Human–Animal Interactions in Disaster Settings: A Systematic Review

This systematic review aimed to assess the current knowledge of human–animal interactions (HAIs) in disaster settings and identify areas for future research. A Preferred Reporting Items for Systematic Reviews and Meta-Analyses search was conducted on three multidisciplinary databases, identifying English-language journal articles published between January 2000 and February 2022 that explored the benefits of and challenges associated with HAI in disasters and emergencies. The review analyzed 94 articles using both quantitative and qualitative methods. The review found a paucity of universal terminology to describe the bidirectional relationship between humans and animals during disasters and a failure to include all animal types in every stage of disaster and emergency management. Additionally, research predominantly focused on the health and well-being benefits of HAI for humans rather than animals. Efforts to promote social and environmental justice for humans and their co-inhabitants should support the welfare of both humans and animals in disaster settings. Four recommendations were developed based on these findings to increase the inclusion of HAI in research, policy, and practice. Limitations of the review included the exclusion of pre-2000 articles and all grey literature, limited research examining different combinations of animal and disaster types, and limited research outside of North America.


Introduction
Disasters, triggered by natural hazards (for example, floods, wildfires, earthquakes, and pandemics), technological hazards (for example, industrial pollution and accidents), and intentional, willful hazards (for example, terrorism and civil wars), affect humans and their co-inhabitants (animals) on our shared planet (Krupnik 2018;Wu and Etienne 2021). Human-animal interactions (HAIs)-also known as human-animal relationships, bonds, attachments, connections, and so on-can provide various health, social, and cultural benefits, supporting human-centered efforts within an entire disaster cycle (Bebber 2019;Farmer and DeYoung 2019;Ranspot 2019). For example, search and rescue dogs seek out those who are still alive under collapsed buildings after an earthquake (Bulanda 2014). Companion animals have been found to support their guardians in dealing with the mental health consequences associated with COVID-19-driven public health disease transmission mitigation strategies, such as quarantine and homeschooling . Undeniably, accelerating climate change has increased the frequency, magnitude, scope, and influences of different disasters, negatively impacting almost all our societal dimensions (for example, physical, social, economic, and health). Developing a comprehensive understanding of disaster-specific HAIs is crucial in shedding light on the multi-sectoral impacts of disasters, promoting human-animal welfare, and enhancing resilience capacity.
For decades, multidisciplinary scholars from fields such as anthropology, psychology, sociology, veterinary science, and environmental science have significantly contributed to the identification of HAI-specific benefits and challenges associated with extreme events in varying environmental contexts, including households, urban and rural communities, zoos, and forests (Wilson 1994;Swabe 1998;Schumm et al. 2013;White 2018;Rodriguez et al. 2021). Yet, knowledge of HAIs in disaster settings lacks clarity, calling for www.ijdrs.com www.springer.com/13753 1 3 a systematic review and knowledge synthesis. The lack of knowledge in this field hinders the integration of animals into disaster research and prevents effective animal-assisted interventions, leading to negative impacts on human-animal welfare and resilience in disaster scenarios. Addressing this knowledge deficit is crucial to developing interventions and policies that enhance welfare and resilience.
Responsively, this systematic review employed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) approach in the examination of HAIspecific disaster literature. This review article portrays the standpoint and synthesizes the current state of empirical evidence regarding HAI research in disaster settings. The knowledge identification and synthesis assist in identifying the research, practice, and policy deficits and generate recommendations accordingly. Hence, this review aimed to answer the following two questions: (1) What is the current HAI-specific knowledge pertaining to hazards and disaster research? (2) How could a synthesis of current HAI-specific disaster knowledge promote future research?

