Abstract
Background:
Sexual minority youth are at more than twice the risk of experiencing homelessness than their peers and both sexual minority youth and youth experiencing homelessness have disproportionate risk for mental health disorder symptoms. Couch-surfing is a common form of homelessness experienced by youth, but research on the relationship between couch-surfing and mental health outcomes, especially among sexual minority adolescents (SMA), is limited.
Methods:
Utilizing a sample of 2,558 SMA (14–17 years old) recruited via social media and respondent-driven sampling, this study explores the relationship between different forms of homelessness (exclusive couch-surfing vs. multiple types of homelessness) and symptoms of depression, anxiety, suicidal ideation, and suicide attempt.
Results:
Nearly 21% of participants experienced any homelessness in their lifetime, with 14% reporting exclusive couch-surfing. All forms of homelessness were associated with large increases in symptoms of anxiety, depression, suicidal ideation and suicide attempt.
Conclusion:
Homelessness – primarily couch-surfing – is a common experience for SMA in this sample. All forms of homelessness – including exclusive couch-surfing – were associated with large increases in depression, anxiety, suicidal ideation, and suicide attempt, emphasizing the importance of services that are available to couch-surfing young people and responsive to the needs of sexual minority adolescents.
Keywords: homelessness, couch-surfing, sexual minority adolescents, mental health
An estimated 3.5 million 18 to 24-year-olds and 700,000 13 to 17-year-olds experience homelessness in the U.S. (Morton et al. 2018) and recent nationally representative samples have found that sexual minority young people have more than twice the risk of experiencing homelessness than their non-sexual minority peers (Cutuli, Treglia, and Herbers 2020; Morton et al. 2018). Sexual minority adolescents and youth (SMA/Y)1 also report disproportionate rates of mental health disorder symptoms, including higher rates of depression, anxiety, and suicidality (Haas, Mays, and Mathy 2011; Marshal et al. 2011; 2013). When combined, disparities in homelessness and mental health disorder symptoms among SMA/Y become particularly marked. SMA/Y who experience homelessness report even higher rates of mental health disorder symptoms and suicidality than SMA/Y without homelessness experiences, including higher rates of self-harm (Moskowitz, Stein, and Lightfoot 2013), suicide attempts (Kidd, Gaetz, and O’Grady 2017; Moskowitz, Stein, and Lightfoot 2013; Noell and Ochs 2001; Rhoades et al. 2018; Van Leeuwen et al. 2006), and symptoms of psychological distress or mental health disorders (Kidd, Gaetz, and O’Grady 2017; Noell and Ochs 2001; Prock and Kennedy 2020; Rhoades et al. 2018).
These disparities may be due, in part, to the unique trajectories into homelessness often experienced by SMA/Y. Qualitative research (Castellanos 2016; Schmitz and Tyler 2017) and work with service providers (Durso and Gates 2012) has found that SMY often experience homelessness due to family rejection. When family rejection is the reason for homelessness, it follows that these youth are more likely to be unaccompanied (that is, they are not experiencing homelessness with the families that rejected them). As such, sexual minority adolescents and youth may be at greater risk for all forms of unaccompanied homelessness, including less visible forms of housing instability such as couch-surfing.
Couch-surfing and sexual minority youth
Couch-surfing – also referred to as precarious housing – is a common form of homelessness experienced by youth. It is 3–4 times more prevalent than unsheltered homelessness among youth and frequently overlaps with other forms of housing instability, including sleeping on the streets and staying in emergency shelter (Curry et al. 2017). In addition to sexual minority youth being overrepresented in the population of youth experiencing homelessness, current evidence indicates that couch-surfers are also more likely to identify as LGBTQ+ (Hail-Jares, Vichta-Ohlsen, and Nash 2020). Service providers have noted a preference among sexual and gender minority youth for couch-surfing—and for engaging in exchange sex in particular—as a way of avoiding dangerous or otherwise non-inclusive shelter environments (Samuels et al. 2018; Showden and Majc 2018). However, couch-surfing does not necessarily reduce exposure to dangerous environments, as couch-surfing has been linked to greater likelihood of being assaulted, having sex with a stranger, and low rates of service utilization relative to other forms of homelessness (Suchting et al. 2020; Tyler, Olson, and Ray 2020).
Antecedents to couch-surfing are similar to those of other forms of homelessness among youth, including family conflict, violence and abuse, being evicted or crowded out, behavioral health issues either of their own or of a family member (McLoughlin 2013), and residing in low-income households (Curry et al. 2017). In one study of young sexual minority men, precarious housing in the preceding six months was associated with growing up without having basic needs met (Krause et al. 2016). In the same study, childhood physical and sexual abuse, an arrest history, and being out to either parents, family members, or teachers were also correlated with precarious housing; meanwhile, being out to friends was associated with sleeping on the streets (Krause et al. 2016).
