Nowhere to Run To: Shortcomings in Japan’s Disaster Evacuation Centers


Despite being prone to major natural disasters, Japan lags its international peers when it comes to the evacuation centers where residents go after one strikes. To address this, the country must create a central government agency to handle disaster preparation and reform the ways it goes about responding to disaster.

Not All Evacuation Centers Are the Same

Japan is known as a nation prone to major natural disasters. From 1995, the year of the Great Hanshin-Awaji Earthquake, through 2018, though, the nation also saw some 4,900 disaster-related deaths—most of these taking place in evacuation centers. Disaster-related deaths, fatalities officially recognized as being connected to major disasters, arise from the physical burden of evacuation or from preexisting illnesses exacerbated by the stresses of that lifestyle. It is clear from the numbers that life in evacuation centers following a disaster represents yet another life-threatening disaster in itself.

Since the Niigata Chūetsu Earthquake of 2004, I have been researching the incidence of pulmonary embolism among evacuees. Commonly known as “economy class syndrome,” this malady involves blood clots forming and lodging in blood vessels in the lungs and elsewhere in the body. Sitting for lengthy periods in cramped positions, as in the economy-class seats of long-haul airlines, brings about poor blood circulation and creates clots in veins in the lower legs. The longer one sits still, the more chance there is that these clots will make their way back to the heart and from there to the lungs. In severe cases, this can lead to death. Following the 2004 Niigata quake, 14 people came down with pulmonary embolism as a result of sleeping in their cars after evacuating damaged homes; 7 of them died. This tragedy brought economy-class syndrome into the public eye as a threat to people in disaster-struck areas.

In 2007, when another powerful earthquake struck offshore from the same Niigata area that had been shaken in 2004, the memory of the earlier disaster remained fresh in people’s memory. Few chose to sleep in their cars then, and there were almost no cases of economy class syndrome caused by cramped conditions in vehicles.

There were, however, disaster-related deaths reported at evacuation centers. These centers were numerous and took many forms, but one thing they had in common was a lack of emergency cots; people slept on the floor, tightly packed together. Venous ultrasound testing of evacuees’ legs found many cases of blood clots in large-scale evacuation facilities, housing more than 100 people, near the quake’s epicenter. And more of these cases were discovered a month after the earthquake struck than a week afterward—this despite guidance being offered at the centers on the importance of engaging in exercise and drinking plenty of water to prevent the malady. This led me to believe that it was impossible to prevent pulmonary embolism among evacuees sleeping on the floor in tight quarters.

My research following the 2008 Iwate-Miyagi Nairiku Earthquake showed a difference in the frequency of clot formation depending on the living environment at various evacuation centers. Embolism incidence was reduced even in centers where people slept on the floor when partitions were used to separate elderly households from younger families with children, or when active steps were taken to boost communication among evacuees, such as by getting people together to make strawberry jam. In crowded centers that were located near noisy, bustling disaster response headquarters, meanwhile, people sleeping on the floor developed leg clots more frequently.

Healthy Environments Mean Healthy Evacuees

The Great East Japan Earthquake of March 2011 impacted a vast swath of northern Honshū, and the evacuation centers that went into action in its wake ranged from mud-soaked coastal locations battered by the tsunami to relatively high-quality living environments farther inland. Ultrasound testing of leg veins across this range of centers showed a high frequency of embolism among evacuees where the tsunami damage was especially severe; the farther from the coast evacuees were, the less likely they were to have clotting.

To elucidate the connection between evacuation conditions and the prevalence of pulmonary embolism, we decided to produce numerical scores corresponding to the living environments in the evacuation shelters, using the Environmental Health Assessment Form for Disaster Shelters created by the US Centers for Disease Control and Prevention, or CDC. The version of this form we used in 2011 evaluated living environments in terms of 60 assessment criteria, including facility, childcare, and health/medical features. The closer a center’s score was to 60, the better its environment was assumed to be.

The chart below shows that among evacuees at centers with scores of 10 or so, the incidence of pulmonary embolism was as high as 50%. When centers scored above 40, however, this incidence rate plunged below 10%. Clearly the higher-scoring centers offered residents healthier living environments, producing fewer blood clots as a result. A more detailed examination of the survey results showed that the so-called TKB features of an evacuation center—its toilet, kitchen, and bed facilities—were particularly important factors in the provision of these ideal evacuation environments. 

Japan Falling Behind

Following the May 2012 Emilia earthquakes in northern Italy, I traveled there to observe conditions in the evacuation centers. The beginning of summer was bringing its first hot weather, and I found the tents equipped with air-conditioning units and beds with mattresses. Large tents were provided for use as cafeterias, and hot meals were prepared in kitchen units installed in mobile containers. Other containers housed the toilet and shower facilities, some of which were even accessible to users in wheelchairs.

In Japan, by comparison, where the standard approach is to sleep side by side on the floor, many local governments do not even bother setting up the cardboard beds distributed by the central government as part of emergency aid shipments following a disaster. True, the days are long gone when evacuees received nothing but hardtack for their meals, but the common meal plan nowadays is rice balls for breakfast, some sort of sweetened bread for lunch, and cold bentō meals for dinner. Half of the food prepared in the Italian disaster kitchens was kept warm with heated water units, ensuring a hot meal for evacuees whenever they came for it. In the United States, too, evacuation center staff strive to prepare hot meals—something that is even the rule in refugee camps operated by the United Nations.

