Background
People with severe mental illness (SMI; comprising disabling psychoses and affective conditions) are over-represented in homeless populations globally. However, the situation is stark in low- and lower-middle-income countries (LLMICs), where an estimated 28–36 million people are homeless and have SMI (Chamie, Reference Chamie2017; Smartt et al., Reference Smartt, Prince, Frissa, Eaton, Fekadu and Hanlon2019). In systematic reviews from high-income countries, the co-occurrence of SMI and homelessness is associated with numerous adverse outcomes, including premature mortality (Fazel et al., Reference Fazel, Geddes and Kushel2014); infectious disease (chiefly tuberculosis, HIV and hepatitis B)(Beijer et al., Reference Beijer, Wolf and Fazel2012); non-communicable diseases (Scott et al., Reference Scott, Gavin, Egan, Avalos, Dennedy, Bell and Dunne2013); co-morbid alcohol and substance abuse; injuries and accidents (Mackelprang et al., Reference Mackelprang, Graves and Rivara2014); and suicide (Arnautovska et al., Reference Arnautovska, Sveticic and De Leo2014). Women and youth who are homeless are at particular risk of sexual assault and exploitation (Goodman et al., Reference Goodman, Koss and Russo1995). SMI is a major risk factor for chronic homelessness (Fazel et al., Reference Fazel, Geddes and Kushel2014).
Despite strong social capital and protective family structures in many LLMIC settings in Africa, caring for a person with SMI can overwhelm informal support networks, exacerbated by precarious household finances, stigma, diverse understandings about mental illness and limited mental healthcare provision (Read et al., Reference Read, Adiibokah and Nyame2009). Harsh conditions on the streets can also trigger new onset of SMI. High-quality evidence from LLMICs is scarce, but most people who are homeless and have SMI have unmet basic needs for water, clothing and food, and untreated medical problems; 30–40% are affected by physical disabilities (Fekadu et al., Reference Fekadu, Hanlon, Gebre-Eyesus, Agedew, Haddis, Teferra, Gebre-Eyesus, Baheretibeb, Medhin, Shibre, Workneh, Tegegn, Ketema, Timms, Thornicroft and Prince2014; Singh et al., Reference Singh, Shah and Mehta2016; Tripathi et al., Reference Tripathi, Nischal, Dalal, Agarwal, Agarwal, Trivedi, Gupta and Arya2013). Beyond basic needs, there is evidence of high levels of unmet needs for social support, relationships and rehabilitation, and of exposure to unjust imprisonment, exploitation, physical and sexual abuse. Few (10%) have ever received mental healthcare, and almost none receive adequate ongoing care.
The evidence base for intervention in high-income countries takes housing as an essential aspect of service provision for people who have SMI and are homeless, particularly the Housing First model (Tsemberis, Reference Tsemberis2011), which improves a wide variety of health and social outcomes (Woodhall-Melnik and Dunn, Reference Woodhall-Melnik and Dunn2016). Other evidence-based health interventions include tailored primary healthcare programmes, care coordination, assertive community mental health treatment and critical time interventions (Hwang and Burns, Reference Hwang and Burns2014). The applicabilityof such models to LLMICs may be limited, given their reliance on access to housing, specialist multidisciplinary workers and government-funded social welfare and healthcare services. Furthermore, factors causing or maintaining homelessness in people with SMI in LLMICs vary and demand different responses, for example, in relation to mental health stigma and discrimination, forced displacement and poverty. A decolonial approach to research is therefore required – generating evidence that is centred on lived experience and embedded in local contexts – to stimulate incremental, organic changes that will positively impact the lives of people who are homeless with SMI (Abimbola, Reference Abimbola2021).
