back to article NHS IT failures mount as GP data system declared unfit for purpose

The towering scrapheap of NHS IT failures may about to rise further, with the increasingly expensive GP Extraction Service IT system deemed not fit for purpose by the government's spending watchdog. Costs for the GPES IT system, which is supposed to extract data from all GP practices in England, have ballooned from £14m to £ …

  1. I Am Spartacus
    Mushroom

    Not fit for purpose?

    If it is late, over budget and not fit for purpose, presumably the contract will allow the client to demand all money paid to be primary contractor (ATOS) to be repaid.

    Wait - I forget that this is a government IT system and they don't do that sort of thing. They just ask the tax payer for more money.

    1. This post has been deleted by its author

    2. Magnus_Pym

      Re: Not fit for purpose?

      The Government can't get those sort of clauses into an IT contract because everybody knows the project will fail before it starts.

    3. Anonymous Coward
      Anonymous Coward

      Re: Not fit for purpose?

      Long ago my old dear helped as a uat tester for a new system in the NHS, she was a mental health nurse of many many years and she noted that the project was bound to fail because nobody in the department could be bothered testing that it did what it was supposed to do. Consultants didn't have time or the will to check it, neither did the doctors or nurses, even the back office staff didn't bother. She noted managers would sign things off as passed when they hadn't been tested and that the 3rd party developer found it almost impossible to get people to turn up to confirm anything so the 3rd party just had to wing it with what they received at the beginning.

      So as bad as some of the 3rd parties may be, most public bodies are their own worse enemies.

      1. Anonymous Coward
        Anonymous Coward

        Re: Not fit for purpose?

        Are we really surprised that NHS Clinical Turkeys will not vote for Christmas by testing systems?

        After all they will lose as a result of such changes.......

        Instead they prefer to continue to use Leeches and Tourniquets frankly for all ailments.

        And I bet they all have private health insurance and don't rely on their own service for help in their hour of need.

        Stop wasting Billions on a broken service; throw all 1.8m on the redundancy scrap heap and lets start over....... £20bn of failed IT projects alone would have put a big dent in the national Debt regardless of the cost (unquantifiable) of the recent NHS structural changes

    4. Mark 85

      Re: Not fit for purpose?

      If the contract is similar those in the States, it's a "cost plus" contract. The refundables are buried and well-fortified due to the "plus" part. So once the buying agency starts making changes and doing the gold-plating... recoverables are zip.

  2. h4rm0ny

    Negligence.

    The people in charge of this debacle should be prosecuted under the law. I used to work in the NHS. I left because of the sheer degree of corruption at the top. The NHS is filled with people who pointed out flaws with this program and others right from the start and said exactly where it was going wrong. Unfortunately they are all at the middle to bottom of the power structure and never listened to.

    I and my project team could deliver better versions of the software components of Careall this for a twelfth the cost. But what would be our chances of even getting to the tendering stage? I will tell you for a fact it is zero.

  3. Anonymous Coward
    Anonymous Coward

    The trouble with Gov buyers is they never ever spec the system properly.

    1. Graham Marsden

      ... probably because they have no idea what they actually *need* it to do.

      Oh, sure, they've got some vague ideas of "wouldn't it be good if it did this..." and, naturally, the Government Approved Contractors will say "Sure, no problems" whilst laughing up their sleeves, knowing full well that the Government has no idea how to write a proper contract anyway, so when they start changing the specification or adding new bells and whistles and it all starts to fall apart as the Contractors say "sorry, can't be done", they'll be in line for another £x million and the whole silly thing goes round and round...

    2. NeilMc

      mmmmm not strictly true......

      they allocate some mental health patients to Spec the system and provide that to the IT Service providers.

      IT Service providers then negotiate with junior NHS staffs who cant spell IT and have not heard of the project.

      Contracts signed

      Now its time for senior NHS staffers to meddle with the project and turn it into a bloated and inefficient system link Homer Simpsons design for a car!!

      Situation normal FUBAR!!!!!!!

  4. This post has been deleted by its author

  5. Captain Mainwaring

    Employ their own consultants

    Perhaps it's time that the NHS and the civil service in general employed their own IT consultants direct. At least this way the spec for the intended system could be carefully and expertly drawn up, before the project goes out to tender. Once the project is underway, any modifications could be fed in via the project consultant who should have the necessarily expertise to negotiate required changes in a meaningful and efficient way. They may have to pay the going rate for such expertise of course, but surely this would be cheaper than the multi million pound cock ups that seem to be commonplace in government IT.

