A response framework for addressing the risks of climate change for homeless populations

ABSTRACT People experiencing homeless have greater vulnerabilities in relation to climate change that require a range of policy and systems approaches. There are two interrelated areas that policymakers can consider in relation to climate change and homelessness: migration and exposure. This synthesis of the available data and expert opinion provides practical information to policymakers, with specific strategies alongside case examples. The data captured here is through systematic reviews, and expert opinion is generated through input from a year-long series of five virtual think tanks. Throughout this synthesis paper, an emphasis is placed on explicitly addressing homeless populations in the policies and plans designed to address climate change-related impacts. Prevention-oriented plans are shown to be more effective in terms of outcomes and cost-effectiveness compared to the more commonly deployed crisis response models. Another key issue considered is the availability of relevant data with which to target policy responses and evaluate outcomes. Data-driven responses tend to be more successful, though relevant data are, to date, lacking for homeless and other marginalized populations. Moreover, effective policy design in this area needs to be intersectional and inclusive, tailored to the needs of local communities and developed in consultation with lived experience stakeholders, including service providers. Policies that ignore local input tend to fail. Prevention-oriented, culturally-situated, and trauma-informed systems and services hold the greatest promise in responding to the severe health risks and inequities that homeless populations face in the climate crisis. Key policy insights Prevention-oriented measures are key, with most focussing on the availability of affordable housing and upgrading housing and living conditions of vulnerable populations. There is a need to include lived experience and input from local communities, especially when designing measures that will impact livelihoods, such as planned migration. Disaster, crisis response, and aftercare plans need to outline explicit measures for homeless populations. There is a need for cross-sectoral alignment of policy and intervention responses. Successful approaches tend to be culturally-situated and trauma-informed.


Introduction
The impacts of global climate change on human health have exacerbated existing health disparities across wideranging social, political, and geographic contexts. This phenomenon has been observed at regional, intra-country, and country-country levels, with greater problems in low-and middle-income regions (Ebi & Luchters, 2021). Population-level health inequities are also exacerbated by climate change, with women, children, the elderly, Indigenous peoples, and their intersections disproportionately impacted (Vickery, 2018;Vinyeta et al., 2015). Homelessness is an important factor in understanding health disparities and represents a nexus in which the lack of adequate shelter compounds the health risks of extreme and changing weather (Kidd et al., 2020).
Furthermore, jurisdictions with large populations exposed to weather extremes experience heightened demand for health and emergency services and associated costs, both in terms of disaster response and ongoing healthcare utilization (Powell et al., 2020). Other considerations include implications for: (i) child protection and justice systems; (ii) the labour market; (iii) public perception challenges regarding how poverty is addressed; (iv) impacts on the quality of life for all in affected environments and, broadly; and (v) losses in the population contributing to the economy (Anderko et al., 2020;Efobi, 2022;Siposne Nandori, 2022).
Despite the clear health risks and implications of climate change for homeless populations, the topic has received relatively little academic attention. This synthesis paper concentrates on the implications of climate change for the health of populations experiencing homelessness. We define the term 'homelessness' as a dimensional phenomenon, ranging from lacking any form of shelter through to various forms of insecure and inadequate housing (FEANTSA, 2017).

Synthesis objective
This synthesis paper outlines policy and service system-level opportunities for addressing the disproportionate impacts of climate change on homeless populationsboth as a driver of homelessness prevalence and as a risk to those most exposed to weather extremes. The intent is to provide policy and practice guidance to decisionmakers from municipal to federal levels with the aim to reduce the prevalence of homelessness and the health risks of those exposed. The emphasis is on prevention-oriented responses while recognizing that until the outcomes of prevention-oriented interventions are realized, crisis response will continue to be required. The paper presents a response framework drawing on recent academic literature reviews and on points discussed in expert think tanks (Bezgrebelna et al., 2021;Kidd et al., 2020). The latter served to validate and expand upon the relatively limited literature-base in this area. To the best of our knowledge, this is the first attempt to capture and consolidate the literature, emerging insights, and to formulate such a set of recommendations. This paper is timely, with the scale of the problem rapidly increasing alongside the political momentum of COP27 and greater policy-level attention to homeless populations during the COVID-19 pandemic.
