key: cord-0805491-7zu3864b
authors: Sakamoto, Masashi; Sasaki, Daisuke; Ono, Yuichi; Makino, Yuko; Kodama, Eiichi N.
title: Implementation of evacuation measures during natural disasters under conditions of the novel coronavirus (COVID-19) pandemic based on a review of previous responses to complex disasters in Japan
date: 2020-10-16
journal: Progress in disaster science
DOI: 10.1016/j.pdisas.2020.100127
sha: e86a1fda98f1ba64ba436539e86939a15fa1b842
doc_id: 805491
cord_uid: 7zu3864b

We aimed to investigate how evacuation measures could be effectively implemented in the event of multiple disasters caused by natural hazards under conditions of the novel coronavirus infection (COVID-19) pandemic, which is rapidly spreading worldwide. We conducted a review of literature focusing on complex disasters, entailing natural disasters in combination of outbreaks or endemics of infectious diseases. Using the Google Scholar search engine, we identified and reviewed 24 papers sourced from academia, governments, and concerned organizations, and associated data on such disasters, commencing with the Great Hanshin-Awaji Earthquake, which occurred in 1995. In light of our review, we developed a summary of correspondences and problems linked to compound disasters involving conjunctions of outbreaks/endemics and natural disasters that could offer insights for developing measures to deal with natural disasters that occur in the context of the COVID-19 pandemic. We subsequently attempted to differentiate the characteristics of evacuation measures relating to COVID-19 from those relating to other infectious diseases using three sets of extracted keywords: (1) surveillance and information sharing, (2) evacuation center environment and stockpiled supplies, and (3) community disaster risk reduction and community leadership. We identified issues relating to evacuation measures that would need to be explored further to improve disaster management and preparedness in the future.

Following the initial occurrence of COVID-19 in the city of Wuhan in Hubei Province, China in December 2019, this highly infectious disease rapidly spread throughout China and the world, prompting the World Health Organization (WHO) to declare a pandemic on March 12, 2020 (WHO, 2020 . As of May 31, 2020, there were 16,851 cases of CIVID-19 infections and 891 deaths in Japan (Ministry of Health, Labour and Welfare, 2020). COVID-19 has a post-infection incubation period of 1-14 days. The main symptoms reported by many infected people are cold-like symptoms, such as fever and dry cough, excessive fatigue, and dysgeusia. Approximately 80% of COVID-19 patients experience spontaneous remission, but 20% go on to develop pneumonia/pneumonitis. Of the latter, the high-risk group comprises not only older individuals and those with certain predisposing risk factors, such as hypertension, diabetes, and/or cardiac diseases but also some young adults, who develop severe illness.

Existing antivirals and related drugs have only provided partial, if any, beneficial outcomes.

Moreover, there are fears that the severe disruptions of economies and daily life caused by the pandemic will be prolonged rather than transient phenomena. Therefore, there is an urgent need to develop and establish effective drugs and vaccines to suppress and overcome the pandemic.

At the time of writing, the rainy season is approaching in Japan, leading to growing concerns about the potential occurrence of natural disasters like flooding coinciding with the COVID-19 pandemic.

The Cabinet Office of Japan has therefore issued multiple notifications providing evacuation guidelines, and various organizations have issued calls for preparations for complex disasters involving a conjunction of COVID-19 and natural disasters (e.g., Japan Academic Network for Disaster Reduction, 2020).

The main concerns at the time of occurrence of such complex disasters relate to evacuation procedures. The first volume of the Preliminary Report on the Spread of Novel Coronavirus Infection, published in 2020 by the Disaster Reduction and Human Renovation Institution and the Novel Coronavirus Evacuation Support Book (Japan Voluntary Organizations Active in Disaster, 2020) both describe concrete methods for operating evacuation centers and measures to be taken for each evacuee. The option of dispersal of evacuees, with suggested evacuation points, is also included in the Recommendations Regarding Evacuation, issued by the Japan Society for Disaster Information (2020).

