‘It’s common sense that an individual must eat’: Advocating for food justice with people with psychiatric disabilities through photovoice

Abstract Background People with SMI have often been excluded in advocacy efforts focused on physical health, health care and health and social policy. Objective Following a Photovoice project focused on barriers to healthy eating and physical activity in urban neighbourhoods, participant‐researchers were invited to present their insights in community advocacy settings. The purpose of this study was to explore the feasibility and participant–researchers’ experience of these community advocacy activities. Design We held four focus groups with the eight participant‐researchers after each community advocacy activity to explore their experience with public speaking, presenting their experiences and advocating. Setting and Participants People with serious mental illness who were overweight/obese living in supportive housing. Analysis approach Qualitative analysis of the focus group transcripts, using a modified grounded theory approach followed by structured coding focused on empowerment, participation and non‐discrimination. Results Participant‐researchers gave three oral presentations of their photographs at a variety of community‐based programmes and settings and participated in a rally to advocate for SNAP benefits. Two themes emerged from analysis: (a) Empowerment (the level of choice, influence and control that users of mental health services can exercise over events in their lives) and (b) Barriers to Empowerment (obstacles to participation and well‐being). Conclusions This evaluation strengthens the evidence that it is feasible for participant‐researchers in Photovoice projects to engage in robust advocacy activities, such as presentations and discussions with local policymakers. During focus groups, participant‐researchers demonstrated realistic optimism towards their roles as change agents and influencers in spite of acknowledged systemic barriers.


| INTRODUC TI ON
People with serious mental illness (SMI) experience significantly worse health than the general US population, a situation that is compounded all too often by homelessness and poverty. 1 10 An analysis by Coleman-Jensen & Nord found that a person with a disability would require more than two-anda-half times the income of a person without disabilities to have the same likelihood of food security. 11 Therefore, limitations of the Supplemental Nutrition Assistance Program (SNAP, also known as food stamps) for low-income Americans in combination with limited neighbourhood access to healthy food 12 can make dietary modification difficult to impossible for people with SMI.
Paradoxically, these circumstances can lead to both obesity and food insecurity in the population. 13 The food insecurity-obesity paradox is poignantly reflected in populations experiencing homelessness, where daily uncertainties regarding food access lead to unhealthy eating patterns that can persist even when housing is available. 14 Food security and social justice are intrinsically related and have given rise to the concept known as food justice. Food justice has become a familiar term in the discourse on food insecurity, and it is used in a variety of diverse frameworks with different interpretations. Following the typologies of Moragues Faus, in this manuscript, we are considering food justice from a modified distributive justice frame. By distributive justice, we understand socio-economic structures, including lack of affordable housing, as the basis of food insecurity and call for policy-level change to increase health equity. 15 A human rights framework holds substantial potential to improve the understanding of the complexity and interrelatedness of these multilevel issues in the public and policy sectors by explicitly linking the problem of poor health in people with SMI to the right to health, the right to food and the rights of people with disabilities. [16][17][18][19][20] There are multiple socio-economic and policy food justice factors limiting the right to healthy food in people with SMI including limited knowledge, skills and experience with food purchasing, meal planning and preparation, and food storage, poor-quality food in institutional settings, as well as experiences of homelessness/marginal housing, and living in areas without access to affordable, high-quality grocery stores. Utilizing a human rights framework is an important mechanism in addressing poor health and discrimination in people with SMI because this framework (a) explicitly acknowledges the role of the social, political and economic determinants of health (SDH), (b) provides a mechanism to address these structural barriers and unequal power relationships at the local, state and national policy levels and (c) calls for a participatory response to the problem as an emancipatory practice as well as an ethical imperative. [21][22][23] We acknowledge the constraints of the researcher-driven context of this project, with further discussion below. Nevertheless, we endeavoured to situate this project in a human rights framework and use a limited participatory action research strategy to assure that the population is included in understanding and addressing the causes and consequences of poor health.
An overview of the project framework is shown in Figure 1.
Inclusion, participation and empowerment are just as much a component of a human rights framework as they are elements of mental health recovery. The terms recovery and empowerment have a diverse variety of meanings for different groups, with sometimes marked tension among definitions by people with lived experience, policymakers and academic researchers. 24 As a result, these terms have experienced a great deal of conceptual drift in a wide variety of disciplines, necessitating a high level of specificity when referring to these concepts as well as acknowledgement of the implicit assumptions in the choice of definitions. 25 With these caveats, for this project, we are using the working definition of mental health recovery by the Substance Abuse and Mental Health Service Administration (SAMHSA) as: 'A process of change through which individuals improve their health and wellness, live a self-directed life, and strive to reach their full potential'. 26 Additionally, we are referring to empowerment in a mental health context using the World Health Organization (WHO) definition; 'the level of choice, influence and control that users of mental health services can exercise over events in their lives'. 27 Despite these advances in policy, people with SMI are still practically disempowered at many levels including (a) the societal/structural level because of stigma, housing instability and barriers to access, (b) the organizational level from being poorly K E Y W O R D S health and human rights, health disparities, obesity, serious mental illness informed, treated and consulted and (c) the individual level from internalized stigma, which is linked to poor health. The public health community must find creative, meaningful and effective ways to ensure inclusion and participation. Authentic inclusion leading to active participation is one way that people with SMI can increase choice, influence and control in their lives and communities. Photovoice is one potential mechanism. 28