    1. h4rm0ny

      Re: Employ their own consultants

      >>"Perhaps it's time that the NHS and the civil service in general employed their own IT consultants direct."

      In my experience with this area, in-house actual work is the last thing that they will do. There will be plenty of in-house people producing paperwork and looking busy, but it will all be just liaising with the third party company(ies) and waffle.

      There are two reason they wont touch actual work in-house. The first is that there is a fair bit of corruption at the upper levels and the aim of a lot of this is to funnel money to outside parties. Oh they will tell people, even themselves, that they're actually spending it for good, but they will still ultimately be funnelling money to their friends. A lot of the SPINE and Connecting for Health was done under Health Secretary Patricia Hewitt, formerly board member of Accenture. Who got tonnes of money for these contracts? Accenture.

      There are a lot of rules and regulations about spending in government. But once you get it out into the third party, accountability goes down the drain, it's essentially a firewall against investigation.

      Which leads to the second big reason which is one of accountability. No-one in upper reaches of the NHS or Department of Health ever risks having the blame for something land on their desk. They ALL have bits of paper that show it's actually not their fault. Doing work in-house, runs directly counter to that. And yes, that even includes specification which you thought was harmless but no - if there's a clear specification done in-house and it turns out to have been flawed and can be blamed for the ensuing disaster by the outside company, then whoever was head of the department or team writing that specification is going to be hung out to dry by the government as an easy scapegoat. So even writing the specification is outsourced where possible. They'll hire an expert consulting company to help draft and develop the specs.

      The one thing you have to understand, if you really want to know why things like this happen, is that the ONE overriding concern of everyone involved on the government side of this, is THEY MUST NOT BE ABLE TO HOLD ME RESPONSIBLE.

      And so, no-one is.

      1. Doctor Syntax Silver badge

        Re: Employ their own consultants

        "The one thing you have to understand, if you really want to know why things like this happen, is that the ONE overriding concern of everyone involved on the government side of this, is THEY MUST NOT BE ABLE TO HOLD ME RESPONSIBLE."

        Quick fix. The Treasury doesn't release funds for any project above £x without having the name of an individual who is held responsible. If cost overruns take a lower cost project up to £x no additional funds are released without their having the name of an individual who is held responsible. For existing projects over £x no further tranches are released without having the name of an individual who is held responsible. And require evidence that the named individuals actually have the clout to exercise that responsibility.

        The Treasury hold the purse strings. They can lay down such conditions if they have the will-power to do so. It might cause ructions elsewhere but that's elsewhere's problem.

        1. Captain DaFt

          Re: Employ their own consultants

          "The Treasury doesn't release funds for any project above £x without having the name of an individual who is held responsible."

          And then you get this from the government office in question:

          Our congratulations to Bob the tea boy, for his promotion to Head of project White Elephant!

          1. Doctor Syntax Silver badge

            Re: Employ their own consultants

            "Our congratulations to Bob the tea boy, for his promotion to Head of project White Elephant!"

            I think you missed "And require evidence that the named individuals actually have the clout to exercise that responsibility."

            1. Stevie

              Re: Bob the Tea Boy

              Heh. I was once roped in to prototype the lifecycle process in my government IT shop by the tiptp mucketymuck. He insisted we get everything signed off as saiod project represented a joint project between two different monolithic government authorities.

              I got past the fact that Bossman demanded I use the wrong process - it was a new project but he insisted we use the MAINTENANCE life cycle procedures. I got past the fact that world + dog wouldn't say anything in one word when twenty would do (any twenty - EULAs made sense compared to some of the submissions by the various people esponsible for deliverables and milestones).

              But the cherry on the top was when the requestor refused point blank to sign off on the requirements document OR write the required statement as to what was needed to bring them into spec.

              Said requestor was, of course, the man who insisted on thois process in the first place.

              To my knowledge I remain the only person to ever go through the process, and in point of fact, since the document was never accepted or rejected, I didn't officially go through it either.

      2. Magnus_Pym

        Re: Employ their own consultants

        The one thing you have to understand, if you really want to know why things like this happen, is that the ONE overriding concern of everyone involved in government is THEY MUST NOT BE ABLE TO HOLD ME RESPONSIBLE.