Method
The PRISMA approach has been widely used, providing evidence-based strategies to support and improve prospective research, practice, and policy/decision making (Page et al. 2021). Accordingly, this systematic review was conducted through the following three phases: literature identification, literature screening, and literature analysis (see Fig. 1

Phase 1: Literature Identification
Animals and disasters are the primary sectors associated with HAI-specific disaster research. Hence, two groups of keywords were developed, namely animals (for example, pet, zoo, livestock, laboratory, and wild) and disasters (for example, climate change, hazards, disasters, crises, and pandemics). In addition, the third keyword group of HAI-related terms (for example, human-animal relationship, bond, attachment, and connection) was added to narrow the focus. The three groups of keywords were cross-searched in three multidisciplinary databases (Web of Science, ProQuest, and EBSCOhost). Three filters were used to establish the search boundary, notably the year range (January 2000 to February 2022), language (English), and document type (journal article). After removing the duplicates, the initial dataset of 334 references was moved to Phrase 2.

Phase 2: Literature Screening
Two researchers collaborated on Covidence (2022), a webbased systematic review management software, to screen the 334 publications. The inclusion and exclusion criteria, such as examining the HAI-specific benefits associated with disaster-related efforts and identifying HAI-driven challenges in disaster and emergency management, were developed to determine the eligibility of each article. The two researchers independently voted on the 334 articles in assessment of the titles and abstracts. Any disagreements were discussed in the team meeting. This phase yielded 94 publications for final analysis. The detailed information of the final reviewed publication list is available on an online data repository of DesignSafe-CI (Heyland and Wu 2022).

Phase 3: Literature Analysis
The two researchers collaboratively conducted the quantitative and qualitative analysis. The quantitative approach focused on univariate analysis, depicting the statistical outcomes of the 94 publications, including the distribution of publication years, the geographic locations of the studies reported in the literature, investigated animal types, examined hazard types, and the stages within the disaster and emergency management cycles. Based on the quantitative outcomes, the qualitative approach thematically analyzed the different themes associated with HAI studies in disaster settings. The two researchers identified the primary themes in each publication and grouped these themes into three categories: HAI-specific terminology and animal types, HAI engagement in disaster and emergency management, and HAI-specific health and well-being benefits.

Results
Detailed information regarding the quantitative and qualitative outcomes is presented in this section to portray the HAI-specific research in disaster settings.

Quantitative Analysis
Between January 2000 and February 2022, HAI in disaster settings steadily gained academic attention. According to Fig. 2, only four articles (one in 2001, one in 2005, and two in 2009) were published before 2010. Since 2010, there has been an upward tendency of HAI academic reporting encompassing 96% (n = 90) of the articles. The global onset of COVID-19 stimulated HAI-specific research, with more than a third of the articles produced as a result (n = 33; 35%). This statistical analysis further examined these 94 studies' geographic locations, their focus on animal types, their hazard types, and stages in the disaster and emergency management cycle. Please note that within each category, one article might cover more than one sub-category.

Geographic Locations
Identifying the geographic locations of the reported studies in the 94 publications provides an understanding of global distribution of relevant research. As shown in Fig. 3, North America produced the most significant number of studies overall (n = 22; 23.4%), followed by Asia (n = 16; 17%), Europe (n = 10; 10.6%), Oceania (n = 9; 9.6%), Africa (n = 6; 6.4%), and South America (n = 4; 4.3%). At the same time, 28.7% (n = 27) of the articles did not specify the continent, either because they compared multiple locations or spoke more generally to global concerns, such as global climate change and zoonotic disease transmission (Esmaeilnejad et al. 2018;Matthew et al. 2022).

Animal Types
Multidisciplinary studies classify animals into five broad categories: companion animals that are often kept at home for companionship (Travers et al. 2017;Protopopova et al. 2021), agricultural animals that produce value or work for the owner (Boni et al. 2009;Wild et al. 2021), laboratory animals that are mainly used for experimental purposes, zoo animals that are kept in community-based facilities for public viewing and education (Hooper et al. 2021), and wild animals that do not typically have frequent contact with humans (Dowsley and Wenzel 2009;McCumber and King 2020). As shown in Fig. 4, some literature focused on a specific category, while 13.8% of the articles discussed multiple animal types (n = 13). Companion animals dominated the literature (n = 39; 41.5%), followed by zoo animals (n = 18; 19.1%), agricultural animals (n = 11; 11.7%), and wildlife (n = 7; 7.4%). No reviewed article focused on laboratory animals and five articles (5.3%) did not specify the animal types.

