Dear Scott,

Test and Protect statistics

As you are aware, my team has been looking into concerns about the treatment of “failed” (or “incomplete”) cases in the Test and Protect statistics set out in the weekly COVID-19 Statistical Report, published by Public Health Scotland (PHS).

Scottish ministers use the statistics from the COVID-19 report to assess how effectively the Scottish Government is managing the pandemic. This includes considering the timeliness of contact tracing in relation to the World Health Organization’s (WHO) criteria for contact tracing: at least 80% of new cases have their close contacts traced and in quarantine within 72 hours of case confirmation. It is therefore extremely important that these statistics are clearly presented.

We have seen the recently updated publication, which I know your team was working on prior to the concerns being drawn to my attention, and consider it to be much clearer in its treatment of incomplete cases. PHS now defines an incomplete case and sets out different reasons a case is classed as incomplete, including providing a table of number of incomplete cases by reason (Table 10). The tables now separate complete and incomplete cases, so it is easy for users to distinguish between the two.

There is now more information in Appendix 2, World Health Organisation (WHO): Contact tracing in the context of COVID-19. This extra information is helpful. However, the publication could be clearer on the impact of PHS choices on the comparability of the PHS figures and the WHO target – for example, on the inclusion or exclusion of incomplete cases, and the impact of the choice of start point for the 72-hour measure.

Overall, the changes made gives users a clearer picture of how the Test and Protect system is performing and better enable them to continue to hold Scottish Government to account. I have copied this letter to Roger Halliday, Chief Statistician, Scottish Government.

Yours sincerely

Ed Humpherson
Director General for Regulation