Introduction
The world has faced numerous disasters, including earthquakes and torrential rains, and more recently, it has encountered an emerging infectious disease disaster in the form of the coronavirus disease 2019 (COVID-19) pandemic. Medical and nursing professionals globally must draw lessons from these disaster experiences and enhance their ability to safeguard lives and health in the face of all disaster types.
In Japan, the 1995 Great Hanshin-Awaji Earthquake (magnitude 7.2) resulted in 6,434 deaths, 3 missing persons, and 43,792 injuries. 1 This catastrophic event prompted the establishment of an acute-stage disaster medical system, including the formation of the Disaster Medical Assistance Team, designed to save lives within the critical 72-hour post-disaster window.Reference Koido, Kondo and Ichihara 2 , Reference Yamaguchi and Tanjoh 3 The 2011 Great East Japan Earthquake (magnitude 9.0) caused 16,079 deaths, 3,499 missing persons, and 6,051 injuries. 4 This disaster presented a multifaceted challenge, as the earthquake triggered a tsunami and a nuclear power plant accident, necessitating the development of a more intricate and long-term health care response.
Based on these experiences, nursing has advanced in practice, research, and education. The Japanese Society of Disaster Nursing was established in 1998, followed by the World Society of Disaster Nursing in 2007, to formalize disaster nursing as an academic discipline. In 2009, “disaster nursing” was incorporated into the curriculum as a required subject in nursing education.Reference Minami 5 , Reference Minami 6 Reflecting on the history of disaster medicine and nursing in Japan, these fields have strengthened organizational resilience against the challenges posed by large-scale disasters.
Studies exploring disaster victims’ behavior from an adaptation perspective include investigations on the adaptation of older adults to evacuation following disasters and the health and adaptation of older adults during floods.Reference Hashimoto, Miyazaki and Ishimaru 7 , Reference Chompunud and Inkaew 8 Research on disaster-affected psychiatric inpatients dates back to 1990, but the systematic evaluation of disasters’ effects on the treatment of psychiatric inpatients remains limited.Reference Stout and Knight 9 Studies addressing the impact of disasters on psychiatric inpatients include fact-finding investigations on fluctuations in patient numbers before and after disasters and studies focusing on disasters and mental health crises, such as suicide and post-traumatic stress disorder.Reference He, Lundy De La Cruz and Olson 10 –Reference Yang, Cui and Sim 13 However, survey-based studies specifically examining the behavior of inpatients with psychiatric disorders appear to be lacking.
This study aimed to examine the behavior of patients admitted to psychiatric hospitals following disasters and to characterize these behaviors.
The findings of this study may contribute to a deeper understanding of psychiatric inpatients’ behavior after disasters, support the identification and improvement of the care they need, and help guide inpatient resource allocation and inform targeted interventions.
Methods
Study Design
This was a qualitative descriptive study.
Participant Recruitment
Nursing staff who worked in psychiatric wards within 1 week of the 2018 Northern Osaka Earthquake or the 2018 July rainstorm were included in this study. Administrators at 2 psychiatric hospitals significantly impacted by these disasters were asked to collaborate, and participants were recruited based on their willingness to participate. However, individuals were excluded if they had experienced severe personal damage during the events, such as the partial destruction of their homes, to mitigate the psychological burden associated with their participation. Table 1 provides a summary of the disasters encountered by the nursing staff, the study participants, and the patients admitted to the psychiatric hospitals, who were the indirect participants in this study.
Summary of disasters experienced by participants
Intervention
Focus group interviews were conducted in psychiatric wards, targeting both acute and chronic care wards. Acute care wards are designed to stabilize psychiatric symptoms, with a typical length of stay of 3 months or less. Chronic care wards focus on rehabilitation for patients hospitalized for more than 3 months. The interviews took place in a private room within the psychiatric hospital to which the participants belonged.