Stages in the Disaster and Emergency Management Cycle
The disaster and emergency management cycle includes four stages, namely pre-disaster preparedness, emergency response, post-disaster reconstruction and recovery, and disaster mitigation (Warfield n.d.). As shown in Fig. 6, the stages of mitigation (n = 40; 42.6%) and emergency response (n = 38; 40.4%) dominated the literature, while pre-disaster preparedness (n = 10; 10.6%) and post-disaster reconstruction and recovery (n = 2; 2.1%) were limited.

Qualitative Analysis
Disasters contextualize HAI studies into a unique landscape to examine the human-animal-specific disasterrelated efforts, such as emergency rescue (McCarthy et al. 2018), animal-assisted therapeutical interventions (Ng et al. 2021), and the engagement of animals in disaster mitigation (Lamb 2019). The following three themes have been identified in the final reviewed articles.

Human-Animal Terminology and Animal Classification
In addition to the HAI terminology, various potentially interchangeable terminologies were found in the final literature pool, such as human-animal bonds (Hill et al. 2020), human-animal relationships (Gurtner and Parison 2021), human-animal connections (Trigg et al. 2016), and human-animal attachments (Every et al. 2016). This diversity illustrates a disjunction of terminological usage, yet the general agreement is that HAI can lead to relationships, bonds, connections, and attachments. Inconsistent use of terminology and lack of clear definitions have hindered cohesive inquiries in human-animal studies in general (Anthony 2003), and in disaster settings in particular. In fact, the literature shows that researchers studying companion animals, when referring to veterinary medicine, prefer using human-animal interactions and human-animal bonds (Ng et al. 2021). In contrast, those studying agricultural animals tend to use human-animal relationships (des Roches et al. 2016). However, animal welfare scholars suggest that humans and agricultural animals can form strong bonds and such animals should not merely be considered for their utility and economic purposes (Anthony 2003). Although these terminologies are easily understandable and exchangeable (Ratschen et al. 2020), using different terms can influence how humans perceive the importance of various dimensions of HAIs, while using commonly agreed upon terminology will avoid potential confusion, inaccuracies, and cultural/linguistic misunderstandings and improve efficiency. Although the disaster-specific classification of animal subjects is easily distinguishable in literature, variation in terminology hinders the acceptance of HAI studies as a unitary field of research. Correction of this discrepancy is vital in hazards and disaster research, since disaster and emergency management are multidisciplinary and multistakeholder engaged, with disasters affecting individuals and communities from different social and cultural backgrounds.
Human-animal interactions in the companionship and agriculture sector are the top two categories in the reviewed studies, especially companionship that has taken place during the global COVID-19 pandemic. The research that investigated HAIs in the context of laboratory animals, zoo animals, and wild animals is relatively recent (Hosey and Melfi 2014). Furthermore, the reviewed studies that involved the laboratory HAIs primarily focused on the interconnections among zoonotic disease, climate change, and human-animal welfare (Wu and Etienne 2021), rather than specifically on the disaster context, such as earthquakes or power outages. These extreme events affect animal welfare and human suffering ), but no research in the reviewed article pool examined this topic. This indicates the need for potential research engaging laboratory, zoo, and wild animals in HAI-specific disaster investigations.
The literature shows that humans attribute varying values to animals based on their species or geographic location. In disaster settings, the level of attachment to animals is associated with their species, with companion animals, namely dogs, cats, and horses, having stronger bonds with humans than small mammals such as birds, fish, and reptiles (Ratschen et al. 2020). For example, studies have demonstrated that dogs and cats can have a significant or moderately positive impact on the mental health of their guardians during disasters, whereas guardians of ornamental fish are relatively indifferent to the effects of their pets on their well-being (Clements et al. 2021). The strong positive impact of companion animals on the well-being of their guardians is well established (Bowen et al. 2020;Bussolari et al. 2021), and Fig. 6 Concerned stages of the disaster and emergency management cycle of the 94 articles on human-animal interaction in disaster settings they can also influence their guardians' risk perception and coping behaviors (Trigg et al. 2016;Heath et al. 2001).
This pattern of behavior may vary depending on the type of animal, as evidenced by Schor and Protopopova's study (2021), which found that during the pandemic emergency response pet services, such as Vancouver's Downtown Eastside pet food bank, which provides services to pet owners of lower socioeconomic status, reported a decline in the number of clientele visits for cats, rabbits, rats, and "other" pets, but not for dogs. In hazards and disaster research, studies primarily explore the economic and health impacts of disasters on livestock and poultry, including chickens, ducks, cows, pigs, and goats, with only a few studies focusing on wild animals (Dowsley and Wenzel 2009;McCumber and King 2020) due to disproportionate devastation of the natural environment caused by disasters.