        FTFY

      3. Sludged

        Re: Employ their own consultants

        I worked at a local authority and the requirement for a large project came up. It was nothing I hadn't done before but management insisted on bringing in a commercial company to do the work. Management referred to them as "third party insurance" I moved on shortly afterward

    2. Brewster's Angle Grinder Silver badge

      Penny wise; pound foolish.

      "They may have to pay the going rate"

      But its tax payers' money -- so we can't charge the going rate!

    3. arrbee

      Re: Employ their own consultants

      They did, back in the 1970s, and a pretty high power bunch they were too.

      Strangely they were one of the first groups to go when the westminster civil service was transformed into the sharp, efficient, go-getting enterprise it is today.

  6. adam payne

    Another NHS IT system over budget and not fit for purpose. That is certainly a huge shock.

    Where exactly was the £40 million spent?

    30 million on consultancy?

    1 million on the tender process?

    5 million on new hardware?

    4 million on making the system?

  7. Terry 6 Silver badge

    Hmm Committee based planning

    Government and local authority schemes ( not just IT) are dogged by the fact that everyone involved wants to have a say in what needs to be included. The project is opened up to all and sundry without fixing its aims first, so that the aims are almost immediately fudged. They all have a different agenda.

    Most will not be in role by the time the thing is even started, let alone implemented, so they aren't trying to get an outcome, just to get a reputation for something peripheral ( "hard-headedness" "business awareness" or whatever) Or just looking for an over well paid job in the private sector.

    The basic "This is what we need to be able to do. This is how we will do it!" is lost before the first page is even out of the printer

  8. Anonymous Coward
    Anonymous Coward

    Where lies the accountability?

    It's easy to blame the contractor for "another failed project", but surely there are measures to protect the buyer in whatever contract they've signed.

    I rather suspect the reality here is that ATOS have delivered to spec, the underlying problem being that the spec provided to the contractor is rubbish, or at the very least keeps changing, hence the ever increasing costs. The ivory tower egos in the NHS aren't therefore able to blame the contractor in a way that would stand up to legal challenge, so it's just a case of either throw good money after bad, or stand up and admit to another failed project and smile through the fact that it's their own negligence that's caused it.

    1. Otto is a bear.

      Re: Where lies the accountability?

      A good summary of pretty much any government IT programme with just about any IT supplier. The degree of culpability is variable, though generally magnified 10 fold by the customer. Don't forget that any government IT programme is bid against a spec which is years out of date by the time the contract is signed. (Not always though, I did one project with a 3 month bid cycle, so the requirements fit was great) . This means the first six months is taken up with unplanned requirement revalidation, the customer won't let you add that to the project timescales and budget upfront. The subsequent changes caused by new legislation and departmental reorganisations then spend six months being agreed. In this time the SI has a whole bunch of people sitting around being paid to do nugatory work because you have to deliver to the contract to get paid on time, failure to do so would mean the smart customer commercial team will go after you for penalties for failure to deliver something they don't want. You can also bet that almost all the Customer and SI team who agreed the contract have moved on, usually within 3 months of contract signature, so no one knows what the original point was. The other gem on these massive programs is that the actual recipients/users of the system have no contact with the designers. The designers have to go through a customer team often made up of contractors or consultants from the consultancy houses who interpret the requirements and user feedback to maximise their usefulness, sorry revenue.

      1. Terry 6 Silver badge

        Re: Where lies the accountability?

        Otto ...Bear

        Yes, and also, for those who have never sat in a committee ( God help me I've wasted a good few days of my life in these) won't realise how these things work.

        But, I can guarantee that in any such meeting there will be several important people who are so out of touch with the users that they have a separate entrance in case they actually meet one. They are looking for a lucrative job outside.

        There will be a couple of "up and coming" less senior managers who are out to advance their careers. They don't care about this project at all. It's just stop along the way.

        There will be some who are totally committed to the project, but formed their ideas 5+ years ago and are trying to plan the world that no longer exists.

        There will also be at least one who is dead against the project, but for all the wrong, or even no, reasons.

        There is often one or two who refuse to admit how much that it should cost and want to squeeze the budget down to below anything reasonable ( or have instructions to do so - see below).

