From Procurement to Consumption: A Model to Understand Nutrition Policy Implementation in Permanent Supportive Housing

Background: Food insecurity has become an increasingly complex public health issue across the United States, particularly among various people battling with current or previous homelessness. This project sought to understand the food system in permanent supportive housing sites (PSH) that serve formerly homeless individuals and to explore the use of nutrition standards, specifically the Food Service Guidelines for Federal Facilities (FSGFFs), in this context. Methods: Participants were members of the administrative staff involved in the food procurement process, food preparation, administrative tasks, and daily operations in a small-intensive program managed by a local nonprofit agency that serves 12 adults over the age of 18 who experience chronic homelessness and persistent mental illness, or substance use disorders, and a second PSH site that helps 41 low-income adults with health conditions experiencing homelessness. The PSH Inquiry Tool (PSH-IT) was developed to better understand the business operations at each site, and the PSH Audit (PSH-A) was created to assess the applicability of FSGFF at each site. Results: Findings suggest that funding mechanisms, staff training, staff capacity, and access to nutrition education were critical barriers to the successful development and implementation of nutrition standards in PSH sites. Furthermore, findings suggest that adaptations to FSGFFs are required before implementation at PSH sites. Conclusion: This report advocates for increased involvement of community stakeholders to support nutrition policy development and implementation, a nutrition policy that impacts all levels of the food system from procurement to consumption, and local, state, or federal policy changes to support improved nutrition in PSH.


INTRODUCTION
principles being that safe and affordable housing is the primary solution to homelessness. 6,7 Permanent supportive housing (PSH) is an intervention that incorporates subsidized housing and voluntary support services for people who have experienced chronic homelessness. PSH beds increased by 20% over the past 5 years across jurisdictions, and congress has invested billions in PSH programs. 8 Despite being a proven solution to chronic homelessness, PSH may not solve the health-related consequences of chronic homelessness. 9 Data have shown that individuals entering PSH may suffer According to the United States Department of Agriculture (USDA), in 2020, over 38 million people in the United States were food insecure. 1 Homelessness compounds the issue of food insecurity, especially for the chronically homeless. [2][3][4] Chronic homelessness is a state of homelessness for at least 1 year or repeated episodes of homelessness in an individual with a mental illness, substance use disorder, or physical disability. Chronic homelessness is associated with several health conditions and premature mortality. [2][3][4][5] The Housing First approach is based on several principles, one of those from a lower baseline level of health than the general population and thus may need additional health care coordination and health services. 3 In addition, studies have demonstrated that individuals in PSH maintain higher rates of food insecurity than the general population and that placement in PSH may not significantly improve health outcomes. 2,4 Moreover, data suggest that existing food procurement and donation networks may be insufficient to meet all of the nutritional needs of those living in PSH. 2 Therefore, understanding the policy, programming, and operational barriers in PSH may improve its impact on health outcomes and its utility as a housing solution. One such barrier to the success of PSH is the current state of nutrition guidelines for sites that serve persons who are food insecure.
Current nutrition guidelines and policies for sites that serve persons experiencing food insecurity are limited in their scope. Research has shown that food pantries, soup kitchens, and food banks play a pivotal role in providing food for those experiencing food insecurity. Therefore, these may be targets for improving access to more nutritious food among persons with food insecurity living in PSH. [10][11][12][13] A recent article identified 42 federal policies on food bank donations; nevertheless, no guidelines addressed the nutritional quality standards for donated foods. 2 Yet, procurement of food is only one component of accessing more nutritious foods. As such, additional policies related to food banks may be needed. 10 Two web-based resources offer information for organizations serving food insecure individuals. Still, when preparing this manuscript (July 2021), the sites had outdated links, a lack of guidance on how to implement recommendations, and no opportunities for technical assistance (TA). 14,15 Technical assistance is a strategy used to build an organization's capacity by providing targeted support to an organization with a need or problem. In addition, a limitation of many of the available resources for PSH sites is that they are not endorsed by an existing agency that can provide TA or enforce standards. Furthermore, many resources cover only one aspect of the procurement-to-consumption process (eg, procuring, preparing, ensuring food safety, marketing foods to clients). One central document that guides nutrition policies and best practices is the Food Service Guidelines for Federal Facilities (FSGFFs) tool created by the US Department of Health and Human Services.
The FSGFFs are specific standards for food and nutrition, facility efficiency, environmental support, community development, food safety, and behavioral design in worksites, organizations, or programs. 16 The goal of FSGFFs is to create healthy food environments such as cafeterias, cafes, grills, snack bars, concession stands, and vending machines in areas that serve large populations of people. The standards included in the FSGFFS were determined by the Food Service Guidelines Federal Workgroup, which comprised 60 representatives from 9 federal departments and agencies. The FSGFFs identify 2 levels of implementation: standard and innovative. These levels are supported by the literature to be advantageous to health and the environment. The standard level is considered widely achievable within food service and is expected, whereas the innovative level is regarded as exceptional performance and is encouraged.
We identified PHS facilities as sites that could benefit from implementing nutrition guidelines. Through discussions with community partners, it was determined that nutrition intervention in areas that serve people who have experienced homelessness is limited. Our goal was to explore the food system, conditions, and capacity in PSH and understand the applicability of FSGFFs in this context. The food system was defined as the policies, procurement, and funding that influence nutrition practices in PSH sites in Cuyahoga County, Ohio.

