Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Government United Kingdom News Your Rights Online

UK Health Service Fears Huge Legal Fight Over Unwanted Contracts 127

DMandPenfold writes "The Department of Health is concerned that Fujitsu, CSC and BT would team up against it in a multibillion pound legal fight, should it decide to scrap the disastrous NHS National Program for IT. Fujitsu walked away from a £709 million contract in 2008, and remains locked in legal wrangling with the government over claims for the majority of the value. Today, MPs urged the government to seriously consider abandoning the program and therefore to consider terminating the remaining CSC and BT contracts, worth £3 billion and £1 billion respectively."
This discussion has been archived. No new comments can be posted.

UK Health Service Fears Huge Legal Fight Over Unwanted Contracts

Comments Filter:
  • by rbrausse ( 1319883 ) on Thursday August 04, 2011 @03:57AM (#36983070)

    I read the summary and "disastrous NHS National Program for IT" reminds me of UK Taxpayers' Money Getting Wasted On IT Spending [slashdot.org].

    both sides are to blame here - government agencies are often really bad in project management and contractors are abusing this...

    • Re:related? (Score:5, Insightful)

      by TheRaven64 ( 641858 ) on Thursday August 04, 2011 @04:11AM (#36983144) Journal

      This is a different kind of problem. The government generates huge sets of requirements and then gives a contract to a company with a track record of failing to meet the requirements, then acts surprised when the project fails.

      The NHS system is particularly irritating, because their current system is a room full of folders containing paper. It would be trivial to deploy a database system to store the more relevant information (although the storage requirements if you want to store x-rays and CT scans get insane) in a way that's easy for doctors to access. The software is only a couple of months work for a single programmer and could be deployed by the existing IT staff. It would then make life easier for everyone involved.

      Instead, the government specifies a system with an insane list of requirements for a huge number of unusual use cases, and then wonders why the project fails.

      • Re:related? (Score:5, Interesting)

        by myurr ( 468709 ) on Thursday August 04, 2011 @04:37AM (#36983302)

        Whilst you trivialise the problem to a degree (scalability and reliability of an NHS sized system is not trivial) it still shouldn't take a small team more than a few months, and a budget in the hundreds of thousands of pounds, to build such a system. This could then be incrementally evolved over time on a budget in the hundreds of thousands per annum (maybe low millions depending on speed of development). You do then have the data entry problem to consider, but that is surmountable for a fraction of the budget of these big IT solutions.

        However that's not how the government thinks. They want to go all encompassing from day one, speccing out a bloated and unworkable unholy mess that the end user doesn't want or need, and certainly doesn't understand, that takes a budget several orders of magnitude more than is required. Then throughout the project more and more people will hear about it and give their input or point of flaws, causing massive amounts of feature creep and confusion, affecting budgets, delivery time lines, and ultimately the quality of the end product.

        As a final anecdote, as a small web agency we once were involved in the build of a website for a London borough. We were in competition with some much bigger agencies, but we went back with a good proposal, some great design concepts, and what we felt was a fair budget. The decision maker loved our whole proposal except for the cost - he actually made us double the cost of the build, simply because that then matched his budget so that it wouldn't be cut the next year (spend it or lose it!) and because it brought it in to line with the bigger agencies (so his managers wouldn't think our offering was less feature rich because it was cheaper). This way of thinking is not unique to the the public sector but is endemic throughout it, and the big suppliers prey upon this.

        • (scalability and reliability of an NHS sized system is not trivial)

          Depends on what you're doing. I have a friend who works in the medical records department at the local hospital. A lot of her job involves running from one end of the hospital to the other because the doctor needs a specific file in 10 minutes and only just realised. A system where every doctor had a tablet PC of some kind and could call up all of the records in a hospital would be really simple to design and deploy. It wouldn't have to cover the entire NHS, just the one hospital. That's only a few TB

          • If one hospital had paid for the development of such a system and specified that they own the copyright on the resulting code, they could have released it under an open license and other hospitals could have used it easily.

            Fraid not for so many reasons - each hospital has its own budget, out of which comes everything from drugs purchases to theatre time to IT systems, so you would end up with one hospital spending the money.

