Health records of 8 million people, never been done.

Despite also having to maintain and serve tons of content, Amazon, John Lewis, Ebay, and a myriad of other companies seem to have achieved such workload.

What he really means is "Noone's ever done it with under the counter agreements with politicians, requiring delivery of the product, AND the employment of thousands of otherwise unemployable types, and a job on the board when they're deselected.

I know a dozen web dev/designers (that's designers in the sense that they design things rather than "decide what it's going to look like" people who've started calling themselves webdesigners,) who could, each of them, to a man, write such a system single handedly.

This is not brain surgery, (not that brain surgery's actually brain surgery either it's just a vocation with an historically artificially high profile, like law,) but the reality is that polticians are stupid worthless dipshits, (with law degrees.)

As a consequence, what do we expect but failure?

Give this man a medal.

... It's not whether or not you can write a program capable of storing and delivering the needed information. You have to be able to convert all the different data from it's current format into the new one.

Ok, ok, ok, you never know. What this man could be referring with "best of breed" and boasting about success could be just an entangled spaghetti of seemingly complementary applications that hold incomplete pieces of information.

Those environments usually have an array of interfaces sending and receiving information all day, with all the usual problems associated with synchronization and integrity.

But yes, each one of those as an individual system is "best of breed", but is a short term fix that does not solve the long term problem.

I'm not saying that the uber-database approach is better, just have seen the "best of breed" failing too many times, usually victim of the "pragmatic" approach. Just gimme something that works in six months, we'll take care of solving the really difficult problems later (read write a lot of interfaces).

The problem is harder than the poster above thinks, but the US Veterans Administration solved it some years ago, and the software is available free, and in a more product-like form from http://worldvista.org

The problem isn't mainly software though.

is that health record data can be reticulated among all health service providers and related parties, with the normal provisos of informed patient consent, etc. This is all about the semantics of the data, we'll assume (heroically) that syntax is not an issue. What applications store it, display it, analyse it, whatever, don't matter a sod as long as it's bang up to date and available to the clinicians, etc, who need it, when they need it without having to twiddle their thumbs waiting for it.

It's people like you that contribute to the negative perception of our industry. I'd like to bet your one of those serial over-promise / under-deliver types.

Lets face it, if you think about it for more than a second, its a very ambitious project and just plain hard.

Amalgamating thousands of different record and data structures from inconsistent digital and paper based sources; even deciding on the scope of a standardised schema would be challenging.

Securing systems, access and records in enviroments prone to data mismanagement and poor IT management. The catastrophic effects of any system or data failure.

Personally, i'd rather not get involved in any project that has lives resting upon it. But hey, if you fancy it.... I hear there's an opening since Fujitsu left.....

Quite apart from the question of whether a single database could work technically, how would it work in terms of clinical governance and patient management? (not sure about Amazon and eBay - records contain a lot of free text and fuzzy data..)

http://shorterlink.co.uk/14497

Different organisations and clinicians have different record needs - hospitals have flat narratives, GPs have highly structured and Coded records, used for cradle-to-grave patient care and QOF - the Quality and Outcome Framework used for performance related pay.

Can the two needs be served by a single record? who will be responsible for maintaining it, who will be liable for errors and what plans are there for managing prescribing - an area risky for patients?

If any of you have any information of how these issues were addressed when the idea of a single record for every patient - the Detailed Care Record model, either Lorenzo - involving all organisations - or Millennium - limited, AFAIAA to a single secondary care Trust - was first mooted - and contracted for - please tell the rest of us!

And I would trust Amazon with my health care records because, hey, they manage book sales and returns well enough so why not?

I would have to point out that at least in *theory* health data management needs a rather higher level of security and availability (to say nothing of a wider distribution system and more complex ACL) than your average bookseller or yard sale management system.

Desperately Ill Person (DIP): "I'm having an emergency..."

Hospital (H): "I just need to ask you to confirm a few things for us. Your mother's maiden name please."

DIP: "Gaaaaasp"

H: "I'm sorry that's not what I have on record. Can you reply to this confirmation email that we'll send to your email account?"

DIP: "Erg."