Outcomes
Before the focus group interviews, a basic attribute questionnaire was distributed to participants, who were asked to respond immediately. Subsequently, focus group interviews were conducted with each group. Semi-structured interviews were employed, with interview items consisting of 5 questions (Table 2). For this study, post-disaster was operationally defined as within 1 week of the occurrence of a disaster (acute phase).
Focus group interview guide
In this study, behaviors were defined as nurse-observed responses to disaster-related stimuli. Behaviors emerging within 1 week after the disaster were interpreted as disaster-related.
Analysis
Qualitative data from the focus group interviews were used to organize patient behaviors after a disaster as units of record based on the similarity of semantic content, and the frequency of these behaviors was counted.
Behavioral characteristics were analyzed from the perspective of adaptation, as disaster crises induce fear, anxiety, tension, and distress in patients. This perspective allows patients’ behavioral responses under severe disaster-related stress to be understood not only in terms of vulnerability but also as part of an adaptive process. Such an adaptation perspective is consistent with the transactional theory of stress and coping proposed by Lazarus and Folkman,Reference Lazarus and Folkman 14 which focuses on coping behaviors in response to environmental stressors, as well as with Roy’s Adaptation Model,Reference Roy and Andrews 15 which conceptualizes human beings as adaptive systems responding to environmental stimuli.
Additionally, the nursing staff were considered capable of objectively observing patients’ unconscious behaviors, and understanding patient adaptation was expected to enable better assessment and provide insights into appropriate care.
Content analysis was selected for this study because it can qualitatively and quantitatively capture post-disaster behaviors, with nursing staff providing objective observations of patients’ unconscious responses.
Ethical Considerations
Prior to the study’s commencement, all participants provided written consent. As the data collection involved recalling disaster situations, care was taken throughout the survey to minimize the risk of flashbacks or other psychological reactions among participants. During data collection, participants were fully informed of their right to withdraw consent. During interviews, we monitored participants’ facial expressions; if excessive tension or distress was observed, the survey was immediately stopped to ensure it did not interfere with participants’ duties. Additionally, a 6-month withdrawal period was allowed following the interviews. For data analysis, personal information was anonymized while maintaining linkability, and a secure table was created for handling this information.
Statement of Ethical Review
This study was approved by the Ethical Review Committee of the Hyogo University of Health Sciences (approval number: 19014).
Results
Participant Flow
Data collection began in August 2019, approximately 1 year after the disasters, and continued until February 2020. The study included 24 participants, whose attributes are presented in Table 3. Eight groups were organized based on the wards to which participants belonged, with 2 to 5 participants per group. Each group interview lasted between 30 and 60 minutes. No participants withdrew from the interviews due to flashbacks or other physical symptoms.
Characteristics of the participants (N = 24)
SD: Standard deviation.
Baseline Clinical Characteristics
Nineteen themes comprising 52 subthemes were identified as behaviors of psychiatric inpatients during a disaster. Table 4 presents the themes and subthemes, while Table 5 lists their frequencies. The theme with the highest frequency was calm and cooperative which included 6 subthemes. Nineteen themes were examined to clarify the behavioral characteristics of patients with mental disorders during disasters from the perspective of adaptation. The behaviors of psychiatric inpatients after a disaster were classified as adaptive (53.1%), maladaptive (22.2%), and unclassifiable (24.7%) (Table 6). Additionally, 9 of the 19 themes classified as maladaptive were further interpreted as themes specific to patients with mental illness or general disaster-reactive themes, accounting for 56.8% and 43.2%, respectively (Table 7).
Behavior of psychiatric inpatients after a disaster: themes and subthemes
Behavior of psychiatric inpatients after a disaster
Behavioral characteristics of psychiatric inpatients after a disaster from the perspective of adaptation
Behavioral characteristics of psychiatric inpatients after a disaster from the perspective of maladaptive behavior
Outcomes
As shown in Table 5, the most frequent behaviors of psychiatric inpatients during disasters were calm and cooperative (24.7%) and adjusting to an inconvenient lifestyle (22.8%), each observed in more than 20% of cases. All other behaviors occurred at frequencies below 10%.