            Which means that they would want to use it as a profit centre with regard to other hospitals, so they would sell it to other hospitals. Unfortunately, working practices between hospitals (hell, between departments within hospitals) are very different, so the package would have to be heavily customised for

        • Re:related? (Score:5, Insightful)

          by Sad Loser ( 625938 ) * on Thursday August 04, 2011 @06:50AM (#36983972)

          I am a senior doctor in the NHS and am one of many trying to unravel some of this unholy mess to work out which bits are workable.

          The obvious stuff - own a basic infrastructure, use open standards, manage contracts tightly and locally, encourage a diverse IT culture within and outside hospitals and use competition to drive down price and drive up performance - this just didn't happen. As the parent says - a centralised system specified by obsessive compulsive people who don't touch patients and with an irresistible urge to gold plate everything.

          The NHS doesn't even own the N3 network - it rents it off BT.

          We are tied down with a vast number of closed systems that will cause untold unhappiness, waste and frustration in years to come - my hospital is about to go live with CERNER, which has a Windows 3.1/ 'visual basic by a first year programmer' look and feel. It takes >30 seconds to authenticate every time you want to do anything (often)! this alone will steal many hours of medical and nursing time waiting.

          They as the parent says, the contracts were poorly specified, carved up by the usual management consultancy clowns and their mates, and then just left to fester.

          Unfortunately, the people running the whole thing were not equipped with the mental or managerial experience to make it work. There was one head of IT, Richard Grainger, who might have had a chance at doing it properly from the off, but was brought in too late when the carve up had taken place, and ran away as fast as he could. The rest is history.

          What they could have done differently?
          1. read ' the mythical man month'
          2. pay someone to re-engineer VISTA in c++/ c# / java
          3. get some people in who are successful doctors, not just the nearest beardy muppet who doesn't want to touch patients any more.

          COI: IAANHSD
          • I'd be interested it getting involved, as a Haematology SpR and an interest in open-source and open-standards. I imagined the NHS would be the perfect poster-child for an open-source project funded by the government to create a hospital system, possibly running on linux (if not the terminals, then at least the backend), which could then be used possibly by others in developing countries. It would then be modular, and updateable, and wouldn't result in the NHS relying on one piece of proprietary software,

        • It's a shame these big contracts are being allowed to tarnish all the achievements which have been made. Digital x-rays, scans etc enjoy 100% coverage across the UK; consultants can get a second opinion from someone 100 miles away in minutes, instead of sending x-rays in the back of a taxi to another city.

          Electronic referrals from GPs (family doctors) for hospital treatment are in the tens of thousands per day and GP2GP record transfers for people changing doctors are becoming widespread. All NHS sites are

          • digital radiology works, but is generally a standalone system and poorly integrated.

            GP to GP transfers - well that would have happened anyway.

            Lorenzo is totally dead in the water. Involved in product testing of modules in last 3/12 - doesn't even get to first base. hopelessly broken.

            Yes CERNER Millennium works, but is a maladapted dinosaur, with the same evolutionary potential.
          • by Rich0 ( 548339 )

            Aside from software issues, one of the major issues the suppliers had was trying to be too helpful; every hospital will insist they are somehow unique and by pandering to every possible requirement the scope of the software build simply exploded.

            Vendors do this all the time - since usually it is in their interests to do so. Most contracts are time-and-materials, so the more you shoot yourself in the foot, the more money the vendor makes.

            Why would an analyst for the vendor fight some bigwig doctor over the feature they want which will add six months and six million dollars to the project? First, they tick off a stakeholder who is used to getting what they want. Second, they make six million dollars less. Third, they make the project get six mont

            • Not the case for CSC though. They're on pay-per-deployment. When trying to deploy a single-instance system shared between many organisations with minimal tiered config scope creep is bad. Very bad.
        • he actually made us double the cost of the build, simply because that then matched his budget so that it wouldn't be cut the next year (spend it or lose it!)

