NHS IT: it was just never going to work

My husband’s business partner, an IT whizz, was offered a consultancy on the NHS project as soon as it began in 2002. It’s a painful memory: “I was getting £2,000 a day. After four weeks, I resigned.” Wait, wait, wait: £2,000 a day is £400,000 a year for a 40-week year. “I resigned because it was never going to work. You could see from Day One it was never going to work. In the room I was working in, there were about 25 consultants, all working a four-hour day. Nobody in the room had any medical knowledge or experience. There wasn’t one person who had any professional understanding of what the system was meant to do.

“One of the reasons it wasn’t going to work was that it was text-only. I was taking my daughter to the dentist a lot at the time. He would take X-ray pictures of her teeth and jaw, and put them on his screen, then he’d build up a 3D picture.”

Some of you will know far more about this than I do.

Are they actually serious, that this system was always going to be text only?

15 thoughts on “NHS IT: it was just never going to work”

  1. My first question would be ‘Which part of the national programme’ but regardless the answer to the above would pretty much be no. If images are required then the relevent software would support them e.g. PACS.

    The key problem with the national programme was that its one size fits all approach was never going to work.

  2. Doesn’t surprise me, Tim. Vast computerisation projects run by external consultants seldom deliver what the users really need. It’s another example of smaller being definitely more beautiful.

    Oh, and £2000 per day – welcome to the wonderful world of IT consultancy. Not unusual for someone with technical skills that are in demand.

  3. Not unusual for someone with technical skills that are in demand.

    In my experience it’s even less unusual for someone with a BMW, a sharp suit and the ability to “talk the talk” – without the slightest smidgeon of technical ability…

  4. Actually, that article is nonsense about the text thing, really.

    Let’s say that you have a database of events, people coming to their hospital and having stuff done. So, at day 1 you have a system where the doctor can record that you have fractured a femur.

    Here’s what you do to get images in:-
    1. Buy a shitload of storage (you’re going to need it as images are much bigger than the rows in a database table).
    2. Create a mechanism to get the X-Ray images into that storage area (they’re already on a computer so you’d need to put in their patient ID and then transfer across via an upload).
    3. Put a new column on your database table to record the location of the new X-Ray image (or if you’re going to have multiple images, a new table to link all the images related to an event).
    4. Change your retrieval system to include show it.

    Done similar things before.

    And the reason they probably didn’t do that in 2002, storing images and sending them around a network was chuffing expensive back them.

    The real problem with NHS IT (like most government IT development, and quite a lot of corporate IT development) is that you can’t do large software development with contracts. Users can’t visualise solutions, the IT consultants don’t have any experience of the problems. And you don’t gain any economies of scale using these companies because IT development isn’t a factory process, so you might as well have your own staff.

    I forget who it was that hung around here, but he worked for the Scottish NHS and they built their computer system for a fraction of the price by starting small and building it up and doing it internally.

  5. from what I can see, the requirements are fairly basic. I am being deliberately simplistic here, though 🙂

    To book hospital appointments, operations etc – there are loads of such programmes around. Every hotel or online booking system has one!

    To avoid carting my file from my doctor’s surgery to the hospital every time I see a surgeon or specialist, all you need is some form of email and some form of standard input document – such as an MS Word template

    So that my GP and any other person can inspect X-rays, MRI etc, without the need to courier them from place to place, all you need is Youtube, in essence.

    From what I gather, the whole project built from the bottom up, starting from scratch and building bespoke programmes for each element, where viable alternatives already existed in the commercial domain.

    There was also an obsession with secrecy and security. I have registered for a system so that I can order repeat prescriptions online. The security and complexity of the system from the user end is much greater than that imposed by my bank for online banking. It takes longer to use than just phoning the surgery for another repeat prescription. And if I phone, I do not have to supply a PIN, my NHS number and a password!..It is nuts!

  6. Frances Coppola,

    Oh, and £2000 per day – welcome to the wonderful world of IT consultancy. Not unusual for someone with technical skills that are in demand.

    No, £2000 a day for people with technical skills that are in demand is very unusual. I work in Microsoft .net (the sort of stuff these people use) and jobs in the city where they demand the very best people and must have experience of things like back office pay something like £500/day.

    As Pogo correctly observes, £2000 is what you pay for a sharp-suited guy from one of the consultancies who barely know their way around a computer.

  7. “As Pogo correctly observes, £2000 is what you pay for a sharp-suited guy from one of the consultancies who barely know their way around a computer.”

    …then it’s high time I went out and bought a sharp suit!

  8. nick,

    …then it’s high time I went out and bought a sharp suit!

    The consultant doesn’t get to see most of that. The consultancies pay about the same as everyone else.

  9. “No, £2000 a day for people with technical skills that are in demand is very unusual”

    Perhaps. But in 1993 I was being charged out at £1100 a day…

  10. I am expecting a service engineer to visit my place of work next month. We will be charged about $4,500 per day. And if previous experience is much to go by, I will know more about his job than he does.

  11. Recording things like x rays may well be practical.
    But what of the real world where I get a bit of a twinge in my back at night. How do you photo that?
    And it might well be the early omens of a neoplasm but probably is not.
    As you get older localised ‘twinges ‘ are almighty common.

  12. A lot of GP’s still use unix/vt100 terminals. My GP has recently “modernised” and replaced their VT100 terminals with PC’s running VT100 emulation software. (For non techs – VT100 = Text based).

    As this project is so old, I assume the initial idea was to support legacy systems (text based), then later the project morphed to upgrade legacy hardware (insane cost) to a min specification to support a graphical system.

  13. Government IT gear is notoriously outdated. And doctors, despite their field supposedly being a scientific one, are notoriously bad at tech (this is true, a fortiori for MPs and civil servants.)

    If you do not have end-users interacting with developers every step of the way, then you get a system which is not fit for purpose and has to be scrapped or patched. If they are involved, it takes a will of iron to avoid mission creep, which means the system is never finished. Both outcomes are ruinously expensive.

  14. It wasn’t going to be text only (Picture Archiving and Communications Systems has been deployed), but it is true that – in the beginning at least – that the people who were supposed to use the systems were not consulted.

    “Why it was just never going to work” is because they did pretty much everything that will cause project failure:

    Lack of clear link between the project and the organisation’s key strategic priorities, including agreed measures of success.
    Lack of clear senior management and Ministerial ownership and leadership.
    Lack of effective engagement with stakeholders.
    Lack of skills and proven approach to project management and risk management.
    Lack of understanding of and contact with the supply industry at senior levels in the organisation.
    Evaluation of proposals driven by initial price rather than long term value for money (especially securing delivery of business benefits).
    Too little attention to breaking development and implementation into manageable steps.
    Inadequate resources and skills to deliver the total portfolio.

    – the most common causes of government IT project failure, NAO / OGC

    They were warned about this from the beginning.

  15. Actually, the imaging part was the one bit which did (does) work.

    The rest was always going to be a catastrophe, and the usual suspicious characters duly cleaned up at the taxpayers expense.

Leave a Reply

Your email address will not be published. Required fields are marked *