NHS IT must spend a fortune to save a fortune, says McKinsey The NHS must bolster its IT budget by up to £8.3bn over the next five years – in order to produce savings of possibly £13.7bn, said the management consultancy McKinsey. Get this right and the prize is huge; the £30bn NHS funding shortfall – estimated at £30bn by 2020-21 – could be cut by a quarter. But, get it wrong ... The …
No, but then has any management consultancy or SI, but then McKinsey generally stay well away from actually delivering anything. McKinsey provide gravitas more than anything else, and as a lot of senior politicians have worked for them, they are generally believed.
The sad fact is that the NHS CTO/CIO function should already know that, but most trusts have appalling IT based on cheap and old kit.
Queue some comments on the money wasted by Spine.
The NHS record for IT projects costing >£1million is woeful, with only 25% delivering at all.
For my institution, the successes with appointment saving come from the staff phoning those with appointments the day before (or the friday before the monday) to confirm that they will attend. This reveals the many who did not receive any appointment letter or got a letter with an error on it. Don't expect a new IT system to deliver any savings over the current ones - it is a combination of how they are used, and the psychological/disease profiles of the patient population that limits their effectiveness.
It'll be a total cock up because it always is. Of course a large chunk of these non-savings will also go to McKinsey for coming up with the idea of making non-savings in the first place.
I don't work for the NHS and never have but I happen to be familiar with one piece of software they use. Essentially it keeps track of patients undergoing a specific range of tests. When the project was set up the NHS went out to tender for writing the software plus five years of maintenance, the tender process was done quite well but by someone who didn't understand software or at least the way licensing works because the software remained the property of the company that wrote it.
Now fast forward five years and it's time for a maintenance contract renewal, by law this has to go out to open tender. The trouble is no one other than the original company can realistically tender for the contract as they are the only ones in the world with the software. Anyone else who wanted the contract would have to write the software for free as there will never be the budget to write it again. So much for competition.
My question to the person now running the project: why didn't the original tendering process insist that the software was open source or at least owned by the NHS? It seems it's just not done that way.
Whilst ive been working for the NHS migrating off xp (!) ive found dozens of old obscure bits of software for medical puposes - usually to integrate medical equipment with computers written by small , often long dead companies. Often they've all been buying each other out and changing name all the time. It seems to me different trusts have buying the same (Crappy) software time and time again when they could easily have bought the tiny backstreet company.
especiaslly if they have BILLions to chuck around
You have to remember that specialised medical equipment (especially those with robust technologies; autoclaves for instance) fall into the same category as fixed industrial plant: they're made to last by companies that disappear once they've been sold to everyone that wants one, they are integrated into the fabric and require demolition work to get out, and - most importantly - they still work! Unfortunately that usually means that controlling software HAS to be what the manufacturer supplied and has to run on the manufacturers choice of platform and will never be updated.
Oh, and it's easy to keep up to date when you have billions, it's the little bit of a couple of hundred (or less) software that's hard to get funding for.
"they could easily have bought the tiny backstreet company."
But that would have involved headcount, HR and whatnot. HR would then probably end up making the only people who understand the code redundant because they don't fit into NHS pay scales - or pissing them off so much the just leave.
> My question to the person now running the project: why didn't the original tendering process insist that the software was open source or at least owned by the NHS? It seems it's just not done that way.
Two main reasons:
1) The cost of supporting future changes.
If ownership transferred to the NHS upon go live, then the developer has no way to accurately determine support costs as the NHS might change something significant a couple of months in. And since the NHS are required to only go with fixed-price contracts, that's a very risky position for the developer to be in.
2) Intellectual property
The developer might have implemented some whizzy way of booking an appointment (or whatever) that they want to retain rights to. (Or at least not advertise to their competitors.)
> If ownership transferred to the NHS upon go live, then the developer has no way to accurately determine support costs as the NHS might change something significant a couple of months in. And since the NHS are required to only go with fixed-price contracts, that's a very risky position for the developer to be in.
But they change things anyway, and get royally ripped-off because of the way the contracts are written. Government negotiators in general (and NHS in particular) don't seem to have learnt this depite having had the lesson several times.
> The developer might have implemented some whizzy way of booking an appointment (or whatever) that they want to retain rights to. (Or at least not advertise to their competitors.)
Not relevant - if it's developed by a contractor for a client, the IP belongs to the client unless the contract explicitly says otherwise. And why would it?
This would be fine if it were a case of spend £500k, save £1m, 16,000 times over. If it didn't deliver after the first few such projects you could cut your losses and run. If it did work, it would pay for itself in short order and not need much in the way of actual upfront investment.
But somehow I expect it's not that simple...
What's your alternative - to leave it broken?
It's clearly not an impossible task, to build a functional IT system for healthcare. The failure must have had a cause. If the cause can be identified and a programme run under best practice principles it should therefore deliver.
From talking to people who were involved it seems a large part of the problem was a continually changing set of requirements - major decisions being overturned on a regular basis. Wary that such a thing was likely the suppliers contracted on a time and materials basis rather than on a cost for the job.
