Diagnosis

COVID-19 can provisionally be diagnosed on the basis of symptoms and confirmed using reverse transcription polymerase chain reaction (RT-PCR) testing of infected secretions or CT imaging of the chest.

On 29 September 2020, a letter in Lancet highlighted the increasing likelihood of overestimating of COVID-19 incidence as more asymptomatic people are included in RT-PCR testing with consequent "misdirection of policies regarding lockdowns and school closures," noting that the false-positive rate in the UK is currently unknown, with "preliminary estimates … somewhere between 0·8% and 4·0%". The letter called for "stricter standards … in laboratory testing, … and pretest probability assessments … including symptoms, previous medical history of COVID-19 or presence of antibodies, any potential exposure to COVID-19, and likelihood of an alternative diagnosis."

Viral testing

The standard test for presence of SARS-CoV-2 uses RNA testing of respiratory secretions collected using a nasopharyngeal swab, though it is possible to test other samples. This test uses real-time rRT-PCR which detects the presence of viral RNA fragments. As this test detects RNA but not infectious virus, its "ability to determine duration of infectivity of patients is limited." Positive tests have been shown not to correlate with future excess deaths.

A number of laboratories and companies have developed serological tests, which detect antibodies produced by the body in response to infection. Several have been evaluated by Public Health England and approved for use in the UK.

On 22 June 2020, UK health secretary Matt Hancock announced the country would conduct a new "spit test" for COVID-19 on 14,000 key workers and their families in Southampton, having them spit in a pot, which was collected by Southampton University, with results expected within 48 hours. Hancock said the test was easier than using swabs and could enable people to conduct it at home.

The University of Oxford's CEBM has pointed to mounting evidence that "a good proportion of 'new' mild cases and people re-testing positives after quarantine or discharge from hospital are not infectious, but are simply clearing harmless virus particles which their immune system has efficiently dealt with" and have called for "an international effort to standardize and periodically calibrate testing" On 7 September, the UK government issued "guidance for procedures to be implemented in laboratories to provide assurance of positive SARS-CoV-2 RNA results during periods of low prevalence, when there is a reduction in the predictive value of positive test results."

In October 2020, the US diplomats and security officials raised a warning for Las Vegas against the use of Chinese test kits donated by the United Arab Emirates. In partnership with Group 42, Nevada was receiving donations of hundreds of thousands of test kits costing from between $15 million to $20 million, which raised concerns around patient privacy and test accuracy. Earlier, there were doubts that UAE's Group 42 could get access to US citizens through test data.

In November 2020, research showed that breath analysis could make the "rapid identification" in seconds for coronavirus possible.

Imaging

Characteristic imaging features on chest radiographs and computed tomography (CT) of people who are symptomatic include asymmetric peripheral ground-glass opacities without pleural effusions. Many groups have created COVID-19 datasets that include imagery such as the Italian Radiological Society which has compiled an international online database of imaging findings for confirmed cases. Due to overlap with other infections such as adenovirus, imaging without confirmation by rRT-PCR is of limited specificity in identifying COVID-19. A large study in China compared chest CT results to PCR and demonstrated that though imaging is less specific for the infection, it is faster and more sensitive.