        And any number of others who are there because they have to be, and don't care either way. But may well be following a line that they have been told they must support.

        And all of these will try to add their own pet idea, grind an axe they brought with them or just be desperate to cover their own a***.

      2. RyszrdG

        Re: Where lies the accountability?

        Pretty good and generic root cause analysis for most programmes that I have been asked to recover. It needs real steel cojones to get the sponsors, customers and suppliers to face up to issues at an early enough stage to take timely corrective action. Usually it is far to late to fix or even halt without blood on the floor - usually it becomes a hunt for someone to blame; unfortunately it is usually the most junior that has to fall on his sword rather than the real culprit.

  9. Anonymous Coward
    Anonymous Coward

    Would have thought

    The basic spec would be the system should not fall over to the exent that fax machines have to be used. There is certainly blame to be placed at Atos feet (and other contractors) as they profit from spec changes, including the bloody obvious being missing from a spec so thay can charge extra to provide it after they have won the lowest bid.

  10. Anonymous Coward
    Anonymous Coward

    I don't actually understand how they can get it so wrong

    Maybe because I'm not close enough to it, but I cannot for the life of me fathom how they could screw this up.

    First of all, if you use a permission structure that has been around for, well, about 15 years or more there would not be an issue with privacy because you could enable the people (you know, "the entities who have rights, certainly because the costs come out of their taxes") to take that decision locally and have it recorded - challenge 1 solved. You build in overrides which are controlled and audited for the mandatory reporting of illnesses that may escalate into epidemics, and presto, control and distribution.

    Secondly, if you devise a fairly flat back end all you have to do is to ask the suppliers of GP practice IT to develop an interface wedge, which leaves you to develop an API test framework and a certification regime which is cheap due to its simplicity.

    I think the first step that needs to happen there is to rip the consultants out of this scam scheme - I worked for such an outfit and the main aim appears to be getting is specified wrongly so they can clock up change control charges. Next step is to get some people involved who actually have a clue - the problem is finding some that are not that hell bent on lining their own pockets. That's probably the real challenge of Government IT overall.

    1. Little Mouse

      Re: I don't actually understand how they can get it so wrong

      "I cannot for the life of me fathom how they could screw this up"

      There's no incentive for either side to get it right. There's too much money to be made from failure.

      On one side of the equation you have incompetent Public Sector fuckwits who can only justify their position and salary within the organisation because they make "Lots Of Important Decisions" about "Important Contracts" involving "Large Amounts Of Money". These people stick around the project for only as long as it takes "Someone Else" to sign the crappy deal that they have brokered, before moving on to the next disaster-in-waiting. You can't fire them and no-one else wants them. Like most people, they don't want to be unemployed or make less money, so they stay. Forever. Doing more and more damage.

      On the other side, you have the highly experienced contractors (same ones every time) who know how to play this particular game very well indeed. The worst-case scenario for them is that they actually deliver what was asked for, for the agreed amount of money. There's a whole load more money to be made for all the "extras" that were never originally specified, for running over budget and getting the customer to stump up for the overrun, or for having the customer actually terminate the contract if they're not hapy with the progress. Some of those penalty & termination clauses are pure poetry, and only ever work one-way.

      "Maybe because I'm not close enough to it" - Just be grateful for small mercies.

  11. Anonymous Coward
    Anonymous Coward

    Same names or different names?

    We rarely get to see the names of the civil servants spear-heading these programmes.Is there a small clique of bad programme managers or is it just "muggin's turn" to take on a bound-to-fail IT project?

    If ( big "if") HMG follows its own PRINCE2 methodology, then there should be a clearly identified single person who was project executive. Name? (There should also be one or more "senior users" on the project board representing the end-users. Names?) One of the primary functions of the project board is to pull the plug on failing projects early, rathe than ride them into the ground after they have spent billions.

    1. Anonymous Coward
      Anonymous Coward

      Re: Same names or different names?

      Bab bad bad boy (girl) you are asking all the wrong questions about accountability. Just see IDS - when in shit, just lie.

    2. RyszrdG

      Re: Same names or different names?

      Methodology, shmethodology - the root cause is always people not method.

  12. x 7

    The sad thing is that this functionality already exists off-the-shelf

    Four products jointly hold around 90-95% of the UK GP clinical systems market.

