[Amygaz]

It's not a question of cost at this point. It's a question of availability of reagents and consumables, materials to make those, and time it takes to ramp up. The supply chain is like a supertanker.

At first we didn't have enough test, but that ramped up quickly, then we didn't have enough RNA extraction reagent, now we don't have enough swabs, and soon is the collection media that will be missing.

There has been a call for more protective equipement. But as the situation gets worst, beds space, ventilator, and then staff are going to be the bottleneck.

If you google you'll find stories of people helping their local hospital by 3D printing parts for for respirators, and swabs. There's nothing cost effective about that, it's not fast, but it gets the job done. If you were to scale that, you would be competing for the same materials than manufacturers using injection molding or plastic dipping are.

[ksrm]

> At first we didn't have enough test, but that ramped up quickly, then we didn't have enough RNA extraction reagent, now we don't have enough swabs, and soon is the collection media that will be missing.

Where are you getting this information from?

[icegreentea2]

For RNA extraction: https://cen.acs.org/analytical-chemistry/diagnostics/Shortag... for an example.

For swabs - here's an example in action up in Ontario Canada: https://www.ctvnews.ca/health/coronavirus/ontario-limits-who...

No one is out trying to actively hide this information, but its spread does seem a little uneven.

I also don't want this to be fearmongering. Capacity and supply will grow (hopefully fast enough). Alternative solutions will be validated. We will keep bumping into problems, and we'll keep solving them (cause we'll need to).

[tonetheman]

We do not have tests.

https://covidtracking.com/data/

We have tested 150K people.

We have 327M people in the US, we have tested 0.00046% of the population.

We do not have tests.

[sp332]

Yeah but about HALF of those tests were done in the last TWO days! Capacity is coming online very rapidly now and while we are starting from behind, we will not be so short on tests for long. Roche alone is shipping 400,000 tests per week and Abbot Labs is on track to perform one million tests per week in the next couple of weeks. https://arstechnica.com/tech-policy/2020/03/america-is-final...

[shalmanese]

The important number is the ratio of tests:true case count. You need ~1000 - 10,000X tests/case to do true contact tracing because you want to a) blanket the country with tests and test every single suspected case even if 99% of them end up not being COVID 19 and b) proactively quarantine every person that a positive person has come into contact with and then test each of those people every single day for 14 days until they come up negative. If a single of those people come up positive, then you want to proactively quarantine all the people THAT person came into contact with, rinse and repeat.

It's great that tests are finally ramping up but the growth in testing can never exceed the growth in cases if cases are doubling every finite number of days. The choices are a severe lockdown where R < 0.5 until the true case count drops to near 0, a non-severe lockdown where R ~1.2 - 1.8 (Hubei's lockdown had an R of 1.3 in the early days until they figured out centralized quarantine could bring it down to 0.4) and you just continue the lockdown indefinitely, or "herd immunity" where R stays at 2.4 until pretty much everyone in the country gets sick.

Once you get your true case load down below your test load, then you can think about using non-pharmaceutical interventions (universal masking, universal temperature screening, mandatory hand washing when entering into a gathering space etc.) along with rigorous contact tracing to ensure that any future outbreaks remain small and contained.

[sp332]

I agree with wanted to blanket the country in tests, but you don't need 1000 tests to do contact tracing for one person, even normally, let alone during social distancing.

You don't need to test people every day for 14 days.

[shalmanese]

You do because as soon as that person is confirmed positive, then you need to contact trace and quarantine that person's contacts. Every day that goes between a confirmed positive and a contact trace, the more that person's contacts are walking around, spreading it asymptomatically and making the cluster even more out of control.

[mjewkes]

0.046%

You didn’t multiply by 100 to change from per unit to per cent.

Still bad though.

[tonetheman]

awwwww man ... i suck at math. thanks!

[DennisP]

The FDA just approved an antibody test. One blood sample, results in fifteen minutes. The manufacturer thinks they can produce enough for 200K tests/day.

https://www.biospace.com/article/releases/20-20-bioresponse-...