Mathematical modeling of the COVID-19 prevalence in Saudi Arabia

The swift precautionary and preventive measures and regulations that were adopted by the Saudi authority has ameliorated the exponential escalation of the SARS-CoV-2 virus spread, decreased the fatality rate and critical cases of COVID-19. Understanding the trend of COVID-19 is crucial to establishing the appropriate precautionary measures to mitigate the epidemic spread. The aim of this paper was to modifying and enhancing the mathematical modeling to guide health authority and assist in an early assessment of the epidemic outbreak and can be utilised to monitor non-pharmaceutical interventions (NPIs). Both ARIMA model and Logistic growth model were developed to study the trend and to provide short and long-term forecasting of the prevalence of COVID-19 cases and dynamics. The data analyzed in this study covered the period between 2nd March and 21st June 2020. Two different scenarios were developed to predict the epidemic fluctuating trends and dynamics. The first scenario covered the period between 2nd March and 28th May when the first peak was observed and immediately declined. The analysis projected that the COVID-19 epidemic to reach a peak by 17th May with a total number of 58,534 infected cases and to end on the 4th August, if lockdown were not interrupted and folks followed the recommended personal and social safety guidelines. The second scenario was simulated because of the sudden sharp spike witnessed in the trend of the new confirmed cases on the last week of May and continue to escalate till the time of current writing-21st June. In the 2nd scenario, the analysis estimated the epidemic to peak on 15th June with a total number of 146,004 infected cases and to end on 29th September, 2020 with a final size of 209,607 (185,757 to 244,310) infected cases, assuming that the NPIs will be maintained while new normal life is resumed carefully. ARIMA and Logistic growth models showed excellent performance in projecting the epidemic prevalence, trends and dynamics at different phases. In conclusion, the analysis presented in this paper will assist policy-makers and health care authorities to evaluate the effect of the NPIs applied and to size the resources needed to manage different phases and cope with the final size of the epidemic estimates and to impose extra precautions.