The basic reproduction number of SARS-CoV-2 in Wuhan is about to die out, how about the rest of the World?

The virologically confirmed cases of a new coronavirus disease (COVID-19) in the world are rapidly increasing, leading epidemiologists and mathematicians to construct transmission models that aim to predict the future course of the current pandemic. The transmissibility of a virus is measured by the basic reproduction number ( R0 ), which measures the average number of new cases generated per typical infectious case. This review highlights the articles reporting rigorous estimates and determinants of COVID-19 R0 for the most affected areas. Moreover, the mean of all estimated R0 with median and interquartile range is calculated. According to these articles, the basic reproduction number of the virus epicentre Wuhan has now declined below the important threshold value of 1.0 since the disease emerged. Ongoing modelling will inform the transmission rates seen in the new epicentres outside of China, including Italy, Iran and South Korea.

INTRODUCTION

The appearance of a new infectious disease is always a complex phenomenon, especially if it becomes pandemic. Globally, infections by SARS‐CoV‐2 that causes COVID‐19 are rapidly growing, and they extended very fast with transmission chains throughout the world since the first case was detected in the Chinese city of Wuhan in December 2019. Imported cases and secondary cases have been reported in more than 1 436 198 confirmed cases globally. On 11 March 2020, the World Health Organization (WHO) declared COVID‐19 a pandemic and called for governments to take urgent actions to change the course of the outbreak.

An infectious disease outbreak can be characterised by its basic reproductive number, known as , which represents the average number of secondary infections generated by each infected person. If is equal to 1 or less, this indicates that the number of secondary cases will decrease over time and, eventually, the outbreak will peter out. If it is higher than one, the outbreak is expected to increasingly transmit infection to secondary cases, indicating the need to use control measures to limit its extension.

As governments and WHO work together to treat infected people and control the spread of the hitherto unknown SARS‐CoV‐2, several mathematical modelling groups in the China, United Kingdom, Europe and United States have rushed to estimate the basic reproduction number and predict the spread of SARS‐CoV‐2 infections and cases of COVID‐19 disease. These groups used different approaches as illustrated in Table 1 with estimates hovering between 0.32 and 6.47 in Tables 2 and 3. These differences are not surprising, as there is uncertainty about many of the factors go into estimating , such as different methods for modelling, different variables considered, and various estimation procedures.

TABLE 1

Description of estimation methods with list of used abbreviations

ID	Methods	Method description with its abbreviation
1	SIR model3, 4, 5, 6, 7, 8, 9	It is a compartmental model in epidemiology that divides an infectious disease into three parts: Susceptible‐Infectious‐Removed (SIR), which is represented as a dynamical system in mathematics.
2	SEIR model10, 11, 12, 13, 14, 15	Susceptible‐Exposed‐Infectious‐Removed (SEIR) model which is another type of compartmental model which differs from SIR model by adding exposed part that represents the delay time of infected by virus and apparing symptoms (latency period).
3	MSIR model 16	Maternally derived immunity‐Susceptible‐Infectious‐Removed (MSIR) compartmental model that babies got protection from maternal antibodies.
4	MSEIR model 16	It is the same as the model MSIR by joining Exposed component and becoming Maternally derived immunity‐Susceptible‐Exposed‐Infectious‐Removed (MSEIR).
5	SEIHR model17, 18	Entering the Hospitalized class to SEIR model to obtain: Susceptible‐Exposed‐Infectious‐Hospitalized‐Removed (SEIHR).
6	SEIAR model 19	A modified SEIR model with another movement class of compartmental model known as Asymptomatic, to get: Susceptible‐Exposed‐Infectious‐Asymptomatic‐Removed (SEIAR).
7	SEQR model 20	Incorporating the quarantine policy to a mathematical model and obtaining Susceptible‐Exposed‐Quarantined‐Removed (SEQR) model.
8	SIRD model 21	It is the SIR model that addresses the removed class with recovered and dead class to be Susceptible‐Infectious‐Recovered‐Dead (SIRD) model.
9	SUQC model 22	In this model, the infectious class of transmission model is separeted as un‐quanrantined, quarantined and confirmed infected. The model is named Susceptible‐Unquanrantined‐Quarantined‐Confirmed (SUQC) model.
10	SIQR model 23	Its modified SIR model with considering quarantine, Susceptible‐Infectious‐Quarantined‐Recovered (SIQR).
11	S E1E2I1I2HR time‐dependent model 24	It is a mathematical model focusing on the effects of medical resourceson transmission of COVID‐19, stands for susceptible S (t), pre‐stage exposed E1(t), post‐stage exposed E2(t), infected with mild symptoms I1(t), infected with serious symptoms I2(t), hospitalized H(t) and recovered R(t) individuals.
12	SIDARTHE model 25	It is a mathematical model that designed to show transsimssion between different stages in infectious disease. The abbrevation refers to: Susceptible‐Infected ‐Diagnosed‐Ailing‐Recognised‐Threatened‐Healed‐Extinct (SIDARTHE) model. In this model, being infected is dividing into 5 types as: undetected asymptomatic infected, detected asymptomatic infected, undetected symptomatic infected, detected symptomatic infected, and infected with detected life‐threatening symptoms; whereas the removed class in compartmental model is classfied into recovered and dead.
13	Exponential growth9, 26, 27, 28, 29, 30, 31	It is a model that varies exponentially with the time by a specific rate.
14	Generalized growth model 32	It is the growth model with two parameters: (r) represents the growth rate parameter with (p) that is the scaling growth rate parameter. Whenever P = 1, the generalized growth model returns to exponential growth and if 0 < P < 1, then it is sub‐exponential (polynomial) growth.
15	Logistic growth model 33	It is a mathematical model that starts exponentially but it gets stabilized due to the capacity of population.
16	Bayesian estimation method 34	It is a paramter estimation method that deals with paramters as random variables in a statistical model.
17	Fudan‐CCDC model 12	Developed model for the growth rate and CCDC stands for Chinese Center for Disease Control.
18	Least square based method 35	It is a procedure to best fit data in statistics.
19	MCMC method 36	Markov Chain Monte Carlo (MCMC) method. In this technique, the posterior distribution of a desired parameter can be found.
20	Maximum Likelihood Estimation30, 37	It is a method used to estimate parameters with knowing their distributions.
21	Phenomenological modelling 33	Statistical method for modelling.

TABLE 2

The basic reproduction number ( ) from the published articles in Wuhan

ID	Researcher	Date	Location	Methods	Ro Est.	Ro (%95 CI)
1	Imai 38	18 January 2020	Wuhan	Epidemic trajectories	2.60	(1.50‐3.50)
2	Li et al 39	22 January 2020	Wuhan	Exponential growth	2.20	(1.40‐3.90)
3	Majumder et al 33	26 January 2020	Wuhan	Phenomenological modelling	2.55	(2.00‐3.10)
4	Park et al 40	24 February 2020	Wuhan		2.20
5	Read et al 11	1‐22 January 2020	Wuhan	SEIR	3.11	(2.39‐4.13)
6	Shao et al 12	16‐February‐20	Wuhan	SEIR model and Gamma distribution	3.12
7	Shao et al 12	16 February 20	Wuhan	Fudan‐CCDC model	3.32
8	Tuite et al 29	24 January 20	Wuhan	Disease transmission model	2.30
9	WHO 2	22 January 20	Wuhan		1.95	(1.40‐2.50)
10	Wu et al 41	25 January 20	Wuhan	Markov Chain Monte Carlo methods	2.68	(2.47‐2.86)
11	Zhang et al 19	27 January 2020‐10 February 2020	Wuhan	SEIAR model	2.88
12	Zhao & Chen 22	Before 30 January 2020	Wuhan	SUQC Model (Stage I)	4.70
13	Zhao & Chen 22	After 30 January 2020	Wuhan	SUQC Model (Stage II)	0.75
14	Zhao & Chen 22	After 13 Feb 2020	Wuhan	SUQC Model (Stage III)	0.47
15	Wang et al 24	23 January 2020	Wuhan	SE1E2I1I2HR time‐dependent model	2.71