Table 6 can be summarized briefly as: 53.1% of behaviors were adaptive, 22.2% maladaptive, and 24.7% unclassifiable, with adjusting to an inconvenient lifestyle (22.8%) being the most frequent adaptive behavior and calm and cooperative (24.7%) classified as unclassifiable.
Table 7 shows that 56.8% of maladaptive themes were disorder-specific, including attempting elopement from ward and displaying mental health reactions, whereas 43.2% reflected general disaster-related responses, such as avoidance of discussion regarding the disaster and being afraid of aftershocks.
Taken together, these findings demonstrate that while psychiatric inpatients exhibited a wide range of behaviors during disasters, adaptive responses were predominant. Notably, calm and cooperative and adjusting to an inconvenient lifestyle were the most frequent, each exceeding 20%. Although maladaptive themes were also observed, more than half were disorder-specific, whereas the remainder reflected general disaster-related reactions. Together, these 2 themes accounted for nearly half of all observed behaviors, whereas each of the remaining themes individually represented less than 10% of observations.
Limitations
This study had several strengths and limitations. We investigated the behavioral characteristics of patients admitted to psychiatric hospitals after a disaster from the perspective of adaptation, a topic not previously studied. This research provides valuable insights to help nursing staff understand patient behaviors and improve care during disasters. However, certain limitations stem from the research methodology and the interpretation of results. As the disasters studied were limited to earthquakes and torrential rains, the generalizability of the findings is restricted; the type and severity of a disaster can significantly influence the behavioral characteristics of psychiatric patients. Additionally, focus group interviews were conducted 1 year after the disaster. This delay was due to ethical considerations, as recalling the disaster experience could pose psychological risks to participants, and their psychological safety was prioritized. However, we cannot deny the possibility that the interviewees’ memories were unclear after 1 year, and a few patient behaviors may not have been described.
Some wards were exposed to direct disaster-related environmental impacts, such as strong shaking, and patients and staff may also have experienced vicarious trauma through disaster-related television news and information provided by families. These direct and indirect influences may have affected observed behaviors and psychological responses.Reference Piras, Usai and Contu 16
Objective indicators such as the number of “code green” responses, administrations of emergency intramuscular medication, and the use of seclusion or restraint were not systematically collected during the disaster period. This limitation is acknowledged in the present study, and these indicators are positioned as important issues to be addressed in future research in order to more objectively evaluate ward functioning during disasters.
In this study, behaviors were defined as nurse-observed responses to disaster-related stimuli. Behaviors emerging within 1 week after the disaster were interpreted as disaster-related. Since ward data prior to the disaster were not collected, direct comparisons with non-disaster conditions are not possible. Therefore, these behaviors are considered suggestive of reactions to disaster-specific stimuli, but causality or exacerbation cannot be definitively determined.
Nurses’ own experiences during the disaster, such as extended shifts and restrictions on returning home, may have influenced their observations and records of patient behaviors. This potential bias should be considered, as nurses’ circumstances may have shaped their perceptions of patients’ responses.
Discussion
Situational crises such as disasters can cause environmental maladaptation, exacerbate psychiatric symptoms,Reference Yamaguchi, Toda and Kouchi 17 and induce psychiatric manifestations such as auditory hallucinations and delusions in patients with mental disorders. Based on these characteristics, psychiatric patients are generally considered more vulnerable to disasters than the general population.Reference Sakairi and Kusaka 18
Based on the distribution of patient behaviors identified in this study, the importance of targeted care and vigilant observation during disasters is highlighted, providing guidance for nurses’ practice in psychiatric wards. Adaptive behaviors accounted for 53.1% of all behaviors, suggesting that initial approaches should follow the basic principles of Psychological First Aid (PFA),Reference Kim and Han 19 aimed at promoting adaptive behaviors and recovery, including ensuring psychological safety during patient engagement. Among maladaptive behaviors, 56.8% were specific to patients with psychiatric disorders, necessitating targeted care based on professional observation and clinical judgment. Behaviors for which adaptation or maladaptation could not be determined, such as calm and cooperative, represented 24.7% of all behaviors, the most frequent among the 10 themes, indicating the need for careful monitoring of patients’ mental state. These approaches serve as practical guidance for nurses in psychiatric wards during disasters.
