BuzzyBuzz

Everything you want to know just here you will find

Coronavirus Antibody Tests—What They Tell Us #JAMAMedNews

/
/
/
4 Views

Dozens of antibody tests for COVID19 have hit the market over the last month, and this wave has been met with both optimism and confusion So let's get into what these tests are and how they might (or might not) be useful

By now there are a lot of different tests There's variation among them, but they're all testing for the presence of two types of antibodies in blood: IgM or IgG, or both Although timelines vary and people respond differently to infection, tests might detect these antibodies by 2 or 3 weeks after symptom onset So these tests are not generally meant to diagnose active infections But they might be able to tell you if you had a prior infection

So let's say you take one of these tests, and it comes back positive Does that mean that you're now immune? [Schuchat:] At the individual level, we all want the antibody measurement to be meaningful You know, each of us wants to know, did I already have this infection? Do I have to worry about it? And there's a couple of problems with that One big problem is that a lot of these tests haven't been validated yet A key issue is that we don't know what their actual sensitivity and specificity are, or in other words how likely are false negatives and false positives

And even a highly accurate test can turn up a false positive [Schuchat:] You know when a condition is rare, if the prevalence of infection for people like me is really low, then a positive might be more likely to be a false positive than a true positive The other big problem is that even if you get a true positive result, meaning you did have a prior infection, you can't be absolutely certain that you're not going to get reinfected with COVID19 While most scientists agree that you'll have some immunity, they're not sure how effective it will be or how long it will last These uncertainties mean that, for now, relying on an antibody test to tell you if you can skip wearing a mask is not advisable

But some of these risks that are associated with making individual-level clinical decisions based on antibody tests are not as pertinent to population-level serology And it's at that population level that antibody tests are more likely to play a big role over the coming months Well designed sero-surveys can estimate how many people have been infected– that's something called seroprevalence And seroprevalence is really important to quantify when talking about relaxing stay-at-home orders This is also where the concept of herd immunity comes in

[Walensky:] So let's hope and pretend right now that we believe that if you have antibody that that will bridge you to the vaccine th at we all hope will be here in a year and a half Then the question is how much of the population do we really need to have been infected before we start feeling like we can stand behind a little bit of herd immunity? The answer to that question relates to an epidemiologic term called R naught If you took an infected person and put them in a completely susceptible population, how many additional people would get infected? That's the R naught [Walensky:] So the projections or the thought about the R naught for SARS-CoV-2 is that it's somewhere around two to three I have seen estimates more recently that it could be as high as five or six

Advertisement

But let's pretend right now it's two or three You can calculate your herd immunity by having an R naught The calculation is 1 minus 1 over the R naught And it tells you how many people need to be immune to keep the epidemic in a steady state Let's break this down

If the R naught for SARS-CoV-2 is 2, that means the seroprevalence needs to be 50% to prevent a rise or spikes in infections If the R naught is 3, then the seroprevalence needs to be 67% And if the R naught is closer to five or six, you're really looking at a percentage in the 80s A quick caveat: the calculation and thresholds I described are nowhere near foolproof, but they do help us get a sense of the extent to which we can rely on herd immunity to curb COVID19 And looking at some of the first serosurveys with published results we are nowhere near 50 to 60% seroprevalence

[Schuchat:] In the New York State sample, which included New York City, they found about 15%, a bit higher in New York City But in Santa Clara county, it was just a couple percent And that is information that to an epidemiologist or a policy person suggests: we are vulnerable to a lot more infections, cause the virus is still out there, and most people haven't gotten infected yet So mask-wearing and social distancing are probably not going anywhere anytime soon Multiple serosurveys are already underway, and many more are gearing up

Scientists are deploying different sampling methods, like neighborhood door knocking or testing donated blood An unbiased serosurvey would test a random and representative sample of a population, which is hard to do But if done well, these serosurveys could help answer key questions about COVID19, like to what extent cases have been undercounted, or how have different demographic groups been affected So antibody tests are going to stay in the limelight over the next year, and (fingers crossed) they will help us make evidence-based decisions

Source: Youtube

This div height required for enabling the sticky sidebar

%d bloggers like this: