Ancestors

Written by Hadley Beeman on 2024-10-21 at 22:15

Lots of discussion about digital health records in the NHS today. Modernisation, and privacy. But here’s the thing — we DON’T need a massive new database. Here’s how we do it. 1/

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Written by Hadley Beeman on 2024-10-21 at 22:16

I see my GP, that generates notes on my GP record. My GP has those.

My GP sends me for blood tests at the local hospital. The local hospital has those results.

I get referred to a specialist in secondary care. That specialist’s trust has my info from those interactions. 2/

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Written by Hadley Beeman on 2024-10-21 at 22:17

This is how things work.

This is also how the web and internet work. For example. I open my email client, and it pulls down — in real time — email from multiple accounts stored in different organisations. 3/

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Written by Hadley Beeman on 2024-10-21 at 22:18

There is NO technical reason why I couldn’t have an app (say, the NHS app) that pulls all my info from everywhere I’ve been seen. There is no reason my GP (who needs info arranged differently to me) couldn’t have software that does the same. 4/

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Written by Hadley Beeman on 2024-10-21 at 22:19

There is no technical reason that specialist — who might be across the country — couldn’t open their software and pull down my latest blood test results, ordered by my GP at home. Live. In real time. Giving them up-to-date info AND saving the cost of repeating the tests. 5/

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Written by Hadley Beeman on 2024-10-21 at 22:19

This is a SOLVED PROBLEM. We know how to do this. It’s a big transformation, and requires some national infrastructure (so the software knows where to find my info across the NHS network), but it’s eminently do-able. 6/

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Toot

Written by Hadley Beeman on 2024-10-21 at 22:20

This approach:

• gives everyone real-time data to make decisions on

• leaves the “source” of the data in charge of their DPA(GDPR) responsibilities

• mimics reality. Which is always the best system architecture. 7/

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Descendants

Written by Hadley Beeman on 2024-10-21 at 22:21

So please, enough calls for one monolithic database. We know those are exorbitantly expensive, complicate life for all users, stifle innovation at the software layer and create massive privacy and security risks. Think: web and internet. 8/8

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Written by Amelia Bellamy-Royds on 2024-10-22 at 16:54

@hadleybeeman How are you imagining that a health provider's software would know where to get your records from, if not via a central database? Are you expecting that every patient would have a steady primary care provider who would be forwarded references to every specialist & pharmacist & urgent care visit?

Not trying to be contrarian. I understand the risks of large central databases. And experience in Canada is that projects to create them tend to be boondoggles. But I can't see how else.

=> More informations about this toot | More toots from AmeliaBR@front-end.social

Written by Hadley Beeman on 2024-10-22 at 17:31

@AmeliaBR You’re a web person, right? I would think of it as an analogy to DNS. What central “routing” would we need for those use cases? I’m not sure yet whether I’d recommend a central look-up table type thing, or if we’d want something more distributed like DNS root servers. But I the danger of tying each person to a specific provider — like a GP — is that it turns the GP practice into a sys admin. Which isn’t in their core skills.

=> More informations about this toot | More toots from hadleybeeman@w3c.social

Written by Amelia Bellamy-Royds on 2024-10-22 at 17:36

@hadleybeeman OK, so you're seeing a kind of bare-minimum central database that would hold the references to other records, but not the content of them?

That's probably a reasonable compromise. If the central database has a data breach, it would have a list of which health providers I've used, but not my actual details. And it would be a lot easier to implement than a single database that needs to handle every possible type of medical file.

And definitely agree, GPs shouldn't be responsible.

=> More informations about this toot | More toots from AmeliaBR@front-end.social

Written by Hadley Beeman on 2024-10-22 at 17:46

@AmeliaBR Yes, exactly. But I like how you’re thinking this through — I think that’s precisely what we need to do to work this all out.

=> More informations about this toot | More toots from hadleybeeman@w3c.social