In the following sections, we summarize the literature, outline methods and the development of the response framework, and present the key findings. The results are summarized and represented in a figure (see Figure 1). The response framework provides practical information to policymakers, with specific strategies alongside relevant case examples. We conclude by discussing the implications of the response framework presented here.

Climate change, migration, and homelessness
Climate-driven migration is a major problem globally. It is a driver of an array of problems from individual to system levels, and is associated with homelessness and compromised housing (Luetz, 2018). Migration in this frame has been most extensive in low-income countries. Alongside the problems that accompany migration, the poorest are, in many instances, unable to migrate and forced to subsist in profoundly degraded environments (Leichenko & Silva, 2014). All major climate change impacts have been associated with migration, including rising sea levels, increased temperatures and aridity, fires, flooding, and storms (Black et al., 2011;Carrico & Donato, 2019;Cattaneo et al., 2019;Shen & Gemenne, 2011). Direct impacts of weather extremes (e.g. compromised housing, water insecurity) are moderated by existing infrastructure and system coordination, such as that aimed to mitigate health impacts that in turn influence outcomes related to homelessness   Table 1). Emerging evidence suggests that the drivers of migration and displacement are interrelated, with climate change being an additional push factor rather than an exclusive cause (Kalin, 2015).
The ways in which migration is associated with homelessness depends on the specific climate impacts (e.g. acute vs. slow-onset), infrastructures, health risks, inequities, type of migration (planned vs. unplanned), and risks of violence. Economic challenges, intensified by COVID-19, and increasingly expensive housing markets worsen this situation, as migrants face substandard living conditions and with international migration complicated by border closures (Chamie, 2020). Rural-urban migration has been the most extensively documented with respect to homelessness. This risk increases when climate shocks influence agriculture, with a lack of education and training being key contributors to rural-urban migrant homelessness (e.g. Luetz, 2018;Nawrotzki et al., 2017). Urban-urban migration has received less attention and has been difficult to attribute international migration to climate change-related events due to the lack of consistent data. However, substandard living conditions in refugee camps suggest an association between country-country, climate-driven migration and homelessness (Stapleton et al., 2017).
Overall, the evidence base related to homelessness as a function of climate-driven migration is nascent. Longitudinal and representative data are lacking with much of this modest literature relying on cross-sectional associations and commentaries (Bezgrebelna et al., 2021;Kidd et al., 2020).

Climate change, homelessness, and exposure
The exposure risks that climate change represents for populations experiencing homelessness occur in parallel with, and as a result of, the migration risks described above. There is more evidence available on the topic of exposure, though it remains modest, relying extensively on cross-sectional associations and with the least information available regarding the most-impacted populations (e.g. low-and middle-income countries; Bezgrebelna et al., 2021;Kidd et al., 2020). Lacking adequate shelter, these populations are acutely exposed to weather extremes and their sequelae (Liu et al., 2016;Marcotullio & Schmeltz, 2021;Nguimalet, 2018). Weather exposure compounds intersectional adversity, including factors common amongst homeless populations, such as various forms of discrimination (e.g. gender-based, racism), violence exposure, and compromised physical and mental health (Vickery, 2018). Consequently, health and social impacts experienced as a result of exposure are linked to various vulnerability factors prevalent within populations experiencing homelessness (Managan et al., 2014;World Health Organization, 2021).
There is a wide array of examples of these exposures and their deleterious impacts, including extreme heat and cold, disasters such as flood and drought, food and water insecurity, disease, and air pollution (e.g. Hughes et al., 2016;Kabir et al., 2016;Kinney, 2018;Lane et al., 2018;Liu et al., 2016;Marcotullio & Schmeltz, 2021;Nguimalet, 2018;Nyiwul, 2021;Schimpf & Cude, 2020). Primary (e.g. heat) and secondary (e.g. water insecurity; disease) risks are moderated by a range of factors (e.g. chronic illness, social infrastructure, heat island effects) that influence outcomes (e.g. Lane et al., 2018;Nguimalet, 2018) (Table 1). These outcomes include illness morbidity and mortality, injury, violence exposure, and mental health issues, exacerbated by exposure to traumatic events. Also observed are increases in emergency service use and infrastructure burden (Dodgen et al., 2016;Gronlund et al., 2018;Sorensen et al., 2020). The COVID-19 pandemic has likely resulted in greater prevalence of weather-exposed populations and further compromised health of those who are exposed (Ejiogu et al., 2020).
