key: cord-0828382-xv7274vn authors: Mahmud, R.; Patwari, H. M. A. F. title: Estimation of the Basic Reproduction Number of SARS-CoV-2 in Bangladesh Using Exponential Growth Method date: 2020-09-29 journal: nan DOI: 10.1101/2020.09.29.20203885 sha: 9a8a593fdf60d8b2f7cb80e3de0f82cae88b3f1e doc_id: 828382 cord_uid: xv7274vn Objectives: In December 2019, a novel coronavirus (SARS-CoV-2) outbreak emerged in Wuhan, Hubei Province, China. Soon, it has spread out across the world and become an ongoing pandemic. In Bangladesh, the first case of novel coronavirus (SARS-CoV-2) was detected on March 8, 2020. Since then, not many significant studies have been conducted to understand the transmission dynamics of novel coronavirus (SARS-CoV-2) in Bangladesh. In this study, we estimated the basic reproduction number R 0 of novel coronavirus (SARS-CoV-2) in Bangladesh. Methods: The data of daily confirmed cases of novel coronavirus (SARS-CoV-2) in Bangladesh and the reported values of generation time of novel coronavirus (SARS-CoV-2) for Singapore and Tianjin, China, were collected. We calculated the basic reproduction number R0 by applying the exponential growth (EG) method. Epidemic data of the first 76 days and different values of generation time were used for the calculation. Results: The basic reproduction number R0 of novel coronavirus (SARS-CoV-2) in Bangladesh is estimated to be 2.66 [95% CI: 2.58-2.75], optimized R0 is 2.78 [95% CI: 2.69-2.88] using generation time 5.20 with a standard deviation of 1.72 for Singapore. Using generation time 3.95 with a standard deviation of 1.51 for Tianjin, China, R0 is estimated to be 2.15 [95% CI: 2.09-2.20], optimized R0 is 2.22 [95% CI: 2.16-2.29]. Conclusions: The calculated basic reproduction number R0 of novel coronavirus (SARS-CoV-2) in Bangladesh is significantly higher than 1, which indicates its high transmissibility and contagiousness. Coronavirus disease 2019 (COVID-19) is a highly infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1] . COVID-19 was first identified in December 2019, Wuhan, Hubei Province, China [2] . Eventually, it resulted in an ongoing pandemic [3] . Up until the 25 th of August 2020, more than 23.9 million cases were confirmed across 213 countries and territories. The death toll is high, with more than 820,000 deaths. Moreover, more than 16.4 million people have recovered from it [4] . However, as COVID-19 has spread out across the world, Bangladesh is not an exception. Ever since 8th March, the first case spotted, there has been a report of 299,628 confirmed cases, 4,028 deaths, and 186,756 recoveries from COVID-19 as of 25 th August 2020 [5] . SARS-CoV-2 infects the host's respiratory system. It spreads between people when they come to close contact with an infected person (symptomatic or not) by inhaling small droplets produced by coughing, sneezing, or talking [6] . Most of the mathematical models on how a contagious disease spreads on a population are based on the basic reproduction number (R0). R0 defines the secondary cases generated by an infected person over his entire period of infectiousness in a completely susceptible population [7] . The significance of R0 lies within its threshold. If R0 > 1, the infected population number will grow. And if R0 < 1, the number will decrease [8] . In the study of epidemiology, R0<1 is the favorable ratio. An epidemic with high basic reproduction number R0 is potentially a civilization-ending threat. Several control methods can be taken to minimize the R0. We experienced numerous highly infectious disease outbreak in the last two decades. Globalization has enabled us with high connectivity. However, this fruit of civilization could become the doom for us instantaneously without proper control measures. We must know what we are up against to implement proper control methods. And R0 gives an explicit idea about the transmission of the virus. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.29.20203885 doi: medRxiv preprint There are many popular methods for estimating the R0 value. For this research purpose, we shall use the exponential growth (EG) method. Using the EG has adequacy because researchers found from the simulation that it is less prone to bias [9] . In the EG method, we need generation time (or generation interval) GT, to formulate R0. GT defines the time interval between symptoms recorded on an infected person and a secondary case [10] . Data used in this paper is the daily confirmed cases of infection of SARS-CoV-2 in Bangladesh (from the 1 st day of the outbreak to the 76 th day). The data were gathered from https://ourworldindata.org/coronavirus-source-data. This website provides real-time data and statistics on various topics. The use of the basic reproduction number R0 is one of the most important concepts in the study of demography, ecology, and epidemiology. In epidemiology, R0 defines the secondary cases generated by an infected person over his entire period of infectiousness in a completely susceptible population [7] . R0 expresses the transmissibility of a disease. The basic reproduction number is the common buzzword during a pandemic. Because the nature of the virus is explicit with it. There are many existing methods for estimating R0. Among them are Exponential Growth (EG), Maximum Likelihood estimation (ML), Sequential Bayesian method (SB), the Time-Dependent method (TD), and so on. For this particular paper, we shall apply the EG method. Also, the EG method requires Generation Time (GT), also known as Generation Interval, of the virus for the estimation. The count of cases increases exponentially in the early stages of an epidemic. The exponential growth (EG) model is a simplified model. If something grows with a consistent rate, then it is said . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.29.20203885 doi: medRxiv preprint to be an exponential growth. The basic reproduction number R0 can be interpreted indirectly from the exponential curve of epidemic growth. As reported by Wallinga and Lipsitch, 2007 [11] , the exponential growth of an epidemic at the early stage is linked to R0 as the following manner, Here, M = moment generating function of GT distribution and r = exponential growth rate. The daily confirmed cases data are integers. So Poisson regression is used to fit the value of the growth rate, r. Generation The optimal time interval is calculated to be from day 2 of the outbreak to day 44. Hence, the optimal R0 is calculated in the optimal time interval. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 29, 2020. To analyze the sensitivity, we estimated R0 of SARS-CoV-2 for the same data using previously reported generation time of MERS and SARS coronavirus 7.6 and 8.4 with a standard deviation of 3.4 and 3.8, respectively [13, 14] . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.29.20203885 doi: medRxiv preprint We found that the value of R0 varies with the change of generation time. Using the EG method, for generation time 3.95, the value of R0 is estimated to be 2.15, and this value is almost doubled and estimated to be 4.07 for generation time 8.4 (Figure 1 ). The analysis shows that R0 is sensitive to generation time. Hence, generation time needs to be estimated accurately to estimate R0 more precisely. The estimated value of the basic reproduction number R0 of SARS-CoV-2 in Bangladesh using the exponential growth method is ranging from 2.22 to 2.78. This result is quite similar to the previously reported R0 using the same method in China 2.24 and 2.90 [15, 16] . This estimated value of R0 is significantly greater than 1, which is evidence of the high transmissibility and . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.29.20203885 doi: medRxiv preprint contagiousness of SARS-CoV-2 in Bangladesh. In this study, we found that R0 shows some sensitivity to generation time. Thus, a more precise estimation of generation time is needed to estimate R0 with more accuracy. The estimation of R0 can be used to estimate other transmission dynamical parameters of SARS-CoV-2 in Bangladesh. It also can help to determine an optimum strategy for vaccination. From the epidemiological study, we have, This study has several limitations too. First is the quality of data. We collected data from a public resource. The quality of data is not ensured. There is a possibility that some confirmed cases remained unreported. This can affect our estimation. Second, the reported values of generation time vary from 3.95 to 5.20. This is not quantified precisely. In summary, using the exponential growth method, the basic reproduction number R0 of SARS-CoV-2 is estimated. R0 has shown some sensitivity to generation time. This study can be helpful for further understanding of the transmission dynamics of SARS-CoV-2 in Bangladesh. The estimated R0 is significantly larger than 1, which is evidence of its high transmissibility and contagiousness. Either vaccination should be done, or more preventive control measures should be taken to reduce the epidemic size. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.29.20203885 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.29.20203885 doi: medRxiv preprint Naming the coronavirus disease (COVID-19) and the virus that causes it World Health Organization. Novel Coronavirus -China World Health Organization. WHO Director-General's opening remarks at the media briefing on COVID-19 -11 2020. 5. Worldometer.info. Bangladesh Coronavirus cases Q&A on coronaviruses (COVID-19) Pandemic potential of a strain of influenza A (H1N1): early findings A brief history of R0 and a recipe for its calculation The R0 package: a toolbox to estimate reproduction numbers for epidemic outbreaks A practical generation-intervalbased approach to inferring the strength of epidemics from their speed How generation intervals shape the relationship between growth rates and reproductive numbers Estimating the generation interval for coronavirus disease (COVID-19) based on symptom onset data Hospital outbreak of Middle East respiratory syndrome coronavirus Transmission dynamics and control of severe acute respiratory syndrome Preliminary estimation of the basic reproduction number of novel coronavirus (2019-nCoV) in China, from 2019 to 2020: A data-driven analysis in the early phase of the outbreak . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprintThe copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.29.20203885 doi: medRxiv preprint Statistical methods in medical research, 2, 23-41.. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprintThe copyright holder for this this version posted September 29, 2020. . https://doi.org/10.1101/2020.09.29.20203885 doi: medRxiv preprint