key: cord-1009129-faojb5hb authors: Nguyen-Tran, M.-D.; Abozaid, A. A.-F.; Tai, L. L. T.; Minh, L. H. N.; Le, Q.-L.; Nguyen, H.-D.; Dao, K.-L.; Le, H.-H.; Huy, N. T. title: Cluster analysis of epidemiological characteristic features of confirmed cases with the novel coronavirus (COVID-19) outside China: a descriptive study date: 2020-07-01 journal: nan DOI: 10.1101/2020.06.28.20142000 sha: b1a02c0e79b6b0593387a11b59a63e23b2892b47 doc_id: 1009129 cord_uid: faojb5hb Background: Novel coronavirus COVID-19 has caused significant global outbreaks outside China. Many countries have closed their borders with China and performed obligate protective procedures, however, this disease was still rising worldwide. In this report, we aim to identify transmission patterns from China to other countries, along with describing the disease control situation of countries. Methods: We retrospectively collected information about infected cases with COVID-19 from WHO situation reports, official notification websites of health ministries and reliable local newspapers from each country. Descriptive and cluster analysis was performed to describe the transmission characteristics while the logistic regression test was used to estimate the risk factors for the occurrence of an infected individual with an unknown source. Results: A total of 446 infected cases were recorded from 24 countries outside China until 12 February 2020, with the number of reported infected cases were doubled every 3.08 days (range from 2.6 to 3.9). Besides the spread from China, the transmission was originated from sub-endemic countries (Japan, Thailand, Singapore, Malaysia, France, German). Out of 6 countries got occurrence of an infected individual with unknown source and possible potential factors contributed to this occurrence was a time of epidemic circulating, number of patients and number of clusters when the occurrence still has not happened, and notably, the unreported situation of Chinese tourists information. Conclusions: The situational reports of each country about COVID-19 should be more detailed mentioning the transmissions routes with keeping contact tracing of the unknown cases to increase the control of this disease. We have measured the reliability of our sources using The Journal of the American Medical 144 Association (JAMA) four benchmarks published in 1997. The four criteria involved the 145 authorship (the information of authors or writers and their affiliation should be provided), 146 attribution (the sources have to list their references prominently), disclosure (ownership, funding 147 source, advertisement and conflict of interests should be obvious for readers) and currency (the 148 date of publishing and updating contents should be posted) 17 . 149 All data have been acquired from online sources and governments` websites without any direct 151 contact data from patients. 152 We defined the transmission clusters epidemiologically into four main types, A, B, C, and D, 154 based on the infectious source and risk of an outbreak. A cluster was classified into group A if the 155 primary pathogen had a history of traveling to China or have been in contact with foreigners 156 abroad outside countries, who were later confirmed SARS-CoV-2-infection. Because it was 157 impossible to determine who exactly spread the disease in the case of an infected group returned 158 from China, each of the patients in the group would be considered as an independent primary 159 pathogen and established an independent cluster. Group B involved clusters that patients without 160 traveling to China but got local contact with Chinese tourists. Due to the information of Chinese 161 tourists was not disclosed and such tourists were not reported as infected cases, we considered 162 clusters in group B to be at higher risk of an outbreak than group A. Groups C and D, at more 163 risk, included clusters where the source of infection could not be identified. While patients in a 164 cluster C were involved with a specific area, such as a church, a conference, etc., information of 165 patients in clusters D remained completely unknown. 166 The doubling time in our study is defined as the time required for the number of reported cases to 167 double. It was calculated based on the formula: N = N0.2 . Because we only could record the 168 date of diagnosis (not the date of expose), the value hence only partially reflected the rate of 169 disease transmission. 170 Descriptive demographic and cluster analysis have been conducted inclusively for this dataset. 172 The missing data have been considered as unreported due to some governments didn`t reveal 173 these data to protect the personality of their patients. Logistic regression then was performed to 174 determine relationships between potential factors and the occurrence of an infected with 175 unknown source, with the significant value chosen was < 0.05. 