‘Hello, you’re not supposed to be here’: homeless emerging adults’ experiences negotiating food access

Abstract

Objective

We aimed to examine the food-seeking experiences of homeless emerging adults (age 18–24 years) in a US urban context.

Design

The study used a qualitative descriptive design, combining semi-structured interviews with a standardized quantitative measure of food insecurity. Interview data were coded using constant comparative methods to identify patterns across and within interviews. Emerging themes were confirmed and refined through member checking.

Setting

Buffalo, a mid-sized city in the Northeastern USA.

Subjects

A sample of thirty participants was recruited through community-based methods. Eligibility criteria specified that participants were aged 18–24 years and did not have a stable place to live. The sample was demographically diverse and included participants who were couch-surfing, staying on the streets and/or using shelters.

Results

Participants’ food access strategies varied across their living circumstances. Common strategies included purchasing food with cash or benefits (reported by 77 %), using free meal programmes (70 %) and eating at friends’ or relatives’ homes (47 %). Although 70 % of participants received Supplemental Nutrition Assistance Program benefits, several reported access barriers, including initial denials of eligibility due to being listed on a parent’s application even when the participant no longer resided in the household. Participants described a stigma associated with using food pantries and free meal programmes and expressed preference for less institutionalized programmes such as Food Not Bombs.

Conclusions

Given endemic levels of food insecurity among homeless youth and young adults, policy modifications and service interventions are needed to improve food access for this population.

Food insecurity, defined as lacking access to adequate nutritious foods and/or being unable to obtain foods in a socially acceptable way^{( 1 , 2 )}, is a major concern for homeless populations. Prior research on homelessness and food insecurity has focused primarily on single adults and families in unstable (e.g. overcrowded or transient), substandard or unaffordable housing^{( 3 – 7 )}. For example, a recent study of homeless mentally ill adults in Vancouver, Canada found that nearly two-thirds of the sample were food insecure^{( 7 )}.

Fewer studies have addressed food insecurity among homeless youth, defined as young people up to age 24 years who sleep outdoors, in shelters or couch-surf (temporarily stay with friends, relatives or acquaintances)^{( 8 )}. Limited prior research has examined the occurrence of food insecurity in the context of homeless youths’ food acquisition practices in Canada^{( 9 , 10 )}, Australia^{( 11 – 13 )} and the USA^{( 14 )}. Building on a previous ethnographic study^{( 9 )}, Tarasuk et al.’s^{( 10 )} survey of homeless youth in Toronto classified youths’ food acquisition practices as routine strategies (e.g. going to free meal programmes, panhandling, stealing, or getting free leftover food from restaurants) or desperation strategies (e.g. putting off other expenses or payments to buy food, eating discarded food, going to friends’ or relatives’ homes for food, or trading sex for food). Food insecurity persisted for youth across their use of multiple acquisition strategies, with 73 % of male participants and 85 % of female participants classified as severely food insecure. A related analysis^{( 15 )} found that regardless of whether youth purchased food or obtained it from free meal programmes, their nutritional intakes were well below recommended guidelines.

Studies conducted in the Australian cities of Adelaide^{( 11 , 13 )} and Sydney^{( 12 )} reported homeless youth using a similar range of acquisition strategies as noted in the Canadian research. These studies also noted that although use of free meal programmes was among the most common food acquisition strategies, stigma often functioned as a barrier to access. For example, Booth^{( 11 )} described shame or embarrassment as a primary reason for youth avoidance of free meal programmes, along with not knowing about the location or existence of these services.

Food insecurity and food acquisition strategies among homeless youth in the USA are vastly under-researched. We located one prior study on this topic^{( 14 )}, which reported that approximately one-third of participants in a Midwest-based survey sample were food insecure, per a three-item measure. The authors classified participants’ food acquisition strategies as deviant, which included panhandling, stealing, trading sex or taking food out of a dumpster, or non-deviant, which included purchasing food or obtaining it from friends, family or social service programmes. Use of deviant strategies was associated with greater likelihood of food insecurity.