The Great Hanshin-Awaji Earthquake, which occurred in January 1995, prompted an evaluation of countermeasures implemented during the simultaneous occurrence of large-scale disasters and infectious diseases. A group of researchers who implemented -A study on construction of health risk management system for dental health care in large-scale disaster," hereinafter referred to as the J o u r n a l P r e -p r o o f Journal Pre-proof Study Group, reported that aspiration pneumonia was prevalent among evacuees, especially older evacuees, who were physically and mentally exhausted after this earthquake (Study Group, 2009 

Each of the above-mentioned studies focused either on the particular natural disaster examined or on the associated pandemic. In light of their finding that only a few studies have linked disaster risk reduction to biological hazards, such as COVID-19, Zhang and Shaw (2020) have called for studies that incorporate various aspects of complex disasters (pandemic response, recovery, and long-term development). To date, there have been no attempts to compile and systematically organize information on past occurrences of complex disasters, entailing a combination of natural disasters and infectious diseases and to discuss evacuation measures that specifically relate to COVID-19, distinguished from other infectious diseases. While organizations have provided recommendations, and individual local governments have published evacuation manuals (e.g. Koyama et al., 2020) , a notable gap in the literature concerns measures relating to evacuation and the operation of evacuation centers in the case of a complex disaster in which COVID-19 and natural disasters occur simultaneously.

Therefore we aimed to conduct a systematic review of various available studies and reports on natural disasters and infectious disease epidemics. Accordingly, we compared evacuation procedures during natural disasters in Japan, which are frequent occurrences, in the context of common infectious diseases and subsequently extracted specific issues that would apply to evacuees during the COVID-19 pandemic.

We conducted a search of the Google Scholar database using -disaster‖ and -infection‖ as keywords for identifying and extracting reports and papers focusing on disasters associated with infectious diseases in Japan, commencing with the Great Hanshin-Awaji Earthquake in 1995. Our search yielded a total of 24 academic and non-academic documents and papers on measures adopted to handle outbreaks of infectious diseases at evacuation centers and other places in the event of disasters. As of the end of May, 2020, we have been unable to find any literature that focused J o u r n a l P r e -p r o o f Journal Pre-proof on COVID-19 and evacuation during a natural disaster using the keywords "COVID-19" and "evacuation."

The WHO recommends the adoption of systematic measures relating to the following steps: (1) rapid assessment, (2) prevention of infections, (3) surveillance, (4) disease outbreak control, (5) disaster management, and (6) evaluation aimed at controlling the spread of infectious disease during disasters (Hashimoto, 2013) . However, these measures are difficult to implement in a situation where damage is so extensive that medical institutions and administrative organizations are also at risk of being affected because the medical support staff at hospitals are themselves potential victims of the disaster. Surveillance was implemented 20 days to one month after the 2011 Great East Japan Earthquake and five days after the 2018 Kumamoto Earthquake (Kawamura et al., 2017) . Although systematic measures and mechanisms for enabling quick responses have greatly improved, in the event of a large-scale disaster, they may not function during the initial post-disaster stage.

Surveillance entails various constraints, such as the preparation time required, disruption of reporting procedures, and an excessive burden placed on staff members at the evacuation centers. In particular, many small evacuation centers do not have permanently stationed medical professionals, and in most cases, the victims' situations are not monitored. In the event of a large-scale disaster, it is necessary to develop and strengthen human resources for implementing countermeasures to control infectious diseases because not only specialists, such as public health nurses, are required to deal with infectious diseases within evacuation centers but also other local government personnel have important roles to play (Kunishima and Kaku, 2012) . Furthermore, a system should be established for disaster preparation that enables cooperation and information sharing with specialized organizations, such as local government departments, medical institutions, and universities during times of non-crisis.

There are three key factors that define the epidemiology and pathophysiology of infections: pathogens, hosts, and environments, of which environmental factors reportedly have the greatest influence on the risk of infectious disease outbreaks occurring after natural disasters . The deterioration of hygiene conditions in disaster-affected areas and the forced confinement of many victims within small spaces in evacuation centers increase the risk of transmission of infectious diseases. Therefore, a countermeasure focusing on maintaining a sanitary environment within evacuation centers is critical. Kamase and Saida (2010) have pointed out that there would be differences in the storage locations of stockpiled goods, the maintenance of records at evacuation centers, and the preparation of J o u r n a l P r e -p r o o f information for evacuees, depending on whether evacuees have previously experienced a disaster.

