Dr view of antibody testing

amother

Thu, May 21 2020, 6:41 pm

Hi, Everyone,

As the COVID-19 pandemic continues to unfold, antibody testing will become increasingly important.

In theory, antibody tests can tell us who has been infected with SARS-Cov-2 already, and thus who may be protected or immune from future infection.

In practice, however, there are two problems with antibody testing—at least as it currently stands:

Most tests aren’t accurate enough, especially for individual use
We don’t yet know if having antibodies protects us from infection, or for how long
In this email, I’ll focus on problem #1 above, and I’ll address #2 in a future message.

The topic of antibody testing is complex and can be difficult for people without a background in science or medicine to understand, so I’m going to keep it as simple as possible here.

I’m also going to focus on whether antibody testing is valuable for you, as an individual, rather than on whether it’s valuable as a screening and management tool on a population level. Those are two different questions—with different answers.

The accuracy of an antibody test depends on several factors:

Is the testing methodology reliable?
Are the results reproducible?
Are the chances of a false positive/negative low enough?
Most currently available coronavirus antibody tests fail on all of these fronts.

The majority of test kits on the market today haven’t been approved by the U.S. Food & Drug Administration (FDA)—or any third party, for that matter. So, we don’t know if the methodology they use is reliable or their results are reproducible.

But even the tests that have been approved by the FDA and independently verified often aren’t accurate enough to be useful.

The ability of an antibody test to determine whether someone has been infected with SARS-Cov-2 is determined by its sensitivity and specificity.

Sensitivity is the ability of a test to correctly identify those who have been infected (true positives), while specificity is the ability to correctly identify those who have not been infected (true negatives).

In a perfect world, a test would have 100 percent sensitivity and specificity, and would thus correctly identify both true positives and negatives all of the time.

But that is not the case with coronavirus antibody test kits (nor is it the case with most antibody tests).

What’s more, even if a test has relatively high sensitivity and specificity, it can still return a disturbingly high rate of false positives and false negatives when the overall percentage of people who’ve been exposed to a pathogen like coronavirus is still low.

This is where it gets complex, so let’s use an example. Say only 5 percent of the U.S. population has been infected with SARS-Cov-2. Then let’s say we have an antibody test that is 95 percent sensitive and 95 percent specific. Ninety-five percent sounds pretty good, right?

Not so much.

The biggest risk with coronavirus antibody tests is a high chance of a false positive, since that would lead people to believe that they’re immune, when they’re not. This might result in behaviors—like not being as careful with hygiene or social distancing—that increase the risk of infection.

Imagine a hypothetical group of 500 people. With a 5 percent infection rate, that’s 25 people that are truly infected. But if the test has only 95 percent specificity, this means that in that same group of 500 people, it will return 25 false positives. So, if you got a positive result with this test, there’s only a 50 percent chance that it’s accurate. Basically a coin toss!

Where I live, the infection rate is more like 1–2 percent, so antibody testing would be even more inaccurate.

(If you’re having trouble grokking this, check out this article in Scientific American. It has a helpful visual that makes all of this easier to understand. This video from ProPublica is also worth watching.)

Over time, antibody testing will become more accurate.

Earlier this week, a German lab called Euroimmun AG introduced a test kit that has been approved in the United States with 100 percent specificity. This means that there is no chance of a false positive. You can see a list of all SARS-Cov-2 test kits that have been approved for diagnostic use in the United States and other countries on the Center for Health Security website.

And, as the infection rate of the population increases, even the existing tests will become more useful.

In that same hypothetical group of 500 people above, with a test with 95 percent specificity, if the infection rate is 25 percent and someone tests positive, they have an 86 percent chance of actually being positive—vs. only a 50 percent chance when the infection rate is 5 percent. 86 percent isn’t perfect, but it’s a lot better than 50 percent!

I hope this has helped you to make sense of a confusing subject.

For what it’s worth, I’m holding off on an antibody test for myself or my family until I have access to a more accurate test with a specificity of 100 percent (preferably).

In health,
Chris

You received this email because you are subscribed to Chris Kresser General News. You may update your preferences to choose the types of emails you receive at any time.

Chris Kresser LLC, a Nevada LLC / Kresser Institute | 810 E. Sahara Ave | Suite 402 | Las Vegas | NV | 89104 | USA