by a range of idiosyncratic IT systems for requesting and viewing test results. Moreover, the ability to transfer records between different parts of the NHS is underdeveloped to say the least.
In its report, the Commons committee acknowledged that the original intention of creating a single electronic care record was still a "worthwhile" goal. I agree entirely, but things have moved on considerably since that original aim was first agreed.
In the recently "liberated" and "listened-to" NHS, the new name of the game is "competition and collaboration". Personally, I can't see how either of these two improvement levers can be deployed effectively unless there is a single care record in place to facilitate them.
Why? Well, for there to be meaningful competition, patients will need the theoretical ability to transfer access to their electronic notes easily between different providers – both between different primary care clinics and between the primary, community, secondary and tertiary sectors.
Equally, for care to be truly integrated, it is essential that professionals working across all of these different sectors – and indeed in social care – should have the ability to read and write in a single set of electronic notes. Only then will the real benefits of integration be fully realised, namely the avoidance of duplication and a reduction in the number of "gaps in care".
All of this suggests that the need for a single electronic record is more pressing than ever. So it’s rather a problem that the Department of Health has announced it cannot deliver them.
What to do? Well, perhaps the time has come for a radical re-think. One sweeping change might be to mandate the use of open-source software across the health service. Yes this would be disruptive in the short term, but it could promote inter-operability between different parts of the NHS because the open standards would be published openly.
At the same time, going open source could help contain costs for a number of reasons. Firstly because open source software is free to download and use. Secondly because it can be cheaper to customise and adapt. But most importantly because the use of open-source software should promote a more competitive market in IT support for the NHS by avoiding so-called vendor lock-in.
This fully open-source option is the route being taken by the national health service of Jordan, which is currently rolling out the VHA's open-source system, “VistA” across the whole country – linking every primary care clinic and all 43 hospitals across the land.
When the National Programme for IT was first launched, a delegation visited the VHA in Washington but chose not to implement their open-source IT system here. Has the time come to revisit that decision?
This blog is also available to read on the Health Service Journal website.
Comments (9)
The type of technology, whether open source or proprietary, is not the issue for health records. IT is more about people, leadership, and relationships more than the technology itself. Open source solutions may be part of the solution (in my health area they already are) but they still require major development work and configuration. Rather than focus on an unachievable single care record, the emphasis would be best placed in local health information exchanges, creating clinically valuable virtual health records across multiple healthcare settings.
With a project as large as IT in the NHS open standards are vital. That *should* be open for the simple reason that no one would be able to connect to anyone else without it! Yet this is not happening. There are 250+ NHS trusts, 150 PCTs, ten SHAs, ten health observatories, yet each one uses different standards for their web site - almost 500 ways of doing the same thing! Why does it matter? Well if you are a patient with "choice" you will want to read the information on the trust websites in your area top compare them. If each trust has presents different information in different ways then patients will not be able to find the information they need. Using standards makes it easier for users to navigate a site, and that is especially true for people who have poor sight or use a screen reader. If we cannot get standards across trust websites, how do you expect there to be standards between trust IT systems? Moving to open source is another issue altogether. The open source movement is especially strong in the university sector for good reason: there is no profit motive. The same might have been true for the NHS since the hard work of one trust could be shared with other trusts in the area and everyone benefits: that is collaboration in the NHS. In the new NHS with competition where every provider is in competition with every other provider, collaboration suffers. As you say, information in healthcare is very valuable, but providers do not own the data; they do, however, own the systems they build around the data. This gives them an advantage over the other providers, do they want to share that advantage? Of course not! Competition is all about them gaining such advantages. Also, software is costly to develop (people who eulogise open source don't seem to understand that), making NHS software open source will require that there is a central budget for development, otherwise no trust would want to share their expensively developed code. We are moving away from anything "central". This is the ethos of the new NHS. So the chance of systems becoming open source is zero. Nice idea for the original NHS, but unworkable in the new NHS. However, once the new NHS has failed (as it will) we may have a second chance ;-)
Thanks for your comment, Andrew.
I take your point but I suppose I'm slightly more optimistic: if they can manage to introduce VistA single care records across the whole of Jordan, I don't particularly see why we can't do it here.
Thanks for your comments, Richard. All good points.
Two quick thoughts:
1. In the case of VistA, the US Federal Government has paid for the development of the code.
2. While we wait for our second chance in England (!) maybe the other UK nations could lead the way...
We presented vista to the UK on the 6th of September, 2001 in Kensington, UK for the Open Source Health Care Association (OSHCA). We really shocked the attendees with the power of VistA. It must be remembered that VistA is a process which we are anxious to export. It is not 1 solution for everybody. It is a process that needs to be used by the people who will be use them. They need to adjust VistA to fit them. This is what VistA was built for, to be adapted to the new environment it will be used in. Also take a look at mumpster.org for the other half of the presenters of VistA in England, Rick Marshall. Best wishes; Chris Richardson
Thanks Chris.
For the uninitiated, a good introduction to VistA is a book by Phillip Longman called, "Best Care Anywhere: Why VA Health Care Is Better Than Yours"
Also, just picking up again on Richard's point about the development costs of single care records, see this blog from the Wall Street Journal a couple of years ago: Free From Uncle Sam: VA Software for Hospital EHR Systems
The web thing above is a digression, but it is a semantic web question. I also tend to think that a language for interchange an discussion among computers of elements of whatever a medical record turns out to be is philosophically better than picking a single system, but if one has a national health service that needs computerising then VistA is well worth looking at. As I have been, since the planning of the Kensington meeting. There are several business models for making money from Open Source software, and as well as those there are good reasons for hospitals to share. Examples of benefit coming back to Trusts from the other approach are, I think, not conspicuous.
Adrian; Good to hear from you. Haven't seen you since Kuala Lumpor, and Molly Cheah's little party there. I gave a presentation on why the VistA installation in Lahore, Pakistan failed. I found out recently that the group that keeps things going continued to work on it and have installed one of the new VistA install packages (Astronaut) and they have hooked into the VistA community and have been helped to get the VistA install back up in Lahore. We are all cheering for them and hoping they succeed. We are hoping to have them work with the community. As it is, I have sent Geraint a copy of my "Support and Familiarization disk". It has a lot of additional information that most folks who will be running VistA should have. The nice thing is that it only takes a single PC to run VistA. See if Geraint will make a copy of the disk for you. Geraint, make as many copies as you would like.
To those who don't know me, I'm a Family Physician educator and have been using VistA and it's predecessor, DHCP, since 1983. My view is that during those 30 years the US Veterans Administration has managed to produce one of the finest electronic health record systems in existence and it is ridiculous that we don't take advantage of this FREE software. To be sure, it needs tweaking and customization for individual sites. And yes, it is written in MUMPS - a language that your hotshot computer science types will tell you is archaic. But... if you check out the Google group "Hardhats" you will find available a community of users who know this software intimately and will assist anyone wanting to implement it. You should also know that MUMPS, far from being archaic, is a schema-free, non-SQL data system - the latest hot thing in data-handling. It was just way ahead of its time, and not by accident is the basis of Epic, the system used by Kaiser in the US. Finally - and this is critical - a volunteer organization, WorldVistA, has available a free, open-source version of VistA that is far less VA-centric than the official VA version and is in the process of being further adapted to general use. I find it truly sad that the direction of the NHS has veered off into business-oriented territory with the tired old "competition will bring out the winners" ethic coming to the fore. Cooperation, the antithesis of competition, is what will bring us the health record system and health care outcomes that we all want.
Have your say