Why we need Information Sharing Orders.
Today, the Foundation for Information Policy Research (FIPR), in a report commissioned by the Joseph Rowntree Reform Trust, have released a survey of the 'database state'. The headline-grabbing part of this report is the number of databases they suggest are 'almost certainly' illegal under the Data Protection or Human Rights acts.
Well, to give credit where it is due, the provisions in the Coroners & Justice Bill that BCS campaigned against would have made them legal. That was, I think, the point. Governments cannot flout their own laws and survive, and if FIPR are correct, then the C&J Bill would have been legislating the status quo.
It's more important than that though. Beyond the status quo, there is a political imperative to reform public services and deliver on the benefits of information technology. Sharing and reusing information is fundamental to many of the aspirations of the current - and future - government.
When it comes to this e-government malarky, the basic choice is to do it properly (under current law), do it badly, or not do it at all. Doing nothing, while being time-tested government practice according to Sir Humphrey, is anathama to politicians. Doing it well may be prohibitively expensive. Imagine having to bother every citizen of the UK to try and get consent to re-purpose their existing information across so many databases. Imagine the re-engineering of databases required to track purpose and consent. Imagine the cost of a meta-system that tracks and updates information across multiple databases, propogating changes, updating errors, and maintaining a full audit.
Wow, that would be expensive - it would make the NHS National Programme for IT look like small potatoes.
So maybe there is a real logic to this; maybe doing it badly is the lesser of all the evils...or maybe I'm giving people too much credit, and it's just a bad idea. Those of us who have in our time presented business cases to senior managers can at least empathise with the civil servants tasked with coming up with a thrifty but effective plan. Would you like to be the one that suggests spending £100Bn just to do the same thing, only 'properly'?
Before I start getting flamed as a government apologist, I think it is worth reminding people that BCS opposed the information sharing provisions in the C&J Bill. The BCS view is that it is in the longer term significantly less expensive financially, in public trust and public misery, to do it properly now. No ifs, no buts.
Yet I think it is important to understand the motivations and pressures on those involved. It's hard to see things from the perspective of government when not in it, and without understanding it is very difficult to engineer the positive consensus required for real action. In the meantime, BCS will continue to promote and develop good practice, while taking action to course-correct the...less advisable...suggestions.
Comments (9)
Leave CommentImagine what happens when you're asked to check what they've got on you and it's confused with somebody else (this did happen to my nephew in the NHS system). Do they trust you to sort out the rubbish for them? Do they think you're a criminal trying to destroy the evidence? Do they wipe the other person's data from your mind? And what about all the people like me who won't give consent? Is it assumed "must have something to hide; everything to fear"? It really isn't feasible to do it well, is it?
Report Comment
I would love to see the amalgamation of databases into a nomalised approach. Unfortunatley I'm looking at one system that is to be superseded by another and the field names I have on both don't instantly correlate and no one in my investigations has provided a cross-reference document. If it's not 'easily' available for this small system then just imagine the headache trying to confidently validate the reference on table2 is the same as the reference on table1. Due to resource issues, I'm going to have to wait to see if my analysis is correct and I've referenced the correct data. ok - my problems are simplified but magnify the issue just 2 fold with millions of table entries and lots of stakeholders, some pushing a deadline, some not available and the number of problems will undoubtably increase.
Report Comment
There are techniques, technologies - and even science - for managing national-scale complex data structures. It's not easy, but it's do-able. Building on legacy systems is another thing. BCS members have practical experience of how complex this is, but it's hard to explain to those who haven't seen it. One of the common issues is 'the workaround'. The horrendous breaches of security and data protection that people on the ground end up using to fix system-wide issues. Same 'ol same 'ol!
Report Comment
It's one of those catch 22's, the Government cannot be seen to be "Doing Nothing" nor "Doing it Badly" nor "Doing it well if excessively expensive" ................ What's the answer then! Perhaps moving those goal posts again: For once I would actually give the politicians some credit as I believe they were attempting this with the Coroners & Justice Bill malarkey but unfortunately for them, and fortunately for us they got spotted: They will try again I am sure! .................... I for one would love the job of merging all their data bases into one, with the right access rights and controls on each piece of data, it would certainly take me past my retirement age and way way beyond: Just getting all the various departments to agree on what the âend result should be, would be a task in itself & a bit like trying to get kids in a kindergarten to agree on only have one type of sweet going forward, but it could be funny watching it from the outside, perhaps another "reality" hit TV series on "Yes Ministers" ....................... Still on the serious side, the report on "survey of the database state" was fairly comprehensive and really damming on the current state of play and going forward I fully agree they only real option is to do it properly and will save money in the long run: ............................. Doing it right is really only common sense and though in the "commons" I don't see a lot of sense coming out of the government: Surely an IQ test could be developed to test politicians common sense, which they would have to pass before they could stand for election, but then getting them to agree on its format, contents and how the data should be held just would make it worse, therefore I would be back to the start another "Catch 22":
Report Comment