Indigenous populations, globally, face unique risks at the homelessness-climate nexus. These challenges concern the geographic location of Indigenous communities, Indigenous over-representation amongst homeless populations, and cultural considerations with respect to relationships with the land (Vinyeta et al., 2015). These impacts represent yet another facet of the legacies of colonial repression and violence (Vinyeta et al., 2015).

A complex systems framework for understanding the health risks of homelessness
The health implications of climate change among the homeless are best considered within a complex systems framework (e.g. Negev et al., 2019;Sterman, 2012). There are a number of indirect threats that occur in parallel with more obvious climate-homelessness impacts. Violence exposure is one example. Weather-related disasters and extremes heighten the risk of interpersonal violence for unhoused populations alongside larger scale regional and national conflicts (Kidd et al., 2020). The resources available in affected countries are another systems consideration. Infrastructure and healthcare limitations, alongside housing and employment shortages, serve to compound the risks faced by unhoused populationsall of which are greater in countries that face the most severe impacts of climate change (e.g. Shemsanga et al., 2010).
The complex systems lens also has implications for the processes that accompany the climate-homelessness nexus over time. For example, a negative feedback loop occurs upon an individual's loss of housing due to climate pressures (Sterman, 2012). Once housing is lost as a critical determinant of health and stability, problems in all areas compound exponentially over time. This negative feedback loop has implications for intervention. In such a system, as risks increase exponentially, the cost and complexity of intervening increases exponentially the further systems engage along the time course, with prevention being far more efficient than crisis response. A complex systems perspective also emphasizes a focus on leverage, which necessitates high quality data. In a resource-constrained problem area such as homelessness, targeting interventions at points of leverage in the system (where the benefits of investment are maximized) is an important strategy (e.g. Mahadevia et al., 2020). However, data is needed to identify leverage points as they are not always readily apparent. Data in the frame of climate change-homelessness is currently lacking (Bezgrebelna et al., 2021). Hence, the recommendations in this synthesis emphasize responses that are mindful of the complex systems involved. This is important, as poorly designed and failed policy in the area of homelessness can compound public and political apathy (Brinegar, 2000).

Methods
The climate change-homelessness response framework was developed in two steps. First, an evidence synthesis drew upon two scoping reviews of the peer-reviewed academic literature on this topic. One review focused on climate change implications for homelessness, especially as they relate to health and social outcomes (Kidd et al., 2020). Twenty-six papers were identified, mostly focusing on the US. The databases searched were PsycINFO, Medline, Scopus, and Google Scholar, with the following search terms used: 'homeless*' and 'climate' or 'weather' or 'environment*'. Themes from the identified papers were charted and summarized thematically (Braun & Clarke, 2006). This scoping review highlighted the need for a broader framework with which to conceptualize the issues touched upon across very diverse sets of evidence and commentaries. Second, to address broader contexts of housing precarity (beyond 'homelessness' per se), a systematic review of reviews was conducted (Bezgrebelna et al., 2021). Fifteen reviews (five systematic and 10 non-systematic) were identified mostly with a global or North American focus. For this review, the databases searched were PubMed, PsycINFO, Web of Science, and Google Scholar, with the following combination of search terms: 'poverty' or 'impoverish*' and 'climate' or 'weather' and 'hous*' or 'homeless*' and 'review' or 'synthesis'. In addition to these reviews, a grey literature search (unpublished) was also conducted to further inform the discussion of the key topics, with clear delineation of information sources. The content analysis procedure was similar to that of the first review. For further details on review methods, please refer to the original papers (Bezgrebelna et al., 2021;Kidd et al., 2020).
This evidence synthesis was complemented through input and feedback from a global network of experts. Expert input was obtained through a year-long series of five virtual think tanks that progressed from a broad scoping and defining of issues through to more concentrated discussions of response options organized along the major themes of migration and exposure. Themes were derived from consensus (MB and SK) through a review of meeting recordings with notes taken (Braun & Clarke, 2006). A review of emergent themes in subsequent meetings allowed for member-checking opportunities. The expert group was diverse, with representation from various geographic regions (Australia, Brazil, Canada, Kenya, New Zealand, Switzerland, the UK, and the USA), disciplines (academic, healthcare provider, policy advisor), and expertise domains (disaster response, city mitigation planning, public health). During the first, most open-ended think tank, the question of which key areas to focus on was discussed. The themes of migration and exposure were developed as a result and the subsequent think tanks focused on these two themes. We had also been advised that, due to the time zone differences, it was best to hold multiple think tanks at different times. Consequently, more experts were able to join overall, although there were fewer participants in each individual think tank. Additionally, breaking the subsequent think tanks into specific theme areas led to experts choosing to participate in the topic domain most pertinent to their interests. Fourteen individuals contributed to the first virtual think tank and six in each of the subsequent think tanks, with 11 contributing to the framework document. Most of the experts involved were part of the first think tank, with only 1-2 experts being new at the subsequent think tanks.