| Strategies to increase self-advocacy and community participation
Photovoice has been used as an accessible method to amplify the experiences of marginalized populations and bring hidden social justice issues to light using photography. The Photovoice approach was first described in Wang and Burris' seminal work 28 and grounded in Freire's emancipatory educational methods. 29 Photovoice methodology provides cameras to participants to identify community assets and needs, followed by critical dialogue about the meaning of these photographs in the participants' lives. Photovoice has been applied with populations experiencing SMI, 30-32 homelessness 33 and people experiencing food insecurity. 21,[34][35][36] The final goal of Photovoice is social action and political change.
However, Photovoice projects have received some criticism for raising 'false hopes' for policy and social change as well 'a vagueness' in how social action plans have been described and evaluated. 37 Johnston also notes 'concerns for the noticeable lack of documented follow-through actions of attempts at social change and project outcomes'. 37 Sanson, Evans-Agnew and Boutain 38 explored social justice intent in Photovoice projects. Less than half of the studies included in their review described a guiding methodology. All studies reported an increase in awareness of social justice issues in participants and audiences, although few described projects that directly improved or transformed unjust conditions. 38

| Participants/Co-Researchers
Potential participants were recruited from those in the intervention arm of the Peer Group Lifestyle Balance study explained above. 42 To be eligible for the Photovoice advocacy project, PGLB participants had to have completed the 12 weekly core education sessions of the PGLB curriculum at Pathways to Housing PA and Project HOME.
Research coordinators for the PGLB project provided a list of the 21 eligible participants. These participants were briefly informed about the Photovoice project through the PGLB staff and given an informational flyer. All eight participants from Pathways to Housing PA expressed interest and commitment to attending the programme.
Three Project HOME participants initially expressed interest but did not attend the first session despite 2 reminder calls. Therefore, a total of 8 people, all from Pathways to Housing PA, joined the Photovoice project. All eight co-researchers participated in both the photo-taking and discussion section of the project. In the first session of the Photovoice project, participants completed a written consent form and were introduced to the theory of participatory action research.
The study protocol was approved by the institutional review boards at Thomas Jefferson University and the Philadelphia Department of Public Health with a reciprocal IRB authorization agreement from Drexel University. Participants were considered as community coresearchers and will be referred to as such in the rest of the manuscript. All co-researchers asked that their real names be used in the manuscript.

| Photovoice procedure
The Photovoice procedure has been described in detail elsewhere. 43 The following is a summary of the approach. All Photovoice sessions occurred weekly in an agency conference room and were audiotaped and transcribed. The first session was an educational session on Photovoice methodology, photography ethics and camera use. Each co-researcher received a Vivitar ViVi Cam 9112 digital camera and participated in an on-site practice photo-taking session. The staff assisted the co-researchers in downloading their pictures to a computer and projector. Co-researchers practiced explaining their projected practice photographs to the group, using the SHOWeD format. 28 The SHOWeD format asks co-researchers After the first session, co-researchers were given a take-home 'assignment' with the following printed instructions: (a) Think about what in your environment helps or prevents you from losing weight, (b) Take as many pictures as you want, (c) Try to take pictures on both different topics: things that make it easier to lose weight and things that make it harder, and (d) Pick three to five pictures to share with the group at the next session. In weekly group sessions two through seven, we used the same initial assignment questions as a springboard to deeper discussion of identified community issues, strengths and challenges. In later sessions (five, six and seven), participants selected photographs for group presentation, identified common themes and planned for the advocacy phase of the project. and all participated in the discussion with the health department staff, and the university researchers were present to support the community co-researchers; however, the community co-researchers were the primary presenters and discussants.