TABLE 3

The basic reproduction number ( ) from the published articles

ID	Researcher	Date	Location	Methods	Ro Est.	Ro (%95 CI)
1	Anastassopoulou et al 21	11‐17 January 2020	Hubei, China	SIRD model	4.60	%90 CI (3.56‐5.65)
2	Choi et al 17	17 February 2020	Hubei, China	Deterministic mathematical model (SEIHR)	4.26	(4.24‐4.29)
3	Choi et al 17	17 February 2020	South Korea	Deterministic mathematical model (SEIHR)	0.55	(0.51‐0.60)
4	Choi et al 17	05 March 2020	NGP‐South Korea	Deterministic mathematical model(SEIHR)	3.50	(3.47‐3.54)
5	Di Lauro et al 42	02 March 2020	World	Metapopulation model	2.50
6	Hao 16	17 February 2020	World	MSIR	1.50
7	Hao 16	17 February 2020	World	MSEIR	3.50
8	Hellewell et al 43	05 February 2020	World	Branching process model	2.50	(1.50–3.50)
9	Hossain et al 4	13 March 2020	China	SIR (44 days quarantined)	1.40
10	Hossain et al 4	13 March 2020	China	SIR (24 days quarantine)	1.68
11	Hossain et al 4	13 March 2020	China	SIR (10 days quarantined)	2.92
12	Imai et al 38	18 January 2020	Wuhan	Computational modelling of potential epidemic trajectories	2.60	(1.50–3.50)
13	Jung et al 36	08 December 2020	China	Developed exponential growth model and using MCMC techinqe.	2.10	(2.00‐2.20)
14	Jung et al 36	24 January 2020	China with exported cases	Developed exponential growth model and using MCMC techinqe.	3.20	(2.70‐3.70)
15	Ku et al 7	12 February 2020	Anhui, China	SIR after lockdown of Wuhan	3.89	(3.27‐4.50)
16	Ku et al 7	12 February 2020	Beijing, China	SIR after lockdown of Wuhan	3.30	(1.89‐4.32)
17	Ku et al 7	12 February 2020	Chongqing, China	SIR after lockdown of Wuhan	2.22	(1.26‐3.14)
18	Ku et al 7	12 February 2020	Fujian, China	SIR after lockdown of Wuhan	1.66	(0.72‐2.87)
19	Ku et al 7	12 February 2020	Gansu, China	SIR after lockdown of Wuhan	2.30	(1.02‐3.96)
20	Ku et al 7	12 February 2020	Henan, China	SIR after lockdown of Wuhan	3.70	(3.16‐4.25)
21	Ku et al 7	12 February 2020	Hubei, China	SIR after lockdown of Wuhan	4.65	(4.10‐5.15)
22	Ku et al 7	12 February 2020	Tianjin, China	SIR after lockdown of Wuhan	2.17	(1.23‐3.54)
23	Ahmadi et al 44	19 March 2020	Iran	Logistic growth model	4.70
24	Kuniya 35	15 January 2020‐29 February 2020	Japan	Least‐square‐based method with Poisson noise	2.60	(2.40‐2.80)
25	Lai et al 45	11 February 2020	World		2.91	(2.24‐3.58)
26	Li et al 39	22 January 2020	Wuhan	Exponential growth	2.20	(1.40–3.90)
27	Lui et al 30	22 January 2020	World	Exponential growth	2.90	(2.32‐3.63)
28	Lui et al 30	22 January 2020	World	Maximum Likelihood Estimation	2.92	(2.28‐3.67)
29	Luo et al 13	13 February 2020	China (except Hubei)	Develped SEIR model.	1.17	(1.15‐1.16)
30	Luo et al 13	13 February 2020	Hubei Province, China	Develped SEIR model.	1.49	(1.48‐1.51)
31	Majumder et al 33	26 January 2020	Wuhan	Phenomenological modeling	2.55	(2.00–3.10)
32	Meng et al 8	12 February 2020	China (except Hubei)	Devloped SIR Model	2.81	(2.72‐2.93)
