Symposium on disaster-related deaths after the Fukushima Daiichi Nuclear Power Plant accident

Reportedly, the health effects after the FDNPP accident were caused by both the direct effects of radiation exposure, as well as various social and living environment changes, including the implementation and lifting of the evacuation order. These secondary health effects were not merely short-term as they were shown to be long-term and diverse [9]. Specifically, these effects could be categorised ranging from days to weeks, months to years, and beyond.

First, using a day-to-week scale, the study reported an increased risk of death associated with: nursing home and hospital evacuation, living in poor conditions in shelters or home evacuation, deep vein thrombosis and pulmonary embolism caused by evacuation in a car, as in economy class syndrome, and cardiac disease caused by high blood pressure due to stress or withdrawal of medication after a disaster [10–16]. Second, using a month-to-year scale, the increased risk of death was caused by lifestyle changes. Specifically, psychological effects such as post-traumatic stress disorder, alcoholism, depression, and suicide were reported in this study [17]. Exacerbation of lifestyle-related diseases such as diabetes, hypertension, and dyslipidaemia, as well as a decline in cancer screening rates and deaths because of cardiac disease among reconstruction workers, were also reported [18, 19]. Third, using a multi-year scale, various social health problems were reported, including changes in the social environment, changes in long-term care insurance services, disparities and poverty, discrimination and prejudice, aging and depopulation of the community, migration to public housing, and the effects of dealing with long-term evacuation [20].

However, the health effects caused by post-disaster changes in both the society and the surrounding environment should not be attributed to the lifestyle actions of individuals, such as dietary habits and lack of exercise. Frequent changes in society and the surrounding environment after a disaster can severely worsen the health status of disaster victims. Healthy young people have relative adaptability to these changes and individuals who have access to social support can overcome the negative health effects after a disaster. However, people living with disabilities, older adults, and other vulnerable groups who require increased social support may not be able to withstand the recurring environmental changes. The health of these vulnerable populations is more susceptible to deterioration, resulting in the reported deaths associated with the indirect effects of disasters. Thus, a multidimensional view of the health effects caused by disasters is required because of the associated secondary health effects, which often result from multiple factors as the causal relationship with a single effect gradually fades over time.

The Sendai Framework for Disaster Reduction 2015–2030 outlines the goal of the United Nations Disaster Reduction Reduction Agency to reduce disaster fatalities. To achieve this goal, it is pertinent to identify, understand, manage, and reduce disaster risk as a priority action [21]. Objective medical data and informed disaster medical response are necessary to reduce indirect disaster deaths. However, medical data are limited for large-scale natural disasters [22, 23]. The Organization for Economic Cooperation and Development suggests that quantitative analysis is necessary to indicate the economic support measures for disasters per country considering the risk of disaster and its impacts on health [24]. However, it was discussed that the analysis of 'disaster-related deaths', (i.e. deaths indirectly caused by disasters) in the current context of Japan, has several major limitations for medical validity. Consequently, three factors were identified contributing to the problem of validity; namely problems with: the definition of disaster-related death in Japan, the method of application, and the method of certification.

First, is the issue of definition. In Japan, the current definition of a disaster-related death refers to the provision of condolence payments for bereaved families and is synonymous with death because of disaster. In contrast, the US definition of disaster death is defined separately from condolence payments. The Japanese definition based on condolence payments is problematic because it overestimates the number of disaster-related deaths. This symposium found that in some cases, individuals eligible for condolence payments after FDNPP accidents (i.e. certified as disaster-related deaths) were dissociated from indirect disaster deaths. It was also pointed out that there may have been an overestimation of disaster-related deaths.

Second, with regard to the issue of the application method, it was underscored that death first requires an application for certification examination by the bereaved family despite its disaster status. For example, the application process for unclear disaster situations placed a heavy burden on the bereaved families and resulted in more problems in contexts where no bereaved families were present to submit applications, thus contributing to the underestimation of disaster-related deaths. In contrast, it was found that in the US, applications from the bereaved family were not required to determine a disaster death, since the death certificate includes a statement that outlines the cause of death [3].