Clearly, another countermeasure relates to appropriate response of administrators. However, in the event of a large-scale earthquake, it is likely that administrative agencies themselves will be damaged.

In addition, not only the staff of the evacuation centers but also the evacuees must protect themselves against infectious diseases at small evacuation centers because of the lack of human resources. The evacuees are required to be self-reliant. Okada et al. (2014) noted that community empowerment has greatly contributed to the maintenance of healthy environments at evacuation centers.

In light of the above considerations, we extracted three sets of keywords to explore the relationship between infectious diseases and natural disasters that may have been issues during previous disasters. These keywords were "surveillance and information sharing," "evacuation center environment and stockpiled supplies," and "community-based disaster risk reduction and community leadership." In the following sections, focusing on these three sets of keywords, we explore the differences between COVID-19 and other infectious diseases within responses to previous natural disasters and infectious diseases described in the reviewed literature.

The Great Hanshin-Awaji Earthquake that erupted in Japan's Hyogo Prefecture in 1995 caused significant damage in the cities of Kobe and Osaka and left 6,434 people dead. Of these deaths, 5,512

were directly attributable to the earthquake, caused by collapsed buildings that led to crushing and suffocation of victims. Of the remaining victims, 922 died of pneumonia, strokes, and myocardial infarctions induced by the disaster. Approximately 60% to 80% of pneumonia cases in older evacuees at the time of the earthquake were thought to be or diagnosed as aspiration pneumonia. Reports on responses to the Great Hanshin-Awaji Earthquake and on aspiration pneumonia were compiled by a team of researchers (Study Group, 2009 ).

Most cases of aspiration pneumonia that occurred among older, immunocompromised individuals are believed to have been caused by the aspiration of saliva rich in oral bacteria because of reduced oral care during the extensive disaster. It has been reported that immediately after the Great Hanshin-Awaji Earthquake many older individuals were unable to maintain oral hygiene practices, such as gargling, because of an extreme shortage of drinking water or because of an increase in oral bacteria caused by poor cleaning of dentures. According to the findings of one study, oral care that is thoroughly performed, thus preventing an increase in oral bacteria while keeping the oral cavity and denture surfaces clean, can lead to a reduction in incidences of pneumonia among the elderly by 40% (Yoneyama et al., 2001) . However, in the event of large-scale disasters, the failure of the water supply J o u r n a l P r e -p r o o f Journal Pre-proof system results in difficulties in procuring water at evacuation centers. Therefore, there is a possibility that sanitation practices required for effective health management, such as hand washing, gargling, and brushing teeth, cannot be performed. As noted in the report, the issue of privacy relating to denture removal also arose at the evacuation centers. In the absence of water, chewing sugarless gum containing xylitol is considered an effective alternative oral hygiene practice. In addition, mouthwashes, especially -toothpaste sheets,‖ are considered effective, and stockpiles of these items should be maintained at evacuation centers (Yoneyama et al., 2001) .

The Great East Japan Earthquake, which occurred in March 2011, caused extensive damage over a wide area, compelling the evacuation of approximately 300,000 people. Influenza and infectious gastroenteritis spread in the disaster-affected area because the earthquake occurred during the cold weather season when there are frequent outbreaks of such infectious diseases, even in the absence of disaster-related events (e.g., Kunishima and Kaku, 2012) . As this was a large-scale earthquake, medical supplies and equipment fell short in the medical facilities as well as at the evacuation centers.

Within the medical facilities, personal protective equipment, such as alcohol-based disinfectants, masks, and aprons were generally sufficient, but supplies of paper towels, liquid soap, containers for infectious medical waste, and catheters, which are used frequently on a daily basis, were inadequate.

At the evacuation centers, there were concerns about the high risk of infectious disease transmission within deteriorating sanitary environments, given large numbers of people living in close proximity in spaces such as school gymnasiums and halls and the scarcity of water for washing hands, cleaning, and food sanitation caused by the suspension of water supplies., Alcohol-based disinfectants, masks, and Tamiflu stockpiled for influenza outbreaks were procured for the evacuation centers through the coordinated efforts of the prefectural government and medical institutions.