A recent study of youth experiencing homelessness across 16 communities in the U.S. indicates that youth experiencing homelessness for the first time are more likely to be couch-surfing and less likely to be on the streets (Petry et al., 2022). Findings from qualitative studies suggest that youth may first draw upon either their peers or the family of their peers for housing as an “immediately accessible tactic” to help navigate an initial housing loss (Curry et al. 2020; McLoughlin 2013).
Mental health and couch-surfing
Current evidence regarding the mental health of couch-surfing youth is minimal and somewhat mixed. The psychological burdens of couch-surfing include feelings of guilt as an impermanent houseguest imposing on others and a fundamental lack of ontological security that can lead youth to resort to sleeping on the streets (McLoughlin 2013). Severity of self-harm has been associated with an increased likelihood of couch-surfing (Hail-Jares, Vichta-Ohlsen, and Nash 2020) and higher rates of self-harm among youth experiencing homelessness have been observed among LGBTQ+ youth (Morton et al. 2018). Among precariously housed youth, higher levels of depressive symptoms have been associated with identifying as female, more frequent physical abuse, and more neglect (Tyler, Johnson, and Melander 2010). However, although some evidence finds couch-surfers are more likely to have mental health diagnoses or symptoms compared to other youth experiencing homelessness or to housed youth (Hail-Jares, Vichta-Ohlsen, and Nash 2020; Hail-Jares et al. 2021), other research reported that couch-surfers indicated fewer days depressed when staying with a friend or romantic partner when controlling for all other sleeping locations (Tyler, Olson, and Ray 2020).
The Current Study
Research into couch-surfing as a distinct form of homelessness is relatively new and understudied. While existing findings suggest sexual and gender minority youth are overrepresented among couch-surfers, more research is needed to understand the correlates of couch-surfing and the mental health outcomes associated with this form of precarious housing. This exploratory study aims to deepen our understanding of couch-surfing experiences among sexual minority adolescents, including which adolescents are more likely to experience couch-surfing as compared to other forms of homelessness, and how homelessness and couch-surfing in particular are associated with mental health outcomes.
To expand our knowledge in this area, this study utilized a large, nationwide community sample (N = 2,558) of SMA recruited via social media advertising and respondent-driven sampling. This sample represents a new and unique contribution to the literature because much of the current research with sexual minority youth experiencing homelessness has not focused exclusively on adolescents despite the key importance of this developmental time frame, has used samples of youth already experiencing homelessness, and has often been qualitative in nature. Moreover, this unique sample offers insights into how the experience of homelessness among SMA differs regionally throughout the U.S. Prior research suggests that distinct patterns in sociodemographics, homeless histories, and behavior profiles among youth across different communities hold important implications for national policy responses to youth homelessness (Bowen et al. 2017; Ferguson et al. 2012; 2010). Urbanicity has also been cited as an important factor to consider, as youth experiencing homelessness in rural areas indicate a greater reliance on couch-surfing, less proximity to youth services, and greater disconnection from education and employment (Morton et al. 2018). Given the dearth of information in this area, this paper is not driven by specific directional hypotheses; instead, these exploratory analyses aim to answer the following research questions:
- Among SMA, what are the correlates of homelessness?
- Do these correlates vary for those who experience couch-surfing only vs. couch-surfing and other types of homelessness?
Do SMA with homelessness experiences report higher rates of mental health disorder symptoms and suicidality and does the relationship between homelessness, mental health disorder symptoms and suicidality vary by type of homelessness experience (exclusive couch-surfing vs. other types of homelessness)?
Method
These analyses used baseline data from 2,558 SMA who were part of a longitudinal study of 14–17 year-old cisgender adolescents. A national community sample of SMA was recruited via targeted social media advertising (Facebook, Instagram, YouTube) based on geography and urbanicity to purposefully recruit adolescents from across the United States in both urban and rural areas. A brief screener determined study eligibility (aged 14–17, identified as cisgender, provided a U.S.-based ZIP code, and reported a sexual attraction other than heterosexual or straight). The larger longitudinal study included cisgender, sexual minority adolescents only as its aim was understanding experiences of sexual minority-specific stress and behavioral health during adolescence. To ensure data integrity, several checks for fraud (e.g., duplicate email address or contact information, screening out on first attempt and re-entering with false responses to get through the screener) and data quality (e.g., unrealistic survey completion times, low validation scores based on attention check measures, or decline to answer numerous questions) were completed before respondents were included in the finalized baseline data. Participants were given the opportunity to refer up to three other adolescents into the study. Participants received $15 for completing the baseline survey and could earn another $10 for each of the three people they referred to the study. The study was granted a waiver of parental consent to protect youth who may not have disclosed their sexual orientation to their parents/caregivers; all participants provided online assent prior to completing the survey. All study methods were approved by the authors’ University Institutional Review Board (IRB#: UP-17-00538) For more details on study methods, see Schrager et al. 2022.