It is important for research to explore the unique contexts of homeless youths’ food acquisition strategies and food-related service-seeking experiences, taking account of age, developmental context, and local and national geographic factors. Studies focusing on other aspects of service systems, such as housing, have suggested that homeless youths’ service use patterns and attitudes are distinct from those of older homeless individuals. For example, homeless youth vary in the extent to which they identify with the term ‘homeless’^{( 16 )}. A strong sense of independence and the perceived stigma of homelessness limit some homeless youths’ willingness to utilize housing services, particularly shelters^{( 17 , 18 )}. Past studies have also noted that homeless youth who are travellers (a subgroup of homeless young people who frequently migrate between different geographic regions) may have distinct patterns of risk and health behaviours and service use^{( 19 , 20 )}. It is not known how these factors might impact homeless youths’ use of services for food access.

Even within the youth population, age and developmental differences may affect young people’s experiences and outcomes. Homeless youth in the emerging adult age range (age 18 to 24 years) face particular vulnerabilities. Because they are legally considered to be adults, these young people no longer qualify for the health and social service programmes they may have accessed as minors^{( 21 )}. With limited formal or informal supports in place, homeless emerging adults are at risk for a number of adverse outcomes, including sexually transmitted infections, substance abuse and mental health concerns^{( 21 , 22 )}. Previous research on homeless youths’ food access included both emerging adults and younger youth but did not discuss findings by age, excepting Tarasuk et al.^{( 10 )}, who did not find a statistical difference in food insecurity prevalence for 16–18-year-olds in comparison with 19–24-year-olds.

Food access for homeless emerging adults is also shaped by the availability of local services and the nature of national and local policies that govern food assistance. For example, the location, service hours and operational policies of free meal programmes can vary significantly by locale and affect the willingness of vulnerable populations to use these services^{( 23 , 24 )}. Further, social welfare programmes that facilitate food access differ between countries, and state and local implementations of national policies also vary. In the USA, the Supplemental Nutrition Assistance Program (SNAP) is the primary policy vehicle for food assistance. SNAP is administered by the US Department of Agriculture (USDA) in partnership with state and local government. US households that qualify for SNAP according to income-based guidelines and meet other eligibility requirements receive a monthly benefit amount (via a programme-specific electronic debit card) that can be used to purchase a variety of foods. In fiscal year 2017, the maximum monthly benefit for an individual was $US 194^{( 25 )}. SNAP cannot be used for certain types of food purchases, including hot prepared foods. We could not locate any studies discussing homeless young adults’ use of SNAP.

To address these gaps in the literature, the present qualitative study aimed to illuminate a detailed picture of US homeless emerging adults’ food-seeking experiences. Our aims centred on two intersecting research questions: (i) What strategies do homeless young adults use to access food?; and (ii) What role does focused food assistance (e.g. SNAP, free meal programmes and food pantries) play in facilitating homeless young adults’ access to food? An implicit aim in answering these questions was to inform policy and services to address barriers to food access for this population.

Methods

Study design

The present study used a qualitative descriptive research design^{( 26 )}, combining use of a standardized quantitative measure of food insecurity with semi-structured interviewing. As a research methodology, qualitative description emphasizes the comprehensive and accurate detailing of an event or phenomenon, in language that is rooted in the data itself^{( 26 , 27 )}. Thus, following our research questions, the study sought to provide a nuanced and data-driven description of homeless emerging adults’ food access strategies and experiences. Given the limited previous theoretical and empirical work on food access among homeless emerging adults, this methodology was selected as appropriate for the study’s aims.

Setting and sample recruitment

The study took place in Buffalo, a mid-sized city in upstate New York, USA. Sample recruitment occurred between July 2015 and January 2016, employing a community-based strategy to recruit youth with diverse living and service-seeking experiences. Although the first study participants were recruited at a resource centre for homeless and at-risk youth, additional recruitment tactics included word-of-mouth participant peer recruitment, posting pamphlets and posters in a variety of local public venues, and directly approaching young people on the street at two outdoor and one indoor programme that provided free meals. Among the final sample of thirty emerging adults, 40 % were recruited at the resource centre, 33 % were recruited by the research team on the street or at free meal programmes, 20 % found out about the study through word-of-mouth and 7 % called in response to posted flyers.

