Ministers kill off failed £12.7bn NHS IT revamp GPs and hospitals have been told to look locally for IT help as the government finally spikes a £12.7bn nationwide NHS computer system. The failed project intended to store everybody's records, institute a national email system for the NHS and make X-rays and prescriptions available electronically. It was touted as the world's …
The irony of Conservative NHS IT policy is that they want to do in the NHS what no large business would do: have local procurement for infrastructure where there should be large economies of scale by using the same products across the entire organisation.
However, the big problem with Labour's Grand IT Plan was that it was a Grand Plan instead of doing what a (good) company would do with acquisitions: integrate and then migrate. Given that each region, hospital or even department has systems that make them like a separate company trying to do everybody at once was a really bad move.
A national health service would certainly be an interesting idea, but we don't have one.
Within the regions the General Practices are each a separate firm, and this is good.
So the problems of connectivity, and of handling whatever a medical record may be (it isn't as obvious as the people who thought they were going to deliver the contracts may have thought) are a lot more like the problems of connecting a lot of networks and hosts in the Internet.
Seriously, how hard can it actually be to build a database and network infrastructure capable of storing the client data and making it available across the varying parts of the NHS, with suitable application layer(s) on top?
What has cost £12.7bn and yet not really delivered?! It's just plain unbelievable.
that the NHS forgot to include in the scope of the project. Training, revising redundant manual procedures, basically changing the whole culture of admin and clerical work associated with the NHS which employs over 2 million people - that's about £600 per year per employee on re-training and skills dvelopment. TDon't believe the hype
The systems all work (have done for years) and constantly come in under budget. Productivity has increased more than 10 fold.
Hope that helps
Many people told those in charge it would be an expensive disaster.
The government did its usual - opened to 'competative tender' but set the rules such that only a very small number of huge and very expensive international consultancies could even bid, it then chose one of them and gave them a stack load of money, a very very small proportion of this money was spent on a bunch of Indian developers, a massive amount was siphoned off into massive wages and bonuses for 2 or 3 individuals 'at the top'.
It was never going to deliver. And the 'new idea' will also end up with similar restrictions so that services won't come from small or medium British companies but again from huge global consultancies who will once again outsource to India. This is the way UK government procurement works - for everything - set the rules to ensure that NO British company can ever be involved, then bemoan the 20+% unemployment and pretend they are 'trying' to do something about the shrinkage that has occured every year since I was born in the number of jobs, the wages and prospects for workers in this country. [I, with a degree my father never had, working in what was supposed to be a 'growth industry' (computers) and successful at my job have a lower standard of living than my father has ever had]
"It is no longer appropriate for a centralised authority to make decisions on behalf of local organisations. We need to move on from a top down approach and instead provide information systems driven by local decision-making."
If only governments would extend this same logic to themselves ...
I don't think that there is necessarily a problem with centralised decisions, or the same major systems being used across the board. However what became known was that there was no consultation with the final users; no-one looked at what was already in place. In fact the whole systems analysis phase seemed to be done with no consideration of what was actually needed.
I'm interested to see what comes out of the cancellation. Since the "new" digital x-ray idea, my local doctors could no longer see digital X-rays on their computer, as they had done for years. Instead, the doctors either had to go to the hospital department that took the x-ray to see it on their system, or ask for a FAX (kid you not!). The old systems where each consultant could bring up the x-rays no longer worked.
It has always been more cost effective to use local service providers. The trusts who try to go it alone will simply be squeezed even more by the real costs.
Never trust a politician - he'll make out you've got what you want when in reality he's got you by your short and curlies. The trick is not to let you know
I wouldn't want to on the NHS payroll working for IT right now, but I hear there are loads of SME startups looking for staff ;-)
It's amazing how politicians can take something as simple as CfH's interoperatability tool kit and spin it up into something that's going to save £12bn. I've never heard so much political waffle and willy waving.
It's not as if this is even news. Contracts were re-negotiated last year with the LSPs and ITK was introduced under the previous Government. It shouldn't come as a suprise that local health authorities are allowed to develop their own IT systems providing they interact with central systems such as SPINE and N3 using certain interoperatabily standards - that's what they've been doing for the last 10 years.
There has never been an insistance that local health authorities use IT systems provided by local service providers appointed by NpfIT. It's just that it makes more sense operationally and cost wise.
"We need to move on from a top down approach and instead provide information systems driven by local decision-making."
Err, wont this be worse if you have a bunch of local managers with no/little IT experience deciding contracts to small contractors/suppliers? Wont this lead to many small expensive problems rather than one big expensive problem?
Surely it should be contractually defined by small group of IT people (and staff, no managers except for the cost) for the whole of the NHS then outsourced to a local software warehouse with well defined scope and requirements? No? Am I in a dream world?
Onto the next project failure at huge expense to the public....
...on a side note, is it not true that conservatives always create white elephant projects that last about 10 years and go tits up with one of the future governments, but not before a few people have made a huge sum of money...?