Engaging Human-Animal Interaction (HAI) in the Entire Disaster and Emergency Management Cycle
As illustrated by the quantitative analysis outcomes, most studies focused on natural hazards, especially biological hazards, such as COVID-19 (Grajfoner et al. 2021 (Aamodt 2018), and toxic exposure (Čapek 2010). Only one article (Tambo et al. 2018) applied HAIs to the examination of human-made conflict events and epidemics (Lassa fever) to inform emergency response practice in Nigeria. The distribution of HAI-specific studies among the three hazard categories indicates that HAIs have been involved in the full spectrum of hazards and disaster research. In addition to COVID-19, the study of HAI engagement in other types of natural hazards, technological hazards, and intentional, willful hazards is still developing. Furthermore, companion animals, like dogs and cats, centered in most reviewed studies (Morgan et al. 2020). For instance, Every et al. (2016) argued that companion animal guardians' risk perceptions and emergency response behaviors have been significantly influenced by their companion animals, including ignoring, delaying, or refusing evacuation orders in wildfires and floods in Australia if they have to leave their animals alone. The same results confirmed the finding in the previous section that livestock, laboratory animals, zoo animals, and wildlife were rarely addressed. Animals have increasingly been included in disaster and emergency management legislation and policies worldwide (Squance et al. 2021). Historically, it has been reported that animals exhibit abnormal behaviors before earthquake events due to their notice of geophysical changes (Woith et al. 2018). Although more research is needed to substantiate sound scientific evaluations, these animal behaviors have been anecdotally helpful warning signs of impending natural hazards and can be exploited for pre-disaster planning. Beginning in 2005, companion animals were considered an essential factor in pre-disaster preparedness (Farmer and DeYoung 2019). One year after Hurricane Katrina (2005), the U.S. federal Pets Evacuation and Transportation Standards Act (PETS Act) was signed into law, recommending the inclusion of pets in emergency response planning (American Veterinary Medical Association n.d.). However, even after several years of development, Darroch and Adamson (2016) argued that including animals in pre-disaster preparedness planning still remains in the initial stage.
Furthermore, during the emergency response stage, immediately after disasters hit, rescue animals have been shown to assist rescue teams in life-search and life-saving efforts. For instance, Frieda, a Labrador retriever, who was deployed throughout her career by the Mexican Navy several times immediately following earthquakes in Mexico, detected the bodies of 52 individuals (12 of whom were still alive) (Jenkins 2017). The human-animal bond has a significant impact on disaster survivors' behaviors. Particularly, Brackenridge et al. (2012) discovered that 38.5% of Hurricane Ike survivors who did not evacuate cited not wanting to leave their companion animals as their main reason. The limited number of animal-friendly shelters was the main reason that people refused to be separated from their animals during Hurricane Matthews in 2016 (Farmer and DeYoung 2019) and the same concerns triggered animal loss in the 2005 Hurricane Katrina (Zottarelli 2010). In addition, overloaded animal-friendly shelters threaten the health and wellbeing of staff (Jacobs and Reese 2021). Similarly, transportation and logistics problems related to relocating animals and keeping them safe are the most prominent reasons that animal guardians risk their own lives in a disaster context. Animals' supporting function is a determining factor in the disaster recovery process. Since many animal guardians are willing to risk their own lives to save their animals behind the evacuation line, in return, animals can help their guardians better cope with short-term and long-term post-disaster recovery. The positive impact that companion animals have on the psychological health and longevity of their guardians has been well established (Bowen et al. 2020;Packer et al. 2021). Most reviewed studies examined HAI-specific disaster mitigation strategies associated with zoonotic transmission within the human-animal-environment triangulation (Seimenis and Battelli 2018;Asaaga et al. 2021). Positive human-animal bonds improve pro-animal attitudes, including general attitudes toward animal protection, in both disaster and non-disaster scenarios (Gu et al. 2021). Thus, individuals with stronger human-animal bonds are more likely to support animals affected by disasters and engage their animals in disaster-related efforts. This illustrates a new direction for HAI-specific disaster meditation interventions within non-epidemic disasters.

