Hacker News Reader

Comment by alpb | Hacker News Reader

Worth noting, this research is n=590. It is disappointing to see 10 months into this there’s still research conducted that's super local (likely a single strain) and low N count. I would expect we have diagnosis and symptom data for at least hundreds of thousands of cases just in the US at this point?

View on HN

It is somewhat frustrating that big data is widely used to nudge us into buying the right detergent, but somehow not to keep us alive.

There are a lot fewer rules and regs around the sale of detergent, than in the world of medical data / experimentation / research.

Oh really, imagine the counter-factual universe where we do. Where does the data come? You log on to HN and hear about how they determined that diabetes is associated with diet soda by linking your store purchases with medical records. Construct a hypothetical comment. What does it look like?

Imagine the front-page of the NYT in that counter-factual world. Imagine the outrage on Twitter. Now, choose to inhabit that counter-factual world by starting the company that does this. Good luck.

This can be done without being done by an insurance company to optimize profit margins.

For example, it's the kind of thing a pilot program in a country with socialized medicine could achieve.

> country with socialized medicine

What is socialised medicine?

Do you mean a country with affordable medicine? No need to invent some weird new names for it, we already have a perfectly fine word for that:

First world country.

If medicine is not affordable where you live, it’s a developing nation I’m afraid.

I agree, I'm Canadian :p

I'm just highlighting another advantage of public healthcare.

There’s an impressive big data symptom analysis going on in the UK through the Covid Symptom Tracker app create by a company called Zoe. They’ve been doing trials comparing reported symptoms with PCR testing.

There were many papers that came out over 4 months ago with more people and other approaches, all with similar conclusion [1-3]. Not sure why this paper is getting a lot of press.

[1] https://www.nature.com/articles/s41591-020-0916-2 [2] https://elifesciences.org/articles/58227 [3] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7386529.1/

Are there restrictions to be enrolled as a participant in these researches? I still do not understand why cannot a large group of people be studied given its disastrous imapct.

I'm just an armchair speculator so my opinion might even be slightly negative value but it appears to me that there just isn't any coordinated leadership at the national level that has tried to implement or expressed interest in implementing that.

Sounds like every nation is panicking with only political games in mind. Nobody to clearly support science with large funds.

One of the difficulties is that the countries that have good, science-led national leadership have concentrated on stopping the spread (just thinking about Finland with its national health system for instance). Countries with chaotic and uncoordinated responses, like the US, are too chaotic and uncoordinated to collect good data for research as well. There are not coordinated, science-led countries that let a lot of people get it for research purposes. Countries like Italy and the UK that got overwhelmed just didn't do this type of systematic research because they were busy. In the US some large insurance systems are doing research, but they're not necessarily incentivized to share their knowledge or data.

Reply on news.ycombinator.com