Measures
Demographics.
Demographic characteristics (age, race and ethnicity, sex at birth, sexual orientation, and socioeconomic status) were assessed with items created by the authors. The race and ethnicity item had six response options (Native American, American Indian, or Alaska Native; Asian or Pacific Islander; Black or African American; White; Latino or Hispanic; and race and ethnicity not listed); respondents could choose all categories with which they identified. Participants who chose multiple racial and ethnic categories were coded as multiracial. For analytic purposes, this variable was collapsed into six categories (White, Latino or Hispanic, multiracial or multiethnic, Black or African American, Asian, and another race and ethnicity). Participants were asked “What was your sex assigned at birth?” Response options were “male” and “female.”
Sexual orientation/identity was assessed by asking an open-ended question, “What would you say is your sexual orientation or identity?” The research team used existing literature, prior work with sexual identity variables, and a range of responses on this question to code these open-ended responses into closed-ended categories for analysis. The responses were coded as gay, lesbian, bisexual, pansexual, complex or multiple identities (e.g., gay pansexual, bisexual lesbian), queer, straight or mostly straight, asexual, and others (e.g., demisexual, agrosexual). For analytic purposes, we collapsed these into three categories: Gay/lesbian, bisexual/pansexual, and another sexual identity. To assess socioeconomic status, we asked: “Are you eligible for free or reduced-price lunch at school? (If you are no longer in school, please answer based on the last year you were in school).” At the time these data were collected, two states (Hawaii and Iowa) provided school lunch at no cost to all students, so respondents from those states were excluded from regression analyses (n=33).
Homelessness Experience.
The authors assessed experiences of homelessness or housing instability by asking: “Have you ever had to spend the night somewhere other than your home, because you had nowhere else to stay?” For these analyses, an affirmative response to the item was referred to as “ever experienced homelessness.” For those SMA who reported a lifetime experience, a follow-up asked for all the locations where they had ever stayed during one of these events, with response options of youth or adult shelter; in a public place, such as a train, subway or bus station, restaurant, or office building; on public transportation (like riding a bus, subway, or train all night); in an abandoned building or squat; outside in a park, on the street, on the beach, on a rooftop, or some other outdoor space; with someone you didn’t know (a stranger); with friends, extended family, or acquaintances (couch-surfing), or some other place. Two separate variables were created from the location item. The first was a 4-category variable that divided respondents into 1. No lifetime homelessness, 2. Lifetime homelessness that was only couch-surfing, 3. Lifetime homelessness that included both couch-surfing and other types of homelessness, and 4. Lifetime homelessness that did not include any couch-surfing experiences. The second variable was an indicator of couch-surfing only vs. other types of homelessness among only SMA with lifetime homelessness.
Parental Rejection.
Parental rejection was assessed through two items from the Sexual Minority Adolescent Stress Inventory (SMASI) (Schrager, Goldbach, and Mamey 2018). Youth who responded yes to either “My mother (or female caregiver) does not accept me as LGBTQ” or “My father (or male caregiver) does not accept me as LGBTQ” in their lifetime were coded as having experienced parental rejection.
Mental Health Outcomes.
Symptoms of depression were measured using the Center for Epidemiologic Studies Depression Scale Short Form (CES-D-4), which contains four items assessing the frequency of depression symptoms during the past week. Items include “I felt lonely” and “I had crying spells.” Participants responded on a Likert scale with response options ranging from 0 (rarely or none of the time [less than 1 day]) to 3 (most or all of the time [5–7 days]); items are summed (0–12; Melchior et al. 1993) and are presented as a continuous sum score.
Symptoms of anxiety were measured using the Generalized Anxiety Disorder 7-item scale (GAD-7), assessing the frequency of anxiety symptoms during the last two weeks. Item examples include “feeling nervous, anxious or on edge,” “trouble relaxing,” and “feeling afraid as if something awful might happen.” Participants responded on a Likert scale with response options ranging from 0 (not at all) to 3 (nearly every day); scores were summed (range = 0–21; Spitzer et al. 2006)) and an indicator of “moderate to severe anxiety” was created based on the sum score.
Suicide attempt.
