Estimating the reproduction number of COVID-19 in Iran using epidemic modeling

Background

On December 31, 2019, the novel coronavirus (COVID-19) emerged in Wuhan, China [1] and the potential of transmitting virus via human travelling caused to spread the novel corona virus to other Chinese provinces as well as the other countries [2].

Iranian governments reported the first confirmed cases of novel coronavirus (COVID-19) infections in Qom, on February 19, 2020 (30 Bahman 1398) [3].

Tehran, Guilan and Markazi were the other provinces confirmed to be infected by coronavirus during the next days of the outbreak. After one week Iranian governments confirmed that 18 out of 31 provinces of Iran were infected by the novel coronavirus. After two weeks the number of reported cases increased to 2922 people of whom 92 cases died and 552 cases recovered [3].

The aim of this study is to estimate basic reproduction number (R0) of COVID-19 in Iran. For this purpose, we use the reported data by Iranian governments and develop an epidemic model to fit on the real data by tuning the model parameters. We then use the model parameters’ values to calculate the basic reproduction number of COVID-19 in Iran.

Methods

We employ the SIR (Susceptible-Infected-Removed) epidemic model to represent the spreading process of COVID-19 in Iran. SIR model is an epidemic spreading model proposed by Kermack and McKendric [4]. We estimate the parameters of model to get a best fit on reported data of COVID-19 outbreak in Iran. The differential equations of the SIR model are given as [5]:

Where S(t), I(t) and R(t) are the number of susceptible, infected and removed people. We assume that removed cases are the summation of recovered and dead cases. β is the infection rate, and γ is the remove rate which is the inverse of infectious period.

The basic reproduction number, R₀, is the average number of the secondary individuals in a complete susceptible population infected by a single infected person during its spreading life [6]. When R₀ > 1 virus spreads through the population and when R₀ < 1 the outbreak will stop due to decreasing the number of new cases. According to Eq. (2) when the number of infected people increases we have:

In a complete susceptible population which S(t) ≈ N the Eq. (5) can be written as the following:

According to Eq. 6 and definition of R₀ we have:

Due to the infection of all provinces of Iran, N is assumed to be the population of Iran (81800269 in 2018) which was obtained from the World Bank [7].

We model epidemic spreading in the period from February 21, 2020 to April 21, 2020. Table 1. Shows the number of reported infected cases and reported removed cases by Iran ministry of health and medical education (MOHME) [3]. MOHME reported the first two cases of COVID-19 infections and their death on 19 February, 2020 which is indicated as day zero in Table 1.

Results

The Runge-Kutta method is used to resolve the SIR model’s ODEs to fit model on reported data. Numerical simulation is conducted in MATLAB and we tune the parameters β and γ to have the best fit plot on reported data plot. We then estimate the basic reproduction number by dividing β on γ. The SIR model fitting results are shown in Fig. 1 Through Fig. 4. The red and green plots are model plots and the blue and pink plots are reported data plots. As shown in Fig. 1, we estimated that the R₀ to be 4.86 in the first week of outbreak which is significantly larger than one. We then set β = 0.585 and γ = 0.13 to fit model to reported data of the second week and we obtained R_t = 4.5 which is smaller than the first week. In Fig. 3 we obtained R_t = 4.29 for the 13^th day to 16^th day of outbreak. Fig. 4 shows that the reproduction number significantly decreased to 2.37 between March 6, 2020 to April 3, 2020, which could be the consequence of self-quarantining and reducing working time by government. As Nowruz occurred on March 20, 2020, many businesses as well as government offices closed and the government restricted travelling between provinces for almost three weeks. As a consequence of this policy it can be seen at Fig 5. that the reproduction number declined to one between April 4 and April 6, 2020. Fig. 6 shows that since April 7, 2020 the reproduction number has been less than one which indicates that disease will decline.

• Download figure
• Open in new tab

Fig. 1:

Plot of COVID-19 outbreak on the first week

• Download figure
• Open in new tab

Fig. 2:

Plot of COVID-19 outbreak on the second week

• Download figure
• Open in new tab

Fig. 3:

Plot of COVID-19 outbreak from the 13^th day to 16^th.

• Download figure
• Open in new tab

Fig. 4:

Plot of COVID-19 outbreak from the 16^th day to 44^th.

• Download figure
• Open in new tab

Fig. 5:

Plot of COVID-19 outbreak from the 45^th day to 47^th.

• Download figure
• Open in new tab

Fig. 6:

Plot of COVID-19 outbreak from the 48^th day to 62^th.

Discussion

In this study, we employed SIR epidemic model, to fit the reported data in Iran and estimate the basic reproduction number of COVID-19 in Iran.

Previous studies estimated basic reproduction number between 2.2 and 4. Chen et.al [1] showed that the basic reproduction number of COVID-19 was 3.58 using the reported data in Wuhan City, China. Read et.al [8] showed that the R₀ of COVID-19 was between 3.6 and 4. Zhao et.al [9] estimated R₀ of COVID-19 to be between 2.24 and 3.58. Other researchers estimated R₀ to be 2.68 [10], 2.6 [2] and 2.2 [11].

The simulation results of this study indicated that the R₀ of COVID-19 was 4.86 in the first week of the outbreak which was significantly larger than one. The schools and universities closed in the first week and in the second week the effective reproduction number decreased to 4.5. In the second week the working time reduced and effective reproduction number decreased to 4.29 from 13^th day to 16^th. Comparing to the other researches, the results showed that R₀ of COVID-19 in Iran in the early stages were higher than the previous estimated basic reproduction number [1][2][8][9][10][11].

In the next weeks of outbreak people decided to stay at home spontaneously and the government reduced the working time. The effective reproduction number decreased to 2.37 between March 6, 2020 and April 3, 2020. In Nowruz, the government restricted travelling and many businesses closed for three weeks. Hence, the reproduction number was reduced and it has been less than one since April 7.

Conclusion

Basic reproduction number of COVID-19 is an important parameter which is used to estimate the risk of COVID-19 outbreak. We estimated the basic reproduction number (R₀) of COVID-19 outbreak in Iran using SIR model by fitting the model to official reported data. According to the results, it is observed that reproduction number of COVID-19 in Iran in the early stages 4.86, 4.5 and 4.29 are higher than the previous estimated [1][2][10].

Self-quarantine and control strategies implemented by government reduced the effective reproduction number of COVID-19 below one. However, it is necessary to continue social distancing and control travelling to prevent causing a second wave of outbreak.

Abbreviations

Covid-19 coronavirus; basic reproduction number; Iran.

Authors’ contributions

ES conducted the experiments and analyzed the results and wrote the manuscript; SS collected the reported cases data and review and edited the manuscript. All authors read and approved the final manuscript.

Funding

The authors received no specific funding for this work.

Availability of data and materials

Information related to the study is in the manuscript.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.