33	Muniz‐Rodriguez et al 32	19‐29 February 2020	Iran	Generalized growth model	3.60	(3.20‐4.20)
34	Muniz‐Rodriguez et al 32	19‐29 February 2020	Iran	Growth model with doubling times which is equal ln (2)/r where r is grwoth rate.	3.58	(1.29‐8.46)
35	Park et al 40	24 February 2020	Wuhan		2.20
36	Read et al 11	1‐22 January 2020	Wuhan	SEIR	3.11	(2.39–4.13)
37	Remuzzi et al 27	08 March 2020	Italy	Exponential growth	3.00	(2.76‐3.25)
38	Riou et al 31	18 January 2020	China	Computational modelling of potential epidemic trajectories	2.20	%90 CI (1.40‐3.80)
39	Rocklöv et al 14	21 January 2020‐19 February 2020	Diamond Princess Cruise Ship	SEIR Model	3.70
40	Shao et al 12	16 February 2020	Wuhan	SEIR model and Gamma distribution	3.12
41	Shao et al 12	16 February 2020	Hubei (without Wuhan)	SEIR model and Gamma distribution	3.01
42	Shao et al 12	16 February 2020	China (except Hubei)	SEIR model and Gamma distribution	3.04
43	Shao et al 12	16 February 2020	Beijing	SEIR model and Gamma distribution	3.25
44	Shao et al 12	16 February 2020	Shanghai	SEIR model and Gamma distribution	3.24
45	Shao et al 12	16 February 2020	Wuhan	Fudan‐CCDC model	3.32
46	Shao et al 12	16 February 2020	Hubei (without Wuhan)	Fudan‐CCDC model	3.37
47	Shao et al 12	16 February 2020	China (except Hubei)	Fudan‐CCDC model	3.34
48	Shao et al 12	16 February 2020	Beijing	Fudan‐CCDC model	3.27
49	Shao et al 12	16 February 2020	Shanghai	Fudan‐CCDC model	3.31
50	Shen et al 15	12 December 2019	Hubei Province, China	By SEIR simulation	4.71	(4.50‐4.92)
51	Shim et al 46	26 February 2020	South Korea	Exponential growth	1.50	(1.40‐1.60)
52	Sugishita et al 5	14 January 2020‐28 February 2020	Japan	SIR Model	2.50	(2.43‐2.55)
53	Sugishita et al 5	11 March 2020	Japan	%35 reduction of basic reproduction number (2.5), 0.65*2.5 = 1.625, by voluntary event cancellation	1.62
54	Tang et al 10	23 January 2020	China	SEIR Model	6.47	(5.71‐7.23)
55	Tang et al 18	03 February 2020	Shaanxi Province, China	Developed SEIHR Model	1.69
56	Tapiwa et al 34	14 January 2020‐27 February 2020	Tianjin, China	Bayesian estimation method	1.59	(1.42‐1.78)
57	Tapiwa et al 34	21 January 2020‐26 February 2020	Singapore	Bayesian estimation method	1.27	(1.19‐1.36)
58	Traini et al 3	20 February 2020‐11 March 2020	Italy	SIR Model	3.40
59	Tuite et al 29	24 January 2020	Wuhan	Disease transmission model	2.30
60	Wang & You et al 47	17 January 2020‐8 February 2020	Hubei, China	Exponential growth	3.49	(3.42‐3.58)
61	Wang & You et al 47	17 January 2020‐8 February 2020	Hubei, China	Exponential growth (After including control measure)	2.95	(2.86‐3.03)
62	Wang et al 48	27 February 2020	China		2.75	(2.00‐3.50)
63	WHO 2	22 January 2020	Wuhan		1.95	(1.40–2.50)
64	Wu et al 41	25 January 2020	Wuhan	Markov Chain Monte Carlo methods	2.68	(2.47–2.86)
65	Wu et al 9	10 February 2020	Henan, China &China(without Hubei)	SIR Model	2.44
66	Wu et al 9	16 February 2020	Hubei, China	SIR Model	6.27
