Estimating the proportion of COVID-19 cases detected by testing
A report on different methods used to estimate the proportion of cases identified through testing during the pandemic, providing insight into how the coverage of testing progressed.
Applies to England
Documents
Details
The report describes the development of a metric to estimate the proportion of COVID-19 cases detected by the testing service. Policy changes are plotted along the timeseries of the metric to explore whether any patterns emerge. Trends and relationships between the proportion of cases detected and other data sets were explored, including testing volume and prevalence.
The most important factor influencing the proportion of cases detected appears to be the level of testing. In addition, there was some alignment between peaks in polymerase chain reaction (PCR) use (which, in contrast to asymptomatic lateral flow device (LFD) testing, was more focused on symptomatic testing) and peaks in the proportion of cases detected. Notably, after the widespread introduction of LFD tests, the proportion of cases detected was consistently higher than it was previously.
There was alignment between the volume of contacts and the detected cases estimated. However, this could be more a function of these 2 measures being intrinsically linked than a reflection of a causal relationship.