Participant eligibility was established through two criteria: (i) participant was aged between 18 and 24 years; and (ii) participant did not have a stable place to live. This intentionally broad definition of homelessness included participants in living situations such as shelters, couch-surfing and places not designed for human dwelling (e.g. sleeping outdoors or in a garage). Those who were in transitional or permanent supportive housing programmes were excluded from the study. Sample recruitment continued until the research team agreed that a theoretical saturation point had been reached, meaning that interviews did not appear to yield new findings or themes^{( 28 )}. The research team came to this consensus based on their review of the interview data, field notes and memos maintained throughout the data collection and analysis process.

Procedures and measures

Most study interviews took place at the youth resource centre, which was centrally located and accessible via multiple transportation routes, and were conducted by the principal investigator or a research assistant. Following recruitment, participants met individually with the principal investigator or research assistant to review study procedures and provide verbal informed consent. Participants then completed a demographic questionnaire, which included the Household Food Insecurity Access Scale (HFIAS), a standardized measure of food insecurity that has been used in previous research with homeless populations^{( 4 , 29 , 30 )}. As in Holland et al.’s study^{( 30 )} establishing the HFIAS as a preferred measure of food insecurity in homeless populations, HFIAS items were worded to apply to individuals rather than households (e.g. ‘Did you worry that you would not have enough food?’).

Following the questionnaire, participants completed semi-structured interviews, which were audio-recorded and transcribed for reference and coding. Interviews averaged 40 min in length and included questions about participants’ food access strategies and challenges related to food, as well as questions about participants’ housing and living conditions, health and future plans. Participants were provided $US 20 cash as remuneration for their time. All study procedures were conducted in accordance with prior approval by the institutional review board at the University at Buffalo.

Data analysis

Data obtained from the questionnaire were catalogued electronically in a spreadsheet and tabulated to display basic descriptive sample statistics. The transcribed interviews were input into the qualitative data analysis software Atlas.ti version 7.5 for coding. The principal investigator applied an open coding approach to generate primary codes from the interview data, using the constant comparative method to inductively identify patterns emerging across and within interviews^{( 28 )}. The principal investigator consulted with a second research assistant to synthesize descriptive themes from the primary codes generated in the open coding phase. Field notes and structured observations conducted by the principal investigator and the first research assistant served as an additional data resource for contextualizing the findings. To verify and refine the emergent descriptions, the principal investigator and second research assistant conducted a member check with six homeless emerging adults, including two individuals who participated in the primary data collection phase of the study and four who met the eligibility criteria but did not participate.

Results

Overview of sample

Descriptive statistics for the study sample (n 30) are available in Table 1. Half of the participants identified as male. The sample was racially diverse, including participants identifying as African American (50 %), White (33 %), other/multiracial (17 %) and Latinx (13 %). The predominant method for obtaining housing among participants was couch-surfing (63 %), with another 30 % of participants staying in overnight shelters or places unfit for human habitation. Six participants (20 %) were travellers passing through the area.

Table 1.

Description of the sample of homeless emerging adults (n 30) recruited from Buffalo, NY, USA, July 2015–January 2016

Characteristic	n or Mean	% or sd
Gender
Male	15	50
Female	13	43
Other gender identities: transgender (female-to-male), gender fluid	2	7
Race
African American	15	50
White	10	33
Other race/more than one race	5	17
Latinx	4	13
Sexual orientation
Heterosexual	23	77
Bisexual	4	13
Other sexual orientation: queer, asexual	3	10
Age (years), mean and sd	21	2
Traveller	6	20
Primary housing
Couch-surfing	19	63
Overnight shelters or places not meant for human habitation	9	30
Other (rooming house, mixed locations)	2	7
Has children	11	37
High school diploma or GED	17	57
Currently employed	6	20
Receiving SNAP benefits	21	70
HFIAS category
Food secure	2	7
Mildly or moderately insecure	4	13
Severely insecure	24	80
Monthly income from all sources ($US), mean and sd	280	355

GED, General Educational Development; SNAP, Supplemental Nutrition Assistance Program; HFIAS, Household Food Insecurity Access Scale.

