Coronavirus (COVID-19): modelling the epidemic (issue no. 37)

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


Technical Annex

Table 1. Probability of local authority areas having more than 50, 100, 300 or 500 cases per 100K (14 – 20 February 21). Data updated on 2 February.
LA P (Cases > 500) P (Cases > 300) P (Cases > 100) P (Cases > 50)
Aberdeen City 0-5% 0-5% 0-5% 5-15%
Aberdeenshire 0-5% 0-5% 0-5% 5-15%
Angus 0-5% 0-5% 25-50% 50-75%
Argyll and Bute 0-5% 0-5% 25-50% 50-75%
City of Edinburgh 0-5% 0-5% 0-5% 5-15%
Clackmannanshire 0-5% 25-50% 75-100% 75-100%
Dumfries and Galloway 0-5% 0-5% 15-25% 25-50%
Dundee City 0-5% 0-5% 5-15% 25-50%
East Ayrshire 0-5% 0-5% 25-50% 75-100%
East Dunbartonshire 0-5% 0-5% 50-75% 75-100%
East Lothian 0-5% 0-5% 5-15% 25-50%
East Renfrewshire 0-5% 0-5% 50-75% 75-100%
Falkirk 0-5% 0-5% 25-50% 75-100%
Fife 0-5% 0-5% 0-5% 15-25%
Glasgow City 0-5% 0-5% 25-50% 75-100%
Highland 0-5% 0-5% 0-5% 15-25%
Inverclyde 0-5% 0-5% 15-25% 25-50%
Midlothian 0-5% 0-5% 25-50% 75-100%
Moray 0-5% 0-5% 25-50% 50-75%
Na h-Eileanan Siar 0-5% 15-25% 50-75% 75-100%
North Ayrshire 0-5% 0-5% 50-75% 75-100%
North Lanarkshire 0-5% 0-5% 25-50% 75-100%
Orkney Islands 0-5% 0-5% 25-50% 50-75%
Perth and Kinross 0-5% 0-5% 15-25% 50-75%
Renfrewshire 0-5% 0-5% 25-50% 50-75%
Scottish Borders 0-5% 0-5% 5-15% 15-25%
Shetland Islands 0-5% 0-5% 15-25% 25-50%
South Ayrshire 0-5% 0-5% 25-50% 50-75%
South Lanarkshire 0-5% 0-5% 50-75% 75-100%
Stirling 0-5% 5-15% 50-75% 75-100%
West Dunbartonshire 0-5% 0-5% 50-75% 75-100%
West Lothian 0-5% 0-5% 75-100% 75-100%

Epidemia

Since May 2020 the Scottish Government has been providing results computed using the Bayesian semi-mechanistic model proposed in Flaxman, S., Mishra, S., Gandy, A. et al[16].

The model is based on a self-renewal equation which uses a time-varying reproduction number R to calculate the infections. It uses the reported number of deaths to back-calculate the infections as a latent variable, with regard to the set of prevalent governmental restrictions at the time. Then the model utilises these latent infections together with probabilistic lags related to SARS-CoV-2 to calibrate against the reported deaths since the start of the pandemic. A detailed mathematical description of the original model is available[17].

The Scottish Government has recently implemented a new approach to running this model; a bespoke package Epidemia[18], developed by the Imperial College team, which has been used to model the epidemic in Scotland this week. Although some differences exist between the previous and new methods (notably a change in sampling frame), the methodology and assumptions remain broadly the same as for the Flaxman paper.

Contact

Email: modellingcoronavirus@gov.scot

Back to top