Past year suicidal ideation and suicide attempt were measured using items from the Youth Risk Behavior Survey (Cleary 2000) asking, “During the past 12 months did you ever seriously consider attempting suicide” and “During the past 12 months, how many times did you actually attempt suicide?” For the attempt question, response options were “0 times,” “1 time,” “2 or 3 times,” “4 or 5 times,” and “6 or more times.” For analytic purposes, suicide attempt was dichotomized; participants who indicated one or more past year suicide attempts were coded 1 and those with no attempts were coded 0.
Analytic Methods
Descriptive statistics are presented for all variables. Descriptives revealed very few respondents who had experienced homelessness but never couch-surfed (n=46; 1.8%); given the small sample size, these youth were excluded from analyses examining correlates of homelessness type and relationships between homelessness type and mental health outcomes. Given this exclusion, regression models therefore use a 3-category variable for type of homelessness: 1. No homelessness, 2. Couch-surfing only, and 3. Couch-surfing and other types of homelessness.
Regression analyses were used to examine relationships among correlates of homelessness, types of homelessness experiences, and mental health outcomes. Linear or logistic (binomial or multinomial) regression analyses were utilized based on whether the outcome variable was continuous, binary, or categorical. Regressions analyses assessed: 1) correlates of any homelessness experience and type of homelessness experience among all respondents; 2) correlates of couch-surfing experience vs. any homelessness experience among only those experiencing homelessness; 3) associations between type of homelessness and mental health outcomes (anxiety, depression, and suicide attempt) among all respondents; and 4) associations between couch-surfing only and mental health outcomes among only those experiencing homelessness. Linear regression effects are presented as β, binary logistic regression results as odds ratios (OR), and multinomial logistic regression results as relative risk ratios (RRR). All models adjust for parental rejection and demographic variables. Analyses were conducted in Stata 16.
Results
The Sample
As shown in Table 1, participants were 15.9 years on average (SD = 0.97); 64.3% assigned female at birth; and 60.6% White, 14.5% Latino or Hispanic, and 7.8% Black or African American. Most adolescents identified their sexual orientation as bisexual or pansexual (48.3%), followed by gay or lesbian (43.1%) or another sexual orientation (8.6%). In this sample, 39.5 percent of participants reported that they were eligible for free or reduced-price lunch at school (excluding adolescents in HI and IA). Nearly 21% of respondents had ever experienced homelessness, with 14.0% reporting only couch-surfing experiences, 5.1% reporting both couch-surfing and other types of homelessness, and 1.8% reporting homelessness that did not include couch-surfing. Among SMA who had experienced homelessness, 66.9% reported couch-surfing only.
Table 1.
Demographic Characteristics, Homelessness and Mental Health among Sexual Minority Adolescents (N = 2,558)
| % (n) or M (SD) | |
|---|---|
| Demographics | |
| Age | 15.9 (0.97) |
| Race and ethnicity | |
| White | 60.6 (1,550) |
| Latino or Hispanic | 14.5 (370) |
| Multiracial or multiethnic | 8.5 (217) |
| Black or African American | 7.8 (199) |
| Asian | 6.4 (164) |
| Native American, American Indian or Alaskan Native | 2.3 (59) |
| Sex at birth | |
| Male | 35.7 (913) |
| Female | 64.3 (1,647) |
| Sexual orientation | |
| Gay or lesbian | 43.1 (1103) |
| Bisexual or pansexual | 48.3 (1,236) |
| Another sexual orientation | 8.6 (221) |
| Eligible for free or reduced-price lunch at school | 39.5 (1,004) |
| Parental rejection | 36.8 (942) |
| Ever experienced homelessness | 20.9 (528) |
| Type of homelessness | |
| None | 79.1 (1,994) |
| Couch-surfing only | 14.0 (353) |
| Couch-surfing and other forms | 5.1 (129) |
| Homelessness without couch-surfing | 1.8 (46) |
| Among those with homelessness | |
| Couch-surfing only | 66.9 (353) |
| Mental Health | |
| Moderate to severe anxiety (GAD-7) | 62.6 (1,601) |
| Depression symptoms (CESD-4 score) | 6.32 (3.41) |
| Suicidal ideation (lifetime) | 43.2 (1,105) |
| Suicide attempt (lifetime) | 16.2 (414) |
Homelessness Experiences: Places of Stay
Table 2 presents the places of stay reported by SMA with lifetime homelessness experiences (SMA could choose as many categories as applied to them, so responses are not mutually exclusive). The most commonly reported homelessness experience by far is staying with friends, extended family, or other acquaintances (couch-surfing), with 92% reporting that this described at least one of their homelessness experiences. Nine percent reported staying with a stranger, 8% stayed in an outdoor place, and 8% in a shelter.
Table 2.
