Hobbes, our cat

Can you feel it, now that spring has come?
That it’s time to live in the scattered sun.

Waiting for the sun. Waiting for the sun.
Waiting for the sun. Waiting for the sun.

This is the strangest life I’ve ever known.
— Waiting for the Sun, The Doors

[I’m writing this on Friday, April 17. Everything in this post might be false in a few days, and conditions might be different in places other than Los Angeles. Keep up to date by checking your local health department website.]

So far, Los Angeles has been fortunate. Hospital censuses are stable. Hospital beds and ventilators continue to be available to those who need them. We have not (yet) seen the numbers of deaths and levels of suffering endured in New York. In a county of 10 million, we have counted just over 11,000 cases and just under 500 deaths. The number of new daily cases hasn’t been increasing. We’ve flattened the curve. Today, for the first time, the number of deaths is less than the day before. Things might be on the verge of improving, so we habituate to the danger and grow restless. Our kids (especially the extroverts) complain that they want to visit friends. Parents stare at the calendar and wonder if they can make tentative summer vacation plans. All of us have become children in the back seat on an interminable road trip. “Are we there yet?”

As a possible roadmap out of eternal isolation our nation turns its lonely eyes to testing. In this post I summarize the current status of testing for Covid-19 and what we can expected in the short-term.

Testing for virus

Tests for the novel coronavirus have become more widely available. In LA County over 67,000 individuals have been tested. This test checks for the actual presence of viral particles on a swab, which is collected from the patient’s nose or mouth. This is used to determine if the patient is currently infected with the virus that causes Covid-19.

The City of Los Angeles is offering free drive-through testing to anyone with any symptoms consistent with Covid-19. The tests are performed only by appointment, which can be scheduled on the website. Doctor’s offices, including ours, also have tests. So if you don’t feel well, call your doctor or make an appointment for drive-through testing through LA City.

The main drawback of these tests remains their sensitivity of about 70%, which I discussed in my last post about testing. It means that 3 out of 10 patients with Covid will test negative. So, while positive results are very reliable, negative results are not.

Antibody testing

Antibody tests (also called serology tests) are blood tests which detect antibodies produced by the patient’s immune system in response to infection. It’s important to note that an antibody test can not be used to diagnose or exclude a current Covid infection. That’s because antibodies can take several days to appear after the onset of illness. A negative test may simply mean that the patient has Covid but hasn’t made antibodies yet. A positive test may mean that she had Covid weeks ago and has recovered.

We have tests for antibodies to lots of other viruses, like HIV and herpes and measles. Unfortunately, antibody tests for the novel coronavirus are not yet ready for prime time. A few private labs have marketed tests to patients that have very poor validation. The problem with these tests is high false positive rates. That means that the test may give a positive result if the patient has antibodies to some old coronavirus strain that she contracted years ago when she had a mild cold. More reliable tests are being developed but are not yet available. Cedars-Sinai is one of the many medical centers working on a serology test, and we will offer it in our office as soon as we can. But that’s still at least a few weeks away.

So the anticipation for a reliable antibody test runs high, for several reasons. First, we know that many cases of Covid-19 are very mild and those patients don’t seek medical attention. This has led many to speculate that there may be a huge population of people who’ve already had the disease, recovered, and are now immune. They could conceivably get back to work, take the lead in patient care, and stop social distancing. They could also provide the frequently mentioned but infrequently understood herd immunity, by which the vulnerable are protected by being surrounded by people who are immune.

The second reason people are looking forward to an antibody test is that a lot of us (including yours truly) went through a miserable flu-like illness in January or February. It’s very tempting to hope that we had Covid-19. After all, even though cases of Covid weren’t officially discovered in Los Angeles until later, given the amount of travel between LA and China, and given what we know about asymptomatic spread of Covid, couldn’t community transmission have been happening in LA much earlier? If some of us have already recovered from Covid, then presumably we are immune, with all the potential resultant benefits I already listed above.

The problem with these lines of thinking is the problem with all wishful thinking. We tend to overestimate the likelihood of positive outcomes and we tend to underestimate the outcomes we’d rather not have to consider. It would be great if half the population was already immune and it would be great if I already paid my dues when I was sick in January. But that doesn’t make it so.

