key: cord-288432-n2y9cunc
authors: Liu, Kun; Ai, Siqi; Song, Shuxuan; Zhu, Guanghu; Tian, Fei; Li, Huan; Gao, Yuan; Wu, Yinglin; Zhang, Shiyu; Shao, Zhongjun; Liu, Qiyong; Lin, Hualiang
title: Population movement, city closure in Wuhan and geographical expansion of the 2019-nCoV pneumonia infection in China in January 2020
date: 2020-04-17
journal: Clin Infect Dis
DOI: 10.1093/cid/ciaa422
sha: 
doc_id: 288432
cord_uid: n2y9cunc

BACKGROUND: The unprecedented outbreak of 2019-nCoV pneumonia infection in Wuhan City caused global concern, the outflowing population from Wuhan was believed to be a main reason for the rapid and large-scale spread of the disease, so the government implemented a city closure measure to prevent its transmission considering the large amount of travelling before the Chinese New Year. METHODS: Based on the daily reported new cases and the population movement data between January 1 and 31, we examined the effects of population outflow from Wuhan on the geographical expansion of the infection in other provinces and cities of China, as well as the impacts of the city closure in Wuhan in different scenarios of closing dates. RESULTS: We observed a significantly positive association between population movement and the number of the 2019-nCoV cases. The spatial distribution of cases per unit outflow population indicated that some areas with large outflow population might have been underestimated for the infection, such as Henan and Hunan provinces. Further analysis revealed that if the city closure policy was implemented two days earlier, 1420 (95% CI: 1059, 1833) cases could have been prevented, and if two days later, 1462 (95% CI: 1090, 1886) more cases would be possible. CONCLUSIONS: Our findings suggest that population movement might be one important trigger for the transmission of 2019-nCoV infection in China, and the policy of city closure is effective to control the epidemic.

In December 2019, an unprecedented pneumonia outbreak caused by a novel coronavirus, namely 2019-nCoV, emerged in Wuhan, the capital city of Hubei Province in China [1] . Similar with severe acute respiratory syndrome (SARS), the outbreak was highly suspected to be linked to the wild animals in the seafood market, although the definitive source was not clear yet [2] .

As of January 31, 2020, the infection has been transmitted to all the provinces in China and a few other countries. Epidemiology evidence showed that most of the cases outside Wuhan had a history of living or travelling to Wuhan, and human-to-human transmission route was possible [3] , which might be the reason for a rapid increasing rate of infection across the country and globally [4] .

Considering the person-to-person transmission and the large travel volume during the traditional Chinese New Year (the largest annual population movement in the world), it is expected that the population movement would lead to further expansion of the infection, so the government imposed a lockdown on Wuhan City at 10:00 am on January 23, as well as some other cities later on [5] . However, an estimated 5 million individuals had already left Wuhan for the holiday or travelling, some of which rushed out after the lockdown announcement [6] . In addition, the novel coronavirus is infectious during the incubation period and when the symptoms are not obvious, which is likely to make the huge floating population potential sources of infection [7] . Therefore, it is reasonable to hypothesize that the population transported A c c e p t e d M a n u s c r i p t 6 from Wuhan may have a significant impact on the potential outbreaks in other parts of China.

Recent studies on the novel coronavirus pneumonia focused more on its etiology [8, 9] , transmission route [10, 11] , and epidemiological characteristics [12, 13] , there is still a lack of investigating the relationship between the migrating population and the outbreak, which is of great importance for making intervention policies. Thus, we conducted this study with the following objectives: 1) to evaluate the impacts of the population movement on the spatial transmission of the 2019-nCoV cases at the provincial and city levels in China; 2) to estimate the potential outbreak risk at areas with the population outflowed from Wuhan; 3) to evaluate the effectiveness of the city closure measures on the epidemic control.

The data on the daily number of 2019-nCoV pneumonia cases from January 1 to 31 were derived from the real-time update of the China Health Commission (http://www.nhc.gov.cn/), 2019-nCoV epidemic report on the websites of Phoenix and Dingxiangyuan. The diagnosis and definition of the case have been described elsewhere [2, 3] . In brief, a confirmed case was defined as a pneumonia case that was laboratory confirmed 2019-nCoV infection with related respiratory symptoms and a travel history to Wuhan or direct contact with patients from Wuhan.

A c c e p t e d M a n u s c r i p t 7

As the city closure took place at 10:00 am on January 23, 2020, and the incubation period of the infection was considered to be about 3-7 days [14] , we obtained the daily index of population outflow from Wuhan and the proportion of the daily index from Wuhan to other provinces and top 50 cities, from January 1 to 31 in 2020, the information was retrieved through the Spring Festival travel information of China released by Baidu Qianxi. The data came from Baidu Location Based Services (LBS) and Baidu Tianyan based on location and traffic information systems, which could provide real-time dynamic information on regional population outflow. Data of Baidu Qianxi was freely available to the public (http://qianxi.baidu.com).

