Estimating SARS-CoV-2 variant fitness and the impact of interventions in England using statistical and geo-spatial agent-based models

Abstract

The SARS-CoV-2 epidemic has been extended by the evolution of more transmissible viral variants. In autumn 2020, the B.1.177 lineage became the dominant variant in England, before being replaced by the B.1.1.7 (Alpha) lineage in late 2020, with the sweep occurring at different times in each region. This period coincided with a large number of non-pharmaceutical interventions (e.g. lockdowns) to control the epidemic, making it difficult to estimate the relative transmissibility of variants. In this paper, we model the spatial spread of these variants in England using a meta-population agent-based model which correctly characterizes the regional variation in cases and distribution of variants. As a test of robustness, we additionally estimated the relative transmissibility of multiple variants using a statistical model based on the renewal equation, which simultaneously estimates the effective reproduction number R. Relative to earlier variants, the transmissibility of B.1.177 is estimated to have increased by 1.14 (1.12-1.16) and that of Alpha by 1.71 (1.65-1.77). The vaccination programme starting in December 2020 is also modelled. Counterfactual simulations demonstrate that the vaccination programme was essential for reopening in March 2021, and that if the January lockdown had started one month earlier, up to 30 k (24 k-38 k) deaths could have been prevented. This article is part of the theme issue 'Technical challenges of modelling real-life epidemics and examples of overcoming these'.

Keywords: SARS-CoV-2; agent-based model; geo-spatial model; statistical model; vaccinations; virus variants.

Figure 1.

Nationwide and regional epidemic time series. (a,b) The time series of the ABM ABC posterior of cases and deaths, respectively, along with the actual data [49]. (c) The time series of the model posterior of the fraction of cases by lineage along with the raw COG-UK data [41]. (d) The weekly cases per 100 k population in each of the nine English regions (electronic supplementary material, figure 3) along with the actual data [49]. The model correctly captures the shape of the curves in each region despite not containing region-specific dynamic parameters. (Online version in colour.)

Figure 2.

Geographic spread of the Alpha variant. The ABM ABC posterior of the fraction of new cases that were the Alpha variant, shown over time and English region. The left set of panels show the nine regions of England, with COG-UK data as black dots and the model posterior in colour (the median is represented by a thick line and the central 90% of the posterior with shading; the median for every region is replotted in every panel with faint lines to highlight the timing difference between regions). The timing of the sweep in each region is accurately predicted by the ABM (despite it not containing region-specific dynamic parameters), with the exception of the South West, where it occurred about two weeks too early in the model. The right set of panels show the model results at finer geographical resolution (ULTAs) at the four time points indicated. (Online version in colour.)

Figure 3.

Scenario analysis: each row contrasts the central modelled scenario (in which interventions were as in reality) with a different counterfactual scenario. The left column shows weekly cases per capita over time, with the actual scenario in grey and the counterfactual in blue. The central column shows total deaths over time, with the actual scenario in grey and the counterfactual in green. The right column shows the difference in total deaths by region from December 2020 to May 2021 for the counterfactual scenario compared with the actual scenario. In the top row, the counterfactual scenario considered is no vaccine programme. In this counterfactual, deaths would have been higher and a new wave would have occurred following the easing of lockdown in March 2021. The second row shows the counterfactual scenario in which the third lockdown was brought forward to December 2020 but only lasted a month. The results demonstrate that this would have just delayed the timing of the second wave by a month. The third row shows the counterfactual scenario in which the third lockdown was brought forward to December 2020 and lasted the same length of time. The results demonstrate that 30 k deaths might have been prevented because the spread of Alpha would have been postponed until after vaccination of the most vulnerable, and that these prevented deaths would have been concentrated in the South East and London where the Alpha variant dominated first. (Online version in colour.)