Optimizing SARS-CoV-2 vaccination strategies in France: Results from a stochastic agent-based model

The COVID-19 pandemic is a major global societal, economic and health threat. The availability of COVID-19 vaccines has raised hopes for a decline in the pandemic. We built upon a stochastic agent-based microsimulation model of the COVID-19 epidemic in France. We examined the potential impact of different vaccination strategies, defined according to the age, medical conditions, and expected vaccination acceptance of the target non-immunized adult population, on disease cumulative incidence, mortality, and number of hospital admissions. Specifically, we examined whether these vaccination strategies would allow to lift all non-pharmacological interventions (NPIs), based on a sufficiently low cumulative mortality and number of hospital admissions. While vaccinating the full adult non-immunized population, if performed immediately, would be highly effective in reducing incidence, mortality and hospital-bed occupancy, and would allow discontinuing all NPIs, this strategy would require a large number of vaccine doses. Vaccinating only adults at higher risk for severe SARS-CoV-2 infection, i.e. those aged over 65 years or with medical conditions, would be insufficient to lift NPIs. Immediately vaccinating only adults aged over 45 years, or only adults aged over 55 years with mandatory vaccination of those aged over 65 years, would enable lifting all NPIs with a substantially lower number of vaccine doses, particularly with the latter vaccination strategy. Benefits of these strategies would be markedly reduced if the vaccination was delayed, was less effective than expected on virus transmission or in preventing COVID-19 among older adults, or was not widely accepted.