67	Yang et al 49	26 January 2020	China	Transmission model	3.77	(3.51‐4.05)
68	Zhang et al 19	27 January 2020‐10 February 2020	Wuhan	SEIAR model	2.88
69	Zhang et al 37	16 February 2020	Diamond Princess cruise ship	Maximum Likelihood Estimation	2.28	(2.06‐2.52)
70	Zhao & Chen 22	Before 30 January 2020	Wuhan	SUQC Model (Stage I)	4.71
71	Zhao & Chen 22	After 30 January 2020	Wuhan	SUQC Model (Stage II)	0.75
72	Zhao & Chen 22	After 13 February 2020	Wuhan	SUQC Model (Stage II)	0.48
73	Zhao & Chen 22	Before 30 January 2020	Hubei (without Wuhan)	SUQC Model (Stage I)	5.93
74	Zhao & Chen 22	After 30 January 2020	Hubei (without Wuhan)	SUQC Model (Stage II)	0.60
75	Zhao & Chen 22	Before 30 Jan 2020	China (excluding Hubei)	SUQC Model (Stage I)	1.52
76	Zhao & Chen 22	After 30 January 2020	China (excluding Hubei)	SUQC Model (Stage II)	0.57
77	Zhao & Chen 22	Before 30 January 2020	Beijing	SUQC Model (Stage I)	0.88
78	Zhao & Chen 22	After 30 January 2020	Beijing	SUQC Model (Stage II)	0.52
79	Zhao & Chen 22	Before 30 January 2020	Shanghai	SUQC Model (Stage I)	3.62
80	Zhao & Chen 22	After 30 Jan 2020	Shanghai	SUQC Model (Stage II)	0.51
81	Zhao & Chen 22	Before 30 January 2020	Guangzhou	SUQC Model (Stage I)	1.20
82	Zhao & Chen 22	After 30 Jan 2020	Guangzhou	SUQC Model (Stage II)	0.50
83	Zhao & Chen 22	Before 30 January 2020	Shenzhen	SUQC Model (Stage I)	5.93
84	Zhao & Chen 22	After 30 January 2020	Shenzhen	SUQC Model (Stage II)	0.53
85	Zhao et al 50	10‐24 January 2020	China	Exponential growth	2.24	(1.96‐2.55)
86	Zhao et al 50	10–24 January 2020	China	Exponential growth	3.58	(2.89‐4.39)
87	Zhuang et al 26	31 January 20	Republic of Korea	Exponential growth	2.60	(2.30‐2.90)
88	Zhuang et al 26	05 February 2020	Republic of Korea	Exponential growth	3.20	(2.90‐3.50)
89	Zhuang et al 26	05 February 2020	Italy	Exponential growth	2.60	(2.30–2.90)
90	Zhuang et al 26	10 February 2020	Italy	Exponential growth	3.30	(3.00‐3.60)
91	Giordano et al 25	20 February 2020‐12 March 2020	Italy	SIDARTHE model	2.38
92	Giordano et al 25	16 March 2020	Italy	SIDARTHE model (Public health care)	1.66
93	Hamidouche et al 51	21 March 2020	Algeria	Mathematical model (Alg‐COVID‐19)	2.55
94	Klausner et al 28	21 February 2020‐20 March 2020	Israel	Exponential Growth	2.19
95	Sahafizadeh et al 6	28 February 2020	Iran	SIR Model	4.86
96	Sahafizadeh et al 6	7 March 2020	Iran	SIR Model	4.5
97	Sahafizadeh et al 6	14 March 2020	Iran	SIR Model	4.29
98	Sahafizadeh et al 6	18 March 2020	Iran	SIR Model	2.10
99	Tian et al 20	prior to 23 January 2020	Anhui, China	SEQR model (Phase I)	2.97
100	Tian et al 20	23 January 2020‐6 February 2020	Anhui, China	SEQR model (Phase II)	0.86
101	Tian et al 20	after 6 February 2020	Anhui, China	SEQR model (Phase III)	0.57
102	Wang et al 24	23 January 2020‐6 March 2020	Wuhan	S E1E2I1I2HR time‐dependent model	2.71
103	Crokidakis, N 23	26 February 2020	Brazil	SIQR model	5.25