    In no particular order

    INPS Vision

    Emis Web

    Microtest Evolution

    TPP SystmOne

    All four products have the ability to extract data in the way required. Web and SystmOne can interchangeably access each others data. A quick kick up the rear end would rapidly make INPS and Microtest join in (though both are suffering decreasing market and are rapidly becoming irrelevant)

    The data from Emis Web and SystmOne are both held in secure data warehouses, making the extract and compilation easy. Evolution data is held locally, while Vision is split - some sites local, some hosted off-site by INPS. Again, access to the hosted data is easy

    Evolution data is held locally, but Microtest say that networking this across sites to allow this kind of extract is easy

    So......all it needs is a bit of an effort, for the NHS to stop trying to do it themselves, give the four companies a nudge in the right direction and let them come up with a working system. At the same time interlinking the four systems would make the Summary Care Records totally irrelevant.

    It just needs the filters put in place to allow the four sets of data to be harmonised. The filters exist - it just needs the willpower

    1. Ilmarinen
      FAIL

      I know nothing of NHS IT systems, but what you write seems likely true.

      As an involantary patron and sometimes user of this Soviet style monolith that seems to be the HNS, I don't actually want it to be spending money on "extracting" my data, whether to a "secure" data warehouse or anywhere else. If a doctor wants to engage the services of a surgeon/consultant/whatever on my behalf and send them information, let them make their own arrangement as they see fit, providing it's secure (same as if I send a report/PO/whatever to a supplier or client).

      What I *would* like to see however is for the people responsible for this current mess (and assorted other messes) to be, in the words of a now sadly deseased school teacher "shot at dawn for a fortnight".

      Or even just once, to encourage the others (and because they deserve it).

      1. x 7

        the data extraction is not to find individuals, but to look for trends regional and national. Partly to enable resource prioritisation, but also to find new trends: disease outbreaks become visible more quickly, success (or not) of treatment regimes can be tracked, regional differences in survival rates can be seen. And rogue GPs who kill patients intentionally or otherwise can be rapidly picked up.

        Theres nothing sinister behind it, and at present its already going on in some CCG areas to recognise local trends and priorities. It just needs expanding to a larger scale.

        Cross-availability of records between institutions is a different matter altogether and realistically is years away in most places, as the hospitals tend to use software that completely incompatible with the GP clinical systems. Theres a trend to make data available between local surgeries to allow for out-of-hours dross-coverage, but so far not much beyond that except for the very sparse national summary care records

  13. Anonymous Coward
    Anonymous Coward

    I don't think the blame lies entirely with the GP Extraction Service. The systems that GP use to make their notes on work in differing ways, using different clinical coding systems and return data with wildly differing quality. I have heard one market leader's data described as a rat's nest in a meeting with them. That was four years ago and I think the problems still exist now. Its pointless to even try to extract data from the GP systems until there is a standard way of recording notes. That means standardising the systems and devoting a great deal of resources to educating GPs in how to record notes electronically. Once you have reasonably consistent data then it is worth trying to extract it. If you put brown sticky stuff in, you get brown sticky stuff out.

  14. Anonymous Coward
    Anonymous Coward

    sorry there is no NATIONAL in the NHS

    The issue is that we no longer have a NATIONAL health system we have local GPs that are then part of a local groups that are part of over 200 Commissioning bodies that then can get services form totally separate group of bodies that run the hospitals who themselves compete and therefore waste resources. and all these layers and separate organisations or groups of organisations use separate non interoperable systems. so there is NO communication, NO Cooperation, and a Post code lottery of care provision.

    NO private company could function with such an organisational structure and internal competition, and so many vested interests and such a high % of managers and such a low % of staff actually providing the core function of the business.

    The NHS needs a root and branch restructure done by someone without a vested interest or a history of failure and someone who needs a BIG budget and the power of God over the decisions.if this means bulldozing the current setup and starting again from scratch going back to Aneurin Bevans principals so be it.

    It would "probably" cost less in the long run and may actually be a sustainable model IF it is properly specified in the beginning with a detailed organisational and communication structure and plan properly implemented.

    The NHS may become a working operating Government function with a successful IT project roll out........ sorry day dreaming again :-(

    Secretary of State for Health - needs to be made Responsible for failures of care or data breaches with the sanction of serious Jail time if there is a failure or data breach.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like