Rapid COVID Testing Is Not Where It Needs to Be

Robert D. Glatter, MD; Michael J. Mina, MD, PhD


January 19, 2021

This transcript has been edited for clarity.

Robert D. Glatter, MD: Hi and welcome. I'm Dr Robert Glatter, medical advisor for Medscape Emergency Medicine. COVID-19 has killed more than 320,000 Americans so far. What was evident from the outset of the pandemic was the lack of a clear national policy and strategy related to testing as a way to reduce transmission of the virus.

Here to discuss this dilemma and solutions is Dr Michael Mina, assistant professor of epidemiology at the Harvard T.H. Chan School of Public Health. Welcome, Dr Mina.

Michael J. Mina, MD, PhD: Thanks very much. I'm happy to be here.

Glatter: It's great to have you here. Over the past several months, you've been a very strong advocate for embracing a new testing strategy using rapid tests in the United States and abroad. You've also made a sound argument that the technology for doing these rapid, convenient, and inexpensive tests in our homes already exists and that the widespread use of such an approach could even be more crucial than a vaccine program.

Could you explain to our audience, for those who may not be familiar with your research or approach, what your central message is regarding how we should be testing the public in order to rein in the pandemic at this point in time?

Mina: Absolutely. The idea is really that we can stop transmission through stopping infections immunologically, and that's where a vaccine comes in. We could stop transmission from just asking everyone to treat themselves as though they are positive, by physical isolation and physical distancing. Finally, we can stop transmission by getting people to know when they are positive so that they can do more targeted quarantining and isolation of themselves on a needed basis in the sense that if you know you're positive, you can self-isolate and stay isolated until you become negative.

Now, that sounds obvious. Of course, we've been telling people that if you're positive, don't go out and see people, and so on. The problem is that most people who are positive don't know it. The type of testing we're doing just isn't frequent enough for a virus like this. When people are transmitting the virus to each other, they're generally not symptomatic yet or maybe will never become symptomatic and therefore don't recognize that they even require a test.

The current version of testing we're doing in this country (which has largely been low-frequency, sometimes symptomatic testing, sometimes asymptomatic one-off screening, and surveillance testing) generally has a sensitivity of around 5% to actually catch infectious people. We're using very high-sensitivity tests, but our effectiveness and sensitivity to actually identify infectious people are probably around 5%. We're barely putting a dent in the virus and it's growing much faster than we're reining it in.


We have other options. The option that I have been really pushing for (based on science we've done and published) is to use tests like these instead of getting a PCR test. This isn't a swab. It might look like a swab, but it's actually a coronavirus test. This is a rapid antigen test that gives results in essentially 5 minutes, costs almost nothing to produce, can be scaled to the tens of millions per day, and be delivered to people so that they have access to the test. And it can be frequently used.

The reason why frequency is so important is because when somebody gets exposed and starts transmitting, they're not yet symptomatic. If you're not testing frequently on a routine basis (say, in your home after you brush your teeth), you will not be able to catch yourself before you go on to potentially infect other people.

What I'm calling for is for us to scale up these inexpensive, highly accurate tests when the target is infectious people. They're not going to detect somebody who was infected 2 weeks ago but may still have RNA in them. They're only going to detect people who are infectious now and are at a risk of infecting other people today. That's essentially what these tests can be used for in a very powerful way.

If you have 100 infectious people right now, they will infect around 130 or 150 additional people. All we have to do to control the virus is get those 100 people to infect 99 people or fewer, and we start to get reductions in cases. If we can do that through giving people knowledge of their transmissible status, their infection status, then we're at a much better position to control this virus overall.

Glatter: A point you brought up in your recent op-ed in TIME in late November, and I'm going to quote you, is: "If only 50% of the population tested themselves in this way every 4 days, we could achieve vaccine-like herd effects."

That kind of struck me, the concept of getting herd effects from rapid testing. Can you expand on this?

Mina: Exactly. So far, we haven't really attempted to get herd effects. We've kind of danced around the question of why we are testing in this country. We never really answered it. There's this idea that we're testing in order to contact-trace, but if we're only able to realistically find 5% of people who are even worth contact-tracing, then we're not having much effect.

