You are hereBlogs / dr-no's blog / Nice Try, But No Cigar

Nice Try, But No Cigar


Posted by Dr No on 24 February 2014

care.data_.jpgThe front door, or rather back door, tactics of the shadowy Health and Social Care Information Centre have achieved a sort of slow-burn blowback over the last few weeks. Kicked off with a junk mail leaflet that aimed to be funky but turned out flakey – eye catching shape, double helix on the front, but the helix unravelling on the inside, and written by a Kafkaesque we who never said who we were, in opaque prose that bizarrely got a Crystal Mark for Clarity from the Plain English Campaign – the idea was to have patients default into allowing the NHS to hoover personally identifiable GP medical records into a vast data silo the size of Russia, generally for the purposes of improving care. But that was only part of it. Buried in the flyer, we also had plans to flog off your data, including sometimes personally identifiable so called red data, but only after the strictest approvals, you understand. Or at least we did. Whether the rest of us did was another matter.

Any talk of NHS Inc flogging off red data was always going to be a red rag to the privacy bull. Once we’s price sheet for your data – note the clever use of service charges: the data may be dirt cheap, but the service isn’t - was discovered, along with a scoping paper that openly discusses using the flogged data to enable ‘insurance companies to accurately calculate actuarial risk so as to offer fair premiums to [their] customers’, the balloon went up, and the cigar, if there ever was one, went out. Red faced officials from we admitted to a PR disaster, and announced that the big data turn on, originally slated for the spring, would be put back six months to the autumn.

Like most who have dabbled at the epidemiological end of medical research, Dr No is ambivalent about the so called care.data project. The scientist in Dr No knows that general practice data is pure research gold. Unlike hospital episode statistics, which are more often than not generated from data entered by low paid operatives, much general practice data is entered by none other than doctors. And then there is the comprehensive nature of the data, across time and population, unbeaten in the world. Truly, the data cup overfloweth. If ever epidemiologists were going to have an orgasm, it would surely be on seeing a set of such cups.

But epidemiological orgasms are only part of the story. The data may be pure gold, but what about the container it is to be kept in? Is the vault secure? And then what about rogue traders with access to the gold, or Kafkaesque elements in government who see nothing wrong with helping insurance companies rig their premiums? Whether by accident or design, sooner or later, the vastness of the data means leaks will happen; not might happen, but will happen. Your personal data, once uploaded, stays uploaded. What if, at some point in the future, a grotesque Idiot Duncan Smith style government decides to join up your care data, which will of course include how much you smoke and drink, with your eligibility for NHS treatment? Absurd? But then, isn’t the eye-wateringly arrogant idea that the default position be that your data is uploaded without your consent also absurd?

The PR crash that caused the data pumping on an industrial scale launch to be put back six months tells us a lot. So too does we’s thinly veiled threat in the flyer for those thinking of being disagreeable: ‘Your choice will not affect the care you receive’ – good, so that’s all right then, isn’t it? Hello? And then there is the arrival on the same day as we’s NHS branded flyer of another NHS branded flyer from Specsavers, a reminder of how far the NHS is already in bed with the private sector. But perhaps the most emblematic of all is the bastard name given to the project: ‘care.data’. Call Dr No old fashioned, but the punctuation mark ‘dot’, the full stop, is a mark that conveys not a marriage but a discontinuity. It marks the end of one thought, and is followed by a gap, before the start of a new thought. The link between ‘care’ and ‘data’, it seems to Dr No, is broken.

So, in his whimsical way, Dr No sees the name care.data as emblematic of we’s – the shadowy un-named outfit behind the project – careless approach to our data. The subliminal unravelling helix image on the inside of the flyer carries the same message: that here is a project that sooner or later will unravel. The data.gold may tempt the scientist in Dr No, but the realist in Dr No also knows that all that glitters is not gold, and that, even worse, all too often the pursuit of gold ends in tears. On balance, a centralised database of personally identifiable medical data is not a good idea.

Sorry guys, nice try, but no cigar, not now, not ever.

8 comments:

I think Witchdoctor's MBC will have a fit when she sees that eye on the front of that leaflet..

Something that gets overlooked in the great data market is store cards. They record exactly what you eat and drink and in the near future I could certainly see a merging of that data with data from GP practices. Tesco has been cosying up to GPs for a while, setting up GPs in their stores. It's a bit too cosy for some of us..

Julie - this post's image is adapted from a cover design for Kafka's The Castle. Putting aside the tricky question of how much an image has to be altered to become a new image (and so avoid copyright infringement) DN's purpose was to link, for those who care to make the connection, care.data with Kafkaesque themes.

Dr No suspects that when it comes to eye-balling, WDsBC is capable of supernaturally long stares that can outrun anything we mere mortals can achieve. Tim Kelsey will need to get the Optrex out if he takes on certain cats.