          I genuinely feel sorry for anyone having to work within those constraints, and hope the policies contribute to the (relatively) quick death of their organizations. Here's how my company's budgeting works:

          1. Once a year, the boss asks what purchases we've planned for the next year or so. He uses that to estimate our budget needs.
          2. When we need to make those expenditures, we tell him. If it's a good day in the budget cycle (e.g. we didn't just host a convention a few days earlier), he authorizes the purchase. I
          • Welcome to the wonderful world of government. While I won't pretend that such things don't happen in corporations, eventually it catches it up to them, as you say. In government, you never have to worry about that. It's a wonderful thing when you don't have to depend on people voluntarily giving you money.

      • "t would be trivial to deploy a..."

        Almost every time someone says this about an IT problem, that usually means they don't understand the complexity of requirements, and you'll end up spending 10x as much as you think you will.

        I'm not defending the integrators in this case (we don't know enough about this project to say who is at fault), but there is rarely a large IT project that can be solved as simply as "throw up a data base and...".

        • The point is that a trivial system would address 70% of the needs. A massively complex system would address 100% of the needs. We could cheaply have had the trivial system deployed over a year ago. The complex system is millions of pounds over budget and still not finished.

          Which is the better solution?

          • by mcmonkey ( 96054 )

            Given that these are healthcare-related data with personal identification information, if the 70% covered by the trivial system does not include security, then there's good reason to hold out for the 99.99% solution (no complex system is ever 100%) rather than use the 70% solution in the interim.

      • Instead, the government specifies a system with an insane list of requirements for a huge number of unusual use cases, and then wonders why the project fails.

        The contractors are at fault - they bought into a project that was obviously going to fail. They could have done the decent thing, and told government that the requirements needed scaling back, but all it takes is one of the competing contractors to say it was doable and they would all fall in line. Of course, the contracts are so badly written that

      • It would be trivial to deploy a database system to store the more relevant information

        Who decides what is relevant or not?

        Everything would be recorded anyway, because most doctors are realistic enough to know that they don't (generally) know what is wrong with a patient, but are working to a working hypothesis. So, if that working hypothesis turns out to be wrong and they've discarded the data that didn't conform to their working hypothesis ... then they've got to do it all again. Plus there's serious ri

    • by mikael ( 484 )

      The Register had a discussion on this some time ago. Basically, every health board patient record system had evolved to completely different formats for ranging from basic details like names and addresses to additional information pages on medical conditions. Imagine trying to merge 200+ separate and constantly evolving online job application webpages into a single unified webpage format including portfolios and show-reels, then you'd understand what they have to do. Add to that, the standard need for consu

  • by Richard_at_work ( 517087 ) on Thursday August 04, 2011 @03:57AM (#36983074)

    Summaries actually summarised the article, and not just reposted the first two paragraphs of it...

    (The below is my opinion, not a summary of the article)

    Basically, what has happened is that the Great And Wonderful NHS Computerised Records System has been in the doldrums for so long that we have ended up with a situation where every GP (community doctor for those not in the UK, they run their own clinics outside of hospitals) and every hospital has implemented their own computer records system, with the large majority of them incompatible with each other.

    The only semblance of the NHS wide system to come to light in a customer facing manner has been the emergency care records, which is a computerised subset of your entire record meant to be accessible to every A&E (ER) department in the country - but they still haven't rolled it out to everyone, and it won't be rolled out to everyone it would seem.

    It has gotten to the point where the NHS requirements have changed so much that the contracting companies are now walking away from their contracts because they are being asked to do so much more work under the original commitments.

    This whole thing has been collossally mismanaged from the start, the current government just gets the blame for the result...

    • by Spad ( 470073 )

      The whole NCRS project was doomed from the start; they made the assumption that the best way to make clinical records available across the country was by way of a gigantic central database and the proceeded to design it without consulting any of the users, using smartcards that were obsolete before the project started (so they can't be used for anything else like SSO because nobody supports them anymore) and changing the requirements every couple of weeks. Oh, and there's no fine-grained access control so t

      • HL7, CDA, and a national private network. Problem solved. Yes, it'll be expensive - but you don't need to "define" anything because everything you need is defined already.

        In case you're wondering, I just described New Zealand's entire health system. The central government holds the demographic record, and your GP holds the medical record. Obtaining that record is a simple matter of requesting it from the GP that holds it, who will then deliver it to you via the national health network and it's received

        • by Spad ( 470073 )

          There's already a national private network (N3) so we're halfway there.