| Procedure for advocacy presentations
Following the initial presentation to this group, other opportunities rapidly emerged by word of mouth through the local network of agencies working for food justice. Each invitation was reviewed with the community co-researchers and the group collectively considered each opportunity. The co-researchers decided on the topics to cover at three of the four speaking opportunities. For the Tuesdays with Toomey opportunity, the agenda had been set by the larger organization to focus on cuts to SNAP benefits. These activities are summarized below and in Table 1.
The second presentation took place at the Good Food for

| Procedure for post-advocacy session focus groups
We held focus groups with the participants after each of the four community advocacy activities to explore their experience with public speaking, presenting their experiences and advocating. The four focus groups were semi-structured and used the same interview guide at each session. All members of the Photovoice group were invited to these sessions, attendance ranged from three to seven participants. We opened each focus group with a review of the community activity for those that were unable to attend.
The topics for the focus groups related to concepts of human rights, mental health recovery and advocacy in people with SMI such as empowerment, participation, influence and inclusion. 48

| Participant characteristics
Seven of the eight Photovoice project participants participated in at least one advocacy session. One participant was unable to join the advocacy sessions due to complications with a chronic health condition. Of the seven remaining participants, five were male and two were female. Five were Black and two were White.

| Overview of advocacy projects
Between  Table 1.

| Themes
The

| Being heard
Several participants remarked that they brought a new group of voices to the topic of food justice, leading to the subthemes of Being   where we go and how far, you know. But after seeing the pictures that raised more questions'.

| Barriers to empowerment
Participants In addition, participants spoke of a lack of understanding on the part of decision makers, as well as their ineffectiveness, 'It's a lot of different avenues, but it seemed like with all the different avenues combined together it's still don't work. The system is still broke.
That's what I learned'. (Irwin) Participants also focused down to the issue of food injustice in the system, again, as Irwin remarked: 'I'm just seeing that's a part of being in the system. But still I feel as though a lot of it's unnecessary. It's common sense that an individual must eat'.
The system, referring to city, state and national social service policies, as the major multilevel barrier voiced by the participants.
However, additional important barriers to participation framed the discussion. Co-researchers often express general distrust the sys-

| D ISCUSS I ON
The purpose of this analysis was not only to document the feasibility of advocacy activities arising from a Photovoice project, but also to explore the experience of the participants during these activities. This evaluation adds to the evidence that with people with experiences that contribute to food insecurity, such as homelessness and SMI, want to participate in food justice advocacy activities, such as presentations and discussions with local policymakers, and are met with interest and acceptance by a wide range of audiences. 28,32 Data from the focus groups confirm that the co-researchers saw themselves as active participants in society and provided emergent evidence of choice, influence and control in their everyday lives as individuals and in community, as well as barriers to exercising these powers. We reach this conclusion through the diversity and reach of the advocacy activities and through the structured conversations with the participants reflecting strength and insight into the concept of mental health empowerment. While these were only four activities, and far less than those of other larger photovoice programmes such as Witnesses to Hunger, 21 the success of our project further supports the inclusion of people with experiences of SMI and homelessness in advocacy. The focus on systems as problematic and potentially oppressive is also in synch with the Witnesses to Hunger photovoice study, as well as with others that identify serious problems with participant experiences with public assistance programming. 21,36 The human rights framework provides a useful lens to review the advocacy activities themselves as well as the co-researchers experiences with the activities. The Good Food for All Conference and the Tuesdays with Toomey event provided an opportunity for the co-researchers to directly advocate for the right to health, the right to food and the rights of people with disabilities. [17][18][19][20]60 During the Tuesdays with Coalition meeting provided the co-researchers with the opportunity for participation and advocacy through discussions with policymakers and government officials. In follow-up discussions from these sessions, the co-researchers identified experiences and perceptions that contribute to mental health recovery such as being heard, not being judged, mutual learning and influencing policymakers.
Exclusion, discrimination and oppression are commonplace experiences for people with psychiatric disability. [61][62][63] This project provides evidence that a simple process such as giving people a camera to document and discuss their experiences in a supportive environment can provide marginalized groups with a means to raise awareness about hidden barriers to health and well-being as well as amplify their own voice, strength and self-determination and contribute to their recovery. Papoulias highlights the potential of Photovoice in translatability of health-care improvement research and asserts 'photography becomes a means of translating local concerns into a community "voice" which in turn becomes legible to a wider audience of policymakers and clinicians'. 64 The experiences of our group suggest the preliminary emergence of a community voice.
At the same time, we recognize the limitations of directly addressing distributive justice and agree with recent recommendations that Photovoice be regarded as a tool for mental health policy informing rather than policy changing. 37 As Anthony insightfully remarked, 'I don't think we can change their minds but provoke thought'.
Nevertheless, our evaluation confirms participants' felt the project to be a worthwhile experience, as Stephan summarized, 'yea it was educational on both sides. Learning, you know, from each other'.