Not for the first time, Dr No wrote and posted and then did a sweep round round the web, only to discover what others were saying, including that Telegraph story on not potential but actual use of NHS data to 'refine' insurance premiums. JT has also written an excellent essay on what he calls care dot data, which includes the important question of data quality (GPs probably do do better than underpaid NHS data clerks but the data is still a very long way from perfect, and may even be misleading). Elsewhere, epidemiology's Einstein has adopted a characteristically priapic position. He's certainly got the hots for the data, but whether that article amounts to a premature cigar is not entirely clear.

Meanwhile, Prof B of the London School put in a good word for care.data on the R4 lunchtime news yesterday. Apparently, he's also going to take a pop at Sir Jar's (note the Kelsey connections) HMSRs on File on Four tonight (8pm R4). Looks like things are hotting up in the dry old world of medical data.

Dr No avoids store cards, which he considers to be reverse Trojan cards (they offer a gift in return for smuggling out your personal data). The cosying up of commerce with the NHS is everywhere. Dr No has already mentioned the NHS branded Specsavers flyer which arrived on the same day as did the infamous 'your choice will not affect the care you receive' leaflet. Personal medical data is already collected on the high street and in shopping malls under an NHS banner...

Wonder if you have read this Dr No...
http://www.pulsetoday.co.uk/20005572.article#.Uwx9f-N_sdR

I too am ambivalent as I was of SCR...I opted out of SCR - but I wonder if I was actually opted out...GP practice are very sluggish in letting me know.

Even if I was to trust care.data now, believe all I read and not opt out (which I will) - would an equivalent of Clause 118 be a very real possibility in the future...?

My family and friends have opted out. Chatting to a GP friend, I asked if any kind of indicator came up on the computer screen to show this patient has opted out. Nothing on screen, also though the whole thing is said to have been put back for 6 months, doesn't mean the data is not being uploaded. Would that be true ?

Anna :o] - Dr No had seen that Pulse article. As one commenter points out, there may have been only 31 approvals, but approvals can cover tens of thousands of individuals. Nonetheless, of the four highlighted examples, three are academic, and one is regulatory (albeit Can't Quite Cope, so one does wonder). Dr No and no doubt most others are fully behind good research properly done. Indeed, as Prof B helpfully pointed out the other day, we have been doing such research for decades to good effect. What Prof B didn't point out, presumably because it took the wind out of his pro care.data argument, was that we have been doing this research without care.data for decades. We don't need care.data to do good research. There is even, Dr No suspects, a risk that gaining access to such unsorted overwhelming quantities of data will impair the research signal to noise ratio. We may find ourselves drowning in precise answers to lots of wrong questions, yet lacking even approximate answers to a few right questions.

The care.data project has a large element of obfuscation to it. Legitimate academic use is mixed up with commercial access. And Dr No thinks you are absolutely right about about the clause 118 risk (clause 118 is JC's post-Lewisham hospital closure clause). Who knows what this or a future government may legislate for? Once your personal medical data is out, it's out, the gene genie's out of the bottle, Pandora's box is open, you name it. For thousands of years wise folk have warned us that letting some things out is not a good idea.

Anna sans :o] - others have noted that the six month pause echoes the FF pause over the HSCB. Whether the data pumping has already started 'in readiness' is a worrying point. As your GP friend may have added, GPs are in something of a quandary. On the one hand, their professional duty is to confidentiality, yet their legal duty is to obey orders at all times, specifically 'Under the powers of the Health and Social Care Act 2012 (HSCA), the Health and Social Care Information Centre (HSCIC) can, under certain circumstances, require Personal Confidential Data (PCD) from GP practices without seeking patient consent' (sic). At least one GP has already been 'hit with [a] contract notice over plan to opt all [of his] patients out of care.data', as Pulse put it. One can quite easily imagine that the 'team' that sent the contract notice would see absolutely nothing wrong and a lot right with getting ahead with the government's care.data agenda. Apparently 'the team' later admitted it had got ahead of itself...but that is precisely Dr No's point.

Discovered via a comment on Pulse - Brilliant! The future if care.data goes ahead:

http://www.youtube.com/watch?v=RNJl9EEcsoE

Greetings from an old sparring partner known to GCHQ, but aren't we all?

http://www.redmolotov.com/catalogue/tshirts/all/gchq-always-listening-ts...

Editor's note: single typo corrected. Excellent to hear you are alive and kicking!

Editor's note: editor's typo corrected!


Add a comment...

Will show as anonymous if no name added

If added, your name will be a link to the address you enter

If left blank, first few words of comment will be used

• Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li>
• Web page addresses and e-mail addresses turn into links automatically
• Lines and paragraphs break automatically
 

NOTE: Dr No's spam filter can be somewhat overzealous. If your comment has been wrongly rejected, Dr No apologises, and asks that you let him know (via Contact Form in side-bar). Many thanks.