        • by mcmonkey ( 96054 )

          HL7, CDA, and a national private network. Problem solved. Yes, it'll be expensive - but you don't need to "define" anything because everything you need is defined already.

          I was just thinking the same thing, particularly in response to the posters writing that GPs and hospitals have gone forward with their own computerized system that are now not interoperable.

          For those who don't know, HL7 [hl7.org] includes, among other things, an XML schema for health care information. Let each office or organization build their own system--with a list of 'best practices' from the NHS to reduce reinventing the wheel--and use the existing standard for inter-org communications.

      • Don't get me started on SSO, my wife changed rotation yesterday, and despite not going outside of the deanery, she still had to submit all the same paperwork yet again, and pick up no less than four usernames and passwords for hospital systems...

        Four.

        Four, for crying out loud. Without ever having to leave the one building to use them all.

  • ... whose palms were greased to secure the signature of those dodgy contracts in the first place?
  • by tebee ( 1280900 ) on Thursday August 04, 2011 @04:13AM (#36983160)

    Maybe governments should start writing contracts that only pay up if a usable systems s delivered at the end of it ?

    OK know this is a gross oversimplification but at least it would give the people doing the work some decent motivation to make sure it did actually work in the end.

    I was brought in as a capacity planner on a former NHS computerization contract about 30 years ago. After 3 months there s was obvious to me that what the were doing, the very silly way they were doing it was not going to ft on the IBM mainframe they had specified to do this.

    On pointing this out to them I was told that some very highly paid consultants had said it was going to work and who was I, a lowly contractor, to question their wisdom even though this was the job they brought me in to do.

    I was asked to produce some pretty pictures and my contract was not renewed.

    • I was involved with a UK government contract years ago which followed that sort of model. It was for a system which would interconnect offices across the whole of the UK, and there was a clause that payment wasn't released until at least ninety-odd percent of the user base had access to the system. As it happens, the job got canned part-way through, leaving the prime contractor in the position that they'd spent a lot of money, but not received any payment.
    • Maybe governments should start writing contracts that only pay up if a usable systems s delivered at the end of it ?

      Sure, you can do that if you're willing to pay ten times as much for the work. And write a complete spec that never changes during the course of development.

    • Maybe governments should start writing contracts that only pay up if a usable systems s delivered at the end of it ?

      For which they'd have to pay an order of magnitude more for, possibly more than just trying multiple times and ditching the failures.

      And of course no one is going to sign up for it unless the requirement are written in stone at the start. Good luck with getting that to happen.

  • Once again, this proves anything that needs to get done, gets done, privately (doctors implementing their own electronic database) without the need of government. The government's version is more costly, inadequate, corrupt, full of nepotism and fraud. The private system does what needs to be done without the heavy hand of government, better, cheaper, faster. And all without the threat of force.

    This reminds me a lot of the essay I, Pencil: My Family Tree [econlib.org]. Anything that needs to be done can be done bett

    • Once again, this proves anything that needs to get done, gets done, privately (doctors implementing their own electronic database) without the need of government.

      Except it doesn't do what needs to be done, only the easy part of what needs to be done. It's fine as long as I only fall ill close to home, but if I need to see a doctor when I'm at the other end of the country, well fine, I can see a doctor, but they won't have access to my medical records.

      • You're missing the point. When it becomes necessary to have your records moved around like that, and the need outweighs the cost, it will happen naturally by private hands. It is the path of least resistance. Anything else is going to be fulfilling needs that are not needed (unnatural), like the government providing an education to people who are starving to death.

        Nobody is going to know the needs of a system like this better than the people who are running and implementing the system not some governme

        • by Spad ( 470073 )

          Anything else is going to be fulfilling needs that are not needed (unnatural), like the government providing an education to people who are starving to death.

          Yeah, fuck those guys!

        • You're missing the point. When it becomes necessary to have your records moved around like that, and the need outweighs the cost, it will happen naturally by private hands.

          And what would the driver for that be? I have the need, the doctor bears the cost. Sure, the the people who are running and implementing the system know the needs of the system better than some government bureaucrat, but they have no incentive to meet them. Your sort of free-market libertarianism doesn't seem to have an effective mechanism for dealing with negative externalities.