Few participants were formally employed and monthly income across the sample was very low, ranging from $US 0 to 1300 with a mean of $US 280 (sd=$US 355). For 70 % of the sample, income was limited solely to money from family or friends or from informal or illicit income-generating activities, such as busking, selling drugs, doing hair and makeup, or collecting bottles and cans for cash refunds. Participation in SNAP was common, with 70 % of participants reporting receipt of SNAP benefits, averaging $US 197 (sd=$US 42) per month. Per the HFIAS categories, the majority of participants (80 %) were considered to be severely food insecure, with the remainder categorized as mildly or moderately food insecure (13 %) or food secure (7 %).

The state of food access for homeless emerging adults

We asked all participants ‘What do you usually do about food?’ and ‘Where do you eat most of your meals?’ In response to these questions, Table 2 lists the strategies participants used to access food and the number of participants who discussed using the strategy. Most participants mentioned using more than one strategy, depending on a number of situational factors. We synthesized participants’ responses by prevalence of use, as follows: (i) purchasing food with cash or SNAP benefits; (ii) using free meal programmes or food pantries; and (iii) alternative strategies, including eating in friends’ or relatives’ homes, scavenging food, eating at shelters or stealing food.

Table 2.

Food access strategies used by homeless emerging adults (n 30) recruited from Buffalo, NY, USA, July 2015–January 2016

	Participants using strategy
Food access strategy	n	%
Purchasing food	23	77
Using free meal programmes or food pantries	21	70
Eating at someone’s house	14	47
Scavenging food from the garbage or asking strangers for food donations	6	20
Shelter food	5	17
Stealing food from grocery stores	2	7

Purchasing food with cash or benefits

The most frequently mentioned strategy was purchasing food. Given participants’ very low incomes, SNAP (typically referred to as ‘food stamps’ within the sample) provided a critical means of obtaining food for those who received this benefit. Participants, particularly those with relatively stable couch-surfing arrangements, expressed appreciation that SNAP enabled them to purchase a variety of foods. ‘I go shopping about twice a month or I just go big shopping all the way, like on one day,’ described one woman, who was couch-surfing with relatives while applying for public housing (Participant 23, age 22 years, mild/moderately food insecure). Being able to store and prepare food where they stayed enabled these participants to make the most of their SNAP benefits. Some would combine their benefits with others in the household. ‘My mom gets a lot of food stamps and then I get food stamps, so we always full in the house,’ remarked a woman who alternated between living with her mother or friends while looking for her own place (Participant 24, age 18 years, severely food insecure).

Although more than two-thirds of the sample were currently receiving SNAP, some participants reported barriers to applying. Per SNAP guidelines, ordinarily youth under the age of 22 years who do not have a separate address are included on their parents’ applications. Recognizing the obstacles this could present for homeless young people, in 2013 the USDA issued a memo clarifying that lack of fixed address should not preclude any youth or adult from accessing SNAP, and that ‘youth who are not living with their parents are not required to apply as part of their parent’s household’^{( 31 )}. According to participants’ comments, these policy provisions were not always explained when they applied for benefits at the local Department of Social Services office. Several mentioned that they were initially denied benefits when they applied, due to being listed on their parents’ applications even when they were not living there. As one woman explained:

‘Downtown [the main office of the county Department of Social Services], they would need to listen to the situation more. I was told that I had to go back home with my dad because my dad had food stamps and everything opened in my name. They said they weren’t able to open another case for me, and that I was denied. That was difficult.’ (Participant 6, age 20 years, severely food insecure)

This participant reapplied and eventually was approved for SNAP. Participants expressed concern that their parents or guardians had a vested interest in keeping them on the household application, even if the participant was no longer living at home, as this increased the household’s benefit. ‘I would go down [to apply for SNAP] and they would tell me like she [participant’s mother] has to write a note saying that she doesn’t take care of me anymore,’ described one woman (Participant 1, age 20 years, severely food insecure). ‘Who is going to do that?’ This participant did not receive SNAP. Another participant stated that not getting his share of food or SNAP benefits from his mother was a factor in him leaving home: ‘And then basically I moved out of my mom’s house because she wasn’t trying to give us the food stamps, let us eat properly, so that’s when I turned to the streets. Did a couple crimes’ (Participant 15, age 20 years, moderately food insecure).

