ACADEMIA Letters Global Variation in the Basic Reproduction Number of COVID-19 Renate Thiede, University of Pretoria Nada Abdelatif Inger Fabris-Rotelli Raeesa Manjoo-Docrat Jennifer Holloway Charl Janse van Rensburg Pravesh Debba Nontembeko Dudeni-Tlhone Zaid Kimmie Alize le Roux The rapid spread of the COVID-19 pandemic caught the world unprepared. The World Health Organisation released a statement on 9 January 2020 reporting that a novel respiratory disease had been identified by Chinese authorities, originating in Wuhan [1]. By 17 January, China recorded 62 cases, and 3 travellers had exported the disease, with two diagnosed in Thailand and one in Japan [2]. At the time of writing, 219 countries have been affected by the disease, with the total death toll estimated at over 2 100 000 [3]. The need to understand the dynamics of disease transmission is crucial in order to combat it effectively. The basic reproduction number (R0) is the average number of susceptible individuals that are infected by a single infected individual [4], and is a critical parameter for modelling disease transmission [5]. It is best measured in the early phase of the outbreak, before control measures have had time to take effect, and when most of the population is susceptible. Herein, we investigated and visualised the spatial variability in estimates of R0 obtained at the start of the pandemic by conducting a systematic literature review. We further explored the relationship Academia Letters, July 2021 ©2021 by the authors — Open Access — Distributed under CC BY 4.0 Corresponding Author: Renate Thiede, renate.thiede@gmail.com Citation: Thiede, R., Abdelatif, N., Fabris-Rotelli, I., Manjoo-Docrat, R., Holloway, J., Janse van Rensburg, C., Debba, P., Dudeni-Tlhone, N., Kimmie, Z., le Roux, A. (2021). Global Variation in the Basic Reproduction Number of COVID-19. Academia Letters, Article 1956. https://doi.org/10.20935/AL1956. 1 between start-of-pandemic R0 and the development level of a country, as quantified by the UN Human Development Index (HDI) [6] and their economic development status as classified by the World Bank [7]. Limiting our investigation to the start-of-pandemic R0 provides insight into the influence of the HDI on transmission under normal circumstances, and allows for a valid comparison between countries without the potentially confounding effects of nonpharmaceutical interventions. Estimation of R0 has proved complex [8] and depends on the extent to which the disease characteristics are understood, which is challenging in the case of a novel disease. Various systematic reviews for COVID-19 illustrated a great variation in values of R0 in the early stages of the pandemic, such as the reviews of [9] and [10] which obtained values ranging between 0.6 and 14.8. The idea of spatial variation is mentioned in the review of [11], which states that R0 varies from place to place, but does not investigate spatial variability. Furthermore, none of the above reviews considered the possible influence of socio-economic development on R0. To investigate spatial variation in R0, we sourced publications from January to June 2020 in PubMed, LitCOVID and WHO COVID-19 databases. Peer-reviewed English-language papers were included that provided R0 estimates for the beginning of the pandemic. A total of 81 studies were included. For each study, the value of the estimate, country under study and publication month were extracted. The HDI was obtained from the UN website [6] for each included country. The studies covered 65 countries across 5 continents. The median start-of-pandemic R0, both national and local, experienced a shift as more estimates became available. We represent this shift by publication month (Figure 1). For each month, the median was calculated based on all the papers that had been published up to and including that month. This should not be interpreted as a change in the R0 value over time, as most of the R0 estimates were calculated based on data from the beginning of the pandemic. Rather, this reflects a change in the understanding of the disease over time as the pandemic started at each location. This may be due to various factors, such as more sophisticated estimation techniques or improving epidemiological knowledge. The estimates published in January and February were all obtained for China, and follow a similar pattern, ranging between 2 and 4. The estimates in January were all local, in particular calculated for Wuhan. For February, it should be noted that an estimate of size 14.8 for the Diamond