‘NHS’ data grab

I believe the Covid-19 Data Store is already something close to this, and it’s also what Palantir have built for the NHS - in a “you give us data, we’ll build the thing for £1” deal, which was probably a bad deal as the data was infinitely more valuable than the modelling AI. And the NHS version is that you can “zoom in” to the individual healthcare details - they claim this is needed.

I’d rather not have it, personally.

Maybe I’d feel differently if I had experienced or was experiencing a rare disease that doctors needed to share knowledge on to figure out how to treat it, but I’m not and I believe that the line of “data sharing helps the NHS provide you with better care” is mostly nonsense as run-of-the-mill stuff isn’t treated any differently.

If you break your leg, you will be referred to a hospital for an X-ray on the leg and to get a cast. It’s not rocket science for that to be done, and no extra insight is going to be gained from the hospital being able to see that you had a bout of pneumonia in 2007.

But what if you are one of many people that had historical pneumonia and have since had broken limbs aka, pattern. Who are those people? - match. Where do they live? - match. What medication do they take? - match. Is it near a retired energy generation plant? - match.

This level of data is absolutely necessary in our progress as a race. Look at 2020 for the absolutely-incomprehensible-levels of side-swiping that nature (or, conspiracy theories considered) can unleash.

I’m not intending to be argumentative here, but big data can reap big outcomes. It’s the control of the big data that needs fixing. Blockchains again then…

6 Likes

This seems fair to me.

I’m entirely for the data but I’m entirely against fake anonymisation and said data being sold to Palantir and Google. And so have very much opted out.

3 Likes

I mostly agree with you, and I’m happy to engage in this discussion as it’s a fascinating topic that doesn’t get discussed enough.

I totally agree with you on this, especially.

My point is really that, at the moment, due to a mix of the technology not really being there (the technology is possible, but not well implemented in the NHS) and the NHS’s very poor track record on data security, I don’t believe the balance tilts in favour of me personally sharing that data widely.

And even if I did, it would be almost impossible to glean anything since I don’t take any medication regularly, hardly ever use a GP, and don’t have an extensive medical history of different things anyway - so little to “join up” in terms of dots.

Would I want the dots joined up if it could produce something? Possibly, but only in a very private way, and I wouldn’t want that to crowd out the opportunity to speak to a doctor about it without the “recommendation” of the system clouding their judgement, in case the system was wrong. I think doctor-patient relationship is very important and I wouldn’t want that to be impeded upon too much.

With things like family history of genetic diseases, just having the patient mention that already gives some of these positive effects. I don’t know if extensive data analysis really would be that transformative.

2 Likes

But what if you’re unconscious and they are about to use a drug on you that you’re incredibly allergic to?

The end result, for me, would be OK as I have no known allergies anyway but I like to avoid the need for drugs where I can, to avoid side effects.

Being not sure, they would therefore use the drug only if needed, but if they did I would be fine.

You can choose to share just this kind of information anyway with a Summary Care Record (with basic information).

I believe in being an informed patient when it comes to this kind of stuff, so if you have been prescribed a medication which is known to interrelate badly with other medications, or whatever, then maybe you should have SCA at least on. Otherwise, I don’t think there is much of a risk.

Me too. So there’s a pattern there - how do the (perceived) ‘heathly people’ exist - location, age, limited medical history - links to immediate and extended family health history? It’s a massive historical resource to help us all through the unchartered territory of a wellbeing future.

But it must be anonymized.

1 Like

I would guess that location does have a lot to do with it, as some areas of the country are poorly catered for and subsequently overall health is probably worse there.

If not caught early, most health conditions do get worse so early treatment is key.

There’s also the point that some people’s health condition may have been missed or misdiagnosed, for whatever reason, so won’t show up. There will also be people who are silent patients, who don’t realise they have high blood pressure, etc but probably could benefit from treatment.

I haven’t bothered to opt out because the data is probably already out there. I hardly ever go to the GP/hospital (maybe twice a decade), but my details are probably already available to VirginCare and other contacts.

Unfortunately it’s likely that data will leak through a third party contract or one of the thousands of NHS mini projects that they get local web development agencies to build.

I agree with this.

And I’ll point out one thing that stands out in the case of this transfer is that the NHS Digital already has a secure access portal which addresses a lot of the concerns around potential risks. It’s just that, in most cases they don’t use it and just sell the whole data load in bulk, and the DoH is refusing to make usage of the secure portal mandatory. For privacy campaigners this has been one of the biggest issues around the transfer.

3 Likes