Introduction
Children with medical complexity (CMC) need daily medical care and require special support during disasters. 1 In developed countries, the number of high-risk deliveries has been increasing owing to late marriage and childbirth. According to the World Health Organization (WHO), the neonatal mortality rate in high- (and higher middle) income countries has improved by > 70% from 1990 to 2020, reflecting significant advances in perinatal survival. 2 Consequently, the number of CMC has been rising. CMC are a vulnerable population that requires special support during disasters. International frameworks, such as the United Nations, the Sendai Framework for Disaster Risk Reduction 2015-2030, 3 and the Sphere Standards, emphasize the importance of prioritizing the protection of persons with disabilities, chronic illnesses, and those requiring medical care. 4 Therefore, disaster preparedness and response for CMC are crucial issues in the field of international humanitarian assistance and an essential aspect of public health.
Parents or guardians provide most routine medical care for CMC and are supported by home-visit nurses and other health care professionals. During disasters, human and material resources become significantly limited, making it difficult to maintain care.Reference Mace and Doyle 5 Many medical institutions prioritize treating disaster-affected people over CMC. This prioritization is fundamentally grounded in the ethical norms of utilitarianism that inform crisis standards of care during catastrophes. Under severe resource scarcity, standard medical care shifts toward producing the greatest good for the greatest number of people, thereby organizing care to save the maximum number of lives possible.Reference Elmer, Atkins and Daya 6 Therefore, ensuring safe and continuous care for CMC under such conditions is vital.Reference Matsumoto, Nakai and Koga 7 –Reference Smith, Donaldson and Mitton 10 According to Tanaka et al., only 12% of CMC were evacuated to designated shelters and none to welfare evacuation shelters during the Great East Japan Earthquake, with caregivers hesitating to evacuate owing to concerns such as noise from suction devices.Reference Tanaka 11 In Japan, the Cabinet Office has promoted the establishment of welfare evacuation shelters to accommodate vulnerable populations. The 2021 revision of the Guidelines for Securing and Managing welfare evacuation shelters introduced a system to publicly designate “specified welfare evacuation shelters” that can accept identified vulnerable individuals and their families. 12
However, in practice, only a limited number of welfare evacuation shelters can accept direct evacuation. Although several surveys have explored the difficulties faced by caregivers of CMC during disasters, including their feelings and perceptions of shelters, little is known about why welfare evacuation shelters capable of directly receiving CMC remain limited, what challenges facility operators face, and what preparations would enable them to accept such children.
According to a survey by the Japan Pediatric Society on the status of individual evacuation plan development for children with medical complexity (CMC), plans have been created for only about 5% of those requiring mechanical ventilation.Reference Asami, Ishiwada and Iwaihara 13
Individual-level evacuation planning and drills for CMC have been actively explored in regions like Nagano Prefecture, primarily addressing personal preparedness.Reference Hoshino, Kobayashi and Kobayashi 14
Building on these foundations, this project integrated field simulation outcomes directly into regional disaster prevention plans and municipal manuals, addressing identified challenges at a systemic, community-wide level.
While this simulation targeted two municipalities to build an operational model, it was designed as a pilot to establish the framework necessary for supporting individual evacuation planning for all 122 CMC across the prefecture.
This paper presents the activities undertaken in Tottori Prefecture and illustrates how these efforts have led to policy changes within the prefecture.
Methods
Study Setting and Context
Tottori Prefecture, located in the western region of Japan, had 122 CMC at the time of this initiative. While the prefecture maintains precise registry data on the number and location of CMC, acute care hospitals capable of admitting these individuals are limited, with many serving as designated disaster base hospitals.Reference Yamada, Ohno and Shiota 15 Approximately 200 general welfare evacuation shelters had been designated across the prefecture, primarily utilizing facilities for older adults and special needs schools. However, none were specifically equipped to accommodate CMC, leaving many caregivers and supporters to rely on hospitals as default evacuation sites. In disaster scenarios, however, accepting non-patient evacuees creates significant operational challenges for hospitals already prioritizing emergency medical care.
Project Implementation and Manual Revisions
To address this gap, Tottori Prefecture and Tottori University Hospital initiated a collaborative project in 2020 to establish and operationalize specialized welfare evacuation shelters for CMC. Direct interventions were implemented across 2 municipalities. A multidisciplinary project team, including medical professionals, engaged with local government authorities to clarify the specific vulnerabilities of CMC during disasters.
