Chickenboy
Posts: 24520
Joined: 6/29/2002
From: San Antonio, TX
Status: offline
|
quote:
ORIGINAL: Lokasenna
More on antibody tests (serological testing).
Why Antibody Tests May Not Be the Answer
https://nymag.com/intelligencer/2020/04/antibody-tests-wont-end-social-distancing-anytime-soon.html
I wanted to keep my distance from this thread due to the political vitriol in it. I'm replying to your post, but not suggesting that you have been the origin of it. But since you posted this article, I need to chime in and say my piece about this and other ballyhooed 'health passports' and the like.
The article is accurate, albeit incomplete regarding the issue with giving someone a clean bill of health due to antibody presence. I've heard all manner of pundits saying that we should 'restart the economy' after someone has shown that they're immune to the agent (e.g., they've seroconverted). Of course, absent vaccination, the only way someone can seroconvert to a novel agent is by previous exposure to the 'wild type' or 'field strain' virus.
Let's make some assumptions, as we have incomplete information. Let's assume that the 667,000 'cases' in the United States are in fact really-o truly-o cases. 100% specificity. Now we need to assume a figure for the number of Americans that have / had the disease that were not tested for the virus for whatever reason. It's unknown really. Say we say double that 'confirmed' figure? So, about 1.3M cases domestically.
Let's also assume that everybody that has had the virus will mount an effective and detectable immune response. We know this isn't the case-there have been some instances of infected and recovered children that don't have detectable antibody titers. Very worrisome if this is widespread. Still, I'd wager that "most" people that were sickened will mount an effective immune response. Let's say 100% for the sake of argument.
The problem with waiting to restart the economy until everyone seroconverts is just a numbers game right now.
First off, the tests (or at least the ability to test large volumes of people outside of small studies) doesn't exist yet. As the article said, it may take a year to get one out there. Then there's the matter of testing and interpreting the value of those tests relative to the population at the time.
In a country of 340M people, using the assumptions above, we have a prevalence of about 0.35%. Well outside the confidence interval of predictive value for most 'quicky' serological assays. Here's what CDC has to say about use of serology to diagnose influenza:
Routine serological testing for influenza requires paired acute and convalescent sera, does not provide results to help with clinical decision-making, is only available at a limited number of public health or research laboratories and is not generally recommended, except for research and public health investigations. Serological testing results for antibodies to human influenza viruses on a single serum specimen is not interpretable and is not recommended.
We don't know what titer is protective or detectable for COVID-19 recoverers. We don't know what test we would use for detection of COVID-19 seroconverters. ELISA (as opposed to VN, CF or other serologic assays) is probably the best way to test large numbers of people, but nobody has anything like that out there yet.
A very small subset of our population will be exposed to this virus. From a public health serosurveillance perspective, this is actually more problematic than having "50-70%" of the people exposed to it over time-we won't 'get there'. So we'll have a very small number of people that have had the virus. Trying to find that needle in a haystack will require us to mass test everybody with a test that will not yield interpretational meaning. And it'll take us >1 year to get there anyways.
Earlier in this thread (first few pages?) I reiterated that diagnostic laboratory testing is best used as a decision making tool. If X result then Y decision. And you decide the next way you want to go based upon the test. But when the test is not available, real time decisions need to be made anyways. As Cap mentioned, testing delay devalued the use of the PCR test in clinical treatment. Patients that were very likely positive were treated as presumptive cases and dealt with accordingly-they didn't wait for a test before they could be treated. If the test doesn't help you make efficient decisions in real time (or near real time), then you go with your best guess.
For these technical reasons, we cannot wait to reopen the economy until we have a serologic 'health passport' for American workers. In an ideal world, we could flip that switch and have instant and universal knowledge. But the amount of time and the value of the information gleaned from such testing would not be worth returning to a job market that has withered on the vine.
We've got to do the best we have with the 'tools' available to us, IMO. Waiting for 'testing to improve' is a recipe for economic disaster.
< Message edited by Chickenboy -- 4/16/2020 8:38:26 PM >
_____________________________
|
Printable Version
RE: OT: Corona virus - 
New Messages
No New Messages
Hot Topic w/ New Messages
Hot Topic w/o New Messages
Locked w/ New Messages
Locked w/o New Messages
Post New Thread