This new idea, which is to achieve population-level herd effects through frequent, accessible, rapid testing, can actually serve almost in lieu of a vaccine. We know, for example, that with the vaccine, we need to vaccinate maybe around 50% or 60% of the population. If we get there, we can actually stop spread across the whole of the population — not immediately, but essentially transmission will drop over a number of weeks or months and we would actually gain control.

Vaccines aren't the only way to do that at a population level. If we can have enough people empowered to know their infection status on a routine basis... A terrific example would be if somebody wakes up and brushes their teeth and uses one of these tests two to four times a week, which takes about 30 seconds to use and is very easy to do. If you were to do that, we would have enough people knowing their infection status every day so that we could effectively stop enough transmission so that, on average, 100 people go and infect fewer than 100 people.

That's what I call "herd effects." While it's individual-level responses, it ends up protecting even the people who are choosing to completely ignore the tests. Some people just won't want to do it, they don't have any interest, or there might not be enough tests, and so on. But if we had 50% of a community utilizing these tests on any given day or in any given week, then on average we would have a reduction of transmission at the overall population level such that the outbreak, instead of exponentially increasing, starts to exponentially decay.

That's the danger and scary part of epidemics, but also the benefit. They're either exponentially going up or they're exponentially going down at some rate. If we can just get that to be an R of 0.9 instead of 1.3 — meaning 90 people get infected from 100 people vs 130 people get infected from 100 people — then after just a few weeks, 100 people become 50 people become 20 people. Instead, where we are right now is that after a few weeks, 100 people become 130 people, 180 people, and before we know it, we're at 400 new infections per day after just a few weeks of just an R of 1.3.

The whole goal is just to get R below 1. The test doesn't need to be perfect. The number of people using the test doesn't need to be perfect. You can have a test with 70% sensitivity used by 50% of the people, and that alone would get R below 1 and start driving down cases instead of up.

Glatter: How do you propose to collect data? Much of the country is averse to having data gleaned from them, and obviously this is an issue. How do we measure the success of such a program?

Mina: There are a few ways. First, we have to really take a step back and ask, why do we want the data? We have trained ourselves throughout this pandemic, and others, to assume that if we don't have the data as a public health entity, then we can't act. That's because we're used to requiring the data of an infection so that we can know where to put contact tracers and allocate resources.

If everyone's using these tests, then we can actually cut contact tracing out of the equation. They don't need to be there anymore, and we don't actually need to routinely collect the data. If 50% of a community is using this test twice a week, then we would have more people automatically testing themselves on a routine basis and more rapidly than contact tracers would ever be able to contact or get to in time. We could cut out contact tracing altogether; if we really have approximately 50% of people using these tests once or twice a week, we would still be doing better than contact tracing.

That's just to set the stage that reporting becomes much less important. Reporting is sort of this thing that we want, because as public health people, we live on data. This isn't a "public health official" emergency. This is a public health emergency, which means that we need to put the public in charge. The officials in charge can only act at the speed of the public. If we can get the public to act quicker on their own without the officials being in charge, then that automatically gives a major benefit.

It doesn't stop there. We can get reporting of these tests. It doesn't look like it because this rapid test is just a little piece of paper. We can make reporting voluntary. If we don't have 1-2 million people testing every day but 20-40 million people, then voluntary reporting of even just a small fraction of that would give us more data, not less.

Glatter: What do you see as the main barriers to uptake of such rapid testing technology from our public health leaders' standpoint?

Mina: The major barrier is that massive confusion about these tests is really at the level of the FDA. To this day, the FDA continues to consider PCR the only useful gold standard for testing. They compare all of these tests with PCR, which gets very confusing for public health agencies to evaluate how to use these tests. This gets confusing for physicians to necessarily evaluate how to use these tests. The reason it's confusing is because we see the FDA packets and we also see headlines that say this test is only 40% sensitive.

That's because most of the time that somebody is PCR positive, they're postinfectious; they don't have live virus in them anymore. I don't want to call them false positives, but they're not a danger to people anymore. After 8 days of being with symptoms, you still potentially stay PCR positive for weeks. Those end up being the people who come forward when we find people with a PCR test who are asymptomatic or say, "I wasn't feeling so well, so I scheduled an appointment and I'm finally here a week later." We'll find them as positive on a PCR test and we'll ask them to isolate. They don't need to isolate anymore because they're postinfectious.