Disaster responses are not static but evolve over time, as described by Zunin and Myers.Reference DeWolfe 20
In this study, nurses recognized emotional and behavioral changes in patients within 1 week after the disaster, corresponding to the acute phase. In this phase, Psychological First Aid (PFA) is widely recognized as an appropriate framework for providing initial support and for identifying individuals who may require further professional care.Reference Kim and Han 19 As disaster reactions change over time, patients with persistent or worsening symptoms should be assessed and treated by mental health professionals in subsequent phases. 21
However, the results of this study revealed that the behaviors of patients with psychiatric disorders in response to disasters were adaptive and that even maladaptive behaviors fell within the range of normal disaster-response behaviors observed in individuals without psychiatric disorders.
These results suggest that such patients are not merely vulnerable but also possess the capacity for adaptation, highlighting that adaptive behaviors can emerge even under disaster-specific stressors.
Adaptive behaviors included passive behaviors, such as adjusting to external factors of the disaster (e.g., adapting to an inconvenient lifestyle and preparing for evacuation), and active behaviors, such as worrying about the safety of family and home and talking to others about their disaster experiences and feelings. This suggests that patients with psychiatric disorders in disaster situations recognize the critical nature of the event, perform expected behaviors, and demonstrate a potential capacity to adapt in relationships with others.
Generalized disaster reactions may suggest that their perception of the disaster is normal, as are their responses. A survey of psychiatric patients living in the community revealed that 46% understood disaster characteristics and vulnerability and were willing to evacuate.Reference Nakai, Itatani, Nishioka and Hamada 22 Although differences exist when patients are hospitalized, these findings suggest that psychiatric patients may be capable of taking adaptive actions if they accurately perceive disaster situations.
Focus group interviews were conducted 1 year after the disaster. This delay was due to ethical considerations, prioritizing participants’ psychological safety when recalling disaster experiences. However, retrospective memory may have affected the accuracy of recalled patient behaviors,Reference Loftus 23 and the disaster context itself may have influenced nurses’ perceptions of patients’ reactions. Previous studies suggest that recall bias and contextual influences can shape retrospective accounts of disaster-related experiences, and that nurses’ perceptions of patients’ reactions may be influenced by the disaster context.Reference Hofmann, Spiller and Rapaport 24 Accordingly, the findings should be interpreted with an understanding that nurses’ perceptions may reflect both patients’ behaviors and the broader disaster context.
Although non-disaster ward data were not collected, psychiatric inpatients are generally considered relatively stable under medical management.Reference Uçar Küçük, Yıldırım and Çelik 25 Therefore, the observed increases in tension and confusion may reflect the influence of disaster-specific environmental stimuli, such as earthquake tremors and water outages, rather than baseline ward functioning.
Certain maladaptive behaviors were specific to patients with psychiatric disorders. Behaviors such as attempting to go out even when it is dangerous, exhibiting violence, and reporting psychiatric symptoms such as hallucinations and delusions were consistent with those observed in acutely ill psychiatric patients with severe symptoms. These behaviors may indicate that the disaster acted as a stimulus that triggered anxiety and excitement, exacerbating psychiatric symptoms.
As individuals with mental health problems require more evacuation support,Reference Nakai, Itatani, Nishioka and Hamada 22 careful observation and tailored support are crucial for patients with severe mental symptoms.