In our scoping review (Smartt et al., Reference Smartt, Prince, Frissa, Eaton, Fekadu and Hanlon2019), we were unable to identify any rigorous evaluations of programmes for people who are homeless and have SMI in LLMICs. Most identified programmes are run by stand-alone, non-governmental organisations (NGOs), rarely integrated with government provision (Narasimhan et al., Reference Narasimhan, Gopikumar, Jayakumar, Bunders and Regeer2019), thus limiting scalability and sustainability. Crucially, little attention is paid to social needs and the preferences of those with lived experience, who may not wish to return to their communities of origins, due to ‘push’ factors such as shame, a need to escape from conflict, broken relationships, exploitative work, or other forms of abuse and exclusion, or because of ‘pull’ factors related to perceived opportunities and freedoms in cities (Haile et al., Reference Haile, Umer, Fanta, Birhanu, Fejo, Tilahun, Derajew, Tadesse, Zienawi, Chaka and Damene2020).
There remains a crucial gap in evidence and global recommendations of care for people with SMI who are homeless. Any effective response will require coordination and close collaboration of specialist and community-based care across sectors, particularly when family supports are overwhelmed or absent, as well as the meaningful involvement of people with lived experience. This evidence gap indicates a pressing need for an Africa-led partnership of researchers and key stakeholders to conduct high-quality research in this area.
The aim of HOPE (National Institute for Health and Care Research Global Health Research Group on Homelessness and Mental Health in Africa) is to develop and evaluate approaches to addressing the unmet needs of people who are homeless and have SMI living in LLMICs in ways that are rights-based, contextually grounded, scalable and sustainable.
Objectives:
1) Establish a partnership of researchers, implementers, policymakers and people with lived experience.
2) Synthesise global evidence and obtain expert consensus on priority actions and approaches.
3) Work in distinctive settings in Ethiopia (capital city), Ghana (regional capital) and Kenya (capital city and rural county) to:
a. Identify the priority needs and valued outcomes of those with lived experience, and opportunities and challenges for intervention.
b. Integrate global and local evidence to select, co-produce and pilot interventions that target priority needs, in order to investigate acceptability and feasibility.
c. Evaluate the impact of interventions on the human rights and outcomes valued by people with lived experience and generate evidence on intervention costs, implementation processes and outcomes.
d. Pioneer the development of methods and ethical frameworks for future research and interventions.
4) Impact global policy and practice through translation of evidence into a ‘how-to’ guide to adapt and implement programmes in diverse LLMIC contexts.
5) Build sustainable capacity across partners that support South–South and South–North exchange of expertise and mutual learning, and develops individuals, teams, organisations and systems.
Methods
See Fig. 1 for an overview of HOPE. HOPE is structured into work packages (WP): WP1 (project coordination), WP2 (formative phase), WP3 (intervention co-production/piloting), WP4 (implementation/evaluation), WP5 (capacity-strengthening) and WP6 (lived experience, community engagement and research uptake). In HOPE, we have prioritised the involvement of people with lived experience in all aspects of the research and delivery of interventions.
Figure 1. Overview of the HOPE project
Countries and settings
See Table 1 for characteristics of the countries and sites. These sites have been selected because (1) there is an evident need to improve support for people who are homeless and have SMI, (2) they are distinctive contexts in East and West Africa across low-income and lower-middle income countries; and (3) there are opportunities to build on existing programmes and political will. Homelessness has increased across the study countries, with accompanying challenges of poor health outcomes (Elsey et al., Reference Elsey, Agyepong, Huque, Quayyem, Baral, Ebenso, Kharel, Shawon, Onwujekwe, Uzochukwu, Nonvignon, Aryeetey, Kane, Ensor and Mirzoev2019). In each study setting, social exclusion and stigmatisation of people who are homeless is common (UNICEF, MOLSA and Development Pathways, 2019), which intersects with stigma and discrimination experienced by persons with SMI (Forthal et al., Reference Forthal, Fekadu, Medhin, Selamu, Thornicroft and Hanlon2019; Mutiso et al., Reference Mutiso, Pike, Musyimi, Gitonga, Tele, Rebello, Thornicroft and Ndetei2019; Read et al., Reference Read, Adiibokah and Nyame2009). Although all three countries are expanding community-based mental healthcare, these services are not tailored to the complex needs of people who are homeless and are inaccessible due to the reliance on families to bring a person to services and often meet the costs of treatment. Faith-based, civil society and NGOs contribute to crisis responses to basic needs for sub-groups of homeless populations but such efforts are fragmented (UNICEF, MOLSA and Development Pathways, 2019).