Even if they were approved, some participants felt frustrated by the application and recertification process. ‘I don’t like going to DSS [Department of Social Services]. For recertifying for food stamps, I hate it. They’re really rude,’ described one man (Participant 13, age 22 years, severely food insecure). ‘I’ve waited there over four hours just to get food stamps, and that’s supposed to be an hour-and-a-half process.’ Some participants felt that staff were particularly rude or hostile to them due to their young age, expressing frustration that they were tasked with adult responsibilities but were not afforded the respect due to adults.

Others remarked upon the difficulty of getting all the food they needed through SNAP. ‘They don’t give anybody enough food stamps to live off of for a month, like, $194. You just got to spend wisely, really,’ described one couch-surfing man (Participant 21, age 22 years, mildly/moderately food insecure). This participant’s income for the past month was $US 41 received from friends and family, making SNAP a critical resource for him. When SNAP benefits ran out, or for participants who did not get them, young people frequently resorted to using cash to purchase inexpensive packaged foods like chips, candy or ramen noodles. ‘It’s not hard to find quarters; chips are really like 25 cents. It’s normally like all I eat though,’ stated Participant 1. A few participants mentioned purchasing fast-food meals, although many stated that this was usually beyond their budget. A male traveller who did not receive SNAP described his unique approach of purchasing fast food or attempting to barter his labour for food:

‘Usually if I can make enough money I like to go to like cheaper fast-food places or like Subway or something and get something to eat. A lot of times when I don’t have enough money I’ll go to the same places and I’ll ask them like, “Hey, do you have any odd jobs that you need done in exchange for some leftovers of food or something?” and a lot of the times they won’t even bother with having you do something. They’ll just help you out.’ (Participant 19, age 20 years, severely food insecure)

Using free meal programmes or food pantries

Accessing free meal programmes and food pantries was mentioned by 70 % of the sample. This included free-standing food pantries, the small pantry and meal preparation space at the youth resource centre, traditional indoor free meal programmes (soup kitchens) and outdoor free meal programmes such as Food Not Bombs. Some participants relied on these programmes for most of their meals, while others used them only occasionally and typically as a last resort. ‘It makes things less stressful, knowing that I would at least get a sandwich,’ said one participant, describing a van-based free meal programme that she regularly accessed (Participant 6, age 20 years, severely food insecure).

Although food pantries and free meal programmes were a part of most participants’ safety nets, participants described stigma as a deterrent to using these services. As one man described:

‘I might go to some free-food agencies … You know you can just tell how people look. When I walk up the street, I may be carrying a bag. Normally, it may be something that I got from the pantry like some free bread or bakeries or snacks or whatever; and people will just move far away from me.’ (Participant 14, age 22 years, severely food insecure)

‘A lot of people that know me, and it would be making me seem like, hello, you’re not supposed to be here … more embarrassing than not, especially [to be] seen walking out of there,’ stated another participant, referring to a local soup kitchen that he occasionally used (Participant 15, age 20 years, mild/moderately food insecure). Participants commented that most free meal programmes primarily served an older clientele, with whom they did not identify. ‘The whole time I was there [at a local soup kitchen] I’m like, why am I here? … [I’ve] seen like two/three people like my age,’ one man stated (Participant 25, age 21 years, severely food insecure).

For some young people, keeping track of the hours and locations of different programmes was difficult. Participants described spending considerable time walking or taking the bus to get to a programme, only to find they had arrived too late and the programme was closed or had already run out of food. ‘Either you know what time they serve dinner or you’ll just be SOL, so out of luck … you either get with the programme or the programme will have your stomach growling,’ explained Participant 14. Meal programmes were often hampered by limited resources. During one recruitment session at the van-based free meal programme, researchers observed the programme running out of bagged meals after approximately 15 min. Participants who had to go to regular appointments for health or mental health care, addiction treatment or legal issues found it particularly difficult to make use of free meal programmes, as one man described:

They [a free meal programme] are usually open from 12 to 1, and by the time I’m done with my counsellor and my programme – I’m in a methadone programme … I have like three minutes to eat, if I make it there. If everything is on time, the train, the bus.’ (Participant 17, age 24 years, severely food insecure)