The Need for a Centralized Response Capacity

There are three main causes for the low quality of Japan’s evacuation centers. First, the country lacks a central government agency tasked specifically with disaster response measures. The bulk of the disaster response is left up to municipal governments in the affected areas. Second, there are insufficient stores of the supplies needed in the wake of a disaster. And third, there are no organizations in place to provide postdisaster support functions. 

In Japan, postdisaster support mechanisms are not set in motion until a disaster strikes. This is because the country has no central government organ focused on disaster response, and thereby able to prepare in advance; budgetary measures can only be taken upon invocation of the Disaster Relief Act. The Basic Act on Disaster Management, meanwhile, specifies that municipalities are responsible for protecting the lives and property of residents; until municipal authorities formally request aid, it is difficult for the prefectural and national governments to provide it. Without a standardized national approach to providing needed support in the wake of a disaster, disparities in the level of support received by each municipality are unavoidable.

Although the Cabinet Office’s guidelines for evacuation centers recommend the use of cots, following the torrential rainfall and flooding in western Japan in June and July 2018, as well as Typhoon Hagibis in October 2019, numerous impacted communities failed to make use of the cardboard beds that were distributed. Another problem arose when municipal requests for aid from the central government had to go through prefectural governments first, leading to lost time. Communication infrastructure can be shaky in times of disaster, creating an additional risk that such relayed requests might get lost along the way.

Japan’s municipalities are also responsible for preparing stocks of emergency supplies, but these stockpiles tend to be minimal, and the local governments depend on “running stock” for much of it. This term refers to arrangements in which governments enter agreements with supermarkets and distributors, allowing the needed goods to continue flowing through the supply chain to consumers in ordinary times but securing supplies of them on a priority basis when a disaster strikes. This reduces the cost of long-term storage and the need to track expiry dates on food items.

In other developed countries, too, the authorities rely on running stock for most of their emergency food stockpiles. Unlike Japan, though, they also stockpile beds, tents, toilets, and kitchen facilities in sufficient amounts to cover the need of some 0.5% of their populations. In Japan, which lacks such a coordinated stockpiling system, the areas hit hard by Typhoon Hagibis included some municipalities that asked residents to bring their own food, water, and blankets to evacuation centers. Typhoons being an entirely foreseeable form of natural disaster, the governments should have been able to prepare these items well in advance, but the lack of laws spelling out the needed stockpiles led to this chaotic situation instead.

The manufacturers engaged in these agreements tend to sign contracts with multiple municipalities. This increases the risk that supplies will run out when a disaster impacts a broad area. And when transportation routes are thrown into confusion or cut off entirely after a disaster, it will not be possible to obtain the needed supplies on the strength of these agreements alone. Budgetary issues pose another obstacle: Prefectural and municipal governments cannot provide the funding to purchase and stockpile goods from their ordinary budgets, which means that the national government must take responsibility for these tasks. Budgetary measures to create these stockpiles must be implemented in ordinary times, well in advance of a disaster—this, too, underlines the necessity of a national-agency-level entity to handle disaster response.

Government Officials Cannot Do It All

Finally, we must consider Japan’s lack of dedicated postdisaster support organizations. In other developed nations, people undergo training preparing them as disaster responders. They then register with the national authorities, forming a network of people with clearly defined roles who can be tapped to provide specialized support when it is needed most. Cooks, for example, are ready to prepare meals at evacuation centers, and truck drivers to deliver emergency supplies. National governments cover all travel expenses when these people are called up following a disaster, as well as paying the cost of meals and lodging and accident insurance premiums during support operations. Japan, too, must move swiftly to put a system like this in place—especially since local government officials will not provide enough manpower to deal with a disaster on the scale of a major temblor striking the Tokyo area or a massive Nankai Trough earthquake and tsunami.

Another lesson to learn from the foreign examples is the way that support activities in a stricken region are handled primarily by personnel sent in from other areas. This is because the officials from that affected region are likely to be disaster victims themselves, and thus less able to form the core of a postdisaster support workforce. From a humanitarian perspective, it is no noble thing to ask those who may have lost members of their own families and are themselves living in evacuation centers to support other victims; it is rather a violation of their human rights.

The lack of a central disaster-response agency in Japan also means that it is up to emergency-response divisions in municipal governments, the Cabinet Office, and other disparate authorities to handle preparations for disasters in ordinary times. The administrative systems in Japan, where officials are shuffled from post to post every three years or so as they climb the ranks, those who take on these tasks do so for only limited periods, leading to a lack of people with deep experience in the disaster-preparation and -response field. As a result, there are few officials who can be counted on to manage evacuation centers with a confident hand when the time comes; those in charge will follow the instructions in the emergency manual to the letter, showing little flexibility.

Another urgent task for Japan, therefore, is the creation of a system in which administrative officials can receive promotions while continuing to gain valuable experience within the disaster-response area of government. If the nation is to reduce the number of disaster-related deaths, it must be ready to roll out comfortable evacuation centers properly outfitted with sufficient TKB facilities. I urge Japan to create an agency specialized in disaster preparation and response, to do better in terms of its stockpiles of emergency supplies, and to create more robust organizations to handle postdisaster support tasks. Lives depend on it.

(Originally published in Japanese. Banner photo: Evacuees in an elementary school gymnasium on October 13, 2019, in the city of Nagano after the destructive Typhoon Hagibis made landfall in Japan the previous day. © Yomiuri Shimbun/Aflo.)

disaster Evacuation health