Table 1. Characteristics of countries and project sites in HOPE
Conceptualising homelessness
Our starting point for defining homelessness is: spending the night unsheltered or in other places not intended for habitation. We will exclude people who spend their days on the streets, for example, to beg, but who have stable night-time accommodation. However, the concept of homelessness is complex and locally nuanced, so the target population will be operationalised for each study site in the formative phase.
Theoretical framework
The research work in HOPE will follow the Medical Research Council/National Institute for Health and Care Research (MRC/NIHR) framework for developing and evaluating complex interventions (Skivington et al., Reference Skivington, Matthews, Simpson, Craig, Baird, Blazeby, Boyd, Craig, French, McIntosh, Petticrew, Rycroft-Malone, White and Moore2021). We will use ToC as a participatory tool to inform each aspect of the MRC/NIHR approach (De Silva et al., Reference De Silva, Breuer, Lee, Asher, Chowdhary, Lund and Patel2014). Applying realist approaches, we will develop programme theories to inform evaluation. Unmet needs will be framed within a socio-ecological model that seeks to understand and propose interventions to address the individual, family, societal and political-level barriers to social inclusion.
WP2: Formative phase
An extensive formative phase will comprise syntheses of global evidence and best practice alongside primary data collection in each setting: context mapping, ethnography and a cross-sectional study.
Systematic review
Building on our previous scoping review (Smartt et al., Reference Smartt, Prince, Frissa, Eaton, Fekadu and Hanlon2019), we will conduct a systematic review, including grey literature, to identify interventions for people who are homeless with SMI in LLMICs (Smartt et al., Reference Smartt, McPhail, Hanlon, Musyimi and Agorinya2023). We will carry out a narrative synthesis of types of interventions, implementation strategies and evidence of impact.
Delphi consensus exercise
A Delphi consensus exercise will help identify global perspectives on best practices and priorities for interventions. We will invite people with experience of implementing relevant programmes in different regions in LLMICs, identified through the grey literature review and including French-, Spanish- and Portuguese-speaking implementers, and representatives of mental health service user associations, disability rights organisations and the World Health Organization (WHO). In the first round, participants identify interventions/components of programmes that they consider important. These will be consolidated and augmented by emerging findings from our review. In the second round, participants will rank each identified component/strategy based on importance and feasibility in LLMICs. Finally, the anonymised aggregated responses will be fed back to the participants, with further ranking to identify essential, desirable and optional elements of intervention programmes, as well as any ethical concerns.
Context mapping
We will conduct a documentary analysis of relevant policies, laws and plans relating to people who are homeless in our settings, inclusive of documents relating to public health, social care, legal decrees and other municipal documents. We will consult with community leaders and other key stakeholders in administrative positions, NGOs and religious and traditional healing sites. We will establish where people who are homeless and have SMI can be reached, and document the roles of different agencies and existing initiatives. Findings from the analysis will be synthesised using a matrix, informed by an implementation framework to map context (Pfadenhauer et al., Reference Pfadenhauer, Gerhardus, Mozygemba, Lysdahl, Booth, Hofmann, Wahlster, Polus, Burns, Brereton and Rehfuess2017), facilitating cross-country comparisons.