Setting
This study investigated 2 permanent supportive housing sites in Cuyahoga County, Ohio. Site 1, a small, intensive program managed by a local nonprofit agency, serves 12 adults age 18 or over who experience chronic homelessness and persistent mental illness, or substance use disorders. Site 2, a PSH program, serves 41 lowincome adults with health conditions experiencing homelessness. Neither site serves children, families, or individuals who are pregnant.

Design
This was a cross-sectional case study of 2 permanent supportive housing sites in Cleveland, Ohio. Sites were sampled based on their connection to existing agencies that serve people experiencing homelessness and identified need by stakeholders.

Participants (inclusion, exclusion criteria), Recruitment Process
Key participants were members of the staff involved in the food procurement process, food preparation, administration, and daily operations of the sites.

Measures/Outcomes
The PSH Inquiry Tool (PSH-IT) (Appendix I) 17 informed by Koh et al, was developed to explore business operations, staff perspectives on food access, and site needs at 2 permanent supportive housing sites. The PSH Audit (PSH-A) (Appendix II) was developed based on the details of FSGFFs such as prepared foods, packaged snacks, beverages, food safety, and behavioral design. Both sites completed the PSH-IT; site 2 also completed the PSH-A. Site 1 could not complete the PSH-A due to internal staff capacity constraints. Dietary Guidelines for Americans 2015-2020 18 was used as the gold standard for identifying and considering a specific food item to be "healthy."

Procedures
The Case Western Reserve University institutional review board granted human participant compliance approval for this research. Data for the PSH-IT were collected in an interview format where participants were asked to type their responses into the electronic questionnaire while the examiner read the questions aloud. The ojph.org Ohio Public Health Association ojph.org Ohio Public Health Association 5 PSH-A was completed after the PSH-IT on a different day and conducted through discussions with staff and clients, observation of facility spaces, and interviews with staff participants. The researchers selected site 1 and site 2 based on stakeholder interviews and expressed needs by each site.

Statistical Analysis
The case study methodology was used to summarize the data gathered from the 2 sites to understand the food system indepth. 19 The case study approach is an empirical method used to assess an event or phenomena within its natural context. 19 Descriptive statistical analysis was used to explore the data.

RESULTS
The PSH-IT provided information about site operations, staff perspective on food accessibility, and site needs. Data from the PSH-IT showed that neither site was familiar with Federal FSGFFs or Dietary Guidelines for Americans 2015-2020. Moreover, funding varied by location, but client contributions, grants, and private donations were primary sources of support. The local food bank was a primary source of food, followed by local grocery stores, donations of surplus foods from local restaurants, and bulk food suppliers. Both sites moderately agreed that they could request and receive whole grains and low-sugar beverages in the last 6 months, while site 2 strongly agreed that they could ask and receive fresh fruits and vegetables, lean protein, and low-fat dairy products (Table 1). Both sites were moderately satisfied with purchasing healthy food within their budget and accessing healthy donated food from food banks (Table 1). Data showed that both sites considered client dietary restrictions when obtaining food; other considerations included food bank inventory, client desires, and funding limits (Appendix I, Item #22).
Both sites shared that energy-dense and nonperishable food items were more accessible than fresh fruits and vegetables and lean meats and identified a lack of funding as their most significant barrier to accessing healthy foods. Other key barriers to preparing healthy foods were staff training, staff capacity, and staff and cli-ent food education. Lastly, neither site had nutrition standards, mechanisms for determining the nutritional value of meals served, or policies to ensure donations and procurement of healthier food options for residents.
The PSH-A provided information about the applicability of FSGFFs within the PSH context and the degree to which the operations of site 2 aligned with these guidelines. The PSH-A of site 2 indicated that they were performing below FSGFFs suggested standards regarding the provision of fruits, vegetables, and protein. In addition, it was discovered that one limitation to using FSGFFs was staff knowledge and capacity. Site 2 staff did not have the training or ability to identify and report information such as the number of trans-fats, sodium content, calorie amount, or nutritional value of foods, all of which are requirements in the FSGFFs. Moreover, site 2 was below FSGFFs standards concerning food safety and behavioral design (eg, how the presentation of food influences individual food choices). Site 2 needed assistance with establishing a comprehensive food safety plan, developing and implementing a written employee health policy, and working with worksite wellness programs or other organizations to promote healthier options. Also, sites needed assistance with executing FDA Food Codes. 20 The PSH-A of site 2 also showed room for improvement around behavioral design strategies such as using marketing strategies to highlight more nourishing food and beverage items or using product innovations and the inclusion of more nutritious options as a default choice at decision points to encourage healthier choices. Lastly, packaged snacks and vending sections of FSGFFs were not regulated aspects of food provision at these sites. Though the PSH-A was completed at 1 site, the similar food serving, preparation, and procurement practices at both sites would make it challenging to systematically implement the FSGFFs.