The theme of “no upset behavior observed” was identified in this study. This theme was classified as “unclassifiable” because it was impossible to determine whether it was adaptive or maladaptive, as no agitation was observed. People typically react to danger in response to situational crises (i.e., stimuli) such as disasters. The disasters examined in this study—an earthquake of magnitude 6.1 and heavy rainfall at a warning level—provided sufficient stimuli for individuals with normal cognitive functions to perceive and react to as unusual events. Indicators for distinguishing reactive and endogenous conditions have been defined by Jaspers.Reference Jaspers, Hoenig and Hamilton 26 In our data, the theme of calm and cooperative was identified, which suggests that despite sufficient disaster stimuli, no reaction was expressed. This finding indicates the need to consider the possibility of negative symptoms or inhibition when interpreting patient behaviors within the framework of reactive conditions.
Under reactive conditions, there is a temporal link between the experience and the reactive state, an implicit connection between the content of the experience and the reaction, and a tendency for anomalies to become less noticeable over time; when the cause disappears, the abnormal reaction also resolves. The reaction process evolves with experience.Reference Utsumi and Kanemoto 27 Applying this framework, the theme of calm and cooperative suggests that the patient did not react to the disaster. This lack of reaction may stem from negative symptoms associated with schizophrenia, such as decreased motivation, blunted thinking, and diminished ability to perceive and act on reality. While these patients may have recognized the disaster, they could not express their awareness through action. Similarly, patients with depression often report being inhibited, as reflected in the complaint, “I’m going to do it, but I can’t move my body.” Such patients may have been aware of the disaster but were unable to act. These findings underscore the importance of nurses accurately assessing the cognitive functions and behavioral expressions of hospitalized patients on a daily basis.
The results of this study suggest 2 distinct patterns of behavioral characteristics in patients: those who are responsive and adaptive to disasters and those who are unresponsive and/or maladaptive. Understanding this divide may be essential for interpreting patient behaviors after a disaster.
Future challenges for preventing mental disorders include elucidating vulnerability factors and indicators and implementing early responses.Reference Ogura 28 The behavioral characteristics identified in this study can provide evidence to improve early disaster responses for individuals with mental disorders. In recent decades, disasters have become more frequent, severe, and prolonged, evolving into an international phenomenon; disaster research and preparedness must be a thoughtful and coordinated international effort, with disaster nursing serving as an integral component.Reference Wall 29 To help vulnerable survivors, including those with mental illness, overcome disasters and move forward, it is crucial to understand their behavioral characteristics.
Conclusions
In conclusion, 19 behavioral themes consisting of 52 subthemes were extracted from 158 behaviors exhibited by psychiatric inpatients after a disaster. Psychiatric inpatients’ behaviors after a disaster indicate general disaster reactions with adaptive behavioral characteristics, although certain behaviors derived from psychiatric symptoms and illnesses were also observed. These findings indicate that care for psychiatric inpatients after a disaster should focus on the high adaptive capacity of some patients, providing support to promote readjustment. Targeted care should be provided for patients exhibiting maladaptive behaviors associated with psychiatric disorders, and careful observation is recommended for patients without overt agitation. Further clarification of psychiatric inpatients’ behavior may require examination of unclassifiable behaviors.
Data availability statement
The participants in this study did not provide written consent for their interview data to be shared publicly; therefore, supporting data is unavailable due to the sensitive nature of the research.
Acknowledgments
We sincerely thank all the nursing staff who participated in this study. We also appreciate Editage for providing English language editing. During the revision process, generative AI–based tools (ChatGPT and Microsoft Copilot) were used solely for English translation and language refinement of author-generated text. These tools were not used in the study design, data collection, analysis, or interpretation. All scientific content, including the accuracy and integrity of the manuscript, is the responsibility of the authors.
Author contribution
YT was responsible for conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, visualization, writing original draft, and writing review and editing; HO critically revised the article and gave final approval.
Funding statement
This work was supported by JSPS KAKENHI Grant Number JP19K10869.
Competing interests
The authors declare none.