Third, is the issue of certification methods. There is no unified national standard for recognising disaster-related deaths in Japan. Some local governments use certification criteria based on the time of death and the post-disaster period, but the actual operation and certification decisions are left to each local government, which could lead to underestimation or overestimation. However, this type of operation may have worked well in the GEJE and FDNPP accident. That is, the accident was unprecedented in both scale and combination, and it was not clear at the time what form the potential health hazards would take. In fact, there have been cases of deaths occurring more than 10 years after the FDNPP accident in which the effects of the accident have been recognised [25]. Thus, if the criteria for recognising disaster-related deaths had been developed simply by the time of death and the period after the disaster, it may not necessarily have answered a variety of individual cases. However, it is also necessary to make the criteria for certification more reasonable while also emphasising science.

The authors introduced the concept of 'shaking' at the symposium by defining 'shaking' as 'the repeated changes in the social and living environment that worsen health conditions, regardless of the disaster'. For example, in the context of general practice, rather than disasters, one example would be the situation in which the health condition of an elderly person deteriorates due to repeated relocations and hospitalisations. In particular, disasters are a context in which a wide variety of social and living environment changes are likely to occur. After the nuclear accident, health effects associated with exposure to radiation were limited. However, the evacuation and subsequent relocation, and the accompanying changes in social and living environments, such as lifestyle, employment status, and relationships with family members, occurred in succession and caused significant damage to health conditions. What could be notable about the concept of 'shaking' is that it can be applied to various types of crises such as COVID-19. For example, in the COVID-19 pandemic, repeated waves of infection led to behavioural restrictions and lockdowns, resulting in loss of employment and strain on health care resources. They caused repeated changes in the social and living environment, such as changes in employment patterns and lifestyles, changes in family relationships, and changes in the behaviour of patients visiting medical institutions, which worsened their health conditions [26]. As described above, 'shaking' is a concept that broadly extends beyond disasters, and the nuclear power plant accident and COVID-19 can be considered as environments in which it is particularly likely to appear.

This report found that Japan's system for disaster-related deaths was not created originally for future disaster preparedness. Instead, it was created in the historical context of the disaster condolence payments system, which was designed to provide relief to bereaved families.

This differs significantly from the US system, which separates the definition of indirect death from the system of relief for survivors because of disasters. In the US, the cause of death because of a disaster is stated clearly on the death certificate, which is the basis for death statistics. In addition, an application for financial assistance is filed with the Federal Emergency Management Agency of the United States. Moreover, the International Classification of Diseases 11th Revision (ICD-11), is cited for clinical use [3]. In contrast, disaster deaths in Japan are not recorded on either the death certificate or according to ICD-11.

Reducing disaster deaths is an important issue for medical care providers. In particular, many indirect deaths were reported in accidents related to the FDNPP. Thus, reducing these indirect deaths is important and requires quantitative medical evaluation. However, Japan's current system of disaster-related deaths has major limitations for quantitative medical evaluation. Specifically, two factors limit the accumulation of medical data on disaster-related deaths. First, Japan's definition of disaster-related deaths is based on the condolence payments system. Second, there are no diagnostic criteria for deaths caused by a disaster.

Thus, the following should be prioritised in the system of disaster-related deaths from the perspective of the medical community. To adequately improve the accumulation and analysis of past cases of disaster-related deaths, two data collection methodologies were considered for further quantitative evaluation to reduce such deaths. The first is to utilise the current system based on the condolence payments system. This would require improvements in the application process to reduce the burden on bereaved families, the creation of certain certification criteria to increase the validity of the certification, and the collection of patient information (age, gender, pre-existing conditions, welfare information, disaster status, etc) in a unified format. Second, a new system should be created with reference to the US system and other systems. Specifically, the system should be two-step, that is, structured medically for the indirect diagnosis of death that is separate from condolence payments. Thereafter, this is followed by an application for financial assistance based on that diagnosis. Notably, several issues need to be discussed in the future. The decision-making process will be shifted from the current municipal judgment committee to the clinician on site. Thus, both the institutional reforms are necessary for future disaster preparedness. However, historically, Japan's system for disaster-related deaths has its origins in the condolence payment system, which involves a variety of stakeholders, including lawyers, government officials, and others in addition to medical professionals.