Recognizing the stigma of traditional soup kitchens and pantries, participants expressed preference for programmes that felt less institutional. Several participants used the food pantry and kitchen space at the youth resource centre and appreciated its home-like feel. ‘But [as] far as what I like about the food is, it’s like you feel like home. You can just go in there and make you something to eat, you know,’ described one woman (Participant 28, age 23 years, severely food insecure). Others visited the local chapter of Food Not Bombs, which served vegan meals at outdoor locations twice per week. One traveller described why he preferred Food Not Bombs to other free meal programmes:

Food Not Bombs is really easy to find … I don’t like asking for handouts, so you can actually help them. That’s the thing I like most; you don’t have to just come up and take the food, you can be like, “Hey, I’ll take some food this time but where do you prepare the meals, can I come and help you out,” kind of thing.’ (Participant 19, age 20 years, severely food insecure)

Another traveller described a positive experience with a restaurant-like free meal programme he had visited in Oregon: ‘It’s really bright and fresh in there … They come around and give you drinks and things, you know, it’s like sitting in a restaurant, which you don’t ever get to do you know, usually’ (Participant 11, age 25 years, food secure).

Alternative strategies

A minority of participants used alternatives to formal focused food assistance, including eating in the homes of friends or relatives, scavenging food from the garbage, eating at shelters or stealing food. Some participants with relatively stable couch-surfing arrangements, such as a friend or family member with whom they could stay for a few weeks at a time, reported that their hosts provided not only a sleeping place but also meals. One man who was temporarily staying with his fiancée’s grandmother reported that she would cook for the household: ‘She makes a lot of Spanish food … it’s a plus, you know’ (Participant 3, age 20 years, severely food insecure). Other participants had relationships with relatives with whom they could not live but who did provide periodic meals. One woman who was couch-surfing remarked that ‘if I run out of food stamps or anything, I always could go to my mom’s house and eat’ (Participant 24, age 18 years, severely food insecure). Another woman commented on eating at her boyfriend’s grandmother’s house: ‘She doesn’t know how to stop cooking!’ (Participant 29, age 18 years, food secure).

A less common strategy was scavenging food from public places such as dumpsters or garbage cans or asking strangers for food donations, such as leftovers from a restaurant. This strategy was mentioned primarily by travellers. ‘I can sit around and play ukulele and sing songs, and then people will give you food and money,’ one gender-fluid traveller described (Participant 11, age 24 years, food secure). ‘I remember sitting in San Diego sometimes just like piled up Chinese [food] boxes and pizza.’ A non-travelling participant described his strategy: ‘Some places, like Aldi’s or 7–11, they’ll throw out like day-old food that’s still wrapped up … sometimes, they’ll save it for me and my friends; and we’ll come up and we’ll split it up between each other’ (Participant 13, age 22 years, severely food insecure). He also mentioned fishing in a local river as a source of food.

Because most participants did not stay regularly in shelters, eating in shelters was not a regular food source and was discussed by only five participants. These participants mentioned long waits and occasional fights breaking out in meal lines, but also expressed appreciation for being able to get a meal on-site when they stayed in shelters. ‘Sometimes it’s great. Sometimes it’s okay. I would never call this [the food] nasty because I could be hungry. So, I never call it nasty,’ one man said (Participant 4, age 23 years, severely food insecure). Only two participants referenced stealing as a food access strategy. ‘Sometimes … I steal because it’s all I can [do to] eat,’ explained one young woman, who had been couch-surfing for one year and whose monthly income was limited to the approximately $US 50 she obtained from panhandling (Participant 16, age 23 years, severely food insecure).

Discussion

The present qualitative study illuminates homeless emerging adults’ food access practices, including strategies utilized and experiences applying for and using SNAP benefits, food pantries and free meal programmes. While our study reported a higher rate of food insecurity among homeless young people compared with the only prior US-based study we located^{( 14 )}, this difference may be due in part to differing measures, as the prior study used a three-item measure of food insecurity drawn from a USDA-developed scale. The prior study also focused on a younger age demographic of 16–19-year-olds. Some of our findings echo the results of previous research conducted in Canada and Australia^{( 9 – 13 )}, which similarly reported high levels of food insecurity persisting across homeless young people’s use of a range of food access strategies. One Australian study^{( 11 )} noted that many youth perceived a stigma associated with using free meal programmes, which our participants also described.