Utilizing existing shelter operational manuals, both tabletop and field-level evacuation simulations were conducted with local stakeholders. Rather than serving solely as isolated training exercises, these simulations were designed to identify practical operational challenges. The insights gained were systematically analyzed to revise provincial and municipal disaster management and medical response manuals. Through this process, regional policy shifted from hospital-dependent care toward establishing community-based welfare evacuation shelters capable of accepting CMC during emergency situations.
Results
Situation in Tottori Prefecture Prior to the Joint Initiative
Tottori Prefecture, located in the northeastern part of the Chugoku region on Japan’s main island, has a population of approximately 550,000 inhabitants, including 65,000 children, of whom 122 are CMC. Approximately half of these children reside in the western region of the prefecture. Tottori University Hospital is the only facility capable of admitting CMC for inpatient treatment and is a designated disaster base hospital. Prior to the joint initiative, local surveys and individual evacuation planning indicated that many caregivers selected “hospitals” as their evacuation destination (Figure 1). In 2022, Tottori Prefecture issued the “Disaster Note for CMC” to serve as a baseline tool for individual evacuation plans. However, practical progress remained limited. The primary obstacles were that the note relied heavily on caregivers to complete it independently, and documenting complex medical information posed a significant burden for families without professional assistance. This situation underscored the necessity of multi-agency collaboration involving health care, welfare, and municipal personnel to establish actionable evacuation plans.
Outline map of Tottori Prefecture and intervention areas.
This map illustrates the geographical context of Tottori Prefecture, Japan, and highlights the primary areas in which the disaster preparedness project for CMC was implemented. It shows the relative locations of Tottori University Hospital (marked with a red diamond), other disaster base hospitals in Tottori Prefecture (marked with dark blue diamonds), and participating municipalities.

Figure 1 Long description
This map illustrates the geographical context of Tottori Prefecture, a narrow prefecture stretching east to west along the Sea of Japan, and highlights the primary areas in which the disaster preparedness project for CMC was implemented. Tottori University Hospital (marked with a red diamond) is located in Yonago City in the western region. Other disaster base hospitals (marked with dark blue diamonds) are distributed across the central and eastern regions, including Kurayoshi and Tottori Cities.
Key Activities of the Project
A multidisciplinary team consisting of pediatric neurologists who regularly care for CMC, pediatric ward nurses, emergency physicians experienced in disaster medicine, and nurses with Disaster Medical Assistance Team experience was formed at Tottori University Hospital. They collaborated with Tottori Prefecture to launch the Project for the Establishment and Operation of welfare evacuation shelters for CMC. Representatives from the prefecture’s disaster management and welfare departments participated in this project. As municipalities in Japan are responsible for establishing welfare evacuation shelters, the teams reached out to local governments. Consequently, the project was conducted in Hiezu Village (2021-2023) and Daisen Town (2024). As many CMC rely on medical devices for daily living, medical equipment vendors and suppliers also joined the project. Although the intervention was initially planned for one year per municipality, its duration in Hiezu Village was extended to 2 years due to the COVID-19 pandemic (Figure 2).
Project overview.

Figure 2 Long description
A vertical timeline illustrating the chronological progress of the disaster preparedness project for Children with Medical Complexity (CMC) in Tottori Prefecture from 2020 to 2024:
2020: Tottori Prefecture and the Department of Child Neurology at Tottori University discussed disaster measures for CMC.
2021: Project establishment. Activities commenced in Hiezu Village, including preparation, investigation, system development, monthly meetings, knowledge sharing, lectures, and tabletop simulations.
2022: Tabletop Simulation: Implementation simulation aligned with disaster drills, followed by review and debriefing.
2023: Activities commenced in Daisen Town, including preparation, investigation, system development, meetings, knowledge sharing, lectures, tabletop simulations, and implementation simulations aligned with disaster drills. Revision of the Tottori Prefecture Regional Disaster Management Plan.
2024: Revision of the Tottori Prefecture Disaster Medical Manual.
Issues in Establishing Welfare Evacuation Shelters and Use of Local Resources
In Hiezu Village, the project team assessed potential facilities that could serve as welfare evacuation shelters by examining their equipment and location. In Daisen Town, the designated welfare shelter was a private older adult care facility under a pre-existing agreement. However, the facility’s readiness for rapid activation during disasters was limited. Consequently, Daisen Town decided to establish welfare spaces within general evacuation shelters. In both municipalities, disaster management officers understood the relevant manuals and plans, but other administrative personnel were often unfamiliar with them. Therefore, the team reviewed the existing welfare shelter manuals and local disaster management plans. In municipalities lacking such manuals, new manuals were developed, and the project team provided support in drafting and revising them to ensure the optimal use of community resources.