176 177 Demographic characteristic of infected patients 180 We have identified 446 cases from 24 countries confirmed with COVID-19 infection outside 181 China. The median (IQR) number of reported cases doubling time was 3.08 ± 0.35 (range from 182 2.6 to 3.9) days and recorded the highest increase of reports (n = 113) between 10-Feb and 11-183 Feb 2020, mostly from cases of Diamond Princess (Figure 1 ) 18 . Concisely, the most affected 184 country with the biggest number of infected patients was Singapore with a total number of 50 185 since the beginning of the outbreak up to February 12, 2020 ( Table 3 showed a summary of the size and the type of clusters. The biggest cluster recognized 202 was the Diamond Cruise ship, with a total of 175 infected patients 18,24-26 , followed by the cluster 203 in Germany (13 patients), Singapore (9 patients), France, Viet Nam, UK, Korea (6 patients) 204 19, 21, 27 . Notably, Results also have shown that the number of clusters of Group A, B, C and D was 205 respectively 182 (85.0%), 9 (4.2%), 10 (4.7%), 13 (6.1%). When approaching group, A, B, we 206 found that 168 non-transmissible clusters (1-patient cluster), 18 clusters confirmed the local 207 transmission between primary pathogen and local people, and 5 clusters recognized the 208 transmission between local people. 209 At the national level, out of 24 countries infected with COVID-19, the occurrence of an infected 210 individual with unknown source happened in 6 countries (Australia, Germany, Japan, Singapore, 211 Thailand, UAE). Summarily, there were 13 countries that local human-human transmission 212 began, with the number of local transmission (F2 and F3) range from 5.6% in Malaysia to 55.6% 213 in UK from identified primary pathogen (F1). Other information has been summarized in Table 214 . CC-BY-NC-ND 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) The copyright holder for this preprint this version posted July 1, 2020. . Table 5 showed the associated factor with the occurrence of an infected individual with 217 unknown source. Time period from the first case reported until Feb 12 was related with 218 is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 1, 2020. . https://doi.org/10.1101/2020.06.28.20142000 doi: medRxiv preprint time of the outbreak inside China to be 7.4 days on January 22, 2020 6 . The mean doubling time 238 we calculated here was every 3.08 ± 0.35 days outside China (Figure 1) . However, we could not 239 compare both estimates to each other, and other similar outbreaks due to lack of sufficient data of 240 expose time in our sources unlike they had. Therefore, we couldn`t either assess R0, besides, 241 further investigations are needed regarding this concern as this rate was likely to be increased Table 1 ). The primary 252 transmission (Group A) represented the majority that brought the primary pathogen (n = 168). 253 Moreover, we did more subgrouping of these cases to know where most of cases mostly came 254 from and found the majority of them are Chinese people who recently returned from China (Aa) 255 (n = 159), however, some cases recognized from other sub-endemic countries (Ab) (n = 9) as in 256 Japan 38 , Singapore 39 , France 40 , Germany 41 , Thailand 42 and Malaysia 43 . highlighting that the 257 governments should not only focus on checking people from China, but also, in countries that 258 reported COVID-19 infected cases. In addition, we noticed some of the reported cases had no 259 history of travel to China, where the outbreak took place but had been in close contact with 260 . CC-BY-NC-ND 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 preprint this version posted July 1, 2020. The primary cases who infected other persons later (F1) were greatest in Thailand (n = 26), 263 Singapore (n = 24), and Japan (n = 19) which is suggestive of the poor tracking the patients 264 entering the country (Table 4 ). In Thailand, a taxi driver was reported to have been infected on 265 January 28, 2020. The patient had no history of travel to China but was in contact with a Chinese 266 tourist which the first case of a person to person transmission in the country 46 . Moreover, 267 transmission with difficulty in detecting the source of infection, and infections with no detectable 268 source (Group C and D) (either not reported or cannot be detected) was presented in some 269 countries. In Australia, UAE and Germany reported 1, 3 and 1 respectively in these two groups, 270 while Asia recorded the highest number of such cases reaching 11 cases in Singapore, 3 in Japan 271 (Table 4) . These are serious numbers and can be misleading as suggestive of the emergence of 272 the disease in these counties with no transmission from China being reported. However, they 273 were estimated in the early period, so they were not always reflected the current statements as 274 some country may gain the ability to find the links of their cases like Singapore latterly but still 275 give some predictions of future events in other countries like what happened later in Japan and 276 Germany. To note, we try to develop a new way of predictions and rapid judgmental tools of 277 ongoing daily situations report for those countries who care to identify and clarify their present 278 conditions. Therefore, authorities should exert more effort in reporting all cases in an evidence-279 based manner. The earliest case to be reported outside China was in Thailand on January 13, 280 2020 who had been to China lately 20 . Besides, the only patient that died from COVID-19 from 281 outside China was reported in Philippine on February 2, 2020 22 . However, the Death numbers 282 were likely to be increased in the following days because of the continuously increasing numbers 283 . CC-BY-NC-ND 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 preprint this version posted July 1, 2020. . The pathophysiology of this COVID-19 has not powerfully known, but Sohrabi et. al. said that 285 most of the transmission occurred in the symptomatic period with increasing some reporting of 286 asymptomatic transferring cases 47 . Subsequently human to human transmission was reported in 