... which wouldn't have made no-one in the Civil Service look good.
1) Define XML schema for patient health records, and binary stuff like X-Rays.
2) Define a simple REST-type API to query these records and request updates.
Then each hospital adds the REST interface to their existing systems. Result - nationally-available health records.
Cost - buttons.
What you mean is, when you have to make life or death decisions based on a patient's history (such as drug allergies, chronic conditions etc.) in the A&E department you try to pull it from a GP's server that has not been upgraded in 10 years (money spent on them computer-things is less money in pay packet), or has been turned off by cleaners, or has simply failed, over a crappy internet connection.
Critical patient data is needed at the point of most importance, which is in a hospital, not a GP surgery.
What a brilliant idea - keep all the data on PC world Dell servers tucked under the receptionists table. Let's just hope they weren't unplug before everyone went home on the night some clinicial needs to pull up your records to see what medication you are on or allergies you have before he can wack you full of painkillers or administer safe antibiotics.
You do realise that GP's aren't actually employed by the NHS? - they are self-employed contractors, just like most of the cowboys who are employed by the NHS to "work" in IT.
Just as I wouldn't take medical advice from a hospital porter, I wouldn't want my GP being held responsible for the availability and security of my electronic patient records even if they think they can handle it.
The patient records need to be with the patient - it is after all their own data! This is what happens in France. You keep your own medical records. If you don't look after them then it's your problem. My experience in the UK is that you get some medical tests, and then two weeks later you go to see a specialist and you find that some of the essential notes have gone missing. In France you go for a test, and you take the results to your own doctor/specialist whatever yourself, very often on the same day.
1) Interoperability
2) ...
3) Profit
(2, of course, is gradually migrate everybody to a common platform based on open-source software which allows you to spend the massive NHS IT budget on more development instead of more licensing, with the side benefit of making OSS more usable for other people everywhere. Or do nothing and still profit from better interoperability.)
Sounds simple doesn't it. The Secondary Usage Service mentioned has a defined schema (see http://www.datadictionary.nhs.uk/). It isn't enforced when data is submitted leading to the quality of the data being complete and utter crap so you have to have various hacks to deal with incorrect data. You have fields that are supposed to be unique not being so, or data that is too long being submitted for a field that just ends up being truncated, and of course lovely duplicated primary keys. The hospitals, sorry trusts, don't seem to care about this. When contacted they sound surprised and state no one else has mentioned it to them, which means the users of the data don't give a damn whether they have accurate info either. The problem for medical records is careless recording and the stupidity of accepting invalid data rather than wholly the fault of cash vampire IT consultants.
"Sounds simple doesn't it. The Secondary Usage Service mentioned has a defined schema (see http://www.datadictionary.nhs.uk/). It isn't enforced when data is submitted leading to the quality of the data being complete and utter crap so you have to have various hacks to deal with incorrect data. You have fields that are supposed to be unique not being so, or data that is too long being submitted for a field that just ends up being truncated, and of course lovely duplicated primary keys. The hospitals, sorry trusts, don't seem to care about this. When contacted they sound surprised and state no one else has mentioned it to them, which means the users of the data don't give a damn whether they have accurate info either. The problem for medical records is careless recording and the stupidity of accepting invalid data rather than wholly the fault of cash vampire IT consultants."
You make it sound like a slow motion car crash.
was that people doing the requirements capture would go into meetings with large numbers of NHS bods and get different requirements from every one of them. They would then go away and work on the various requirements to try and make sense of them. However, the next meeting they attended would consist of another, different large number of NHS bods who would all have their own requirements, etc., rinse and repeat. In other words, it was the proverbial, failed piss-up in a brewery.
The fact of the matter is that most successful, large, complex IT systems which worked EVOLVED from successful, small, simple IT systems which worked. Starting out to create an enormous, complicated system from scratch is a recipe for disaster but no one in government ever wants to learn because of the lucrative executive directorships available when they're no longer in government.
I worked on this with BT, within a week of being hired with some other poor sods we'd taken a look at the system that was supposed to be happening and we didnt know whether to laugh or cry.
We got the anti virus part of it, specifically to set up, test and configure that so it would automatically push updates, over the *entire* system whenever required - with the intent of it not throwing a total eppy on general principle. Simple enough, and we got a fair level of accuracy (aka, updates doing what they should when they should).
Slight problem though - I might be wrong but NT4 domains and 'Active Directory' is never going to work and thats what they were envisaging when I was working on the system (at £35 an hour). Theres more chance of Obama giving Gaddafi a hug than those two working in concert - but hey, thems paying.
The thing that gets me is that they know to a few hundred pounds how much these total and utter frell ups cost, but we (the taxpayer) who paid for the sodding palaver in the first place never seem to get that money back, whereas if it was a private company this sort of thing would send them into bankruptcy...
Me... bitter? nah
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
