An extended Susceptible-Exposed-Infected-Recovered (SEIR) model with vaccination for forecasting the COVID-19 pandemic in Sri Lanka

Background: After declaration of a Public Health Emergency of International Concern (PHEIC) by the World Health Organization, Novel Corona Virus had been spreading throughout the world despite various degrees of movement restrictions and availability of multiple safe and effective vaccines. Vaccination rate against COVID-19 is one of the infection rate’s main determinants. The role of modelling in predicting the spread of an epidemic is important for health planning and policy making.
Objective: This study aimed to construct a compartmental epidemiological model incorporating vaccination coverage, vaccination rate, vaccine efficacies and applied a computational tool for predicting the evolution of different epidemiological variables for COVID-19 in Sri Lanka.
Methods: We used a dynamic Susceptible-Exposed-Infected-Recovered-Vaccinated (SEIRV) model and simulated potential vaccine strategies under a range of epidemic conditions. The predictions were based on different vaccination coverages (5% to 90%), vaccination-rates (1%, 2%, 5%) and vaccine-efficacies (40%, 60%, 80%) under different R₀ (2,4,6). We estimated the duration, exposed, and infected populations.
Results: When the R0 was increased, the days of reduction of susceptibility and the days to reach the peak of the infection were reduced gradually. At least 45% vaccine coverage is required for reducing the infected COVID-19 population to mitigate a disastrous situation in Sri Lanka.
Conclusion and recommendations: The results revealed that when R0 is increased in the SEIRV model along with the increase of vaccination efficacy and vaccination rate, the population to be vaccinated is reducing. Thus, the vaccination offers greater benefits to the local population by reducing the time to reach the peak, exposed and infected population through flattening the curves. The prediction models will lead to policy relevance despite the significant uncertainty associated with real-time forecasting in complex systems with timely predictions and steadfast reports.