3 ms·

Not sure if NHS has this concept, but in other health systems, the NHS equivalent (an economy that operates a health system o.b.o a government) is a set of serv

Not sure if NHS has this concept, but in other health systems, the NHS equivalent (an economy that operates a health system o.b.o a government) is a set of service providers, with data custodians at the edges. Custodians hold accountability for health information privacy, where service providers are accountable to a custodian. It all rolls up into these entities. Health admins tend to forget that the relationship they are scaling is between physicians and patients, and it is not the government managing the vetrinary system for a person farm. This health system as proxy for public policy issue is dangerous. If you see my previous comment on this thread about objections to data collection, an opposing view of another health tech and policy expert would be really valuable to the discussion.

dkarp 5y ago
That's more or less how it ends up working in the NHS, and I guess it's Conway's Law playing out again rather than being intentional. Health is particularly difficult because it's both data that needs to be accessible by a large number of different people/organisations and also about as sensitive as you can get. Any attempt to make the data more secure necessarily slows access, but making the data more accessible also makes it less secure. Having seen the quality of some of the systems holding this data on the edges, I would rather see a central database and a lot of funding go into the technology of that system. Why? Because I agree with your other post, that technical controls are the only solution. Controls that could make it impossible for the data to be misused or leaked, or at least make access auditable. That's where the research and funds should be spent. Ideally, that could be done by every data custodian at the edge, but I just don't think that will happen. It's easier to solve the problem in one place than in 10s or 100s or 1000s of places. In the UK, there are lots of central data repositories holding pretty sensitive information. I think it's fair to say that the government wouldn't have much problem finding your health data, along with detailed census/tax/internet/phone/travel data, if they decided they wanted to.

dkarp 5y ago

That's more or less how it ends up working in the NHS, and I guess it's Conway's Law playing out again rather than being intentional. Health is particularly difficult because it's both data that needs to be accessible by a large number of different people/organisations and also about as sensitive as you can get. Any attempt to make the data more secure necessarily slows access, but making the data more accessible also makes it less secure. Having seen the quality of some of the systems holding this data on the edges, I would rather see a central database and a lot of funding go into the technology of that system. Why? Because I agree with your other post, that technical controls are the only solution. Controls that could make it impossible for the data to be misused or leaked, or at least make access auditable. That's where the research and funds should be spent. Ideally, that could be done by every data custodian at the edge, but I just don't think that will happen. It's easier to solve the problem in one place than in 10s or 100s or 1000s of places. In the UK, there are lots of central data repositories holding pretty sensitive information. I think it's fair to say that the government wouldn't have much problem finding your health data, along with detailed census/tax/internet/phone/travel data, if they decided they wanted to.