Princess cruise ship was removed as an outlier and is not visualised here. This number may not be representative of COVID-19 transmission, as the disease spread rapidly in the confined conditions of the cruise ship. The influence of super-spreaders would also have been exacerbated by these conditions. Recent studies suggest that super-spreaders are responsible for the majority of infections [12,13]. Academia Letters, July 2021 ©2021 by the authors — Open Access — Distributed under CC BY 4.0 Corresponding Author: Renate Thiede, renate.thiede@gmail.com Citation: Thiede, R., Abdelatif, N., Fabris-Rotelli, I., Manjoo-Docrat, R., Holloway, J., Janse van Rensburg, C., Debba, P., Dudeni-Tlhone, N., Kimmie, Z., le Roux, A. (2021). Global Variation in the Basic Reproduction Number of COVID-19. Academia Letters, Article 1956. https://doi.org/10.20935/AL1956. 2 Figure 1: Distribution of estimates of the start-of-pandemic R0 value by publication month. In March, estimates from the rest of the world became available, including Italy and Japan. A large amount of variability was experienced for this month. In April, 20 countries had estimates. Both local and national estimates evidenced a bimodal distribution, with the greater mass concentrated between 1 and 3. The smaller concentration of mass, between 4 and 6, is largely due to the local estimates for the United States from the paper of [14]. In May, the pattern shifted, with smaller R0 estimates generally. Very few estimates were obtained above 4. By June, most estimates were between 1 and 2.5. The higher estimates were all obtained at a national level, with the top five recorded for Brazil, India, China, Spain, the UK and France. Figure 2 shows the median R0 per country. As of June, most of the estimates of the start-ofpandemic R0 were between 1 and 4, with only the USA (median R0=4.78) and Spain (median R0=4.25) being over 4. Only two were below 1, namely Hong Kong (median R0=0.61) and Singapore (median R0=0.7). Variance could not be measured at a national level due to the small number of estimates for many countries. Of the 65 countries, 36 had only one estimate available per country. From Figure 2, it is clear that the value of the start-of-pandemic R0 differs across countries Academia Letters, July 2021 ©2021 by the authors — Open Access — Distributed under CC BY 4.0 Corresponding Author: Renate Thiede, renate.thiede@gmail.com Citation: Thiede, R., Abdelatif, N., Fabris-Rotelli, I., Manjoo-Docrat, R., Holloway, J., Janse van Rensburg, C., Debba, P., Dudeni-Tlhone, N., Kimmie, Z., le Roux, A. (2021). Global Variation in the Basic Reproduction Number of COVID-19. Academia Letters, Article 1956. https://doi.org/10.20935/AL1956. 3 Figure 2: National medians of start-of-pandemic R0 estimates as of June 2020. and regions. Since the R0 value was estimated at the beginning of the pandemic, this is not a result of varying laws or non-pharmaceutical interventions in the face of the pandemic, but rather some factors related to the ordinary state of affairs in a country prior to lockdown. In order to investigate possible reasons for these differences between countries, we considered the impact of the development level of a country on R0. The UN HDI and the World Bank classification of economies’ development status were used as indicators of socio-economic development. The World Bank classifies countries as developed, developing, or transitioning from developing to developed [7]. Based on the R0 values in the study, R0 has weak or no linear relationship with the HDI (ρ=0.21), making linear regression infeasible. The trends in the data may be studied visually. Figure 3 shows these patterns in the data. No high R0 values were obtained for countries with low HDI. The boxplots show that the median R0 is similar for developing versus developed countries, however the spread of values is negatively skewed for developing countries and positively skewed for developed countries. Additionally, the variability is higher for developed countries. It should be noted that most estimates were obtained using the data of reAcademia Letters, July 2021 ©2021 by the authors — Open Access — Distributed under CC BY 4.0 Corresponding Author: Renate Thiede, renate.thiede@gmail.com Citation: Thiede, R., Abdelatif, N., Fabris-Rotelli, I., Manjoo-Docrat, R., Holloway, J., Janse van Rensburg, C., Debba, P., Dudeni-Tlhone, N., Kimmie, Z., le Roux, A. (2021). Global Variation in the Basic Reproduction Number of COVID-19. Academia Letters, Article 1956. https://doi.org/10.20935/AL1956. 4