Surveillance is one of the measures applied for monitoring and controlling infectious diseases disaster was particularly difficult because of gasoline shortages and traffic disruptions. One week after the disaster, only 4.4% of the evacuation centers were covered. Surveillance was limited for two weeks after the earthquake at a time when the number of occupied shelters and victims peaked (Kim et al., 2013) .

In closed or semi-closed environments, such as evacuation centers, there is an increased risk of infections and their outbreaks. According to Endo et al. (2012) , it is important to implement basic measures, such as non-pharmacological interventions (NPIs) (e.g., hand hygiene, cough etiquette, ventilation, and isolation), early detection of symptomatic persons, rapid test performance, isolation of infected individuals, and monitoring of close contacts. Proper implementation of the NPI in the midst of infectious disease outbreaks, such as influenza and infectious gastroenteritis, in disaster-affected areas includes the conduct of appropriate and accurate educational activities, an example being the distribution of the "eight articles on infection prevention" educational posters. Posters of this type for enhancing NPI literacy are posted at evacuation centers.

According to Kaku and Kunishima (2012) , adequate communication of information was considered the most important issue when dealing with infectious diseases in the aftermath of the Great East Japan Earthquake. For several days after the disaster, all means of communication, including fixed phones, mobile phones, and the Internet, were unusable, including in the city of Sendai, which is the largest city in the Tohoku region. In rural coastal areas that were severely damaged by the accompanying tsunami, mobile phones were not usable for a month after the disaster. Under such circumstances, cooperation is necessary not only among hospitals capable of dealing with infectious diseases in large cities but also among rural communities and hospitals, social welfare facilities, and other relevant agencies. Moreover, evacuation centers and administrative organizations must respond in a unified manner. In particular, the construction of a system that enables the restoration of medical institutions and the transmission and sharing of information, right from the initial stage of recovery when information transmission is challenging, is essential (Kaku and Kunishima, 2012).

In September 2015, Typhoon No. 18 and other heavy rainfall events that occurred in the regions of northern Kanto and southern Tohoku caused levees to collapse and flooding to occur in several areas.

A heavy rainfall emergency warning was issued throughout Ibaraki Prefecture, and a total of 229 evacuation centers were set up in 35 municipalities at the peak of the crisis. In Ibaraki Prefecture, in light of lessons learned from the Great East Japan Earthquake, the surveillance of evacuation centers was incorporated into the prefectural disaster prevention plan. Immediately after the occurrence of the flood disaster, the prefectural government's health prevention division applied its evacuation center 

In the case of the Kumamoto Earthquake, which occurred on April 16, 2016, evacuation centers were set up on the day of the main shock and medical care was initiated the same day. Localized torrential rainfall, measured at 105.5 mm, continued unabated for nine hours on April 21, resulting in the issuance of an evacuation advisory that led to increased numbers of occupants at the evacuation centers. At that time, there were reports of increased numbers of individuals suffering from diarrhea at the evacuation centers in north-eastern Kumamoto Prefecture; testing with a Norovirus rapid diagnosis kit confirmed a positive result for infectious gastroenteritis.

According to Goto and Okamoto (2017) , the sanitary environment of the evacuation center was inadequate, with no partitions in the living space provided by a school gymnasium. People wore their shoes, which were covered in sediment and excrement from the rain, to pass in and out of this space through a corridor. Because of the water shortage and the lack of disinfectants, lavatories could only be cleaned with water stored in buckets, and the wearing of gloves was the only protective measure taken to prevent infections. Thorough hand washing with water and soap was difficult to perform because of the water shortage, and some evacuees did not use soap and water at all to wash their hands.

On the same day that an evacuation order was announced, posters calling for hand washing with water and soap were displayed in the clinic. Further, the wearing of shoes was strictly prohibited throughout the gymnasium. Additional measures were implemented within 72 hours of the increase of J o u r n a l P r e -p r o o f infected individuals, which included isolating patients with symptoms of acute gastroenteritis. While many of the required environmental measures were initiated within 24 hours of the disaster, surveillance could not be implemented because of the lack of staff at the medical institutions and the large numbers of evacuees. If the surveillance had been carried out, individuals with symptoms of acute gastroenteritis would have been identified from an early stage and the spread of the infection could have been prevented. Further, it would have been possible to work with administrative officials at evacuation centers from an early stage (Goto and Okamoto, 2017) .