H: "I'll take that as a yes. You should receive the email in 20 to 30 minutes, but the servers have been acting up a bit today because we were adding the ability to cross-sell insurance. So it might be a bit longer than usual."

DIP: ...

It's not just a case of client demographic data (Such as address, contact information, age, name, date of birth) but also the clients GP and GP history, a history of the clients address, the Health Authority/Electoral Ward associated with the address, the clients immunisation history, appointment history, hospital stays, bed allocation, transfer history, consultant information. The software needs to track which HCPs are in charge of what clients, it needs to show a comprehensive diary for the GP to manage appointments. It has to manage referral data, mental health sections, special needs and associated data. Above all it needs to manage exactly who can access the data, and in some cases, what specific time of day the data can be accessed.

Try doing that for 8 million people.

When it comes to migration data from legacy systems into the new ones, every single data item needs to be checked and verified. An immunisation code in one system must be correctly mapped to an imunisation code in another, for example. Everything needs to be clensed and verified, for all the data mentioned above.

Again, try doing that for 8 million people.

Posting anonymously because I work in this sector, and my bosses have probably read the article.

I've only heard 2nd hand about the "MESS" and the moving goal posts "Fujistu" put in.

I can see what they are trying to achieve and possibly how it should be done..

But with all the multiple layers of information about each person it does get a bit more complex.

So anyone care to actually figure out how to solve it, data integrity & data security being priorities of course, although only part of the over all solution.

I mean don't we all have a unique NHS ID from Cradle to grave anyway, there is a start point.

Speaking as someone who used to work in healthcare IT but escaped before the current fiasco started...

In the beginning was the NHS number.

And it came in 23 different formats, and there were duplicates which had been issued historically.

So a project was created to allocate everyone a New NHS number.

And it ate a lot of time and effort for everyone.

And when it was rolled out... d'you know what happened?

In the early stages... the systems managed to issue duplicates again.

And as far as I know they are still there.

I also used to work in interfacing healthcare IT systems with one another, and was involved in one of the earliest EDI projects in the NHS, to send pathology results from Hospital path labs to GPs.

And just at the point where we had a few pilot hospital sites sending results electronically, using a standard message, to a few GP practices...

Someone at the Department of Health got a bee in their bonnet about the European Standardisation Effort for medical messaging...

...and suddenly started mandating that everyone should implement a new, different set of standard messages - without any additional funding of course - which set the cause of healthcare EDI back a good couple of years.

Then just as that was sorting itself out, some idiot at the Department of Health realised that all this old-fashioned EDIFACT stuff was rather old hat, and there was this new thing on the block called XML, and another argument started over why everyone was using EDIFACT and we should all move to XML as the encoding scheme, quietly ignoring the fact that no message schemas had been defined in XML yet, and we had already changed horses once causing great upheaval.

So the current balls-up is basically just the logical conclusion of that sort of stupidity, with the addition of Tony Blair behaving like an idiot, compounded by the fact that there is now no in-house NHS IT capability since the organisations which used to provide in-house NHS developments were all outsourced to the likes of EDS and CSC many years ago when Virginia Bottomley was the Health Secretary.

Nothing changes...

The NHS is a bottomless pit of money.

"Connecting for Health" are proud to state that the NPfit Contractors only get paid when they deliver, but how many payments have been made for interim software solutions installed whilst waiting for the troubled final solution and who will be picking up the bill to replace the interm software once the product is finally ready considering we have already paid once.

It is a bold project, a waste of public money and an unworkable solution and they are all to blame, the government, the CFH, the PCTs managers on the ground who grab all the money they can from the project, so they do not have to use there own budget replacing old IT equipment, they can waste their budget on something else. The LSPs who have been contracted to hold a monopoly on their area and stifle competition altogether.

Does anyone know if iSoft as an LSP actively promote Emis software, or vica versa? I would doubt it very much. This project was supposed promote competition by splitting up the country into sections, all that it has done is provided a monoply for the LSPs in the areas they have been given.

Now heres an idea, the NHS develops a backend system, but the patient holds their own data at all times, where ever the patient goes the data goes and can be read by the NHS backend. This needs to be worked on, give me 6 billion and I’ll get right on it.