The point is that when we then evaluate these rapid tests, we're including those people in the denominator. These tests should be negative when somebody is no longer infectious. It looks like this test has 40% sensitivity against PCR. If we just change our goal to say this is a public health test; we want our goal not to be diagnostic to know if somebody has any leftover RNA in them, but to know if somebody is infectious today. This has somewhere around a 90%-99%, sometimes 99.9%, sensitivity to catch infectious people.

The FDA has yet to embrace the fact that RNA positivity should not be the gold standard as a metric for whether somebody is currently infected. They have yet to change it, and that is driving almost all of the discussion globally, frankly. The WHO has yet to embrace this. This is simple biology, but we don't have the right people in the room talking about the fact that RNA is not indicative of current infection. It's just indicative of the fact that you have RNA in your nose, which could be 3 months old.

This is actually formally recognized by the CDC, so all we have to do is take the pieces that the CDC has already said and just put two and two together, and you'd come to this understanding. The CDC, for example, says if you're infected, don't test yourself again for 90 days because you might keep testing positive. Those people who keep testing positive, we keep putting them in the denominator when we evaluate the use of this test. That is a real problem.

Glatter: Do you think a change in leadership in the upcoming administration could in some way influence how we potentially change our testing paradigm?

Mina: I think so. I think we are going to get a much more science-based group of leaders. Right now, we've had FDA leadership that generally has kind of continued things status quo, hasn't really been able to adapt, and hasn't been able to really use the current evidence around this virus and other viruses to adapt their strategies and their authorization processes.

I do believe very much that with the new administration, we're going to have science infused into the discussion. I've had many discussions with President-elect Biden's COVID task force and with incoming leadership in the CDC, and all of these individuals are very science minded. They understand complex subjects — not that many people in the current administration don't, but we've also seen this unfortunate intertwining of science-based policy with politics in the current administration.

Frankly, we've seen an administration that has actively gone out of its way to not have testing be a major part of the battle against this virus, choosing, at least at the highest levels, herd immunity as a valid approach to tackling this virus over control measures. That's been devastating and puts us where we are today, which is over 3000 deaths every single day.

Glatter: Absolutely. Your recent article in Science was fascinating. The headline says it all, that a one-size-fits-all approach to testing is not the way this should be organized. In your mind, do you feel that this is about to change?

Mina: I think that things will change dramatically with the new administration. I don't know who is going to be the new HHS secretary or the new FDA commissioner, but I do believe that they will be people who are willing to look with fresh eyes at the situation. [Editor's note: Biden has nominated California Attorney General Xavier Becerra as the next Health and Human Services Secretary.] Everyone in this country can recognize that testing is not where it needs to be. Anyone who knows a little bit about what's going on on the inside recognizes that the vaccines are not getting rolled out nearly as quickly as the White House has made it appear. Because of that, we are all in a pretty big pickle right now, to say it very mildly.

I think that we're going to see a major shift. For example, Dr Rochelle P. Walensky, who is the incoming director of the CDC, has been the chief of infectious diseases at Massachusetts General Hospital in Boston. She's an infectious disease physician so she understands these things. She had some of the earlier studies, along with some of our studies, that showed that frequent testing with these types of tests can be one of the most powerful ways to control this pandemic.

Given the fact that we have an incoming CDC director who has herself been the senior author on numerous studies discussing these tests and has vocally discussed the need for them, I think we should anticipate seeing a pretty big shift in policy.

Glatter: That's great news to hear. I also wanted to talk about an article you coauthored on the economic benefits of COVID-19 screening tests that was quite eye-opening. Trillions of dollars have been spent so far, but we're talking about very small amounts of money to invest in a rapid approach to testing. Maybe you can expand upon that a little bit.

Mina: This is, frankly, one of the most frustrating aspects to me. When you say $1 billion, it sounds like a lot, but $1 billion is less than 0.1% of what this virus has cost us, and it's 0.02% of what this virus is costing us in what we have spent on it so far.

This whole testing program, to get all Americans rapid antigen tests twice every week for a full year — and we won't need it for a full year if we actually rolled it out to all Americans — would only cost $1 billion to scale the tests up and then about $10 billion to produce enough tests to really do that for a full year.

The point is that for less than 1% of the cost of this virus that we have spent so far, we could have a tool that could have been controlling the virus 6 months ago and 1 month from now. Instead, we just keep taking this exceedingly medical, biomedical, industrial complex approach, and we're saying that the only thing worth creating an Operation Warp Speed over is a vaccine or maybe monoclonal antibodies. If we can inject it or adjust it, then Americans are all about it.