The response framework, in turn, was organized around the two major themes of migration and exposure with content drawn from both literature reviews and think tanks, with the ideas described here reviewed and validated through both think tank discussion and through the involvement of several network members in the authorship of this paper. Within each theme, specific problem areas from policymaker perspectives are described alongside potential interventions. Case examples for each intervention are provided. Following these sections, cross-cutting principles are outlined in order to describe approaches and processes that apply to most or all of the responses articulated in the synthesis.

Government and system leader challenges and solutionsmigration
For governments, prevention-oriented responses are both far more impactful and a better investment (see Figure 1). This response theme emerged in both reviews and in think tank conversations. An emphasis on prevention was linked to a need for better data with which to target preventative responses. Reliable, predictive data is needed to identify high-risk populations and the leveraged, evidence-based interventions that they require. These imperatives, of enhancing prevention and the data with which to target prevention, are especially salient for, and challenging in, low and middle-income countries, which frequently rely on financial support from high-income countries (e.g. European Union providing support to Turkey to address migrants; European Commission, 2021). Governments considering this challenge may be motivated by the economic losses that accompany the abandonment of affected jurisdictions, alongside the many downstream problems that displaced and homeless populations represent (see Exposure, below). Further, unplanned migration increases difficulties in the accommodating areas.
In terms of preventing risk (related to climate-driven migration into homelessness), both expert opinion and the academic literature suggest that the most leveraged and impactful interventions concentrate on remediating risks posed by housing that is not climate resilient and that is located in vulnerable geographic areas. Addressing housing quality is, in this context, situated in the broader effort to facilitate staying in place where it is possible and safe to do so (see Figure 1) (Musah-Surugu et al., 2018;Satterthwaite et al., 2020). Intervening to improve housing quality minimizes the economic, social, and health risks of individuals being driven into migration due to low-quality housing. Such interventions can be facilitated by providing financial support to local governments in high-risk areas where climate financing is available to support development and adaptation measures (Musah-Surugu et al., 2018).
The first step in this process involves gathering data (e.g. geographic and climate risk projection, socioeconomic, community response-preparedness, remediation options and associated costs) (Carcellar et al., 2011). Subsequent efforts include engaging pertinent local governments and organizations to raise awareness, co-design and align approaches with existing cultural and organizational models, and to implement programmes and iterate approaches based upon feedback. Such work can include the implementation of mandated or subsidized improvements to low-quality housing (Cramer et al., 2016;Satterthwaite et al., 2020). Successful programmes consider local context and needs and rely on participatory approaches to address housing vulnerabilities (see Cross-Cutting Principles below). Governments and organizations can also implement risk remediation plans that incorporate disaster response strategies as well as adaptation plans that prioritize the most vulnerable. For example, in Ghana, some external climate change adaptation funding is only available if it creates jobs for the rural poor, thus aiding in poverty reduction and addressing climate-related issues simultaneously (Musah-Surugu et al., 2018). Similarly, an analysis of slum upgrading programmes in Latin America indicated benefits for both climate change adaptation and mitigation (Núñez Collado & Wang, 2020).
The assessment and engagement activities described above might also reveal areas where remaining in place is untenable even with available supports. The next step in these instances would be planned migration (see Figure 1). While migration is not ideal, planned migration is far preferable to unplanned migration where families and individuals in crisis are far more likely to become homeless as a result. As with staying in place, local governments and organizations would work in collaboration with communities, relying on existing connections and practices to develop ethical and inclusive approaches to planned migration. These themes arose in both reviews and think tank discussion. Jamero et al. (2019) indicated that this kind of intensive engagement is essential for establishing communication with and support for community members who are unwilling or unable to relocate (e.g. due to poverty, loss of property, cultural beliefs). For example, in small island communities in the Philippines, a relocation program was more successful when in-situ adaptation strategies are applied first. Such strategies provide community members with access to the needed resources and enable them to make the decision to relocate, rather then being forced to do so (Jamero et al., 2019). In a complementary manner, local organizations can play a pivotal role in ensuring broader public understanding, creating opportunities for public debate and cooperation with government plans when they are the best course of action.
