Monaz: Hello, I'm Monaz Mehta, editor in the Cochrane Editorial and Methods department. Screening programmes are in place for many conditions and the COVID-19 pandemic has led to discussions of whether people should be screened for SARS-CoV-2, which is the virus that causes COVID-19. A Cochrane rapid review from September 2020 looks at the evidence for universal screening and we asked lead author, Meera Viswanathan from RTI International in the USA to describe the findings in this podcast.

Meera: Universal screening involves things like checking body temperature or symptoms, doing rapid diagnostic tests, or a combination of these approaches in people who aren’t known to be sick. It can be done in lots of places. You might have your temperature taken or be asked about your symptoms, or even asked to do a quick saliva test, when you enter a building or before or after travelling. If the screening suggests that you may have the virus, it needs to be followed by a cascade of actions, including a confirmatory diagnostic test and the need to isolate, and having your contacts traced and asked to isolate as well. These steps might help prevent the spread of the virus, but only if universal screening approach is beneficial and our review suggests, for the moment at least, that it probably isn’t.
We started the review by searching very widely for eligible studies and found many that have been done over the last six months. We found studies on screening for symptoms, temperature, travel history, exposure to known infected persons, exposure to known or suspected infected persons, a combination of approaches, and one rapid test, known as RT-LAMP. 
Our main results come from 17 studies, that included more than 17,000 patients and showed that one-time screening isn’t accurate. For example, if 100 infected persons are screened, the virus will be missed in between 20 and 100 of them. So, at best, these approaches miss one-fifth of infected persons who could then go on to infect others. And at worst, they miss all of them!
These screening strategies also struggle when it comes to confirming that someone does not have the virus. For instance, if 100 people who don’t have the disease are screened, up to 38 of them will be incorrectly identified as being infected. In other words, more than a third of healthy persons might have to be tested further, isolate unnecessarily and have their contacts traced.
There are also some other things to consider about the studies in our review. Generally, they didn’t report their findings very clearly, so we can’t be sure about the reliability of our findings, and they compared the results of screening against the so-called gold standard test, RT-PCR, which is itself not always accurate. 
So the bottom line from our results is that these screening approaches aren’t very accurate. For them to work in the future, we need at least three key elements to be in place: first, the screening strategies need to be far more sensitive than those evaluated to date; second, there needs to be a robust infrastructure for testing, tracing, isolating, quarantining and so on; and third, the public need to buy-in to screening, so that they are willing to get screened, get tested, give contacts for tracing, and isolate if needed. In the meantime, we should continue to focus on the things that we know to work, such as face masks, physical distancing, adequate, personal protective equipment for frontline workers, and isolation if you have COVID-19.

Monaz: If you would like to read more about the screening studies that have been done, the full version of this review is available free in the Cochrane Library. If you go online and search 'Cochrane review of screening for SARS-CoV-2' you'll see the link to it.