Knowledge Sharing Between Health and Administrative Sectors
Monthly meetings were held to identify the difficulties faced by administrative and welfare officers and to clarify the information they needed from medical professionals. Health care providers delivered lectures on CMC, welfare shelter operations, and infection control. These sessions specified what each stakeholder—administrative staff, welfare workers, and caregivers—had to prepare in advance. To broaden awareness, participation was encouraged among staff directly involved with the CMC and among as many municipal employees as possible. Participation in these seminars and practical simulations is considered to have enhanced mutual understanding between service providers and recipients regarding welfare shelter operations. By clarifying the required knowledge and role allocation before the simulations, the participants were able to respond more smoothly during the exercises (Figure 3).
Collaborative learning session for CMC disaster preparedness.

Figure 3 Long description
A photograph of a lecture and briefing session held in a hall. In the foreground, a mannequin of a child with medical complexity wearing a navy tracksuit is placed on a reclining wheelchair, fitted with medical tubing connected to a ventilator and a medical device bag. Medical professionals explain its posture, device connections, and handling to the audience. Other attendees stand in the background, listening attentively.
Health care providers engage with administrative and welfare staff during a lecture on CMC. These sessions were crucial to foster interprofessional understanding and build capacity for the effective support of vulnerable populations in disaster situations.
Simulation Exercises and Problem Identification
Tabletop and field exercises were conducted using the same scenario to simulate the establishment and operation of welfare evacuation shelters for CMC. The disaster scenario assumed that an earthquake was occurring on a weekday during daytime, causing a widespread power outage. CMC were assumed to seek evacuation, excluding liquefaction, and walking access to the shelters was possible. When the earthquake occurred, the municipal disaster response headquarters decided to open welfare evacuation shelters, and the designated facilities initiated operations accordingly. The target CMC was modeled as a 5-year-old girl with a tracheostomy, on a mechanical ventilator, continuous oxygen saturation and heart rate monitoring, suctioning and airway clearance assistance, occasional oxygen therapy, and gastrostomy tube feeding. Another child with disabilities was included in the scenario, and both were accompanied by caregivers.
Tabletop Simulation
The participants used facility photographs to confirm the damage assessment procedures and arranged zoning and staffing plans on blank maps using figurines. Movements of CMC and information exchange between caregivers and shelter staff were simulated, along with communication with the disaster headquarters. After the exercise, the participants reviewed the issues in the manuals and operations. For instance, in Hiezu Village, public health nurses (PHNs), not facility staff, were assigned to manage reception and triage, as they possess the expertise to assess CMC conditions. Consequently, the shelter’s operational structure was revised to reflect this change.
Field Simulation
The field simulation was conducted in coordination with community disaster drills, ensuring realistic constraints on personnel and resources. A medical care model manikin was used to represent CMC, with actual medical devices employed. Evacuation was simulated from a location 800 m away (representing a home), carrying a manikin, medical devices, 3 days’ worth of supplies, and daily necessities. Communication with the disaster headquarters was conducted using existing Internet Protocol (IP) transceivers, enabling real-time, interactive (2-way) communication among the personnel involved. Facility damage assessments were performed by on-site inspections and photo reporting. In Hiezu Village, emergency power was available, but Daisen Town lacked such infrastructure. Therefore, electric vehicles were provided through disaster cooperation agreements between the prefecture and local companies. In both municipalities, PHNs verbally obtained medical information from the caregivers. Each field simulation lasted approximately 3 h, followed by participant debriefings (Figure 4).
Field simulation for disaster preparedness of CMC.
Participants engaged in a field simulation exercise, testing the operational protocols to support CMC during a disaster scenario.

Figure 4 Long description
Two photographs illustrating an evacuation and care simulation at a welfare evacuation shelter for a child with medical complexity:
Top image: An outdoor photograph of project staff and participants pushing a specialized reclining wheelchair carrying a mannequin along a paved rural road during a disaster drill. A second participant pushes a cart equipped with portable medical equipment alongside them under a clear blue sky.
Bottom image: An indoor photograph showing a care environment at the welfare evacuation shelter during a simulation. A mannequin of a child with medical complexity lies in a medical bed, connected to a ventilator with synthetic tubing. On the floor beside the bed, multiple black emergency transportation bags are lined up, along with portable medical devices, power supplies, and hygiene supplies prepared for disaster response.
Challenges Identified
The post-simulation review revealed several issues:
-
• Space requirements per person were larger than expected.
-
• Mobility of CMC was more difficult than anticipated.