Human-Animal Interaction (HAI)-Specific Health and Well-Being Benefits
There is a growing appreciation for different types of companion animals following the social isolation experienced during the COVID-19 pandemic that altered the human-animal-environment triangulation. All the COVID-19-driven literature strongly supported the bidirectional relationship associated with human-animal welfare. This review highlights the health and well-being benefits during COVID-19 from four sub-themes, as shown in Table 1.
Physical health benefits In technology-reliant societies, HAIs promote better health by establishing a physical activity routine (Applebaum et al. 2021;.  found that owning animals, such as dogs and/or horses, is directly linked to an increase in physical activity and improved health. Among the studied dog guardians, 42% reported walking their dog more than they used to prior to the pandemic , particularly during the lockdowns and periods of social isolation Mueller et al. 2021). Walking and feeding companion animals daily and at specific times helps people maintain a routine and therefore contribute positively to human health Hoffman 2021). Shifting one's attention to care for an animal not only offers a chance to be physically active but also offers a break from ruminating about the unpredictable state of affairs and the future, thus possibly reducing anxiety.
Mental health benefits The perceived emotional support that companion animals offered to their guardians during the pandemic helped people adapt to new social and environmental changes (Bowen et al. 2020;Ratschen et al. 2020). For pet guardians who worked remotely, their companion animals became their primary psychological resources. The HAIs were particularly crucial for individuals living alone, especially among older adults and people experiencing homelessness (Kelemen et al. 2020). A survey of 4105 dog owners found that over half of the respondents were able to reduce feelings of distress through interactions with their dog during COVID-19 . It was also found that the number of new companion animal guardianships dramatically increased during COVID-19 due to the same HAI mental health benefits (Ng et al. 2021). Aside from potentially being able to help individuals cope with psychological distress, including anxiety, loneliness, and depression, at home, HAIs can positively impact employees' work behaviors and attitudes toward others in a group setting (Ng et al. 2021).
Overall well-being Being around and interacting with an animal can provide individuals with a healthy and active routine and foster emotional regulation skills through perceived social support that may help to alleviate feelings of social isolation (Bowen et al. 2020;Ratschen et al. 2020). Human-animal interactions increased among all age groups during COVID-19, especially during the lockdowns and periods of social isolation Mueller et al. 2021). Pets offered their guardians a sense of purpose and improved their well-being (Applebaum et al. 2021). Pet-related activities such as walking, feeding, and petting helped individuals maintain a routine that otherwise may have been challenging due to blurred boundaries between home, neighborhood, community, and work (Hoffman 2021)-a routine change that can be seen as a positive contributor to overall health.
One Health and One Welfare Extreme events provide an opportunity to examine the interplay between humans, animals, and the environment. The literature addressed the specific linkages between health and well-being in this context through emerging conceptual and multidisciplinary models including One Health (Kahn 2021), and One Welfare (Chen and Yuan 2020). One Health seeks to improve human health by integrating the influences of relevant environmental exposures and animal contact (Rabinowitz et al. 2017;Hassani and Khan 2020). One Welfare, built upon the One Table 1 Human-animal interaction (HAI)-driven physical health, mental health, and overall well-being Health framework, calls upon related professionals within this triangulation (for example, environmentalists, human healthcare workers, and veterinary professionals) to collaboratively recognize the contributions of human-animal-environment interplay to human-animal welfare, by holistically encompassing broader social, economic, environmental, and cultural dimensions within natural, built, social, and human environments (Squance et al. 2021;Wild et al. 2021). One Welfare proposes a potential theoretical framework to respond to the increasing number of worldwide zoonotic diseases caused by global climate change and related human behaviors.
Growing evidence suggests that HAIs boost human health and well-being in disaster settings and beyond. To maintain human-animal welfare, a holistic approach must consider human-animal-environment interconnections. Companion animal guardians are the primary caregivers for their companion animals when disasters hit. Different organizations, such as animal protection and animal rescue agencies (for example, Society for the Prevention of Cruelty to Animals, SPCA) have taken on the responsibility for ensuring animal welfare in disaster situations (Gurtner and Parison 2021). These efforts only cover a minority of the co-inhabitants on the Earth. Hence, an inclusive solution must be implemented to engage all animals in disaster and emergency management planning and consider the human-animal-environment triad.