Problems like that are a sign of poor governance or of external interference. Fix that, spend the money, make it better.
The NHS costs a fortune they got that right. And more and more money gets thrown at it but somehow ends up funding management, failed IT projects and labours glorious PFI deals. At what point did the NHS give up on healthcare?
I was amused at the recent report blaming the last 5 years for the NHS to suddenly be rated one of the worst world healthcare systems. It was a problem before the last 5 years and it is hard to pour in more money (and to the right places) when there is a recession and everyone demands some more of the pie. The monolith will consume as much money as we throw at it and still come short no matter what we do.
Eventually the NHS will either collapse due to its excesses or get back to focusing on healthcare and funding healthcare.
It's one of the most efficient and effective comprehensive healthcare systems in the world.
It lacks sufficient funding to operate as well as it could. The UK spends less than our neighbours - almost all the problems would be eliminated by matching those funding levels; 9.6% of GDP in the UK versus 11.6% in Germany and 11.9% in France. A tax rise to reach that level of funding is the cheapest available option.
People moan about the NHS; well there is a lot to moan about. Virtually everybody has a story about things that go wrong, money that's wasted, ridiculous levels of bureaucracy, clueless management etc. etc.
But it's still a pretty damn good service for 99% of the population. The majority of people get treated reasonably efficiently, and at a price that is still way below what many other countries spend. I know that there are numerous things that could be improved; but just occasionally, it might be nice to hear some more of the positive side.
A colleague made a comment that he thought it would be a good idea to change the NHS on Jan 1st next year to a completely private model for the next 12 months. Anything you want, you pay for until Dec 31st. People might then realise the real value of the NHS and appreciate it a lot more.
> People might then realise the real value of the NHS and appreciate it a lot more.
Maybe people would realise that the fortune we pay in tax (sorry, "National Insurance") is poor value! Actually, I think people know the value, it's the cost they have a problem with.
My proposal is that the NHS caps the salary of all non-medical staff (HR managers, accountants, ...) at the level of a newly-qualified Nurse's salary. (No need to go as high as a junior Doctor's salary, definitely.) That way, at least there would be an excuse for the managerial cock-ups they keep making.
We do tests for NHS and private healthcare. I took a phone call yesterday, large nhs hospital after a result. I could confirm it had been analysed and posted to the department a couple of weeks ago.
Gist of callers comment on it not being in records " we don't do digital here, we rely upon patient notes [big folder full of paper]"
I faxed him a copy of the result.
Some areas where a bit of IT spend might help. That said I'm not sure if our clients went entirely digital if we'd be up to sending entirely digital (as opposed to some kind of pdf) reports.
We are where we are 'cos the NHS builds bespoke for just about everything. There is a reason for this and it's a bloody silly one.
There is plenty of good medical software for all aspects of running large healthcare organisations available off-the-shelf. The trouble is that all these ask one simple question up front, globally universal to healthcare[1], whenever adding/updating a patient or their care. How are they paying?
Try to deploy anything like that in the NHS and the unions scream blue murder about "creeping privatisation". Bunch of bloody spanners that they are....
[1] And if the NHS actually asked it, like everyone else, "health tourism" wouldn't even exist. (Heavy hint for the terminally thick: The patient's NHS number goes in this field.)
Really? I'm not sure where you get your information from, but the key pieces of software in the NHS for primary and secondary care are:
iSOFT/CEC Patient Centre (plus a number of other hospital patient management systems)
GE Centricity PACS (plus a number of other radiology RIS systems)
EMIS Patient Management (plus a number of other GP patient management systems)
A number of systems for use in speciality areas, e.g. various operating theatre management systems, Euroking for maternity data, BadgerNet for neonatal data.
All of these are owned by profit-making companies and sold to the NHS for a tidy profit. To my knowledge, they're all mutually incompatible, requiring further expensive "interface" software to do semantic mapping when data is moved from system to system. What's absurd is that the NHS is unable or unwilling to design its own systems to replace the above, and sell them to public and private providers in other countries for profit to help fund our own system.
A health care service provider (which some of the NHS is) and a software development company should not be confused. I've met many "IT" people in the NHS, but they don't have any real developers there - nor should they.
You could stop NHS technology failures by buying software from UK software developers instead of consultancy companies.
That may be necessary but it's not sufficient. You can't fix a broken organisation with technology.
So I would not personally invest with McKinsey on this.
... includes advising Enron on corporate finances.
My personal experience of some things McKinsey did for my company was a questionnaire which was poorly worded, and which I could not guarantee would give the same results twice, took over an hour to complete (in Q1 2014) and we've still not had the 'results'.
They seem all to have firsts in PPE from Oxbridge, never have had a job they actually needed or where they had to clear up their own mess and suffer the actual consequences of their own mistakes and have no knowledge of the pain their 'recommendations' inflict on the people doing the actual work of the organisations they advise.
Any current or former McKinsey staffers reading this, please, please correct me (if I'm wrong).
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
