Monaz: Hello, I'm Monaz Mehta, editor in the Cochrane Editorial and Methods department. The COVID-19 pandemic is creating challenges for diagnosis of the disease, as well as for its treatment and the rapid reviews being prepared by Cochrane are covering both of these areas. We asked the lead author of the September 2020 review of the evidence on using imaging tests to diagnose the condition, Jean-Paul Salameh from The Ottawa Hospital Research Institute in Canada, to describe their findings in this podcast.

Jean-Paul: Several pathways have been proposed for diagnosing and screening individuals who might have COVID-19, but there is uncertainty about the optimal role and use of imaging tests, such as X-rays, CT scans and ultrasound. However, it’s important to know how effective these tests are, not least because they may be faster and more efficient for people working in emergency departments, compared to waiting for laboratory test results.
Our extensive search for research from the past few months allowed us to review the findings of 84 studies, of which 71 assessed patients with a confirmed diagnosis of COVID-19, while the other 13 included patients who might have COVID-19. Patients in the studies underwent both an imaging test and the best available test to diagnose a current infection, the RT-PCR test, to see if the imaging tests were as accurate at determining whether or not someone really had COVID-19.
We were able to find that, for chest X-ray, 82% of the patients with a confirmed diagnosis of COVID-19 had a positive X-ray.
For CT-scans, 93% of the patients with a confirmed diagnosis of COVID-19 had a positive scan. But in the studies in which patients were suspected of having COVID-19, CT-scans gave approximately the same proportion of positive results for patients with and without the disease. 
We were only able to include two studies of ultrasound, so the evidence is much weaker for this type of imaging, but this pair of studies showed no false negative results.
Across all types of imaging, the studies we identified showed large differences in results and the risk of bias was high. For example, studies did not clearly report the criteria used to define a positive test. This means that our results are probably not representative of most clinical situations, where it is not clear who has COVID-19 and who has not (which is why we need to do the tests). It also means that the implications of the current evidence are difficult to judge at this time. However, based on our findings, we believe that one important conclusion is that CT-scans are not capable of making a distinction between patients with and without COVID-19.

Monaz: If you would like to read more about the current evidence, and watch for future updates of this review as the evidence moves forward, it’s available free to view at Cochrane Library dot com. Just go to the website and type in a search for 'imaging and COVID-19' to find it.