Discussion
The quantitative and qualitative analysis outcomes reflect that the human-animal bidirectional connections have not effectively developed despite having gained some attention in hazards and disaster research. Built on the current knowledge, this section discusses some potential approaches to create inclusive HAI-specific research, practice, and policy in disaster settings.

Inclusive Human-Animal Interaction (HAI) Terminology
Although diverse terms warrant closer examination in a specific disaster context, these disciplinary language disparities have become barriers preventing effective global interdisciplinary/transdisciplinary collaboration in hazards and disaster research and practice communities (Peek et al. 2020). A universal term and definition for HAIs is required to promote multidisciplinary and multi-stakeholder engagement throughout the entire disaster and emergency management cycle, efficaciously reflecting the mutual connections between humans and their co-inhabitants.

Inclusive Animal Types
Companion animals have received the most attention in disaster settings. Climate change and disasters have disproportionately affected the agriculture sector. This has gained the animal industry's attention (Tschopp et al. 2010). An inclusive approach is needed to expand the current knowledge about HAIs. This inclusive approach will present a balanced picture of HAIs vis-à-vis all animal types and improve the application of scientific knowledge to advancing HAI-related knowledge, more importantly, promoting environmental and social justice for both humans and their co-inhabitants.

Inclusive Disaster and Emergency Management Cycle
Literature suggests that legislation and policy need to consider transdisciplinary tactics to include HAIs in the entire disaster and emergency management cycle (Darroch and Adamson 2016;Squance et al. 2021). Zoonotic disease prevention was the primary focus of the reviewed literature during the disaster mitigation stage. Most literature focusing on the emergency response stage contributed to the nuanced understanding of companion animals as the motivating factor influencing disaster survivors' risk perception and risk response behaviors (Heath et al. 2001;Every et al. 2016;Trigg et al. 2016;Farmer and DeYoung 2019). During the post-disaster reconstruction and recovery stage, vital social and economic factors impact diverse groups within the human population; these interconnected factors also hinder companion animal guardians' social and health recovery (Schor and Protopopova 2021). A few studies have examined the animal influences on the post-disaster reconstruction and recovery stage and pre-disaster preparedness stage (Trigg et al. 2019). The reviewed literature emphasized the need for a transdisciplinary approach to incorporate HAIs into the entire disaster and emergency management cycle, given the interconnected nature of the different stages.