Primary data collection for formative studies
Primary data collection comprises a focused ethnography (Sangaramoorthy and Kroeger, Reference Sangaramoorthy and Kroeger2020) and a cross-sectional study to understand experiences of homelessness and SMI, population burden, level of unmet needs, and sources of support, preferences, opportunities and potential barriers to intervention. We will carry out participant observation in locations where people are likely to sleep, visit or spend the day. Observations and linked interviews will focus on people who are homeless and have SMI, other members of the community, and their interactions. Interviews and focus group discussions will be conducted with key informants. The specific aims and methods are detailed in Table 2.
Table 2. Overview of methods for formative phase studies and pilot study
a Previously adapted for homeless populations in Ethiopia [Fekadu et al., Reference Fekadu, Hanlon, Gebre-Eyesus, Agedew, Haddis, Teferra, Gebre-Eyesus, Baheretibeb, Medhin, Shibre, Workneh, Tegegn, Ketema, Timms, Thornicroft and Prince2014].
b Previously adapted for Ethiopia and Kenya.
Ethics
Key ethical challenges will now be highlighted: consent, responding to basic needs and safeguarding.
Consent to participate
In the ethnography, gatekeeper permission will be obtained. Aligned with accepted ethics of ethnographic practice, we will not seek formal consent from individuals who are being observed. The purpose of participant observation is to access an understanding of the day-to-day lives of people who are homeless and have SMI which might not be obtained via other methods, such as formal qualitative interviews. The process of obtaining informed consent can disrupt participants’ natural behaviours, actions and responses, including their relationship with the researcher-observer, risking the valuable insights ethnography might add. This was a point of contention in the HOPE team, with concerns raised by the Lived Experience Advisory Group (LEAG; below) about the dignity and autonomy of the person. To address these concerns, we will seek to build rapport with people who are homeless and have SMI over time through regular visits by the researchers and informal conversation, an approach used to engage homeless people with SMI in similar settings (Eaton et al., Reference Eaton, Des Roches, Nwaubani and Winters2015). This will facilitate respectful engagement with the person, allowing them to gain an understanding of the researcher’s role and develop trust. The researchers will be carefully trained and supervised to ensure they respect any desire not to be observed and cease observation if a person communicates discomfort (verbally or non-verbally). For in-depth interviews, informed consent will be obtained.
In the cross-sectional study, we will include people who lack mental capacity to provide informed consent as participants in the study with appropriate safeguards. We argue that the ethical principle of justice is upheld because the study enables us to obtain an understanding of the needs of the most vulnerable people who are homeless and have SMI and design interventions that best meet those needs. The United Nations Convention on the Rights of Persons with Disabilities implies a presumption of equal treatment with others who may wish to participate in research (i.e. having equal legal capacity), with the proviso that the prospective participant’s will and preference is always actively recognised and respected. We will not include anyone in the study if they indicate refusal or if there is any other indication that their will and preference would be not to participate. To identify evidence of the latter, we will seek to speak to a trusted person who is nominated by the person who is homeless and has SMI. That person could be anyone who credibly supports the individual. In the absence of such a person, a mental health/disability advocate will communicate with the person, seek to identify their will and preference and provide permission for the person’s participation. Mental capacity will be assessed by a mental health professional, using a standardised approach (Hanlon et al., Reference Hanlon, Alem, Medhin, Shibre, Ejigu, Negussie, Dewey, Wissow, Prince, Susser, Lund and Fekadu2016). As a further safeguard for study participants who lack mental capacity, we will make systematic efforts to reassess capacity after 2 weeks, during which time we will support them to access mental health services. If the person has regained capacity, they will be invited to provide informed consent to participate and, if they decline, they will be withdrawn from the study.
