A Hybrid Approach Toward COVID-19 Pandemic Modeling in Saudi Arabia Using the Modified Susceptible-Exposed-Infectious-Recovered Model and Open Data Sources

Abstract

The coronavirus disease 2019 (COVID-19) pandemic has caused the world to operate uncharacteristically for almost the last two years. Governments across the globe have taken different control measures to eradicate it. The Oxford COVID-19 Government Response Tracker (OxCGRT) provides open access data for different countries on 20 control measures, including numerous aggregated indices. This paper employs the modified Susceptible-Exposed-Infectious-Recovered (SEIR) epidemiology model to study the COVID-19 pandemic in Saudi Arabia. The modification has been achieved by including control measures and the infectiousness of exposed compartment. A hybrid approach has been used to estimate and incorporate control measures. Initially, a composite control measure has been derived from OxCGRT data to make an attempt to fit the COVID-19 pattern in Saudi Arabia. The derived model has proven to be satisfactory through statistical tests. Nonetheless, the model patterns do not resemble the reported patterns more closely. Hence, a second heuristic approach has been utilized to devise effective control measures from the reported pattern of COVID-19 from the Saudi government agency. A satisfactory model was derived utilizing this approach with successful validation through statistical tests. Also, the model patterns more closely resemble the reported patterns of COVID-19 cases. This hybrid approach proves more robust and ensures the validity of model parameters better. The R naught (R₀) value with the current control measures has varied from 0.515 to 1.892, with a mean value of 1.119, and is presently less than 1. The threshold herd immunity, in the absence of any control measure, is estimated to be 47.12% with an R₀ value of 1.89 and would end up infecting 76.32% of the population. The scenario analysis with gradual partial and complete relaxations up to December 31, 2021, shows that the peaks are likely to occur in 2022; therefore, Saudi Arabia must continue to inoculate its population to eradicate COVID-19.

Keywords: basic reproduction number; compartment epidemiology model; control measures; coronavirus disease 2019 (covid-19); infectiousness of exposed and infectious compartments; oxford covid-19 government response tracker (oxcgrt); saudi arabia; susceptible-exposed-infectious-recovered (seir) model; threshold herd immunity.

Figure 1. Boxplots of OxCGRT control measures for Saudi Arabia

Figure 2. Different indices provided by the OxCGRT and derived control from OxCGRT control measures

Figure 3. Various indices estimated from OxCGRT aggregated indices, derived control measure, and effective control measure

Figure 4. Active cases with points of maxima and minima

Figure 5. Model fitting with OxCGRT derived composite control measure index

Figure 6. Patterns of active cases in the model and the reported data for OxCGRT derived composite control measure index

Figure 7. Model fitting with effective control measure devised from the reported pattern of COVID-19 cases

Figure 8. Patterns of active cases in the model and the reported data for effective control measure devised from the reported pattern of COVID-19 cases

Figure 9. Herd immunity scenario without any control measure

Figure 10. Scenario analysis for gradual relaxations by the end of 2021