We have very few ways to estimate the prevalence of mild Covid in the population so far. A letter to the New England Journal of Medicine published the results of Covid testing for every woman admitted to the labor and delivery ward of a New York City hospital from March 22 to April 4. 15% of the women tested positive, even though only 2% had symptoms. That means in the US city with the greatest concentration of Covid cases, about 15% were actively infected as of two weeks ago. Mathematical models suggest that 29 to 74% of us would have to be immune to confer herd immunity on the rest of the population, that is to give the virus so few available vulnerable hosts that the epidemic fizzles out. That suggests that in all but the hardest hit cities, herd immunity will remain out of reach. The whole point of flattening the curve has been to minimize new cases and keep the numbers as low as possible. Success has meant keeping the demand for beds and ventilators less than the supply. But it has also meant that there are very few of us who are immune.

There are lots and lots of other viruses that I might have had when I was sick in January, including adenovirus, parainfluenza, respiratory syncytial virus, and metapneumovirus (which is the most fun to say). Even though I had the flu shot, I could have also had influenza. I don’t want to have had those other viruses. I desperately want to have had Covid so I could be bulletproof now, but that doesn’t make it so.

I will certainly encourage all my patients to have the antibody test when it’s available, and I’ll definitely have it myself. But we should understand what to expect (and what not to expect) from it. A negative result will definitively exclude a prior infection, but will not exclude a current infection. A positive result will diagnose a past infection with a certainty that depends on the specificity of the test and the prevalence of past infection in the population. (The specificity of any test is the fraction of people who don’t have the disease who test negative.)

Imagine if 1% of the people in LA County have had Covid and are immune. That’s 100,000 people, which is about 9 times the number of people who have had a positive virus test so far. Say a test with 99% specificity is developed. That means for every 100 people who don’t have antibodies who are tested, only one false positive results. So for every 100 people tested you would have 1 true positive (the person who actually had it), 1 false positive, and 98 true negatives. That means that half of the positive results would be false. That would make a positive result not very useful. The only way to decrease the fraction of positive results that are false would be to use it in a community with higher disease prevalence or to have an even more accurate test.

So a positive test might mean you’ve had Covid before, or maybe not. That wouldn’t give you the certainty you need to stop social distancing and potentially expose yourself to the virus.

The serology test attempts to answer a different question for individual patients and for entire populations. An individual patient is asking “Am I immune now? When I was sick last month, was that Covid?” To those questions the test might give a definitive no or a maybe. It’s unlikely that we’ll have a test reliable enough to get a trustworthy yes. But populations ask “What fraction of us have already recovered from Covid?” And if lots of people get tested, that’s a question that we might be able to answer.

But that answer might be disappointing. We might find out that we’re not there yet. We might find out that our herd is not immune. We might find out that this horrible trip doesn’t end until we have a vaccine. Until then, stay inside and stay healthy.

At the beginning of a pestilence and when it ends, there’s always a propensity for rhetoric. In the first case, habits have not yet been lost; in the second, they’re returning. It is in the thick of a calamity that one gets hardened to the truth–in other words, to silence. So let’s wait.
The Plague, Albert Camus

Learn more:
COVID-19: The Worst Days of Our Careers (Annals of Internal Medicine)
Schedule a COVID-19 Test (City of Los Angeles)
Antibody Tests For Coronavirus Can Miss The Mark (NPR Shots)
Health Authorities Roll Out New Coronavirus Tests to Gauge Infection’s Spread (Wall Street Journal)
Prominent scientists have bad news for the White House about coronavirus antibody tests (CNN health)
Coronavirus ‘Immunity Passport’ Stumbles in U.K. (Wall Street Journal)
Universal Screening for SARS-CoV-2 in Women Admitted for Delivery (New England Journal of Medicine)

My previous posts about the novel coronavirus:
Of Masks And Meaningful Measures
Updates From The Socially Distant
Testing, Testing
Novel Coronavirus FAQ Part 2 – Pandemic Hullabaloo
Coronavirus Frequently Asked Questions
Community Transmission Of Novel Coronavirus In LA County
What You Need To Know About The Novel Coronavirus