Experiences of Homelessness among Adolescents Experiencing Any Lifetime Homelessness (n=528; categories are not mutually exclusive)
| % (n) | |
|---|---|
| With friends, extended family, or other acquaintances (couch-surfing) | 92.1 (486) |
| With someone you did not know (a stranger) | 9.3 (49) |
| In a youth or adult shelter | 8.1 (43) |
| Outside in a park, on the street, on the beach or overhang, on a rooftop, or some outdoor place | 8.1 (43) |
| In a public place, such as a train, subway, or bus station; restaurant; or office building | 6.4 (34) |
| On public transportation (like riding a bus, subway, or train all night) | 3.6 (19) |
| In an abandoned building or squat | 4.2 (22) |
| Motel | 2.8 (15) |
| Camping | 0.57 (3) |
| Vehicle | 0.76 (4) |
| Romantic partner | 0.38 (2) |
| Some other place | 4.2 (2) |
RQ1: Among SMA, what are the correlates of homelessness?
As shown in Table 3 (Model 1), correlates of any homelessness experience include older age (OR=1.16; p=0.011), experiencing parental rejection because of their sexual minority identity (OR=2.10; p<0.001), living in the Midwest vs. the West of the U.S. (OR=1.47;p=0.031), and having been eligible for free or reduced price lunch in school (OR=3.01; p<0.001). Asian SMA – as compared to white SMA – were less likely to have experienced homelessness (OR=0.45; p=0.011).
Table 3.
Correlates of Lifetimes Homelessness Experiences, Overall and by Type of Homelessness
| Model 1 (n=2,193) | Model 2 (n=2,150) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| ||||||||||||
| Any Homelessness | Couch-Surfing Only | Multiple Types | ||||||||||
|
| ||||||||||||
| Odds Ratio | P>|z| | [95% Conf. Interval] | RRR | P>|z| | [95% Conf. Interval] | RRR | P>|z| | [95% Conf. Interval] | ||||
| Age | 1.16 | 0.011 | 1.03 | 1.29 | 1.19 | 0.010 | 1.04 | 1.36 | 1.10 | 0.339 | 0.90 | 1.35 |
| Race/ethnicity (white=referent) | ||||||||||||
| Latinx/Hispanic | 0.74 | 0.070 | 0.53 | 1.03 | 0.82 | 0.317 | 0.56 | 1.21 | 0.64 | 0.126 | 0.36 | 1.13 |
| Multi-racial/multi-ethnic | 1.11 | 0.601 | 0.75 | 1.63 | 1.29 | 0.243 | 0.84 | 2.00 | 0.77 | 0.518 | 0.35 | 1.69 |
| Black | 0.80 | 0.298 | 0.54 | 1.21 | 0.77 | 0.301 | 0.47 | 1.26 | 0.79 | 0.493 | 0.40 | 1.56 |
| Asian | 0.45 | 0.011 | 0.25 | 0.83 | 0.40 | 0.023 | 0.18 | 0.88 | 0.64 | 0.378 | 0.24 | 1.71 |
| Native American/Alaskan Native | 1.73 | 0.074 | 0.95 | 3.17 | 1.81 | 0.091 | 0.91 | 3.60 | 1.91 | 0.184 | 0.74 | 4.93 |
| Sexual Orientation (gay/lesbian=referent) | ||||||||||||
| Bisexual/pansexual | 1.06 | 0.628 | 0.83 | 1.36 | 1.00 | 0.986 | 0.75 | 1.33 | 1.17 | 0.473 | 0.76 | 1.82 |
| Another sexual orientation | 1.30 | 0.202 | 0.87 | 1.95 | 1.24 | 0.377 | 0.77 | 1.98 | 1.25 | 0.556 | 0.59 | 2.65 |
| Assigned female at birth | 0.85 | 0.206 | 0.67 | 1.09 | 0.70 | 0.020 | 0.52 | 0.95 | 1.00 | 0.999 | 0.64 | 1.55 |
| Experienced parental rejection | 2.10 | 0.000 | 1.69 | 2.60 | 1.74 | 0.000 | 1.35 | 2.24 | 3.32 | 0.000 | 2.20 | 5.00 |
| U.S. Region (West=referent) | ||||||||||||
| Southwest | 1.16 | 0.438 | 0.80 | 1.67 | 1.41 | 0.127 | 0.91 | 2.19 | 0.84 | 0.575 | 0.45 | 1.56 |
| Midwest | 1.47 | 0.031 | 1.04 | 2.09 | 1.94 | 0.002 | 1.28 | 2.94 | 0.82 | 0.529 | 0.43 | 1.53 |
| Southeast | 1.19 | 0.292 | 0.86 | 1.64 | 1.44 | 0.069 | 0.97 | 2.12 | 0.86 | 0.582 | 0.50 | 1.48 |
| Northeast | 1.12 | 0.517 | 0.80 | 1.56 | 1.38 | 0.113 | 0.93 | 2.07 | 0.59 | 0.096 | 0.32 | 1.10 |
| Lives in urban area | 0.79 | 0.092 | 0.61 | 1.04 | 0.77 | 0.097 | 0.56 | 1.05 | 0.92 | 0.730 | 0.57 | 1.49 |
| Received free or reduced price lunch | 3.01 | 0.000 | 2.40 | 3.78 | 2.64 | 0.000 | 2.03 | 3.45 | 5.40 | 0.000 | 3.47 | 8.42 |
RQ1a: Do these correlates vary for those who experience couch-surfing only vs. couch-surfing and other types of homelessness?