Governments, in turn, can play a key role in providing funding, setting standards, and addressing infrastructure necessary to prepare housing, livelihoods, health and social services, and schooling for planned migration (Ferris, 2015;Hallwright & Handmer, 2021). For instance, Ferris (2015) wrote that successful relocations (e.g. in Vietnam and New Zealand) differ from unsuccessful ones (e.g. in Peru and Turkey) largely due to concerns regarding livelihoods and lack of consultations with the communities involved. For country-to-country migration, governments were described as playing a central role in securing international cooperation and planning, with shared responsibilities including ensuring that new locations are ready and providing support to recent migrants (Negev et al., 2019). As evidenced in Germany, for instance, successful long-term strategies need to include access to healthcare as well as translation services (Hahn et al., 2020). However, many destination countries currently enact only short-term strategies for migrants that may have detrimental effects in the long-term (Oliveira Martins & Strange, 2019).
Where migration is already underway and planned migration has not proven to be possible, the emphasis would then be on early homelessness intervention at known migration stopping points. Data on existing migrant pathways is needed to target appropriate interventions. The role of governments in this situation is to facilitate these assessments, including issues such as language barriers and potential points of connection with local infrastructures (Kaur et al., 2021). The focus is then on engaging migrating populations with appropriate supports (e.g. healthcare, employment opportunities), including the establishment of permanent and temporary housing and supports aimed at homelessness prevention (Mahadevia et al., 2020). Further, host communities should be engaged as well to ensure acceptance of the migrating populations. Governments play a significant role by clearly articulating the policies and supporting both migrating populations and local communities in establishing new ties (Oucho & Williams, 2017).
Of course, providing housing for homeless and precariously housed groups would also significantly improve their resilience to climate shocks. Interventions in this frame would also need to proactively assess and address the violence and trauma that often accompany unplanned migration. While not ideal, prevention efforts at this later stage are still likely to be more impactful than allowing natural migration courses to proceed without intervention, where they often terminate in environments of abject poverty, inadequate shelter, and health crises (e.g. slums, see next section on Exposure).

Government and system leader challenges and solutionsexposure
As with work to address climate-driven migration, exposure to climate change risks has complex policy implications, requiring cross-sectoral responses in systems that are typically fragmented. The policy emphasis in papers and expert commentary about addressing exposure, again, is one of prevention with the objective of preventing further degradation of housing status and reducing weather exposure. Concrete steps include the development of a national housing strategy, enabling and incentivizing cross sectoral and interjurisdictional collaboration, and establishing a prevention-oriented funding strategy with associated targets/indicators and a focus on poverty reduction (Banks et al., 2011;Centers for Disease Control and Prevention, 2020). It is important to recognize the local context for poverty reduction strategies to be comprehensive and effective. For example, in Bangladesh, most such strategies are aimed at supporting rural poor populations, while the urban poor population, which is expected to grow due to climate change, is frequently neglected in prevention-oriented policies and programmes (Banks et al., 2011). Establishing building standards that support risk remediation represent another opportunity, alongside avoiding housing development in areas with climate change-associated exposure (e.g. flood plains) (Mahadevia et al., 2020). As research in India shows, for instance, in heat-prone areas the building standards, especially for low-income populations, need to include heat-reduction techniques and strategies. In moving towards preventative responses, initially there may be the dual requirements of sustaining a temporary shelter plan for crises as more sustained prevention efforts come into effect and reduce the need for crisis services (Biswas, 2020). Important complementary interventions include slum remediation and improvement, improving the quality of temporary street dwellings, warming and cooling centres, and ensuring access to publicly available clean water (e.g. Mahadevia et al., 2020).