DISCUSSION
The threat of food insecurity and homelessness for millions of Americans requires focus from multiple sectors to develop best practices and policies that provide agencies with the tools to supply healthier foods. This project identified intersecting factors at permanent supportive housing sites related to procurement, menu and meal preparation, and consumption of foods that ultimately impact the nutrition of individuals living in PSH (Figure 1). We developed a framework for Nutrition Access Intervention in Permanent Supportive Housing (Figure 1) based on the preliminary themes identified by the PSH-IT, PSH-A, and discussion and observations at each site. The findings in Figure 1 enhance many of the conclusions made in previous literature and provide a cohesive model for improving food access for this population. For example, studies have found that meal delivery programs and increasing the income of those living in PSH could solve food insecurity for this population; aspects of nutrition access that fit into the food acquisition umbrella of this project's framework. 2 Moreover, data show that education is negatively associated with food insecurity. 2 In similar ways, this project found that staff and client education was a barrier to nutrition access and could be addressed under meal preparation and consumption. Specifically, consumption refers to the choices individuals make on what foods to consume at any given time. As suggested in Figure 1, educating clients about nutrition could impact their choice to consume healthier foods. The framework provides sites and their collaborators an opportunity to identify specific areas for intervention to improve nutrition services.
Moreover, policy can create far-reaching systemic changes across the food system and have downstream impacts on multiple areas of the above framework. Previous studies have identified a need to understand the policy and program roadblocks that prevent PSH from substantially impacting health outcomes. 4 This study offers foundational insight into PSH facilities' challenges and the relevance of FSGFFs as nutrition standards. Staffing and funding limitations at PSH can result in uneven, inconsistent, and ineffective implementation of nutrition standards and practices. In addition, minor differences in funding sources observed between site 1 and site 2 in our study may account for differences in access to healthy foods and greater reliance on charitable donations and the food bank. Management and technical assistance on behalf of an existing county, state, or national agency could improve site capacity to implement existing nutrition standards while bridging the gaps between procurement sites and PSH. Technical assistance may involve helping permanent supportive housing sites to display and market healthier options to clients, training staff on the 2020-2025 Dietary Guidelines for Americans (DGA) or educating staff on ways to understand the nutritional value of food. These strategies would address staffing capacity and staff training opportunities identified in our framework. These results are consistent with findings from other studies that suggested that PSH programs could benefit from standardization and improved staff involvement. 3 Technical assistance, existing agencies, community, and academic partners could support PSH sites in developing, implementing, and evaluating new standards that align with the DGA.
The FSGFFs were not designed to be used for sites with smaller population sizes or populations with specific dietary restrictions, which may contribute to difficulties in adapting them for PSH sites. While site 2 was compliant with city and county food safety standards, the disconnect between local policies and FSGFFs standards created variability in nutrition standards, resulting in site 1 underperforming concerning FSGFFs. Moreover, data demonstrated that FSGFFs do not address important aspects of how these sites procure or prepare foods nor provide direction to these sites on ways to improve procurement and preparation. These findings are consistent with previous studies that suggest that foodbanks would be more effective when combined with solutions that address operational resources, access to nutrient-dense foods, and client needs and preferences, all of which are discussed in Figure 1. 10 Alternative nutrition standards could help procurement sites such as food banks and donation partners prioritize the availability of healthier options while also taking into consideration site-level concerns.
Regardless of the policy structure, careful consideration is needed when working in these settings due to various internal and external factors. Using our proposed framework as a guide can allow for more comprehensive approaches to nutrition access in PSH.

PUBLIC HEALTH IMPLICATIONS
First, nutrition policy is essential because it can create farreaching systemic changes in all food system levels. The FSGFFs and alternative nutrition standards have a role in improving policies at food banks and donation partners while also addressing site staffing capacity, staff training, and funding limitations. Secondly, the current system places the onus for changing on individual sites, which is challenging to navigate understaffing and funding constraints. Policy intervention on behalf of an existing county, state, or national agency could improve site capacity to create and implement nutrition standards by regulating nutrition standards for various procurement sites, donation partners, and permanent supportive housing sites while also providing technical assistance and coordinating staff training. Lastly, federal food service guidelines are not designed for permanent supportive housing units that operate on limited budgets and staffing capacity and have unique ways of procuring and preparing foods. The limitations of food service guidelines could be counteracted by making the current city, county, and state policies more aligned with FSGFFs standards. In addition, FSGFFs could be improved by considering funding limitations, staff capacity, and staff training, which could be mediated through the provision of technical assistance.