The daily index of population outflow from Wuhan to other provinces and top 50 cities was obtained by multiplying the daily index of population outflow within 

C is the cumulative 2019-nCoV cases in each province from January 1 to 31:

D is the total index of population inflow from Wuhan to other provinces:

We finally calculated the average number of cases per unit outflow population for each province in China: The net loss index of outflow population caused by advanced Wuhan city closure for each province: 11 A c c e p t e d M a n u s c r i p t 10 Similarly, we evaluated the impacts of one-day and two-day delayed city closure. We took the average index of the population outflow between January 21 and 23 as the daily index of population outflow before the city closure, and used the same calculation method to estimate the index of population outflow within Wuhan increased by the delayed city closure on January 24 and January 25 (the delayed outflow index). We multiplied the delayed outflow index by the average proportion and one corresponding unit to estimate the increased number of cases caused by one-day and two-day delayed city closure of Wuhan for each province in China. (Table 1) . On the contrary, if the closure measures was delayed for one to two days, the number of cases would increase by 722 and 1462, respectively. Our study provided timely evidence for the formulation of efficient strategies to prevent diseases from spreading out. On the one hand, the result could help assess the effectiveness of the prevention and control efforts. For example, the cases in Zhejiang and Guangdong are apparently more than estimated, which indicated a better health emergency response system (i.e. higher detection efficiency) or inadequate isolation, whereas the cases reported in Henan were much lower than expected. Two possible explanations should be considered: (1) strong prevention and control measures had been adopted in Henan; (2) the epidemic in Henan has been underestimated and enhanced screening efforts should be enforced. On the other hand, exploring the association was expected to help identify high-risk areas and guide health strategy A c c e p t e d M a n u s c r i p t 15 formulation [19, 20] . Take Henan as an example, great difference between estimated and reported data may imply a great increase of cases in the future, which required enhancement of the surveillance system and rational allocation of resources [17] . The medicine supply, personal protective equipment, hospital supplies, and the human resources necessary to respond to an outbreak should be always ensured [21] . In addition, this study could be used to guide the assessment of the risk of disease transmission and help raise public awareness. As a large number of infected people had transported to all of 31 provinces, epidemics across the country may be inevitable.

To halt the spread of the epidemic, harsh measures including quarantine and isolation of exposed persons, cancellation of mass gatherings, school closures, and travel restriction were needed to reduce transmission in affected areas. Furthermore, screening of people who have been to Wuhan recently was of crucial importance, especially cities with close ties to Wuhan.

Considering the impact of population movements on the outbreak, the Wuhan government announced the suspension of public transportation on January 23, 2020, with a closure of airports, railway stations, and highways, to prevent further disease transmission [22] . Despite inconsistent reports on the role of the lockdown in halting the disease transmission across China [11, 20, 21] , the unprecedented measure might play an important role in slowing the epidemic spread, especially when an effective vaccine was developed [23, 24] . In addition, to explore the impact of date selection, we estimated the changes of cases when the measure was implemented on different A c c e p t e d M a n u s c r i p t 16 date. The results varied significantly, 1420 cases could be prevented with the measure implemented two days earlier, and the number of cases will increase by 1462 with the lockdown implemented two days later, suggesting that the effect of the lockdown depending on the choice of date greatly, which could provide a reference for the future outbreaks. Since the political and economic effects were not considered, further studies on secondary impacts of the measure, like socioeconomic impacts, were also warranted. Though we estimated that some cases would possibly be prevented if the policy was implemented earlier, it was actually hard to make such a huge decision given the whole picture of the infection was not clear at that stage. The authors believe that the current policy was appropriate at this complex situation.

There were a few limitations of our study. Firstly, we used the index of population outflow to reflect the general real-time magnitude of population movements, so it was not an accurate representation of the actual population flow data.

Secondly, some possible influencing factors, such as socio-economic factors and demographic characteristics, were not included in the analysis because of data inaccessibility. Thirdly, it is assumed that the infected travelers in the population were randomly distributed [25] and that there was no significant difference in the surveillance capability between cities [17] , which would result in some difference between the estimated value and the actual situation. In addition, daily data used in this study was reported infection data, rather than the actual number of incident cases.

In summary, our study indicates that the population outflow from Wuhan might M a n u s c r i p t 

We appreciated the support by National Key R&D Program of China (Grant No: 2018YFA0606200). We thank our colleagues for their careful reading and editing of this manuscript.

The funder had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

The authors declare they have no conflict of interest.

A c c e p t e d M a n u s c r i p t M a n u s c r i p t 23 

New SARS-like virus in China triggers alarm

Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia

A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster

A novel coronavirus outbreak of global health concern

Modelling the epidemic trend of the 2019 novel coronavirus outbreak in China

Mayor of Wuhan: more than 5 million people left Wuhan

Real-time tentative assessment of the epidemiological characteristics of novel coronavirus infections in Wuhan, China, as at 22

The 2019-new coronavirus epidemic: evidence for virus evolution

Homologous recombination within the spike glycoprotein of the newly identified coronavirus may boost cross-species transmission from snake to human

A mathematical model for simulating the transmission of Wuhan novel Coronavirus

Novel coronavirus 2019-nCoV: early estimation of epidemiological parameters and epidemic predictions

The continuing 2019-nCoV epidemic threat of novel coronaviruses to global health-The latest 2019 novel coronavirus outbreak in Wuhan, China

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

New coronavirus pneumonia diagnosis and treatment program

Pneumonia of Unknown Etiology in Wuhan, China: Potential for International Spread Via Commercial Air Travel

Modeling Impact of Temperature and Human Movement on the Persistence of Dengue Disease

Assessment of the Middle East respiratory syndrome coronavirus (MERS-CoV) epidemic in the Middle East and risk of international spread using a novel maximum likelihood analysis approach

Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia

Population Movement Patterns Among the Democratic Republic of the Congo, Rwanda, and Uganda During an Outbreak of Ebola Virus Disease: Results from Community Engagement in Two Districts -Uganda

Forecast and control of epidemics in a globalized world

Nowcasting and forecasting the potential domestic and international spread of the 2019-nCoV outbreak originating in Wuhan, China: a modelling study

Commission CNH. Announcement from the Headquarter for novel coronavirus pneumonia prevention and control

Delaying the international spread of pandemic influenza

Empirical evidence for the effect of airline travel on inter-regional influenza spread in the United States

Predicting the international spread of Middle East respiratory syndrome (MERS)

A c c e p t e d M a n u s c r i p t 20 A c c e p t e d M a n u s c r i p t 22