Also shown in Table 3 are the results of the multinomial logistic regression exploring whether there are different correlates of SMA who experience couch-surfing only and multiple forms of homelessness, as compared to those without homelessness histories (Model 2). The significant correlates from Model 1 remain consistent as correlates for couch-surfing only among this sample – with older age (RRR=1.19; p=0.010), parental rejection (RRR=1.74; p<0.001), living in the Midwest (OR=1.94; p=0.002), and eligibility for free or reduced price lunch (RRR=2.64; p<0.001) associated with increased likelihood, and being Asian associated with a decreased likelihood of couch-surfing only (RRR=0.40; p=0.023). Additionally, couch-surfing only experiences, as compared to no homelessness, were less likely among participants who were assigned female at birth (RRR=0.70; p=0.020). However, when looking at SMA with multiple forms of homelessness experiences, only parental rejection (RRR=3.32; p<0.001) and eligibility for free/reduced price lunch (RRR=5.40; p<0.001) remained consistent with Model 1.
Among those who experience homelessness, what are the correlates of exclusive couch-surfing?
Table 4 (Model 3) shows the results of logistic regression examining correlates of couch-surfing only vs. those with multiple forms of homelessness among those with a lifetime homelessness experience. In this model, parental rejection (OR=0.53; p=0.008) and free or reduced price lunch (OR=0.43; p=0.001) were associated with reduced likelihood of couch-surfing only, as compared to multiple types of homelessness experiences, while those living in the Midwest (OR=2.23; p=0.028) or Northeast (OR=2.06; p=0.045), as compared to the West, were more likely to have couch-surfed only vs. having multiple forms of homelessness.
Table 4.
Correlates of Exclusive Couch-Surfing among SMA Experiencing Homelessness (Model 3; n=427)
| Odds Ratio | P>|z| | [95% Conf. Interval] | ||
|---|---|---|---|---|
| Age | 1.10 | 0.427 | 0.87 | 1.39 |
| Race/ethnicity (white=referent) | ||||
| Latinx/Hispanic | 1.08 | 0.820 | 0.57 | 2.05 |
| Multi-racial/multi-ethnic | 1.60 | 0.290 | 0.67 | 3.83 |
| Black | 0.82 | 0.633 | 0.37 | 1.82 |
| Asian | 0.55 | 0.360 | 0.15 | 1.99 |
| Native American/Alaskan Native | 1.03 | 0.956 | 0.36 | 2.99 |
| Sexual Orientation (gay/lesbian=referent) | ||||
| Bisexual/pansexual | 0.79 | 0.352 | 0.48 | 1.30 |
| Another sexual orientation | 1.02 | 0.958 | 0.44 | 2.40 |
| Assigned female at birth | 0.66 | 0.111 | 0.40 | 1.10 |
| Experienced parental rejection | 0.53 | 0.008 | 0.33 | 0.84 |
| U.S. Region (West=referent) | ||||
| Southwest | 1.62 | 0.201 | 0.77 | 3.39 |
| Midwest | 2.23 | 0.028 | 1.09 | 4.57 |
| Southeast | 1.57 | 0.160 | 0.84 | 2.96 |
| Northeast | 2.06 | 0.045 | 1.02 | 4.16 |
| Lives in urban area | 0.82 | 0.481 | 0.47 | 1.42 |
| Received free or reduced price lunch | 0.43 | 0.001 | 0.26 | 0.71 |
RQ2: Do SMA with homelessness experiences report higher rates of mental health symptoms and suicidality and does the relationship between homelessness, mental health symptoms and suicidality vary by type of homelessness experience (exclusive couch-surfing vs. other types of homelessness)?