I count myself as relatively enlightened on such matters as I work with, but not for, central government - as a consultant not a civil servant. Cross-government databases do indeed worry me but we are a long way from that. I worked in the NHS 9-10 years ago and saw the costs of poor information-sharing in healthcare. Indeed the NHS number which I helped to introduce has still, ten years later, to be made a compulsory component in patient administration systems and would have avoided the issue of identity confusion mentioned above. That's pretty basic! We need to agree what the final destination of a journey towards the industrialisation (there must be a better word but it's Friday evening) of public service information is, of course, but this report seems to suggest that any step in that direction is a bad one without saying how else we might address the very real issues we face today. I attended a large number of workshops, seminars, etc on health information sharing all those years ago and a certain report author spoke at one of them. He stood out from the crowd like an extreme right or left politician, unable to see any validity in the arguments of those elsewhere on the spectrum of opinions. Clearly this was someone on an evangelical mission, almost laughably extremist if he hadn't been taking such a narrow and fundamentalist view on a (literally) vitally important subject. Did he/does he for example know how much the NHS spends on litigation as a result of erroneous treatments and how many of these arose from inaccurate or misunderstood patient information? It's not about the £, it's about the wrong leg amputated, the wrong operation on entirely the wrong part of the body in the wrong patient... real people with real lives changed in a way no £ can put right but several billion (yes, billion) is spent on every year. I didn't know whether to laugh or cry when said fundamentalist was appointed to a prestigious post in a prestigious academic institution a couple of years later. What on earth were they thinking of? And here's the result. When I saw this story break in the Sunday Times, the aforementioned was described as the author and was the only named individual in the article. When my wife drew my attention to the article and I read the sensationalist headline I even scanned it looking for his name...and there it was! I laughed then cried...The old health information arguments recycled and extended but no more constructively than 9-10 years ago. Still, I suppose it's one way of keeping your name on the seminar and conference circuit. Are we, the citizens and/or taxpayers of this country, paying for this "research"? In whose interest? Anyone who really cared for the patients in whose name (amongst others) he speaks and writes would propose constructive alternatives to addressing the healthcare problems we face. Whether we agree or disagree with much broader initiatives such as the identity register and scheme also covered in the report, surely we can see why any discussion which includes healthcare information in its scope needs to consider the pros and cons of information sharing and focus on asking under what circumstances which data should be accessible by whom. I find it insulting to the many people who genuinely tried to engage constructively (and with creativity, tenacity and openness) with these issues over many years that credence and coverage are given to such arrogant fundamentalism by, substantially, one individual. I'm pleased the BCS played its rightful professional role in objecting to the Coroners and Justice Bill, buried in which was a cynical attempt from those at the other end of the spectrum of opinion to gain almost carte blanche data-sharing rights for government. What will it take for us to find or create the middle ground where constructive and creative dialogue can take place and, eventually, disempower the fundamentalist perspectives which seem to dominate discussions in this important area and hold us back from making real progress.
Report Comment
Many thanks for that Nota Bene. There certainly is a middle ground, and at least from my perspective BCS is trying to stay in it. The suggestion that government should get carte blanche is of a similar order of incautiousness as suggesting they should not be allowed to use information. You're also right that there is a lot of ideology involved that can get in the way of a balanced viewpoint. The trouble with an attempt to occupy the middle ground is that those on either side may not like you terribly much...
Report Comment
The problem with the middle ground is that it is defined relative to the two opposing conditions. As these are relative, so is the middle ground. It's lazy thinking to try and occupy it, because by implication it's defined by other people. The press is particularly good at this. Much more appropriate to have an opinion of your own. The problem with the data sharing debate is that we are discussing red herrings. We should ignore the practical issues and discuss the philosophical and political points. By the way, I'm against data sharing, and so, I suspect, is Jacqui Smith. "Nothing to hide,... etc."
Report Comment
Stuart, you're right that if the middle ground is defined by the arguments on either side, attempting to hold that would be a sign of intellectual weakness. Yet I don't believe that is what Nota Bene meant - and certainly not what I think. The middle ground as I see it is an attempt at balancing the benefits and pitfalls. One could take a view that government use of information is by nature coercive, anda necessary evil to be minimised. One could also take the view that information use an at best d sharing in - for example - healthcare is fundamentally beneficial, so secondary negative effects should be ignored. The problem is that neither view presents the full picture...or the middle ground if you will. I'd also suggest that ignoring practicalities and only looking at the ideals is a good way to avoid doing anything, and that too is an extreme.
Report Comment
The primary difficulty here, I think, is in conceptualizing the shared-data-user as simply 'government'. Each branch of government (and its corresponding executive arm) has its own needs and priorities that inevitably shape its approach to information gathering and use. For example, social services have a mission that is centrally concerned with health and well-being. This is a mission that benefits from time to time by being able to ask people to volunteer information about themselves (e.g. prostitution, drug use) that they simply would not divulge if they knew another branch of 'government' would get hold of it. In such cases, the quality and validity of personal information depends on the willingness of the data subject to divulge it. The picture is not just about information-sharing, poor or otherwise, in healthcare. It is about competing priorities in the governance of society and the value of separating those concerns with the technical infrastructure that supports each of them. The report suggests that the degree of availability of cross-governmental data is akin to the sanctioning of its use. What is urgently required is a clear consideration of under what circumstances, and correspondingly by what technical means, it is reasonable to bridge between information resources generated by each branch of government. This, I believe, is the big question for the 'Database State'.
Report Comment
Post a comment