Responding to basic needs
During the ethnography fieldwork, researchers may offer refreshments to people who are homeless during informal interactions as an act of reciprocity and following altruistic norms. For the cross-sectional survey, refreshments will be made available for all people who are homeless in the vicinity, regardless of study participation, so as not to provide undue incentives or coercion to participate. People with emergency medical needs will be supported to access care, working with local healthcare services and community health workers. When safeguarding needs surface, we will respond as detailed below. Participants will be informed about resources in the vicinity where they may access support, e.g. in relation to shelter, feeding programmes, local health services, and support for vulnerable women. For people with SMI, concerted efforts will be made to link them to mental health services (according to their preference). Involuntary mental healthcare will be utilised only in line with the country’s legislation (Ghana, Kenya) or, where no legislation exists (Ethiopia), if the person’s mental state is assessed as posing an imminent risk of harm to themselves or others. In the previous study from Ethiopia (Fekadu et al., Reference Fekadu, Hanlon, Gebre-Eyesus, Agedew, Haddis, Teferra, Gebre-Eyesus, Baheretibeb, Medhin, Shibre, Workneh, Tegegn, Ketema, Timms, Thornicroft and Prince2014), only 2 out of 89 people who were homeless and had SMI received involuntary psychiatric treatment.
Safeguarding
Standard operating procedures (see supplementary file 2) have been developed to support researcher responses to safeguarding concerns arising from (1) chaining, restraint or seclusion; (2) sexual abuse, exploitation or harassment; (3) physical abuse or serious physical health concerns; (4) suicidal behaviour; (5) violent or aggressive behaviour; (6) human rights abuses by professionals or community members; and (7) child trafficking or other child protection needs. Community advisory boards, including persons with lived experience, informed the development of localised protocols.
WP 3: Intervention co-development and piloting
Participatory ToC workshops and intervention selection
We will conduct ToC workshops at the beginning and end of the formative phase. Each ToC workshop will last 0.5–1 day. Participants will include stakeholders identified through the formative work. Workshops will start by developing agreement on the desired long-term outcomes and impacts of the programme, then map out the interventions needed to achieve intermediate outcomes, identifying underlying assumptions and barriers, potential implementation strategies, required resources and inputs, and indicators of success. Global and country-specific evidence and experience arising from the formative phase findings will be integrated into the initial programme theory derived from the ToC workshops.
Interventions with evidence of effectiveness in low- or middle-income countries identified through our reviews and Delphi exercise will be examined for relevance to preferences, unmet needs and feasibility of adaptation (based on formative studies). We will identify where new intervention development is needed (Box 1). This process will initially be undertaken within each country by a small (n = 5–8) working group comprising people with lived experience of SMI, potential implementers and the HOPE country project teams. HOPE consortium members will then review the proposed interventions/approaches and suggest further ways to tailor the interventions to the local context. We will give particular focus to rights-based approaches, potential scalability and sustainability, and identifying the role of peer support.
Co-development of interventions and implementation strategies
Working groups of people with lived experience, implementers and the research team will meet several times to co-produce the interventions, implementation strategies and fidelity checklists. We will describe the interventions according to recommended guidance (Hoffmann et al., Reference Hoffmann, Glasziou, Boutron, Milne, Perera, Moher, Altman, Barbour, Macdonald, Johnston, Lamb, Dixon-Woods, McCulloch, Wyatt, Chan and Michie2014) and catalogue the implementation strategies according to existing taxonomies (Powell et al., Reference Powell, Waltz, Chinman, Damschroder, Smith, Matthieu, Proctor and Kirchner2015) to enhance testing in new settings. The FRAME (Framework for Reporting Adaptations and Modifications Expanded) implementation tool (Wiltsey Stirman et al., Reference Wiltsey Stirman, Baumann and Miller2019) will be used to document intervention adaptations and their justification. We anticipate that the resulting intervention packages for the three study settings will have elements that are similar, although addressed in differing ways depending on the context (e.g. accessing physical and mental healthcare, addressing basic needs, common elements of needs assessment and planning, peer support, individual engagement, addressing social exclusion) as well as elements that may be site-specific (e.g. addressing substance misuse, family interventions, housing). Candidate personnel for delivering and supervising the intervention/s will be existing human resources, such as community health workers, peers (people with lived experience), the NGO sector, social workers, primary healthcare staff and psychiatric nurses.