-
• Securing power sources was essential.
-
• Accurately obtaining medical information verbally was challenging.
-
• Procedures for managing health deterioration were unclear.
-
• The roles of PHNs needed further clarification.
These findings led to revisions of municipal manuals.
Revisions to Manuals and Local Disaster Plans
Following these interventions, both municipalities swiftly revised their manuals. Hiezu Village updated its Welfare Shelter Manual, while Daisen Town developed a new Evacuation Shelter Manual and began preparing individualized evacuation plans for CMC. In 2023, the Tottori Prefecture revised its Regional Disaster Management Plan to include a new section titled “Persons Requiring Medical Care.” The plan stated that local governments must ensure the availability of the necessary equipment for such individuals, emphasizing the importance of securing electricity for medical devices. In 2024, the Tottori Prefecture Disaster Medical Relief Manual was also revised to include “Considerations for Persons Requiring Medical Care,” defining medical care as “mechanical ventilation, suctioning, and other medical procedures,” and requiring municipalities to maintain evacuation support registries and plans. It further instructs local public health centers to coordinate welfare evacuation shelters and transportation for such individuals. These revisions marked a policy shift from a medical institution-centered disaster response to a community-based welfare shelter model.
The details and outcomes of these interventions are summarized in Table 1.
Intervention details and outcomes

Table 1. Long description
Table 1 outlines details and outcomes for 5 key actions for CMC disaster preparedness:
Action: Formed multidisciplinary team (Tottori Univ. Hosp., Prefecture, municipalities, manufacturers) for CMC welfare shelters. / Result: Multi-stakeholder project initiated.
Action: Assessed facilities; revised/created manuals to fill gaps and optimize resources. / Result: Practical manuals and facility plans improved.
Action: Monthly medical-admin meetings/lectures for mutual understanding. / Result: Better staff understanding and smoother responses.
Action: Conducted simulations (tabletop/field) for operational challenges. / Result: Identified key issues (space, power, info) for manual revisions.
Action: Revised disaster plans emphasizing “Persons Requiring Medical Care”. / Result: Policy shift to community-based welfare shelter model formalized.
Note: CMC = children with medical complexity.
CMC, Children with medical complexity.
Discussion
Prior to the project, the reasons behind the limited number of welfare evacuation shelters capable of directly accepting CMC and the operational barriers perceived by shelter staff were poorly understood.
Through interviews conducted during the preparation phase, significant discrepancies were identified between the awareness of medical and administrative personnel regarding CMC. Even within the same occupation, information-sharing was often inadequate, contributing to difficulties in establishing welfare evacuation shelters. Many administrative staff and welfare facility workers did not understand home-based medical care because parents, and not health care professionals, often perform medical procedures. This lack of familiarity led to anxiety about accepting CMC, particularly among workers accustomed to serving older populations. This mirrors the findings of previous studies, such as that by Sorata et al., who reported that nursery teachers responsible for CMC were anxious when medical staff were not constantly available.Reference Sorata 16 Such anxiety is natural even among professionals who regularly engage with CMC, and is therefore understandably greater among those who do not. In our project, health care professionals conducted educational sessions for administrative and welfare personnel, and facility staff visited special needs schools to observe the care settings. These activities deepened understanding and reduced anxiety. Therefore, continuous educational collaboration between health care providers and local governments is essential.
Furthermore, the simulations revealed the pivotal role of PHNs in welfare shelter operations. PHNs possess medical and administrative expertise, allowing them to assess patients’ needs, perform welfare triage, and collect accurate health information. PHNs were not initially involved in these operations in Hiezu Village because disaster management was performed by the General Affairs Division. In contrast, Daisen Town integrated PHNs and welfare departments from the outset, leading to a smoother implementation. These experiences, together with prior national surveys, indicate that municipalities where PHNs are not formally assigned disaster-response duties often have lower rates of individualized support planning.Reference Hosoya, Sato and Sugimoto 17 This suggests the importance of institutionalizing their roles in disaster preparedness.
Additionally, this study emphasizes the necessity of aligning local disaster plans and medical relief manuals with practical realities. In Hiezu Village, the existing manual assumed evacuation first to a temporary site before being transferred to welfare evacuation shelters, which is inconsistent with the National 2021 Guideline encouraging direct evacuation to designated welfare evacuation shelters. Manuals that do not reflect actual operational conditions impede effective response. As municipalities base their plans on prefectural frameworks, concrete descriptions, including space allocation, power supply, and PHN involvement, must be incorporated into these official plans, not merely in the national-level guidelines.