In this review, we summarise the basic reproduction number of multiple published articles for pandemic COVID‐19. Screening from 1 January 2020 to 6 April 2020, yielded 50 articles which estimated the basic reproduction number for COVID‐19. Most of these studies concern China, some of them are from Italy, Iran, South Korea, Singapore, Japan, Israel and Brazil.

Initially, the WHO estimated the basic reproduction number for COVID‐19 between 1.4 and 2.5, as declared in the statement regarding the outbreak of SARS‐CoV‐2, dated 23 January 2020. Additionally, several articles aimed to more precisely estimate the COVID‐19 . A review written by Liu et al compared 12 published articles from the first January to the seventh of February 2020 which estimated for the for COVID‐19 a range of values between 1.5 and 6.68.The authors of the review evaluated the mean and median of estimated by the 12 articles and they calculated a final mean and median value of for COVID‐19 of 3.28 and 2.79, respectively, with an interquartile range (IQR) of 1.16. Zhao and Chen developed a Susceptible, Un‐quanrantined infected, quarantined infected, confirmed infected (SUQC) model to characterise the dynamics of COVID‐19; suggesting that this model was more suitable for analysis and prediction than adopting existing epidemic models. Using daily confirmed cases, they applied the SUQC model to analyse the outbreak of COVID‐19 in Wuhan, Hubei (excluding Wuhan), China (excluding Hubei) and four first‐tier cities of China (only Wuhan considered in Table 1). They found that the reproduction number for all mentioned regions except Beijing, before 30 January 2020, was defined as stage I, for all regions after 30 January known as stage II, even smaller after 13th February called stage III. The article by Kucharski and colleagues combined mathematical modelling with multiple datasets to calculate the median daily reproduction number in Wuhan, within 2 weeks of introducing travel restrictions; this crucial number began at 2.35 and declined to 1.05 throughout December 2019 and January 2020.

In order to understand a measure of transmissibility of the new disease, a lot of preprints and papers were published in the last months (Table 3), modelling various mathematical and statistical techniques, considering different compartment models in epidemiology and analysing its evolution in some countries. In this paper, we highlight the articles' estimates of COVID‐19 , explore the assumptions of the preditive methods of and illustare values of in differing geographic regions.

METHODS

Along with reviewing articles and presenting their computing basic reproduction numbers, the mean; dividing the total of values by their number, of all that calculated by participating finding of it in each. The median, anothor measure of central is found for ungrouped ordering data which returns to the middle number among the whole values by Microsoft Excel 2010. A measure of variability, finally, named the interquartile range (IQR); is computed by dividing rank‐ordered data into 4 parts and finding quarties as follows: is the middle of first two parts and is the middle of last two parts, while is the median and it is the middle of all values as it is mentioned before. IQR, thus, is the difference between and also it found via Excel 2010.

LOESS method is utilised to sketch the curve of values in Wuhan with their range. LOESS stands for local regression; it is a non‐parametric approach that fits multiple regressions in the local neighbourhood. LOESS can be particularly useful when the x‐axis variables are bound within a range. It allows greater flexibility than traditional modelling tools because it can be used for situations in which we do not know which the parametric form of the regression surface is. A regression line (or curve) is fitted to the observations that fall within the window, the points closest to the centre of the window being weighted to have the most significant effect on the calculation of the regression line. It uses nearest neighbour algorithm. However, the predictor variable can just be indices from 1 to the number of observations in the absence of explanatory variables (as in Figure 1). A window of a specified width is placed over the data. The wider the window, the smoother the resulting loess curve. In other words, the size of the neighbourhood controls the degree of smoothing.

FIGURE 1

Smoothed curve showing the R 0 value in Wuhan city in the period from the 12th of December to the 1st of March 2020. The blue line marks travel restrictions starting on 23 January 2020, red line represents R 0, and grey shading represents 95% confidence intervals of the models estimate

The articles are estimated COVID‐19 that were published from 1 January 2020 to 6 April 2020, searched in Science Direct, Google Scholar, PubMed, Scopus and MedRxiv, using the keywords “basic reprodation number,” “ ,” “SARS‐CoV‐2,” and “COVID‐19,” and yielded more than 60 articles. After screening relevancy, 50 studied met inclsion criteria, providing 103 estimaties. The reason for exclusion the rest of them due to have , and instead of with couple of papers written in different languages. However, no research were excluded because of poor quality.

RESULTS

As recently announced by WHO, the virus epicentre Wuhan and its surrounding Hubei province have not recorded new cases of COVID‐19, which shows the researchers' prediction on are on track (Figure 1 and Table 1). Figure 1 presents different estimated values of the in Wuhan city, Hubei province in China in the period between 12 December and 1 March 2020. It shows different estimated values in Wuhan city through the papers reviewed sorted by chronological order; we can see how the reproduction rate smoothed with LOESS regression method shows a decreasing trend over time. It is worth noting that after the control measures were introduced in Wuhan on 23 January 2020, shown by a vertical blue line in Figure 1, the started dropping down, based on the data in Table 1.

