Coronavirus (COVID-19): modelling the epidemic (issue no. 34)

Technical Annex

Wastewater Covid-19 report methodology

Samples from Waste Water Treatment Works (WWTW) in Scotland have been analysed by the Scottish Environment Protection Agency (SEPA) to detect fragments of SARS-CoV-2 virus RNA in wastewater (see section ‘what can analysis of wastewater samples tell us about local outbreaks of Covid-19 infection’). This is reported from lab analysis as gene copies per litre. Raw measurements of this concentration of wastewater RNA are affected by both the size of the catchment area at each waterworks (and hence the population served), as well as the amount of flow into the works (with high volumes of fluid flow diluting RNA values).

Hence, values are normalised by three methods depending on data availability.

• Direct flow: when measurements of flow are available, the raw RNA measurement is multiplied by the daily flow total, and divided by the population served at each site, to produce a daily value of RNA copies per person.
• Ammonia-based estimate: In some cases (especially with the most recent data), flow measurements are unavailable; if however measurements of ammonia concentration are available, they can be used to estimate flow via a statistical model. 
• Simple estimate: when both flow and ammonia measurements are unavailable, flow is estimated as the historical average for that site; when a site has no associated flow data at all, a prediction is calculated based on the population characteristics of the site. For all methods, the normalised figures are measured in daily value of RNA copies per person.

Case data are available aggregated at the Local Authority and Neighbourhood (IZ) spatial scales. Case data are associated with each wastewater sampling site by estimating the extent to which Local Authorities or Neighbourhoods (IZs) overlap with the catchment. To do this, census data from 2011 is used to quantify the population of each LA or Neighbourhood that live within each site’s catchment area. These population counts are then used to weigh case data to produce weighted averages for that site. The focus is on the daily case rate (i.e. the daily number of new cases, scaled to population), with the data presented as 7 day moving averages.

The scaling on the double-plotted graphs is automatically chosen at each site to best align the wastewater data to these local case trends. In consequence, especially high daily observations of waste water may exceed the limits of the plots.

EAVE 2 Study Group – clinical characteristics of S Gene dropout cases

The analysis of the S gene drop out data in the section ‘what we know about the risk of hospitalisation from the new variant’ is based on Lighthouse samples only and the true drop out corresponds to negative on the S gene and Ct^[11] values < 30 for at least one of the OR and N genes. A weak positive is negative for S and Ct >= 30 for both OR and N genes; all other Lighthouse samples are labelled S gene positive. Unknown corresponds to individuals who are tested in the NHS labs attached to the hospitals. Those tested in hospital represent a different population from those tested in the Lighthouse laboratory and many of these individuals tested in hospitals are admitted directly to hospital.

Counts of the individuals testing positive are presented from November 16, 2020. The analysis was carried out using all data report to Public Health Scotland by 5 January, 2021.

The testing data are linked to the EAVE study data of GP clinical conditions for a clinical and demographic description of the individuals testing positive with the S gene deletion in comparison to those who do not have this deletion. The laboratory data and GP data are then linked to hospital admissions and deaths.

A hospital admission is defined as an admission to hospital within 14 days of testing positive for Covid 19. Individuals who tested positive within two days following the hospital admission are also counted. The time from test to hospital admission is the number of days from the date of the sample to the date of admission with individuals who tested positive on the two days following admission defined as having a time from test to admission of zero days. Hospital acquired Covid infections are not included in this analysis. Hospital admission data is largely complete up to the 29 December 2020.