          • And what would the driver for that be?

            I'm glad you asked!

            Your need is the driver . Believe it or not, your doctor is trying to serve please you. Adding value added services like portable records do this. And draw your business away from doctors who don't implement this technology. If a doctor not implementing the technology loses too many patients they either a) implement the technology or b) go out of business. Both courses are totally natural and not compelled through the use of force.

            How many c

            • Your need is the driver . Believe it or not, your doctor is trying to serve please you.

              I'm sure my doctor is -- she seems to be a nice person. But I'm not sure I would be her choice of charity donation.

              Adding value added services like portable records do this. And draw your business away from doctors who don't implement this technology.

              My doctor has pretty much all the business she can cope with, and so has no incentive to draw more business. Besides, I would need to know when choosing a GP everywhere that I might visit in the future, to check that they had data sharing with doctors there. Doesn't work.

              What sort of strange world do you live in where you trust your life with someone who you don't trust with the money you pay them?

              Perhaps you should read the original article, and learn that it relates to the UK, and that I don't (directly) pay my doctor.

              • (or have the costs gone down significantly with Obamacare?)

                Latest I've seen show indicate that costs are going to go UP significantly with Obamacare.

                Of course, part of that is that Obamacare included Medicare cuts as part of the cost-balancing, and the Medicare cuts are unlikely to actually happen (in general, contrary to popular rumour, Medicare cuts in the budget don't happen, because Medicare payouts are low enough now that many doctors won't take Medicare patients) since that would cause even more doct

            • by tazan ( 652775 )
              It hasn't happened yet and it's not likely to on its on. I just had records from my doctor sent to a specialist. It took 6 weeks. There's no incentive for my doctor's office to speed this up. If anything the incentive is to make it more difficult so you won't change doctors.
  • by Anonymous Coward

    I guess no one in the NHS has heard of the term "Minimum Viable Product". Build the simplest thing that works and provides some value to someone, then iterate and improve from there. As the saying goes, "A complex system that works is invariably found to have evolved from a simple system that worked. A complex system designed from scratch never works and cannot be patched up to make it work. You have to start over, beginning with a working simple system."

  • Forbidding the Government to make any contract which it cannot terminate within 3 months of announcing its intention to do so.

    • No, you just need to set realistic contracts - here are some restrictions that I have been discussing with others around the military contracts debacle, where entire programmes can go through development only to be cancelled prior to the purchasing phase, meaning the development is wasted money that has still been spent.

      1. Set a fixed ceiling for contractor-driven budget overruns, something like 115% of original budget. Anything else after that must be covered by the contractor.

      2. Require every change to t

  • Governments do IT very inefficiently, they are also clueless when outsourcing but they think that they're good at it. Vendors have teams who manage deals all the time and a government agency thinks that it can draw a team together every few years and not get skinned by the vendors. It a bit like the hometown team going up against a bunch of pros.
    Internal government IT departments make these vendors attractive because they're monopolies, if the business want to wind down costs that means cutting services, th

  • by Anonymous Coward on Thursday August 04, 2011 @07:05AM (#36984070)

    As someone who was involved with the project from early on...

    The NHS really didn't know what it wanted, it just knew that it kinda wanted some sort of joined up system, and that it had a massive wodge of cash to spend.
    Result? Even when the project was years late, the NHS was STILL delivering requirements.
    Add to that entrenched company's refusing to be a part of the project and working against it from the outside (One of the biggest GP software suppliers did this), good old fashioned stupidity, and a reporting structure that was classically backwards, everyone could see it would have issues.

    The big suppliers are far more astute than government is. They could see several years down the line that the project would get canned, especially if the Tories got in, so they started building to that conclusion to the project (and turned it into a self-fulfilling prophecy).

    One last kick at everyone involved... the GPs themselves. Under the ideas of "privacy", they fought the system wholesale. Despite the system having adequate safeguards in place. The reality is that the system would make it easier to expose bad practice among HCPs, and harder to bury evidence when needed by FOI requests. You can't sell that system to the people who are using it... it would be like making politicians vote for making themselves more transparent. Never going to happen.

Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it.

Working...