Most of the evacuation centers that the Kagoshima Prefecture Medical Association assisted were established within schools. Preparations for classes that began simultaneously made it difficult to use classrooms and health rooms because evacuees and patients were being attended to in these rooms.

Thus, places where individuals with infectious diseases could be isolated were limited. In cases where it was difficult for such patients to return home, hospitalization was considered, and pre-emptive medication was administered to close contacts at the evacuation centers. If a gastroenteritis infection is suspected, regardless of whether the results of the norovirus rapid test are positive or negative, the patient is asked to return home if he or she is able to do so. While all of these measures were taken at designated evacuation centers, it cannot be confirmed from information available with the government whether sufficient measures were implemented at non-designated evacuation centers and at shelters that did not receive relief supplies, including masks and hand disinfectants.

The performance of administrative functions mostly continued because of the efforts of the staff and the clear definitions of their roles established in the regional disaster prevention plan before the disaster struck. In addition, the medical association and the local government had built up a good relationship through pre-disaster collaborative activities, such as home medical care. Consequently, whereas activities relating to surveillance and infection control at the evacuation sites began 20 days to one month after the occurrence of the Great East Japan Earthquake, they began as early as 5 days after the occurrence of the disaster in Kumamoto. What was critically required at that time was not only information sharing among evacuation centers and medical rescue teams but also cooperation among these organizations. In the case of the Kumamoto Earthquake, the -Surveillance in Post Extreme Emergencies and Disasters, Japan version," known as -J-SPEED,‖ was introduced on a trial basis for the first time. It is expected that in future, this tool will enable the number of patients to be visualized through the inputting of information on the victims treated by medical relief teams at the emergency headquarters.

One important characteristic of the Kumamoto Earthquake was that many evacuations occurred at night and many evacuees spent the nights in their cars. Determining the symptoms of the infectious disease among evacuees who stayed in their cars was especially difficult. To address such issues, J o u r n a l P r e -p r o o f evacuees should be alerted, their self-reporting should be facilitated, and the managers of evacuation centers should devise surveillance strategies for them. In addition, because the government was not aware of non-designated evacuation centers, which were consequently omitted during the patrols, it is possible that information regarding the spread of infectious diseases would not have been reported to medical institutions (Kawamura et al., 2017) .

We applied the three previously mentioned keywords as an axis for considering evacuation issues in the context of the COVID-19 pandemic, focusing on differences between COVID-19 and other infectious diseases. Table 1 presents a comparison of evacuation status during previous complex disasters and pertinent issues under conditions of the COVID-19 pandemic. In the case of COVID-19, which is regarded as an infection that is spread through contact or droplets, there is a risk of rapid spread from infected persons. Therefore, a system is required that enables post-disaster surveillance to be carried out simultaneously with the establishment of evacuation centers. 20 days to one month after the Great East Japan Earthquake 5 days after the earthquake in the case of the Kumamoto Earthquake

Continuous implementation of measures without omissions, even for small-scale evacuation centers and evacuees who are dispersed, is essential.

Surveillance was conducted by public health nurses and evacuation center staff using surveillance systems.

A system for ensuring that information reaches evacuees, including those who are dispersed or in small-scale evacuation centers, is necessary.

Sharing information with small-scale evacuation centers was challenging.

Evacuation space 6 m 2 per person is required for securing 2 m of social distance.

Typically, 1.57-2.93 m 2 of space per person was maintained at evacuation centers. The number of individuals who can be accepted at evacuation centers will be about one-third of the number usually accepted, making it more difficult to secure evacuation space.

Because droplets are the infectious agent, securing space for isolation is the most critical issue.

In some evacuation centers established at schools designated for this purpose, it was impossible to secure spaces for isolation.

During pandemics, masks and alcohol disinfectants are in short supply. Therefore, evacuees need to bring these items with them.

Xylitol gum and mouthwash are effective for oral care when the water supply is cut off. Supplies of masks and alcohol-based disinfectants could be maintained through stockpiling. Supplies could be provided to meet the shortage.

Community disaster risk reduction and community leadership (self-help and mutual assistance)

Hand hygiene that extends beyond conventional practices is necessary and is performed according to individual evacuees' judgments and sense of responsibility.