Realist methods
We will conduct a realist synthesis to inform our understanding of what works, for whom, in what circumstances and why (Wong et al., Reference Wong, Westhorp, Greenhalgh, Manzano, Jagosh and Greenhalgh2017). Drawing on cross-country analyses of findings from the formative phase, we will map our initial ToC-based programme theories onto mid-range programme theories and, in collaboration with key stakeholders, develop hypotheses about the important ways that interventions can bring about change in different contexts, i.e. ‘context–mechanism–outcome configurations’. These will then be explored in the pilot phase and used to develop a realist evaluation framework for larger-scale implementation.
Box 1. Candidate interventions for adaptation
HOPE co-investigators have been involved in the development, implementation and evaluation of several interventions for people with SMI in the study countries which have potential relevance to homeless populations. Materials (manuals, training materials) and implementation strategies (e.g. for competency-based training and supervision (Laura Asher et al. Reference Asher, Birhane, Teferra, Milkias, Worku, Habtamu, Kohrt and Hanlon2021)) are available for the following: community-based rehabilitation for people with SMI delivered by lay workers (Asher et al. Reference Asher, Birhane, Weiss, Medhin, Selamu, Patel, De Silva, Hanlon and Fekadu2022; Asher et al. Reference Asher, Hanlon, Birhane, Habtamu, Eaton, Weiss, Patel, Fekadu and De Silva2018), social contact-based stigma reduction interventions (Kohrt et al. Reference Kohrt, Jordans, Turner, Rai, Gurung, Dhakal, Bhardwaj, Lamichhane, Singla, Lund, Patel, Luitel and Sikkema2021), awareness-raising and community outreach for people who are homeless and have SMI, engagement with traditional and religious healers (L. Asher et al. Reference Asher, Birhanu, Baheretibeb and Fekadu2021; Yaro et al. Reference Yaro, Asampong, Tabong, Anaba, Azuure, Dokurugu and Nantogmah2020), task-shared models of integrated mental health care in primary health care (Hanlon et al. Reference Hanlon, Medhin, Dewey, Prince, Assefa, Shibre, Ejigu, Negussie, Timothewos, Schneider, Thornicroft, Wissow, Susser, Lund, Fekadu and Alem2022; Hanlon et al. Reference Hanlon, Medhin, Selamu, Birhane, Dewey, Tirfessa, Garman, Asher, Thornicroft, Patel, Lund, Prince and Fekadu2019; Mutiso et al., Reference Mutiso, Pike, Musyimi, Gitonga, Tele, Rebello, Thornicroft and Ndetei2019), brief psychological interventions (Bitew et al. Reference Bitew, Keynejad, Myers, Honikman, Medhin, Girma, Howard, Sorsdahl and Hanlon2021), brief interventions for substance use disorders (Clair et al. Reference Clair, Rossa-Roccor, Mokaya, Mutiso, Musau, Tele, Ndetei and Frank2019; Hailemariam et al. Reference Hailemariam, Fekadu, Selamu, Medhin, Prince and Hanlon2016; Harder et al. Reference Harder, Musau, Musyimi, Ndetei and Mutiso2020) and gender-based violence (Keynejad et al. Reference Keynejad, Bitew, Sorsdahl, Myers, Honikman, Medhin, Deyessa, Sevdalis, Tol, Howard and Hanlon2020), family interventions (Asher et al. Reference Asher, Hanlon, Birhane, Habtamu, Eaton, Weiss, Patel, Fekadu and De Silva2018; Casey et al. Reference Casey, Krupa, Lysaght, Price, Canes, Ngan, Little, Jayaraman, Ruhara, Mutiso and Ndetei2018), livelihoods interventions (BasicNeeds-Ghana; Lund et al. Reference Lund, Waruguru, Kingori, Kippen-Wood, Breuer, Mannarath and Raja2013), addressing basic needs (BasicNeeds-Ghana), self-help groups and peer support (Cohen et al. Reference Cohen, Raja, Underhill, Yaro, Dokurugu, De Silva and Patel2012; Puschner et al. Reference Puschner, Repper, Mahlke, Nixdorf, Basangwa, Nakku, Ryan, Baillie, Shamba, Ramesh, Moran, Lachmann, Kalha, Pathare, Müller-Stierlin and Slade2019). In the ongoing SCOPE project (Ethiopia), we are developing a toolkit for early identification of people who are homeless and have SMI (Hanlon et al. 2023).
