Publication - Research and analysis

Coronavirus (COVID-19): modelling the epidemic (issue no. 66)

Latest findings in modelling the COVID-19 epidemic in Scotland, both in terms of the spread of the disease through the population (epidemiological modelling) and of the demands it will place on the system, for example in terms of health care requirement.

Coronavirus (COVID-19): modelling the epidemic (issue no. 66)


1. UKHSA has now taken over the role of compiling the consensus from SAGE, based on models which feed into the Epidemiology Modelling Review Group (EMRG).

2. Particular care should be taken when interpreting this estimate as it is based on low numbers of cases, hospitalisations, or deaths and / or dominated by clustered outbreaks. It should not be treated as robust enough to inform policy decisions alone.

3. The cyan bars use Covid‑19 test data and purple bars use multiple sources of data. The estimates produced by the Scottish Government are the two on the left. (Yellow uses confirmed cases from PHS and deaths from NRS; green uses wastewater data).The UKHSA consensus range is the right‑most (red).

4. Based at Edinburgh University, Strathclyde University Aberdeen University and Public Health Scotland.

5. Coronavirus (COVID-19) risk assessment - NHS Digital

6. In January just over 16% of people testing positive were 65 or over, this reduced to just over 5% in mid-April. See issue 48.

7. Clinical at-risk groups refer to individuals with certain underlying medical conditions who are at-risk of Covid-19 related complications and for whom seasonal influenza vaccination is recommended. See issue 23 for more details.

8. Deaths, Cases and Hospitalisations from PHS COVID-19 daily cases in Scotland dashboard.

9. Vaccination and contact data for the 0-17 age cohort is not presented due to the vast majority of this age group not being offered vaccinations and the SCS excluding contacts between children.

10. All scenarios are based on current vaccine roll-out plans and efficacy assumptions.

11. The actual positive tests are adjusted to coincide with the estimated day of infection.

12. Actual data does not include full numbers of CPAP. ICU bed actuals include all ICU patients being treated for Covid-19 including those over 28 days.

13. Four week projections are provided here: Scientific evidence supporting the government response to coronavirus (COVID-19) - GOV.UK (

14. The exceptions to this are Moray, Na h-Eileanan Siar, Orkney Islands and Shetland Islands.

15. Numbers are included in Table 1 in the Technical Annex.

16. Anomalously high values, one in Seafield (Edinburgh) in mid-February (see Issue 40), one in Dunblane in mid-June, and two in Daldowie in January, were removed.

17. The temporary decrease in the number of samples tested continues this week, which impacts on the frequency of observations and the population covered by sampling. This means there should be particular caution in interpreting the maps in Figure 16 and the local authority results in Table 2 in the Technical Annex. In particular, in Perth and Kinross only one sample has been measured from the main site in Perth in the two week period between 7th and 20th August, making the map output unreliable.

18. The black line and red shaded area provide a smoothed curve and confidence interval for WW RNA that is estimated from a generalised additive model based on a Tweedie distribution.

19. Advancements in detection and interpretation practices allow us to identify when outlying results are anomalous rather than indicators of spikes in Covid-19 levels. Table 2 provides population weighted daily averages for normalised WW Covid-19 levels both with and without the outliers removed. See Technical Annex in Issue 60 of these Research Findings for further details.

20. Coverage as at the week beginning 14th August 2021.