Table 5 presents results of models examining how types of homelessness – as compared to no homelessness – are associated with mental health outcomes. Both measured types of homelessness – couch-surfing (OR=2.23; p<0.001) and multiple types of homelessness (OR=2.07; p=0.002) -- are associated with increased odds of symptoms indicative of moderate to severe anxiety (Model 4). Symptoms of depression (Model 5) are also higher for those with couch-surfing only (β=0.14; p<0.001) or multiple types of homelessness (β=0.11; p<0.001), as compared to those without homelessness experiences. Past year suicidal ideation (Model 6) and suicide attempt (Model 7) are also higher for couch-surfing only SMA (ideation: OR=2.04; p<0.001; attempt: OR=3.04; p<0.001) and SMA with multiple forms of homelessness (ideation: OR=3.40; p<0.001; attempt: OR=4.71; p<0.001), as compared to those without homelessness. For depression symptoms, suicidal ideation and suicide attempt, experiencing multiple types of homelessness appears to have a larger effect size relationship than couch-surfing only, though across all these models both types of homelessness experiences are consistently associated with worse mental health outcomes, relative to SMA who do not have homelessness experiences.
Table 5.
Associations between Mental Health & Suicide Outcomes with Types of Homelessness Experiences
| Model 4a (n=2,143) | Model 5b (n=2,140) | Model 6a (n=2,052) | Model 7a (n=2,021) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| ||||||||||||||||
| Moderate to Severe Anxiety (GAD-7) | Depression Symptoms (CES-D) | Lifetime Suicidal Ideation | Lifetime Suicide Attempt | |||||||||||||
|
| ||||||||||||||||
| Odds Ratio | P>|z| | [95% Conf. Interval] | β | P>|z| | [95% Conf. Interval] | Odds Ratio | P>|z| | [95% Conf. Interval] | Odds Ratio | P>|z| | [95% Conf. Interval] | |||||
| Type of Homelessness (none=referent) | ||||||||||||||||
| Couch-surfing only | 2.23 | 0.000 | 1.65 | 3.02 | 0.14 | 0.000 | 0.95 | 1.76 | 2.04 | 0.000 | 1.56 | 2.67 | 3.04 | 0.000 | 2.24 | 4.12 |
| Multiple types | 2.07 | 0.002 | 1.29 | 3.31 | 0.11 | 0.000 | 0.96 | 2.19 | 3.40 | 0.000 | 2.19 | 5.28 | 4.71 | 0.000 | 3.10 | 7.17 |
| Age | 0.86 | 0.001 | 0.78 | 0.94 | −0.07 | 0.001 | −0.39 | −0.11 | 0.78 | 0.000 | 0.71 | 0.85 | 0.76 | 0.000 | 0.67 | 0.86 |
| Race/ethnicity (white=referent) | ||||||||||||||||
| Latinx/Hispanic | 0.78 | 0.085 | 0.59 | 1.03 | −0.02 | 0.419 | −0.60 | 0.25 | 0.94 | 0.673 | 0.71 | 1.25 | 0.92 | 0.675 | 0.63 | 1.34 |
| Multi-racial/multi-ethnic | 1.07 | 0.684 | 0.76 | 1.51 | 0.01 | 0.687 | −0.40 | 0.61 | 1.26 | 0.182 | 0.90 | 1.76 | 1.02 | 0.928 | 0.65 | 1.60 |
| Black | 0.84 | 0.367 | 0.58 | 1.22 | −0.00 | 0.931 | −0.57 | 0.52 | 1.13 | 0.491 | 0.79 | 1.63 | 1.23 | 0.370 | 0.79 | 1.92 |
| Asian | 0.70 | 0.059 | 0.48 | 1.01 | −0.02 | 0.387 | −0.84 | 0.33 | 0.80 | 0.271 | 0.53 | 1.19 | 0.54 | 0.079 | 0.27 | 1.07 |
| Native American/Alaskan Native | 0.56 | 0.057 | 0.31 | 1.02 | −0.00 | 0.942 | −0.95 | 0.88 | 0.81 | 0.506 | 0.44 | 1.49 | 0.94 | 0.876 | 0.45 | 1.98 |
| Sexual Orientation (gay/lesbian=referent) | ||||||||||||||||
| Bisexual/pansexual | 1.06 | 0.567 | 0.86 | 1.31 | −0.02 | 0.504 | −0.42 | 0.20 | 1.20 | 0.081 | 0.98 | 1.48 | 1.05 | 0.734 | 0.80 | 1.38 |
| Another sexual orientation | 1.06 | 0.741 | 0.75 | 1.51 | 0.00 | 0.947 | −0.51 | 0.55 | 1.12 | 0.521 | 0.79 | 1.60 | 0.98 | 0.918 | 0.60 | 1.57 |
