ORIGINAL: Chickenboy
ORIGINAL: geofflambert
Not knowing is a problem. Not testing is not an answer to that problem. Random testing could at least provide us with scientific wild-ass guesses but we're not even doing that. The testing we are doing is ad hoc and incomplete. New York recently added thousands of deaths as being "likely" related to the coronavirus. "Flattening the curve" buys us time, but only if we actually do something with that time, like produce more tests by a factor of 1,000 or more. Just waiting until everyone eventually is infected and the survivors may or may not be immune before we start the world turning again is not an answer. If you're telling me that we just can't test everyone, you're telling me we didn't vaccinate virtually everyone against smallpox or that we really didn't almost eliminate polio. When we go to our doctors our medical record is checked to make sure our vaccinations are up-top-date. We regularly get tested for things like colon polyps and prostate problems. This will all get to be routine eventually. The problem we have to fix is getting to where it's routine sooner, way sooner than we're going at the current pace. We need to get to the place where any doctor anywhere can order tests and get them in a timely fashion from LabCorp or wherever. Diagnosis ---> treatment. Modellers are trying to model what is actually happening. Critics are checking that modelling against invalid data. What's the use in that?
Demanding 'more testing!' for the virus (e.g., PCR testing) from the general populace doesn't matter for stemming the epidemic curve. And (with the exception of the triage example above) it hardly matters for the clinical patient outcome.
All that aside, I do think extrapolation of 'attack rate' data can help clarify a population's probable prevalance or at least get closer to ground truth. But that sort of testing isn't being done anywhere that I can think of. But that's another post for later.
Here is why testing is important. Asymptomatic carriers.
We make predictions about virtually everything using incomplete data. A sample size that is representative enough of a general population to "see" the infection rate on that population can help determine policy around measures to open up the economies of all of these countries. If we don't know more about actual infection rates (Ro if possible but this may not be possible to determine accurately) then we can't take useful measures.Contact tracing once cases reduce will also be needed to slow or shut down new outbreaks.
Right now countries are flattening the curve and beginning to open some parts of their cultures and workplaces.
Should all people be required to wear masks? Which masks? Only surgical? Or more?
Should all people going to work stagger travel times and work hours so that there are less in the office at any one time and less traveling at any one time?
Should restaurants and bars be allowed to reopen with some modification of existing seating or would this be both transmission ineffective and economically difficult (fewer patrons)?
Should schools open as normal or stage weeks for different ages to reduce numbers in the building?
Should all healthcare workers be tested regularly, like each week, to protect them from developing a severe infection?
I can't see any reason why extensive testing would not be an essential part of opening up any country or region after the case curve has lessened to the point reopening is being considered.









