You asked

Please provide the following information.​

Application of estimated false positive PCR % rate to the published COVID-19 positive figures:

​With regard to the COVID-19 PCR test results figures, which are published by the office of national statistics, could you tell me how you apply the estimated false positive percentage rate to the data you receive. This rate of false positives PCR test results has been estimated at between 0.8% and 4.3%.

Is the estimated false positive PCR test percentage rate applied to the gross number of COVID-19 PCR tests taken or to the resulting figure of positive PCR test results or not applied to the data?

You said

Thank you for your request.

Please see Section 5 in our methods article, which details the work undertaken to understand the impact false positives and negatives could have on our estimates.

We do not apply any false positivity rates to our data, but instead have undertaken work to consider the impact on our overall positivity rates estimates from two scenarios using possible test sensitivity and specificity rates.

The results show that when we consider that the sensitivity of the test could lie between 85% and 95% (with around 95% probability) and specificity is above 99.9%, the prevalence rate would be slightly higher but still very close to the main estimate we publish in our weekly bulletin.

If we consider a lower sensitivity rate or around 60% (between 45% and 75% with 95% probability) then the positivity rates are estimated to be slightly higher again, and to both decrease faster and increase faster. This is because the more positives there are, the greater the impact of assuming that we miss ~40% of the true ones.