Inclusive Animal Perceptive
Historically hazards and disaster research is human-centered. Human-animal interaction-specific hazards and disaster studies focus on animals' contributions to human health and well-being. Investigation into animal welfare affected by disasters remains sparse. A study investigating the human-dog relationship during the pandemic showed that new behavioral problems in companion dogs impaired their guardians' quality of life (Morgan et al. 2020). Many who underwent lockdown or quarantine were more likely to report negative behavioral changes in their companion animals (Jezierski et al. 2021). Similar to humans, animals need socialization. Experiencing high levels of stress can also negatively impact animal welfare (Packer et al. 2021); this can reduce the positive influences animals could potentially have on their human guardians. Disaster-driven restrictions, such as uncertainty and evacuation, could amplify the animal's pre-existing behavioral issues (Bowen et al. 2020).
Missing the animal's perspectives, not providing proper treatment of enough socialization can be mainly attributed to the caretakers having an anthropocentric view. The bidirectional relationship associated with human-animal welfare is strongly supported by COVID-19-driven research; however, no objective measures were taken. Based on the human-animal-environment triangulation in which the One Welfare framework is rooted, an inclusive approach to engaging animal welfare, in turn, supports human health and eventually promotes human-animal resilience.

Limitations
This systematic review explicates a synthesis of the current knowledge regarding HAIs in disaster settings-a prominent cross-sectoral field that requires urgent interventions. While a systematic review portrays the landscape of current knowledge and suggests potential research orientations, several limitations are noteworthy. First, both HAIs and disasters studies have long histories, and the outcomes have been published in different venues of academic and grey literature worldwide. The review period excluded the academic publications before 2000, the grey literature, and non-English publications; all the excluded literature also provides valuable knowledge to examine HAI-specific disaster-related efforts from different perspectives. Second, although the review illuminated the brief landscape of HAIs and disasters, due to limited data available on different types of animals, disasters, and related influences, more detailed information and evidence-based strategies are urgently needed to promote effective engagement of HAIs in disaster settings. Since different types of animals and different types of disasters generate different influences on HAIs and require customized interventions (for example, companion animals versus zoo animals in urban floods, as well as laboratory animals and agricultural animals in industrial explosion), these deficits in research mirror the under-researched areas. Third, most articles we reviewed came from North America. Research done in South America and Africa was minimal (Bebber 2019;Egeru et al. 2020), where the problems associated with climate change, disaster, and zoonotic disease were most prominent. Fourth, the reviewed COVID-19-driven studies covered the most vulnerable and marginalized groups, such as children, people with (dis)abilities, low-income people, and people experiencing homelessness. Hence, disasters have disproportionately affected vulnerable and marginalized groups, with their representation in climate-induced tragedies remaining sparse. Therefore, all these limitations call for more studies to contribute to a larger pool of nuanced understanding of how HAIs are affected by disaster settings.

Conclusion
Although gaining attention, HAI-specific hazards and disaster research is still developing. This systematic review used both quantitative and qualitative approaches to identify and synthesize current HAI knowledge, which influences the different disaster-related efforts. The quantitative analysis indicated that there is a lack of universally agreed upon terminology to describe HAI. In addition to companion animals, other types of animals (agricultural, laboratory, zoo, and wildlife) have not been fully included in different stages within the entire disaster and emergency management cycle. The qualitative outcomes not only provided detailed information confirming the exclusion of some types of animals in disaster settings but also demonstrated that current studies have predominantly focused on HAI-driven health and well-being benefits for humans, rather than animals. These outcomes made it clear that an inclusive and transdisciplinary approach is necessary to improve the understanding of HAIs and promote environmental and social justice. This includes establishing a universal definition of HAIs, studying all animal types in disasters, and ensuring that animal welfare is not neglected. Zoonotic disease prevention, risk perception, and social and economic recovery should also be considered in all stages of disaster and emergency management to promote resilience for both humans and animals.
Since disasters affect almost all societal dimensions, these sectoral and cross-sectoral influences shape both the human and animal dimensions of HAI. COVID-19-driven global companion animal guardianship has stimulated HAI-specific disaster research. As the world moves into the pandemic recovery stage and accelerating climate change increases extreme events, prospective research could apply a more inclusive approach to the examination of the various influences that disasters have on bidirectional human-animal relationships, examining the contributions of HAI on diverse disaster-related efforts. These potential research areas will not only support human-animal welfare in disaster settings but also eventually promote social justice and environmental justice for humans and their co-inhabitants on the shared planet.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.