Knowledge dissemination through visiting doctors and education of public health nurses at evacuation centers Autonomous activity Self-governance entailing residents' leadership is challenging to implement in the case of dispersed evacuees.

A hygienic environment can be maintained at environmental centers through forms of leadership existing among residents, such as residents' associations.

Whereas rapid diagnostic immunochromatographic kits are available for influenza and norovirus, they are only partially available for COVID-19. Evidently, rapid diagnosis of COVID-19 infected persons is essential, but the only method currently available for diagnosing COVID-19 is polymerase chain reaction (PCR) testing, which is difficult to carry out at evacuation centers at the time of a disaster. While the establishment of a surveillance system that includes the use of rapid and sensitive diagnosis kits should be a future priority, at present, this system is limited to isolating infected individuals, which is difficult to achieve given that many people are asymptomatic. In the case of infectious diseases, such as infectious gastroenteritis, contact transmission is dependent on evacuation centers' environments, whereas in the case of COVID-19, which is regarded as an J o u r n a l P r e -p r o o f infection transmitted through both contact and droplets, the disease may spread rapidly from the infected person. There were significant delays in carrying out post-disaster surveillance at evacuations centers in the case of events such as the Great East Japan Earthquake. However, in the context of COVID-19, because subjective symptoms may be difficult to identify, an individual may unknowingly transmitted the infection to others leading to its rapid spread. Therefore, it is critical that surveillance is conducted immediately after the occurrence of a disaster and that the stipulated sequence of measures is promptly initiated, such as immediate isolation in the case of a symptomatic individual. To prevent the spread of infection in the event that an individual is symptomatic, evacuees should implement health management on their own, using health check sheets and other relevant tools, and they should record the names of their contacts.

Initiating early surveillance is difficult in the event of a major disaster, as medical and administrative institutions may have been extensively damaged. Nevertheless, the establishment of a pre-disaster surveillance method in collaboration with the concerned individuals at evacuation centers is necessary to detect occurrences of infectious diseases at an early phase. In particular, a strategy of dispersed evacuation is recommended under conditions of the COVID-19 pandemic, while conducting continual surveillance, even when evacuation is dispersed. To ensure efficient information sharing, non-designated evacuation centers and evacuees should be listed simultaneously. A surveillance method based on self-reporting should be designed, and check sheets should be prepared in advance to facilitate responses to surveillance at a later date. In addition, surveillance of evacuees as well as external staff and visitors entering the centers should be considered.

Schools were designated as evacuation centers during previous disasters, and in some cases, it was not possible to secure spaces for isolating individuals with infectious diseases. In the case of the COVID-19 pandemic, the issue of securing space for isolation is critical for preventing the spread of the infection. In general, whereas the area to be occupied by evacuees is calculated using the standard prescription of 1.57-2.93 square meters per person in Japan (Cabinet Office, 2008), the international "Sphere criteria" standard, which provides several guidelines on sanitation and maintenance of the living environments of evacuation centers, stipulates a minimum area of 3.5 square meters per evacuee (Sphere, 2018) . Combined with social distancing of 2 meters, the total area required per person would be 6 square meters according to the international standard.

Accordingly, the number of people who could be accommodated in shelters would be reduced from one-third to a quarter of normal occupation. Evidently, the designated evacuation centers would not be sufficient to accommodate all of the evacuees. Provided that dispersed evacuation is conducted J o u r n a l P r e -p r o o f along with surveillance and information sharing, as discussed above, it would be appropriate. Another potential issue concerns the difficulty of procuring and distributing stockpiled goods, such as masks, tents, and cardboard boxes for separating compartments allocated to each evacuee at evacuation centers in the event of a pandemic. Therefore, countermeasures such as private procurement of supplies should be considered.

According to Okada (2014) , there is an increased risk of outbreaks of infectious diseases at evacuation centers during disasters, and hand hygiene in excess of conventional practices is required as an infection control measure. However, judgments and responsibilities for maintaining hand hygiene lie with the evacuees, who should be self-reliant. Strong "leadership of the people" is reported to have contributed significantly to the maintenance of sanitary environments in evacuation centers in the past (Okada, 2014) . Therefore, raising awareness to prevent the spread of infection within communities and efforts to maintain sanitary environments are essential. The question of whether "leadership of the people" can be demonstrated in the case of dispersed evacuation remains a critical one.