Enabling disaster and crisis response strategies is another important domain where substantial populations experience homelessness. Indeed, while prevention is considered a priority, most research regarding existing practices concentrates on crisis responsereflecting it as the main approach in most jurisdictions at the present time. Emphases include explicitly requiring that weather-related disaster response planning embed considerations for homeless populationsincluding both appropriate funding and direction for first respondersand supporting health and social services (Banks et al., 2011;Handmer & Dovers, 2013). Planning in this regard offsets the risks faced by homeless populations in weather disaster zones, addressing important considerations such as knowing where to locate individuals that require evacuation.
Organizations working with homeless populations also need to be funded and otherwise enabled to coordinate outreach to homeless populations during extreme weather events, such as heatwaves or flooding. Outreach activities include providing access to water, other vital supplies, and culturally appropriate education about weather-related dangers, including warning signs and steps to be taken (Every et al., 2019;Nicolay et al., 2016). More broadly, such outreach can be bolstered through the support of public health messaging to destigmatize the issue and by supporting community responses. For example, de Vera (2019) argues that community-oriented grassroots organizations can have a significant impact by helping to direct resources to those most in need.

International engagement
International engagement was an important recurring theme emerging both in reviews and think tanks, with reviews especially highlighting the importance for high-income countries to support low-and middle-income countries. At the country level, a policy imperative is to develop an international response strategy that addresses both the migration and exposure-related pressures described above. This level of response can recognize the prerogative of addressing the health and social impacts of climate change, where low-and middle-income country populations bear disproportionately higher climate change impacts (Ebi & Luchters, 2021). Governments can move to establish processes and mechanisms through which aid and expertise can be quickly put in place to help address extreme weather exposure emergencies, alongside training and support initiatives to proactively build the experience and skill base of key local stakeholders. Fahey et al. (2016) provides an example of such initiatives, where science-based education is contextualized based on the needs identified by the local practitioners. There is also a need for high-income countries to expand on initial steps to provide development assistance to low-income countries (e.g. Collier et al., 2008). Such support needs to include requirements that the most impoverished and stigmatized populations are addressed, whereby funders monitor the progress while at the same time not putting undue pressure to conform to rigid accountability measures (Basak & van der Werf, 2019). As with 'Loss and Damage' finance for vulnerable populations (see COP27), such support is essential in settings where climate impacts are exceeding the adaptive capacities of local populations and ecosystems.

Cross-cutting practices and principles
Several cross-cutting themes emerged in both scoping reviews and expert input through the virtual think tanks in considering approaches to the climate change-homelessness nexus. These themes can be applied to both migration and exposure pressures. Many authors and experts highlight the need for culturally-situated, trauma-informed practices informed by those with lived experience as well as the perspectives and expertise of local service providers (Andrews & Heerde, 2021;Mukherjee & Sanyal, 2021). For instance, in some cases, such as the Tuvaluan migration to New Zealand, climate change might not be the immediate and direct cause of migration in terms of its impacts on livelihoods (Shen & Gemenne, 2011). However, many Tuvaluan migrants indicate feelings of fear and despair associated with climate change as prominent factors. Such an understanding of lived experience should be incorporated to develop informed response approaches in local contexts.
Policymakers have the opportunity, through funding and governance activities, to build in requirements and accountability related to the degree to which responses index to local socioeconomic, cultural, systems, and geographic factors (Hanratty, 2017;Nursey-Bray & Palmer, 2018;Parrish et al., 2020). Other alignments that can be required include the application of a social justice lens, such as the Human Rights-Based Approach or the Greenhouse Development Rights approach (Baer et al., 2022;Calañas & Urbinati, 2020). Climate justice perspectives such as these put the considerations of the most vulnerable populations at the forefront of policy development (Klinsky et al., 2017). Trauma-informed approaches, in turn, can bring specific attention to those most at risk -Indigenous peoples (e.g. women, people with disabilities, children, and others who face intersecting forms of discrimination and marginalization, including identities around race, religion, country of origin, and 2SLGBTQIA + identities) (Abebe, 2014;Alston, 2017;Barbeau, 2017;Bransford & Cole, 2019;Goldsmith et al., 2021;Potter, 2020;Satterthwaite et al., 2020;Vickery, 2018;Villeneuve et al., 2021;Vinyeta et al., 2015). It is important to involve local organizations and networks, which can enhance community-level engagement and help integrate the local knowledge into the policies developed (Khatibi et al., 2021).
