The daily tally of infections seriously understates the actual number of infections, because if you are sick with Covid19 today but had it any time in the past (even last spring) your new bout is not included in the daily dashboard figures.
ITV News’s political editor Robert Peston said in a tweet that the number of Covid-19 cases being reported each day on the government’s Coronavirus Dashboard “seriously understates” the real number, because people who catch the disease more than once are not counted after the first time.
While it is true that these reinfections are not counted as new cases on the dashboard, this is unlikely to mean that the daily case number “seriously understates” the number of cases each day.
According to figures published by Public Health England (PHE), there have been 23,105 “possible reinfections” of Covid-19 in England up to 4 July 2021. PHE defines a possible reinfection as one person giving two positive tests “at least 90 days apart”.
These are not definite reinfections, however. PHE says: “For a possible reinfection to be categorised as confirmed it requires sequencing of a specimen at each episode and for the later specimen to be genetically distinct from that sequenced from the earlier episode.”
This is often not possible, however, since not all positive samples are genomically sequenced.
The proportion of cases that are suspected reinfections changes from week to week, but the overall rate has been estimated by a PHE epidemiologist at around 1%.
On the day that Mr Peston posted his tweet, 29,173 new cases were reported in the UK on the Coronavirus Dashboard. If a further 1% had been detected as reinfections and added to the total, following the pattern in England, this would have amounted to an extra 290 cases or so.
It is possible that reinfections may become more common over time, if immunity from past infections wanes, or if new variants of Covid prove more likely to reinfect people. (Indeed there are indications that PHE may monitor reinfections more closely in future.)
Reinfections aside, it’s true that the number of cases reported each day does underestimate the total number of actual infections in the real world, since some infections are asymptomatic and not everyone gets tested.
Overall though, it is not right to say that the reported daily case total “seriously understates” the total number of detected cases, as the current evidence suggests it would be only 1% higher if reinfections were included.
Mr Peston has since clarified that his concern was that “the quality of the dashboard infections data was not what it should be because of the policy driven exclusion of those who have been infected a second time”.