We don't take the simple, effective, cheap road, and I just can't understand why. It is the most economically sound approach. It is the only approach that's frankly on the table besides vaccines, which are liable to have to be completely rewritten if this virus mutates, and it will. It will likely mutate around vaccine immunity, whether it's this month or 5 months from now. It will likely happen once we start rolling in the vaccines. [Editor's note: The virus indeed has mutated since this discussion took place, but current vaccines appear to remain effective.] These tests won't be prone to the same problems.

Glatter: One of the things I find frustrating is that there are really no rapid tests that are available truly without a prescription. They're coming online now, obviously. Where do you see the next 6 months? Do you see more of these tests becoming widely available to the public without a prescription?

Mina: Yes, I certainly do. I think I've been pushing that issue a lot. It might end up having to be a directive from the HHS secretary to go around FDA, similar to what Brett Giroir has had to do so far to make these tests available sort of more in congregate settings and still sort of pseudo by prescription, but for asymptomatic screening.

The Abbott's BinaxNOW test costs a dollar or two to make. They sold it to the US government for $5 per test. Once that contract runs out with the US government, it's immediately going to cost $25 per test because it's prescription. They have a diagnostic claim, so they can do it as a prescription.

On the other hand, this rapid test, which costs probably 70 cents to make or less, is selling right now for $30. It doesn't make sense.

I do think that we are going to see a major shift. We're starting to see over-the-counter claims be given. The FDA is really blocking most of those because of some onerous — and, frankly, completely unnecessary — requirements in terms of the data they want to see. It's 100% unnecessary if they just looked at the science, but they haven't yet done it.

This test — I'm not trying to place products — is Ellume's COVID-19 test. It's a great antigen test and I'm supportive of it. This is the first one to get an over-the-counter claim. But to do it, they had to build in Bluetooth into the disposable test. It's got circuit boards and batteries. It's a one-off test that they could have really scaled, and it's simply driven by a rapid antigen test inside. In order to get through the FDA's process, they ended up having to require cell phones and all this stuff, and it drove up the cost.

I do think with the new administration, we're going to see simpler tests come out in much greater numbers.

Glatter: You talked about shipping people a bunch of rapid tests, along with some confirmatory tests in the case of a potential falsely positive test. That seems like a great idea to invigorate the population and start testing. Is that something that you think could happen in the next year?

Mina: Absolutely. I really do. I'm hoping it will happen in the next 3 months. I've been talking with all of these companies that are building these tests. I've been trying to act as a matchmaker to get them into discussions with the decision-makers at the FDA. From their perspective, the FDA is more or less a black box, so I've been really trying to help everyone communicate a little bit better, and help to try to design these and figure out what's the most scalable test.

I think that's one of those things where confirmation with a rapid test just makes sense. You have two different tests. People use this one every day or twice a week, and if it's positive, then you use this other one. For every 100 of these, you need one of these. If we could get both of those tests into people's homes — a 10-pack of one test with one more confirmatory test — then that immediately abrogates a lot of the concerns from laboratory physicians saying, "No way can we have all of these tests go out; all of them will need to be confirmed with PCR."

That's a simple switch that the CDC needs to just flip and say that if we just have people confirm on their own with another simple rapid test right there in their house, we'll get false-positive rates down to less than 1 in 2000. That's a totally reasonable number for a public health tool.

If somebody were to use this every week for a year, they might get one false positive that year. It's really at an individual level; it's not onerous. I think that's one of the most important things that we can do, and I do see it happening in February or March.

Glatter: Well, thank you so much for your insight and your expertise. This has really been helpful to me and to our audience. I think you really put it in a very crystal-clear perspective. Thank you again for joining us, Dr Mina.

Mina: Absolutely. I'm very happy to.

Michael J. Mina, MD, PhD, is an assistant professor of epidemiology at Harvard T. H. Chan School of Public Health (HSPH) and a member of the Center for Communicable Disease Dynamics. He is also assistant professor of immunology and infectious diseases at HSPH and associate medical director in clinical microbiology (molecular diagnostics) in the department of pathology at Brigham and Women's Hospital, Harvard Medical School.

Follow Michael J. Mina on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.