Pilot study
The resulting intervention package will be piloted in each country in a circumscribed geographical area linked to a primary healthcare centre or other relevant facility. We will conduct a single-arm feasibility study over 3 months, focusing on initial assessment and care planning, engagement and early interventions (see Table 2). Based on the ToC and realist synthesis, we will identify process indicators (spanning quantitative and qualitative data) for each intermediate outcome and explore potential mechanisms through which context influences outcomes. We will examine the feasibility and appropriateness of using routinely collected service indicators and identify where project data will be needed as the basis for monitoring and evaluation, and quality improvement. We will initially run training for a small group of implementers. Regular meetings (every 1–2 weeks) with implementers and people with lived experience, supported by the project team, will use plan-do-study-act (PDSA) cycles (Taylor et al., Reference Taylor, McNicholas, Nicolay, Darzi, Bell and Reed2014) to identify and address emerging challenges, allowing iterative improvement of the intervention and approaches. The structures and supports that underpin this co-production approach are the focus of WP5 and WP6.
Costing interventions
Proformas will be designed and tested to facilitate key decisions about our costing approach; for example, determining which intervention components can be reliably estimated at an individual (bottom-up) level and which will require an aggregate (top-down) approach. In semi-structured interviews with implementers, we will gain views on the extent to which additional tasks related to the intervention replace or add to existing activities, to identify possible resource allocation for services or organisational implications that need to be scrutinised in the larger-scale implementation and evaluation study. A service use data collection form will be adapted and piloted.
WP4: Implementation and evaluation
Based on pilot study findings, the country working groups will finalise interventions and implementation strategies. We will then carry out phased implementation and theory-driven evaluation of the intervention package at a larger scale in the formative study sites. A study protocol will be reported separately.
WP5: Capacity-strengthening
We are not aware of a partnership focused on addressing needs of people who are homeless and have SMI in Africa. Tackling this problem mandates use of research methodologies that have not been widely used in global mental health, including ethnography and participatory action research. To be successful, the partnership also needs to expand research capabilities to peer researchers and implementers, build researcher capacity for rights-based work and strengthen capacity for multi-sectoral working that is respectful and inclusive. We will deliver a programme of capacity-strengthening that builds on identified capabilities, is tailored to the different needs across countries and institutions, links directly to the work in HOPE, is designed for sustainability and is evaluated for impact (Hanlon et al., Reference Hanlon, Semrau, Alem, Abayneh, Abdulmalik, Docrat, Evans-Lacko, Gureje, Jordans, Lempp, Mugisha, Petersen, Shidhaye and Thornicroft2018) (see Table 3). Guided by the ESSENCE framework (TDR/World Health Organization, 2016), our capacity-strengthening efforts will be evaluated for impact by collecting the following data: process indicators (participants in training courses and webinars: gender, age and background of participants; uptake and completion of training), satisfaction (anonymous online survey), impact (publications, percentage of first authors from LMICs, with lived experience and/or who are female, grant applications and conference presentations).