| Limitations
The co-researchers were limited in their participation a priori as a consequence of situating this project in a larger federally funded randomized controlled trial. 42 As noted in the introduction, this context also constrained the definition and experience of recovery and empowerment to a more conventional and researcher-based agenda.
The small number of participants in our project functioned well in a group setting and could articulate their views and experiences.
In this way, they may not be reflective of the entire cohort of 314 participants in the parent project. Photovoice projects have been criticized for not adequately evaluating their impacts on the policy level. 45 This is understandably attributed to the complex and longterm nature of policy change. Our own project is certainly limited in this aspect as well. However, the Photovoice group was satisfied to see that a few months after they attended the demonstration, Senator. Toomey did agree to meet with constituents at an in-person town hall for the first time in months.
The principal university investigator for this Photovoice project was involved with the larger project on several levels. She is a co-investigator on the parent project and supervises two of the peer specialists who deliver the intervention. This researcher is also the Director of Integrated Care at Pathways to Housing, and she serves as the primary care physician for a majority of co-researchers in the programme, including some of the co-researchers in this Photovoice project. Clearly, there is a significant power differential between the university researcher and her co-researchers, leading to inevitably to limitations in the nature of participant responses based on perceived social desirability. 65 Indeed, this power differential may increase the likelihood of consenting to participant in the project.
Additionally, throughout the project, the university researchers medical and public health orientation and framework contributed to the interpretation of the findings.
With these limitations in mind, explicit efforts were made to equalize the creation of knowledge among the university and community researchers, for example the facilitators present during the group sessions were involved primarily in giving directions, clarifying goals of the sessions and intervening to allow all group members equal time to present their photographs. In later sessions, some co-researchers took the lead in facilitating the discussions and eliciting input from all members. Additionally, the initial manuscript resulting from the Photovoice project was written and reviewed in partnership with the community researchers. 41 Indeed, Kramer-Roy 66 advocates specifically for health-care professionals' involvement in creative and participatory research projects with disabled and marginalized populations. She asserts, 'making the congruence between their profession and these methods explicit has the potential to enhance the emancipatory role of their profession as well as enabling them to use their professional skills to carry out the research well'. 66 On an individual level, the principal university investigator acknowledges the significant benefit she has received from this project in terms of clinical growth and understanding, as well as academic advancement. While the case can be made of mutual learning between academic and community researchers, the power differential inherently limits the evidence for this assertion.

| CON CLUS ION
We have demonstrated additional evidence that creative and participatory methods, such as Photovoice, are an important approach for helping people living with serious psychiatric disabilities and multiple disadvantages to confidently advocate for the right to health and the right to healthy food. However, this is only a small step; larger, coordinated efforts are needed to truly co-create the conditions for a healthy life in the community for all. Important next steps are to authentically include people with lived experience of SMI in all levels of decision making, development and funding in areas such as (a) efforts to assure access to high quality, integrated medical and behavioural health care including support for wellness services, (b) useful research that will change practice and improve outcomes in the health care of people with SMI and (c) inclusive initiatives and policies to increase affordable healthy food availability and safe places for physical activity in low-income neighbourhoods.

ACK N OWLED G EM ENTS
We are thankful for the support provided for this project by the National Institute of Mental Health (1R01MH104574-01 NIMH, PI Cabassa) and the Diabetes Education and Research Center. We are grateful for the opportunity for on-going partnership with the Pathways to Housing participants and staff.

CO N FLI C T O F I NTE R E S T
We confirm that all of the authors report no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available on request due to privacy/ethical restrictions.