Modelling plausible scenarios for the Omicron SARS-CoV-2 variant from early-stage surveillance

TL;DR

This study used a spatial agent-based model combined with early surveillance data to explore plausible scenarios of Omicron's transmission and immune escape, informing early intervention strategies and highlighting the importance of prior immunity landscape.

Contribution

The paper presents a novel modeling approach integrating early surveillance data to simulate Omicron's outbreak dynamics and assess intervention impacts.

Findings

Model accurately captured initial Omicron outbreak dynamics.

Early interventions would have limited effect due to Omicron's transmission advantage.

Prior immunity landscape significantly influences outbreak outcomes.

Abstract

We used a spatially explicit agent-based model of SARS-CoV-2 transmission combined with spatially fine-grained COVID-19 observation data from Public Health Scotland to investigate the initial rise of the Omicron (BA.1) variant of concern. We evaluated plausible scenarios for transmission rate advantage and vaccine immune escape relative to the Delta variant based on the data that would have been available at that time. We also explored possible outcomes of different levels of imposed non-pharmaceutical intervention. The initial results of these scenarios were used to inform the Scottish Government in the early outbreak stages of the Omicron variant. Using the model with parameters fit over the Delta variant epidemic, some initial assumptions about Omicron transmission rate advantage and vaccine escape, and a simple growth rate fitting procedure, we were able to capture the initial outbreak dynamics for Omicron. We found that the modelled dynamics hold up to retrospective scrutiny. The modelled imposition of extra non-pharmaceutical interventions planned by the Scottish Government at the time would likely have little effect in light of the transmission rate advantage held by the Omicron variant and the fact that the planned interventions would have occurred too late in the outbreak's trajectory. Finally, we found that any assumptions made about the projected distribution of vaccines in the model population had little bearing on the outcome, in terms of outbreak size and timing. Instead, it was the landscape of prior immunity that was most important.