Table 3. HOPE capacity-strengthening activities
WP6: Lived experience, community engagement and research uptake
Mental health service user and advocacy organisations, and representatives from Ministries of Health in each partner country, contributed to the development of the HOPE proposal and were named collaborators on the grant application, in recognition of the fundamental importance of their active engagement in HOPE. A LEAG has been established and meets monthly to provide input into evolving study protocols and standard operating procedures, especially in relation to ethically sensitive aspects of the study. The LEAG feeds back to the HOPE steering committee and investigator group monthly and has directly influenced methods for ethnography and the survey. We are seeking to expand the LEAG to include people who have lived experience of homelessness as well as SMI, supported by empowerment methods such as PhotoVoice (Rai et al., Reference Rai, Gurung and Kohrt2023).
Policymaker collaborators and World Health Organization representatives also join steering committee meetings on a quarterly basis and attend the annual meeting, with interim engagement within countries. In these meetings, priority is given to understanding the country level and global policy context and opportunities for HOPE to achieve impact, as well as keeping colleagues informed about, and engaged with, developing plans and emerging findings. Local community and stakeholder engagement is achieved through multisectoral community advisory boards and national level country management groups. Ongoing engagement and targeted messaging are key aspects of our research uptake strategy, focused on achieving high levels of ownership and the potential for sustainable impacts. We will monitor policy uptake of HOPE outputs.
Discussion
Potential challenges
The HOPE programme is ambitious and faces important structural barriers to its success. These include stigma and the limited availability of social welfare provision and services tailored to the needs of people with SMI who are homeless. For sustainability we are focusing on mobilising existing resources and ensuring fair access, but we will also use findings to advocate for greater prioritisation of this group in resource allocation.
Practice implications
Our inclusive approach to partnership combined with our focus on the preferences and priorities of people with lived experience of SMI and/or homelessness marks an important change to existing practice. The resulting intervention packages will seek to coordinate services around individuals rather than vice versa, with efforts to avoid vertical programming. The importance of multi- and inter-sectoral programming to support recovery of people with SMI has been highlighted (van Rensburg and Brooke-Sumner, Reference Van Rensburg and Brooke-Sumner2023). Lessons learned from HOPE will have relevance for other populations of people with SMI who have complex needs and WHO’s renewed efforts to promote de-institutionalisation of mental health care (WHO, 2022).
Policy implications
The HOPE NIHR Global Health Research Group will speak directly to the United Nations Sustainable Development Goal imperative that ‘no one should be left behind’ in development efforts (Patel et al., Reference Patel, Saxena, Lund, Thornicroft, Baingana, Bolton, Chisholm, Collins, Cooper, Eaton, Herrman, Herzallah, Huang, Jordans, Kleinman, Medina-Mora, Morgan, Niaz, Omigbodun, Prince, Rahman, Saraceno, Sarkar, De Silva, Singh, Stein, Sunkel and Unützer2018), seeking to address the systematic exclusion of people who are homeless and have SMI from key services and societal opportunities. While the imperative for social inclusion of people who are homeless, including those with SMI, is emphasised by the United Nations (United Nations Secretary General, 2023), governments and implementing organisations are hampered by a lack of compelling and fit-for-purpose evidence; and there is a particular evidence gap in LLMICs. In HOPE, we will generate evidence on rights-based, contextually relevant, effective and scalable interventions for people who are homeless and have SMI.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S2045796025000186.
Availability of data and materials
The paper describes planned work and does not make use of any data.
Acknowledgements
CH and AA receive funding support from Wellcome Trust through grant 222154/Z20/Z. CH also receives support from WT grant 223615/Z/21/Z. AA also receives funding from Developing Excellence in Leadership, Training and Science Initiative II (DELTAS Africa II) Grant a joint funding support from Wellcome Trust and UK’s Foreign, Commonwealth and Development Office (FCDO).
Financial support
The research reported in this publication is funded by the National Institute for Health and Social Care Research (NIHR) through the Global Health Research Group on Homelessness and Mental Health in Africa (NIHR134325), using UK aid from the UK Government. The views expressed in this publication are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care or Public Health England.
Competing interests
None.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
Open access
