Rapid-result coronavirus tests aren't as accurate as those processed in a lab, but some experts say they could play a big part in bringing this pandemic to an end. One advocate for more rapid-result testing is Dr. Ranu Dhillon, an instructor and global health physician with Harvard Medical School and Brigham and Women's Hospital. Dhillon spoke with WGBH All Things Considered host Arun Rath. The following transcript has been edited for clarity.

Arun Rath: In broad strokes, can you explain how rapid-result tests could make such a big difference in getting this under control?

Dr. Ranu Dhillon: So essentially, anytime you're dealing with an epidemic where there is no vaccine, you really have two routes to try to control transmission.

One is prevention, and in the context of COVID-19, that's social distancing, wearing masks, many of the messages we've been talking about. The second is identifying somebody who is infectious, capable of infecting another person, as quickly as possible so that the period of time where they might be able to expose others is limited and that person be put in isolation before they're able to infect other people. It's really that second area that we've struggled with because of the fact we've not been able to detect as many cases as we need to and as quickly as we need to to really close that window of transmission.

Rath: And a window of testing that takes even 24 hours to come back sort of robs you of that opportunity?

Dhillon: Yeah, exactly. So the PCR testing that we're using currently to diagnose patients is limited in terms of having a slow turnaround time. The second part of the challenge is also that it can't be scaled to test the millions of people that we really need to be testing in order to identify enough transmission chains and break them off before they spread further.

So the PCR can work well when you're testing a circumscribed number of patients, but when you're really talking about trying to test across a population, which you need to do when the virus is widespread as COVID-19 is, then you need to use some other modality of testing. And going past that, the PCR is also limited, even with some of the innovations that are coming out around saliva samples and other things that are in the news, it's still limited by the fact that you would need to collect samples and take them to a centralized lab.

So for a number of those reasons, PCR, as we experience with Ebola and Zika in the past, can scale to really screen across a population. And that's where these rapid tests come in, that are not just fast in terms of getting results, but can be used very easily, can be decentralized widely, and can also be used frequently. So if you miss somebody at one moment, you catch them at the next moment. In terms of being able to test more than once every few weeks, you can actually test every couple of days with a test like a rapid test.

Rath: You mentioned Ebola and Zika, and I understand you have been involved in rapid result-testing, including with Ebola epidemic. Can you talk about how that makes a difference and how it made a difference there?

Dhillon: Well actually, the tragedy of it is that we were pushing for a rapid test to be used for Ebola, which as with COVID-19, existed at the time of the epidemic, but wasn't used for many of the same reasons why there's been pushback on using the rapid tests for COVID-19 now. And the challenge with Ebola was, as with COVID-19 in the U.S., transmission was widespread, meaning anyone from among the population could be carrying the virus and there's no way to know. The only way to know would be to test. PCR could not scale to do that testing.

We were really pushing for the rapid tests to be used during the Ebola epidemic in West Africa. Ultimately, that did not happen other than outside of a few small scale pilots, and the reason that didn't happen was because of concerns over the accuracy in terms of not having false negatives, not having false positives.

But what was overlooked then is that in addition to accuracy, the ability to test widely -- meaning covering a lot of people -- the ability to test frequently, and the ability to identify people who are carrying the virus and be able to pass it on to others quickly, really reducing that window of transmission, becomes just as important when you're dealing with a population-wide epidemic.

And I think the lens that we've looked at that testing through has been too much about what we would do for an individual patient in a doctor-patient kind of scenario rather than in a public health scenario, looking at an epidemic that's spreading across the population.

Rath: Where you've written about this, you have stipulated that accuracy is an issue with the rapid-result tests. But when it's being done on such a large scale, does that make up for that margin of error?

Dhillon: That's correct, and that's really the thinking behind it. The only way we're going to be able to find enough people who are contagious at a single point in time and be able to isolate them so that they're not infecting and propagating more of the epidemic is to really be able to test widely, get those results quickly, and then isolate people quickly.

And so the test that can do that may lose something in terms of accuracy, but really, the strategy has to be one that is not just about replacing PCR with these tests. It's really, this is an add-on to the test that we're doing now to diagnose people in the hospitals. When you're deploying a test like that, it's really just to be able to have a strategy for how you're going to do it. So you expect that there may be some cases that may be false positives, but you need to have a plan of what you're going to do to be able to confirm those cases and have the capacity set up to do that.

So, as with all things, there is no silver bullet for any epidemic. What there are are tools that if you deploy in the context of a systematic response, where it's choreographed in terms of how something's being used and what the response is going to be for whatever downside that tool introduces, you can then really make a big impact on transmission and then mitigate the downsides of maybe missing some cases or having some people test false positive.

Rath: We've heard about how U.S. military academies will be employing rapid-result testing in an aggressive fashion as they reopen. I'm curious what you think of that in the context, more broadly, of opening schools and colleges this fall.

Dhillon: Yeah, I think when it comes to opening any kinds of facilities where you're going to have large numbers of people congregating, I think first and foremost we have to get the level of community transmission far lower than where we have it in much of the country. No matter what kind of rapid testing or other tools and strategies you have, really those work well once you're actually getting the virus brought down to very low levels.

Second to that, once we get that amount of circulating virus down to low levels, that's where tools like the rapid testing make a lot of sense. You can use it to test a lot of people quickly, as we just described, but the other key will be when you're opening schools and other facilities, you're going be able to test frequently. Should someone become infectious next week, it's not that you're overlooking that they might be carrying the virus because they tested negative the week before.

And in addition to that, if you do have situations where somebody who is carrying the virus initially tests false negative one day, the fact that you'll test them again the next day or a few days later, you'll likely catch them then. And so a strategy of that kind can really make sure that we're catching cases quickly and able to do it in a way where we can start to increasingly reintroduce different parts of our life and our different institutions can start functioning the way they used to.

Rath: We've talked about how we can respond with public health here in Massachusetts, but then there's the broader country. Given how much trouble we've had with coordinating any sort of national strategy for testing as a whole, do you think that what you're calling for -- how possible or likely is that on a countrywide scale?

Dhillon: Dealing with any epidemic that's as widespread as this one, there's going to be no easy or simple solutions. And so no matter what, we need to brace ourselves for doing the hard work of implementing whatever strategies are going to be worth pursuing to actually counter this epidemic.

Going past that, I think a lot of the barriers so far, unfortunately, have been issues marred by politics. I think if we can figure out a way to get around those politics more effectively and actually start having coordination between federal and state governments, having coordination across states as some states are now taking the initiative to do it on their own, I think we can actually go really far in terms of both fast-tracking the deployment of rapid tests like this, as well as actually coordinating in a fashion where we're able to implement effectively on the ground.

There's so much we can learn from each other in terms of states that have done certain approaches well. I think we're missing a lot of that opportunity because of that lack of coordination that you mentioned. But I really think that can turn around, and it's really just a matter of, I think, some changes in terms of how we've been approaching the epidemic politically. And I think going past that, really rising to the challenge.

This is not going to be easy, but it also is something that's very doable. Generations before us have done far greater and more difficult things when they faced similarly enormous challenges.