We regularly nowcast and forecast COVID-19 infections and deaths. This information feeds directly to SAGE sub-group, Scientific Pandemic Influenza sub-group on Modelling (SPI-M) and to regional PHE teams.

The work uses a transmission model (Birrell et al. 2020), data on daily COVID-19 confirmed deaths from PHE (by NHS region and age group) and published information on the risk of dying and the time from infection to death, to reconstruct the number of new COVID-19 infections over time; estimate a measure of ongoing transmission (R); and predict the number of new COVID-19 deaths in different regions and age groups.

Today we have published a new report on our real-time tracking of the COVID-19 pandemic.

Here we have highlighted the latest findings and provided interpretation of what these findings mean. We have also explained our recent model and report changes.

---

Updated findings

1. Our current estimate of the daily number of new infections occurring each day across England is 41,200 (28,500—58,700, 95% credible interval).
2. The daily number of new infections is particularly high in the Midlands (13,500 infections per day). Note that a substantial proportion of these daily infections will be asymptomatic.
3. We predict that the number of deaths occurring each day is likely to be between 205 and 317 on the 5th of December.
4. We estimate R_t to be around 1 in all regions. The probability of R_t exceeding 1 is above 80% only in the South East, while it appears certain that R_t is less than 1 in both the North East and Yorkshire and the North West
5. The growth rate for England is estimated to be down to -0.01 (-0.02–0.01, 95% credible interval) per day. This means that, nationally, the number of infections has stopped growing with clear decreases in the North West and North East and Yorkshire.
6. London, followed by the North West, continues to have the highest attack rate, that is the proportion of the population who have ever been infected, though these have been revised down to 19% and 18% respectively. The South West continues to have the lowest attack rate (4%).
7. Note that the deaths data used are only weakly informative on R_t over the last two weeks and are still occurring in relatively small numbers in some regions. Therefore, the estimate for current incidence, R_t and the forecast of daily numbers of deaths are likely to be subject to some revision.

Interpretation

The plots of R_t over time show a clear downward trend since early-November, with estimates of R_t lower in all regions in comparison to last week. These lower values of R_t are likely to be the combined result of various social distancing interventions, detected through the Google mobility data, and half-term school closures.

The number of new infections is decreasing in both the North East and Yorkshire, and the North West, but continuing to increase in the South East. In all other regions, incidence has plateaued.

We believe that the number of deaths occurring each day has already reached a peak in most regions. However, the number of deaths reported publicly each day may continue to increase for some weeks; these reports are based on deaths identified during the previous 24-hours and also include a number of delayed reports of deaths that had occurred over previous days (for more information see nowcasting COVID-19 deaths).

The lock-down introduced on the 5th of November will have induced changes in contact patterns, which cannot be quantified with any certainty at this point but have the potential to induce a continued decrease in the R_t values and the number of new infections. Further changes will be reflected in the weekly iterations of our model.

Quote from lead researcher – Professor Daniela De Angelis, Programme Leader and Deputy Director

“The pandemic is slowing down: the R_t values are estimated to be around 1 in all regions; the total number of new daily infections has been estimated to decline over the last week to 41,000; and the number of deaths occurring daily is decreasing. These trends likely result from a combination of the social restrictions introduced in October, the temporary decrease in activity over the half-term period, and the ongoing lockdown. Disentangling the contribution of these factors is challenging.

We remain concerned that R_t may not have fallen to a level sufficiently below 1, which strongly suggests that effective measures to control infection rates must continue to be in place after the end of the current lock-down period.”

Model and report changes

1. The definition of deaths has been adapted to include all deaths that occur in individuals who have had lab-confirmed infection within 60 days from the date of their most recent positive test. This definition reflects more realistically the burden of COVID-19.
2. Using observations of improved survival in hospitalised COVID-19 patients, we have allowed the probability of dying following infection with SARS-CoV2 (the infection-fatality rate, IFR) to gradually change over the course of June 2020, with a decrease being estimated.
3. The model uses seroprevalence data on the presence of COVID-19 antibodies in blood samples taken by NHSBT to estimate the levels of cumulative infection within the population over time. As of early June, the NHSBT has been giving a constantly declining prevalence of antibodies, and these data have been curtailed at this point.
4. The model now accounts for a different susceptibility to infection in the under-15s, using information from literature (Viner et al, 2020) suggesting that children less likely to acquire infection when in contact with an infectious individual.

Link to full report: https://www.mrc-bsu.cam.ac.uk/nowcasting-and-forecasting-26th-november-2020/

---

Contact for media enquiries

Alison Quenault – Communications Manager

alison.quenault@mrc-bsu.cam.ac.uk