Beyond these consistencies, our findings also demonstrate nuances of the food-seeking experience that have not been described well in the extant literature. For example, we could not find any previous research discussing SNAP use for homeless emerging adults. Homeless young people’s preference for deinstitutionalized free meal programmes that promote a sense of choice and normalcy is also scantly explored in previous research.

Altogether, the descriptive themes that emerged in our analysis form a picture of food access as a constant negotiation for young people experiencing homelessness. Across the various access strategies that participants used, the daily challenge of procuring food repeatedly thrust participants into spaces and situations where they often felt they did not belong or did not feel welcome. This was reflected in the systematic denials and perceived patronizing or rude treatment that participants reported when applying for SNAP, as well as in the stigma that some participants perceived about using food pantries and free meal programmes. Even in the accounts of participants who ate meals at the homes of relatives or friends was an implicit reminder that spaces that were available to them for eating were not always available for sleeping or other routines of daily life.

These findings convey several policy and practice implications for improving food access for homeless emerging adults. First, despite a high rate of SNAP participation in the sample, participants’ interviews indicate that they were not consistently notified by social service workers of their eligibility to apply for SNAP benefits on their own, per USDA policy. This is an example of street-level bureaucracy, e.g. the idea that frontline workers often ration resources and implement policy in ways that differ from the policy’s intentions, particularly in instances where implementation guidelines are not clear or training and supervision are not sufficient^{( 32 )}.

Although a complete analysis of the factors influencing SNAP access and administration for this population is beyond the scope of the present study, it is apparent that social service workers may need additional training on homeless young adults’ unique eligibility for benefits independent of their parents’ household applications. Given that documenting residence and independence from parental households is often a complex task, young adults with limited experience navigating social services would also benefit from having a designated advocate, such as an experienced case manager, who can assist them in preparing the documentation needed to prove their eligibility. Further research should evaluate SNAP policy implementation for homeless youth (including emerging adults and younger youth) in other geographical locations, as well as examine frontline workers’ interpretation of the policy.

Beyond SNAP benefits, our findings point to a need for flexible and deinstitutionalized ways of facilitating food access for homeless emerging adults. Integrating food and nutrition services with other services in which young people are engaging may help to reduce the stigma associated with traditional soup kitchens and pantries, as well as lessen the challenges associated with seeking transportation between different resources. For example, in response to well-publicized reports of rising food insecurity among college students, more than 500 college campuses across the USA now offer on-campus food pantries^{( 33 )}. Food pantries, ideally with space for participants to prepare simple meals for themselves on-site, could be integrated into other programmes serving vulnerable young people, such as GED (General Educational Development) programmes, drop-in centres and health clinics. This approach was supported by low-income (but mostly housed) youth in a recent report on teens and food insecurity in the USA^{( 34 )}. Meal programmes that invite food recipients to assist in cooking and serving, such as volunteer-run ‘pay as you can’ community cafés and Food Not Bombs, may be more empowering and appealing for many young adults. Lastly, although homeless emerging adults differ in the depth and strength of their social networks, service providers can encourage young people to build and sustain connections with the supportive adults in their lives when possible, as these ‘natural mentors’ may provide access to food as well as other forms of material and intangible assistance^{( 35 )}.

The study’s findings should be considered in the light of several limitations. The sample for our qualitative analysis was selected from a single US city and although it included participants with diverse living circumstances and experiences, it is not known if the findings generalize to other settings. Further, while our sample size of thirty is typical of qualitative descriptive studies, additional insights may be gained through mixed-methods studies that include a survey component with larger samples. Some participants in our study were recruited at free meal programmes, possibly biasing the sample toward participants familiar with these services. Although our measure of food insecurity, the HFIAS, has been used in previous research with homeless individuals, additional psychometric studies are needed to document its validity for this population. Lastly, participants were interviewed at only one point in time. Prolonged engagement through observations and longitudinal interviews would provide further opportunities to examine if and how food insecurity and food access may shift for participants alongside changes in their housing or other circumstances. It is apparent that food access is a persistent challenge for young adults in the context of homelessness, and that further research and service and policy interventions are needed to respond to this challenge in innovative ways.