The dot chart in Figures 2 and 3 stratifies COVID‐19 estimates in the period between the first of January to the 18th of March 2020 by authors in the analysed papers in Table 2. Figure 2 illustrates 68 values over 17 different regions in China. Tang et al show the highest in China based on early outbreak data following the SEIR model, while Zhao and Chen estimated the number to be 0.47, which is the lowest in the entire China through SUQC model, after 13 February 2020.

FIGURE 2

Dot chart showing the R 0 value estimated in the analysing papers coloured by location of interest in China

Figure 3 illustrates 35 values over 10 different countries. Brazil has the highest outside China, estimated more than 5. In Iran, Muniz‐Rodriguez et al estimated a value of about 3.5. Zhuang et al, Traini et al and Remuzzi et al estimated range of basic reproduction number from 2.6 to 3.4 in Italy. Kuniya estimated to be 2.60 in Japan, Hamidouche et al estimated to be 2.55 in Algeria, Klausner et al estimatied to be 2.19 in Israel and Tapiwa et al, estimated to be 1.27 in Singapore. Regarding the Republic of Korea, Choi et al reported a value below 1 on 17 February 2020.

FIGURE 3

Dot chart showing the R 0 value estimated in the analysing papers coloured by location of interest in the global

With available articles regarding in Italy, Iran, South Korea, Singapore, Japan, Israel, Algeria, Brazil and China, we calculated the estimated mean for COVID‐19, with median = 2.73 and interquartile range (IQR) = 1.73. This mean is very close to the upper boundary estimated by WHO but lower than the previous review by Liu et al. However, the average between 2 and 3 seems to have stabilised in recent articles shown in Table 2.

As more results to mention, there are various methods utilised in estimating as listed in Table 1, some of them being special compartmental models which are mathematical models in epidemlogy, while others are statistical models and techniques; whereas some others are mix of mathematical and statistical approaches. More accurately, from 103 findings of , 28 of them estimated it using statistical approaches, reported a range of 1.27 to 4.70 with an average 2.71, and 6 obtained of were found by mathematical models with statistics techniques estimated ranging from 3.01 to 4.71, with an average 3.39, the remaining 66 used mathematical models to estimate calculated a range from 0.47 to 6.47, with an average of 2.69.

CONCLUSION

In the globalised world of today, the evolution of the outbreak and information on COVID‐19 have become available at an unprecedented pace. Still, is not easy to calculate, especially there is much more to know about this new infection. The articles in Table 3, estimated different values of , using results obtained from their respective models. The discrepancies observed among the studies of COVID‐19 depend on a variety of assumptions in mathematical and statistical techniques, namely, the duration of contagiousness, the likelihood of infection per contact and the contact rate. Due to variation in the assumptions and control strategies with time, the intervention measures, such as border control and quarantine in China, reduces from 2.92 to 1.40, voluntary event cancellation in Japan reduced COVID‐19 infectiousness by 35%, social distancing and strict restriction on travelling in Iran during 4 weeks reduced from 4.86 to 2.1 and closing schools and remote working with some basic recommendations in Italy reduced from 2.38 to 1.66. Moreover, the basic reproduction number is continuously modified during a pandemic by accurate assumptions introduced and becomes more reliable as more data and information come to light.

In this article, the potential transmission of the SARS‐COV‐2 virus results in COVID‐19 that is expressible in basic reproduction number is summarised from 50 publishes with identifying their used approaches in finding it across the world. This review found that the estimated for COVID‐19 in the case of Wuhan has decreased below the threshold of 1, and the estimated mean of is around 2.71 for COVID‐19, with a median of 2.73 and IQR of 1.73. Our review coincides with a recent published article by Wang et al, they estimated COVID‐19 to be 2.71 in Wuhan. More reasonable match in their article showed that the epidemic gradually died out from calculating effective reproduction ratio, which is used to measure the daily reproduction number, started from 2.71 as of 23 January, has declined rapidly to below 1 since eighth February 2020 and dropped to 0.06 at 6 March 2020.

Along with new pandemic control measures introducing and treating procedures more mathematically desiged models are required to take account of all factors, in this point of view, the mathematical models are more recommended to be used. All in all, still is not easy to calculate especially there is much more to know about this novel virus.