In this paper, we have presented a comprehensive review of countermeasures taken during previous disasters and outbreaks of infectious diseases and have outlined issues relating to the formulation of evacuation measures required during complex disasters under conditions of the COVID-19 pandemic in preparation for complex disasters. For example, while the evacuation of residents living in high-risk areas is essential, the practice of keeping them concentrated within evacuation centers needs to be revised. Given the capacities of evacuation centers and the need to practice social distancing, a strategy of dispersed evacuation is necessary. In the case of dispersed evacuation under conditions of the COVID-19 pandemic, providing that disaster risks are not high, evacuees' homes and cars may provide alternative shelters to evacuation centers. However, the problems of information sharing, identification of evacuees, and surveillance need to be addressed.

Evacuation methods may also differ depending on the duration of the evacuation (short-term or long-term), so more specific and appropriate evacuation methods need to be developed.

In addition, asymptomatic or mildly ill patients may be treated at home or at accommodation facilities. More specific options for evacuating such individuals need to be considered. To reduce the numbers of evacuees, it is essential to develop systematic approaches and measures for evacuees that are differentiated according to the targeted hazards.

Future studies should focus on examining more concrete evacuation measures. One study could focus on the responses of India and Bangladesh concerning evacuations that were compelled by a cyclone in the Indian Ocean at the end of May 2020. Comparative studies of Japanese and foreign case studies relating to evacuation associated with natural disasters under conditions of COVID-19, ways of running evacuation centers, and various available options could yield useful insights. In addition, guidelines on evacuation methods and shelter management methods specific to COVID-19, which contribute to a reduction of the complex disaster risk, should be prepared on an urgent basis and disseminated to governments at home and abroad, local governments, and other administrative organizations.

This work was supported by the Japan Science and Technology Agency under the SOLVE Project for Sustainable Development Goals (grant number JPMJRX9I3).

This work was supported in part by research grants from the Japan Society of the Promotion for Science (JSPS No. 16H05346) and by the Joint Usage/Research Center, -Research Center for Zoonosis Control, Hokkaido University.

World Health Organization, 2020. WHO announces COVID-19 outbreak a pandemic

Current situation of the COVID-19 and the response of the Ministry of Health

Please be prepared for complex disasters of infectious diseases and natural disasters

Disaster Reduction and Human Renovation Institution

Japan Voluntary Organizations Active in Disaster

Research on Establishment of Health Crisis Management System for Dental Health Care in Large-Scale Disasters" Study Group, 2009. Collection of Reports on Oral Care in Large-Scale Disasters

Infectious Diseases Post Great East Japan Earthquake

Norovirus Outbreak Among Evacuees from Hurricane Katrina

Cholera Surveillance during the Haiti Epidemic -The First 2 Years

Identifying Research Trends and Gaps in the Context of COVID-19

A guide to disaster risk reduction in the event of floods during COVID-19 epidemic

Medical care for infectious diseases in affected areas: what we did during the Great East Japan Earthquake

Support for infection control in evacuation centers following 2016 Kumamoto Earthquakes by Kagoshima Prefecture Medical Association Team

Risk of infectious disease outbreak after major natural disasters

Municipality's measures to large scale natural disasters out of view of infectional disease prevention: differences due to experiences suffering earthquake or not

Infection prevention measures and hygiene precautions among evacuees in emergency shelters

A Study of the Effects of Oral Health Care on the Prevention of Aspiration Pneumonia in the Compromised Elderly Patients

Infectious disease surveillance in Miyagi after the Great East Japan Earthquake

Outbreak of Influenza A at the refuge center after the East Japan Great Earthquake Disaster

Utilization of evacuation surveillance for the Kanto-Tohoku Heavy Rainfall Disaster

Prevalent Gastroenteritis and Infection Control in a Shelter After Kumamoto Earthquake

Reference material on measures for evacuees

The Sphere Handbook

This study was also supported by Pacific Consultants Co, Ltd.We thank Radhika Johari from Edanz Group (https://en-author-services.edanzgroup.com/ac) for editing a draft of this manuscript.

The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper.

The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper.J o u r n a l P r e -p r o o f