The second domain of cross-cutting response themes includes the need to identify and align with existing initiatives and organizations. Such work includes standardizing and aligning terminology, culturally-relevant messaging, and aligning with existing response imperatives and methods while fostering coordinated responses. Such coordination might range from cross-sectoral collaboration in a city through to coordinated international responses, such as through the United Nations programmes (e.g. United Nations Development Programme, 2022). A lack of attention to coordinated responses that are implemented with attention to local cultures and practices are likely to fail. For example, Mostowska (2014) found that social workers engaged with migrants are frustrated with systemic issues, such as strict regulations on who is to receive support and in what form, that interfere with their ability to provide the necessary services to those migrants experiencing homelessness.
The third cross-cutting principle for policymakers involves promoting the collection of high-quality data that can inform a shift to prevention-oriented responses. As discussed earlier, such information can be used to identify points of policy leverage in these complex systems, and provide outcome data to support, monitor, and iterate on responses to improve them over time. For example, homelessness-related data are needed on the types of weather impacts as they relate to specific populations and their socio-geographic contexts (Hackenbruch et al., 2017). These data need to take into account both responding to anticipated gradual, or slow onset, climate change and modelling responses to rapidly emerging crises. Specific topics for study include factors that foster resilience, the nature and outcomes of climate risk mitigation, and adaptation strategies that are inclusive of housing-vulnerable populations, of migration, and of homelessness patterns (ongoing and projected) (Gray & Wise, 2016;Hermans et al., 2020). For example, such studies might consider how precipitation and temperature variation patterns interact with migration and homelessness. Policymakers can promote and fund the development of data infrastructures and risk modelling research that includes homeless populations. This effort will require cooperation and coordination of activities between different levels of government as well as cross-country collaboration (Dale et al., 2020). There will also be a need for the development of cross disciplinary institutes that can address the systemic complexity of the problem and also train the next generation of scientists who will be well-equipped to work in this area (Fahey et al., 2016). An example of such a planning process is the Canadian National Adaptation Strategy (Government of Canada, 2022).

Conclusions and implications
Access to housing represents a fundamental challenge for diverse populations experiencing poverty and climate impacts, and is a key determinant of climate change resilience. Just as intersectionality theory provided a way of organizing and thinking about complexity related to racialized and cultural identities, a lack of adequate housing is a likely point of intersection that can help to inform responses to climate change-related health disparities. The complexity of the climate risk-homeless population nexus and its dynamic systems nature requires evidence and expert-informed response frameworks, such as the one proposed in this paper. Without such guidance there is a risk of generating disjointed, inequitable, and ineffective actions, which could exacerbate global health problems and cultivate stakeholder apathy. Across the many strategies described above, both cross-cutting and those specific to exposure and migration, moving as upstream as possible in the frame of prevention alongside community-level engagement and co-design are fundamental recommendations. While such efforts run counter to predominantly reactive, short-term responses, there are very promising examples to learn from globally. A key point of advocacy will involve creating clear arguments for larger prevention-oriented investments that move away from superficially less expensive and more proactive responses. Indeed, where these problems are already well underway, governments face the challenge of paying for both crisis response and prevention interventions, while waiting for prevention to deliver on the promise of a greatly reduced need for crisis response initiatives. Further, it should be acknowledged that even with greatly increased investment in preventative interventions, the escalating effects of climate change mean that there is still likely to be an increased need for crisis relief services. This is a well-known policy conundrum in the homelessness policy space, and one that has often proven unappealing to governments with mandates that are too brief to demonstrate outcomes from expensive initiatives within the electoral period (Gaetz & Dej, 2017).
The recommendations and considerations presented here must be considered a tentative first draft. While useful information was obtained to articulate problems and design responses, an absence of data in all areas related to the climate-homelessness problem was abundantly clear. Addressing the data problem is essential for validating the points raised here and, most importantly, providing more specific intervention targets and evidence for response strategies. Nonetheless, bringing closer attention to homelessness in climate change responses will likely serve to strengthen the many substantive efforts to mobilize government risk mitigation and adaptation investments. These efforts can show whole-society benefits and work against misperceptions of such work being charity for the poor (i.e. othering; Andreucci & Zografos, 2022). Importantly, bringing attention to the climate change-homelessness nexus can contribute to resilient communities and their ability to more effectively face the climate crisis, by helping to stem the large and growing scale of suffering and mortality faced by unhoused persons.