287 Vietnam 48 , Japan 49 and USA 50 . Some cases are asymptomatic 12,51,52 due to the 2-week incubation 288 period of the virus which makes it hard to detect these cases, further, possible transmission of the 289 virus to other contacts is easy. In addition, a Vietnamese case was reported with normal clinical 290 findings and mild symptoms 53 . We couldn`t report these findings due to non-adequate 291 information in the majority of cases. Nonetheless, we kept tracked to know about the transferring 292 patterns of this disease in different countries, then we found the direct transmission ratio from 293 Chinese people (F1 to F2) has recorded the highest in the UK (55.6%), Vietnam (44.6%) and 294 Republic of South Korea (24%) while the indirect local ratio (F2 to F3) revealed in South Korea 295 (33.3%) and Vietnam (14.3%) can represent a hazard ratio of outbreak occurrence but still 296 remained non-significant (p > 0.05) ( Table 4, Table 5 ). However, the impact of unreported 297 Chinese cases in out of control countries was a significantly high risk of the outbreak (p < 298 0.001). The time of announcement of cases and the number of control patients or clusters are still 299 risk factors to produce pandemic disease (p < 0.05) ( Table 5 ). The cluster size didn`t show a 300 significant this risk attribution (p > 0.05), in spite of that, this assessment has been made in the 301 early stage when the super-spreader factors haven`t grown yet. Consequently, healthcare officials 302 must put restrictive measures at all levels for rapid detection of the virus on people coming from 303 China or those countries which became lastly an epicenter of this disease. 304 Eventually, we would like to recommend the diagnostic measures of the COVID-19 to be more 305 developed. Mostly we used RT-PCR techniques according to the WHO recommendations 54 . 306 . CC-BY-NC-ND 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 preprint this version posted July 1, 2020. In this study, we reported the epidemiological features of the COVID-19 outside China. Further 329 . CC-BY-NC-ND 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 preprint this version posted July 1, 2020. Not applicable. 349 Not applicable. 351 . CC-BY-NC-ND 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 preprint this version posted July 1, 2020. . https://doi.org/10.1101/2020.06.28.20142000 doi: medRxiv preprint All data generated or analysed during this study are included in this published article [and its 353 supplementary information files]. 354 No funding resources were gained for this study. 356 We declare that the authors had no conflicts of interest while conducting this study. CC-BY-NC-ND 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 preprint this version posted July 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 1, 2020. . CC-BY-NC-ND 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 preprint this version posted July 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 1, 2020. CC-BY-NC-ND 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 preprint this version posted July 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 1, 2020. . https://doi.org/10.1101/2020.06.28.20142000 doi: medRxiv preprint CC-BY-NC-ND 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 preprint this version posted July 1, 2020. . https://doi.org/10.1101/2020.06.28.20142000 doi: medRxiv preprint . CC-BY-NC-ND 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 preprint this version posted July 1, 2020. . . CC-BY-NC-ND 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 preprint this version posted July 1, 2020. . is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 1, 2020. CC-BY-NC-ND 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) The copyright holder for this preprint this version posted July 1, 2020. . https://doi.org/10.1101/2020.06.28.20142000 doi: medRxiv preprint A novel coronavirus associated with severe 373 acute respiratory syndrome Newly discovered coronavirus as the primary 375 cause of severe acute respiratory syndrome Identification of a novel coronavirus in patients with 377 severe acute respiratory syndrome Isolation of a 379 novel coronavirus from a man with pneumonia in Saudi Arabia Commentary: Middle East respiratory syndrome 382 announcement of the Coronavirus Study Group Early transmission dynamics in Wuhan, China, of novel 385 coronavirus-infected pneumonia Systematic Comparison of Two Animal-to-Human Transmitted 387 Human Coronaviruses: SARS-CoV-2 and SARS-CoV The effect of travel restrictions on the spread of the 389 2019 novel coronavirus (2019-nCoV) outbreak The Novel Chinese Coronavirus Infections: challenges for fighting the storm Cambodia confirms first case of coronavirus -health minister Finland's first coronavirus case confirmed in Lapland The first 2019 novel coronavirus case in 469 First patient with coronavirus reported in Sri Lanka TheLocal.se. First case of coronavirus confirmed in Sweden. TheLocal.se. 2020 Republic of South Korea: Yonhap News LD) 3 more virus cases bring S. Korea's total to 19 Yonhap News Agency Sanidad confirma en La Gomera el primer caso de coronavirus en España Four more cases discharged; three new confirmed cases of novel coronavirus 493 infection Epidemiology and cause of severe acute respiratory 497 syndrome (SARS) in Guangdong, People's Republic of China CoV) epidemic using open source data Journey of a Thai Taxi 503 D: No information 0: Only Fn patients (patients related to area clusters) 1: Fn and Fn+1 patients