Published 15 April 2020
As temperatures soared over the long bank holiday weekend, we surveyed over 1,500 people - you might have been one of them!
Just over 21% of respondents have been in a household where someone has had Covid-19 symptoms, such as fever, since the start of March.
Some of the discussion on social media suggests that there are a lot of people who think they have had the Covid-19 disease - some as early as November 2019. Unlikely as this might be, no one knows. Of the symptomatic people, less than 6% were tested - of which 35% were positive. The chart below splits this out by the type of role reported.
If these results are applied to the UK, this would suggest needing 14 million tests (to do full households) - even at 100,000 tests per day that would take 140 days. Even if we assume 35% of this wider group are found to be positive that is still fewer than 5 million positive results.* The other 9 million may need retesting in the future. And I bet someone tries to record the first 1 million records in Excel. Testing admitted patients helps to triage, manage infection control and make the best use of PPE.
Testing staff is essential - especially as they might unknowingly spread the disease.
Testing randomly in the community will help us to understand the spread of the disease.
Who to test next is less clear. The point being that antigen or antibody tests for everyone seems like a goal that might not be realistic or affordable. Given 71% of laboratory-confirmed Covid-19 people have a fever, would it not make sense to have temperature checks like they do in many Asian countries? The same ones that have more experience of managing diseases like Covid-19. And have citizens who willingly wear face masks on public transport...
* As my friend Patrick pointed out - there is a much longer discussion about specificity and sensitivity of testing and whether all tests are "equal".