This research demonstrates the possibility of establishing and operating welfare evacuation shelters that accommodate CMC. To achieve this, collaboration between medical professionals and local governments, verification through simulation, and the development of plans and manuals are necessary. In Japan, establishing welfare evacuation shelters remains a “best-effort” obligation based on the Disaster Countermeasures Basic Act, and no mandatory establishment obligation exists, unlike designated evacuation shelters. This means that responding to minorities requiring support, such as CMC, is left to the discretion and prioritization of local government officials.
Many local governments have limited human and financial resources, and in disaster management, priority tends to be given to most evacuees. As a result, responses to CMC, who represent a vulnerable minority, are often delayed, and concrete preparations fail to progress. Even in this study, tangible results were achieved only after a 3-year intervention in 2 municipalities, highlighting the clear limitations of relying solely on the voluntary efforts of local governments.
Given these administrative constraints, it should be emphasized that this simulation was not designed as an evacuation drill for specific individuals but rather as a framework-level simulation to build a community-wide support system for CMC in general.
Therefore, this pilot simulation in the two municipalities focused on testing how local administrations, welfare evacuation shelters, and medical institutions can collectively receive CMC during a disaster. Rather than viewing individual planning as an isolated paperwork exercise, this project established the essential operational foundation needed to support both individual evacuation planning and future community drills across the entire prefecture.
While securing options for sheltering at home—such as pre-arranging backup power sources and medical supplies—is essential, establishing functional welfare evacuation shelters as public assistance is equally critical. When home-based care becomes unviable, welfare evacuation shelters serve as a vital safety net. Therefore, this study focused on operationalizing welfare evacuation shelters so that effective public support is available alongside home-based preparedness when sheltering at home reaches its limits.
Article 11 of the UN Convention on the Rights of Persons with Disabilities requires that “States Parties take all necessary measures to ensure the protection and safety of people with disabilities in situations of risk, including armed conflicts, humanitarian emergencies, and natural disasters, applying international human rights and humanitarian laws to ensure inclusion, access to services, and non-discrimination, moving away from viewing them as merely ‘vulnerable.’” 18 Furthermore, the Sendai Framework for Disaster Risk Reduction 2015-2030 3 emphasizes the need to consider vulnerable groups in disaster risk reduction. Moreover, the Sphere Standards 4 call to specially consider persons with disabilities and chronic illnesses in humanitarian assistance during disasters. Considering these international standards, establishing evacuation shelters for disaster-affected persons requires special consideration, and including CMC should be a “mandatory obligation” rather than a “best effort obligation.”
Policy measures such as clarifying the criteria for establishing welfare evacuation shelters based on the number of CMC, institutionalizing financial support for their operation, and ensuring visibility and progress management of their establishment status are considered essential. Practical initiatives like those shown in this study should not be hindered by the institutional constraint of a “best effort obligation.” Ensuring the safety of CMC during disasters is a responsibility of the national and local governments, in accordance with international humanitarian assistance standards and the “right to maintain the minimum standards of wholesome and cultured living” guaranteed by Article 25 of the Japanese Constitution. 19
Limitations
This study describes the activities conducted in Tottori Prefecture, a rural area with a small population. A prefecture-wide completion rate was not centrally available because individual evacuation plans for CMC were maintained by individual municipalities. The interventions were limited in scale and duration (3 years); therefore, their long-term sustainability and applicability to larger urban regions remain uncertain. Future studies should include follow-up evaluations and broader implementations.
Conclusions
Tottori Prefecture and Tottori University Hospital collaboratively launched a project to establish and operate welfare evacuation shelters for CMC during disasters. Through direct communication between medical professionals and local governments and tabletop and field simulations, critical policy and manual revisions were achieved. These reforms shifted disaster response from hospital dependency to community-based welfare shelter systems. This initiative provides a practical and scalable model for other municipalities to strengthen disaster preparedness for CMC.
Author contribution
YN conceptualized and designed the study, conducted the data analysis, and drafted the initial manuscript. MT secured the funding for this project and provided critical intellectual input. HY, KN, and SY contributed to data collection, logistical coordination, and the implementation of the simulation. MH assisted in the study’s methodology and provided clinical expertise. YM supervised the project and provided final approval of the manuscript. All authors contributed to the interpretation of the results, participated in the critical revision of the manuscript, and approved the final version for submission.
Funding statement
This study was supported by JST Grant Number JPMJPF2301, Japan.
Competing interests
None.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. This study was approved by the Ethics Committee of the Faculty of Medicine, Tottori University (Approval No. 22A039).