| Assigned female at birth | 0.56 | 0.000 | 0.46 | 0.69 | −0.21 | 0.000 | −1.83 | −1.21 | 0.58 | 0.000 | 0.47 | 0.72 | 0.80 | 0.116 | 0.60 | 1.06 |
| Experienced parental rejection | 1.80 | 0.000 | 1.49 | 2.19 | 0.15 | 0.000 | 0.73 | 1.30 | 1.82 | 0.000 | 1.50 | 2.20 | 1.67 | 0.000 | 1.30 | 2.15 |
| U.S. Region (West=referent) | ||||||||||||||||
| Southwest | 0.99 | 0.958 | 0.73 | 1.35 | 0.01 | 0.757 | −0.39 | 0.54 | 0.90 | 0.501 | 0.66 | 1.23 | 1.17 | 0.457 | 0.77 | 1.78 |
| Midwest | 1.11 | 0.493 | 0.82 | 1.50 | 0.01 | 0.720 | −0.37 | 0.53 | 1.07 | 0.650 | 0.79 | 1.45 | 0.96 | 0.842 | 0.64 | 1.45 |
| Southeast | 1.28 | 0.074 | 0.98 | 1.68 | 0.05 | 0.073 | −0.03 | 0.77 | 1.03 | 0.825 | 0.79 | 1.35 | 1.11 | 0.565 | 0.77 | 1.60 |
| Northeast | 0.80 | 0.094 | 0.61 | 1.04 | 0.02 | 0.315 | −0.20 | 0.61 | 0.96 | 0.763 | 0.73 | 1.26 | 1.05 | 0.794 | 0.72 | 1.54 |
| Lives in urban area | 0.96 | 0.728 | 0.75 | 1.23 | −0.05 | 0.016 | −0.82 | −0.09 | 0.73 | 0.010 | 0.57 | 0.93 | 0.80 | 0.160 | 0.59 | 1.09 |
| Received free or reduced price lunch | 1.20 | 0.077 | 0.98 | 1.48 | 0.06 | 0.007 | 0.11 | 0.72 | 1.04 | 0.648 | 0.86 | 1.28 | 1.52 | 0.002 | 1.17 | 1.98 |
Logistic regression
Linear regression
Other correlates of moderate to severe anxiety symptoms in these models include age (OR=0.86; p=0.001), being assigned female at birth (OR=0.56; p<0.001), and experiencing parental rejection (OR=1.80; p<0.001). Other correlates of depression symptoms include age (β= −0.07; p=0.001), being assigned female at birth (β= −0.21; p<0.001), experiencing parental rejection (β=0.15; p<0.001), urbanicity (β= −0.05; p=0.016), and free or reduced price lunch (β =0.06; p=0.007). Other correlates of suicidal ideation include age (OR=0.78; p<0.001), being assigned female at birth (OR=0.58; p<0.001), experiencing parental rejection (OR=1.82; p<0.001), and urbanicity (OR=0.73; p=0.010). Other correlates of suicide attempt include age (OR=0.76; p<0.001), experiencing parental rejection (OR=1.67; p<0.001), and free or reduced price lunch (OR=1.52; p=0.002).
Table 6 presents analyses of the relationship between couch-surfing only and mental health symptoms/suicide attempts only among those with homelessness experiences. There were no differences in anxiety or depression symptoms based on type of homelessness experience among those with lifetime homelessness (Models 8 and 9), but exclusive couch-surfers had a lower likelihood of suicidal ideation (OR=0.58; p=0.036) and suicide attempt (OR=0.59; p=0.031) than SMA with multiple forms of homelessness.
Table 6.
Associations between Couch-Surfing (vs. other types of homelessness) and Mental Health & Suicide Outcomes
| Model 8a (n=425) | Model 9b (n=426) | Model 10a (n=413) | Model 11a (n=403) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| ||||||||||||||||
| Moderate to Severe Anxiety (GAD-7) | Depression Symptoms (CES-D) | Lifetime Suicidal Ideation | Lifetime Suicide Attempt | |||||||||||||
|
| ||||||||||||||||
| Odds Ratio | P>|z| | [95% Conf. Interval] | β | P>|z| | [95% Conf. Interval] | Odds Ratio | P>|z| | [95% Conf. Interval] | Odds Ratio | P>|z| | [95% Conf. Interval] | |||||
| Couch-Surfing Only | 1.00 | 0.989 | 0.56 | 1.79 | −0.04 | 0.469 | −0.99 | 0.46 | 0.58 | 0.036 | 0.35 | 0.96 | 0.59 | 0.031 | 0.37 | 0.95 |
| Age | 1.00 | 0.979 | 0.77 | 1.30 | 0.02 | 0.660 | −0.26 | 0.41 | 0.75 | 0.021 | 0.59 | 0.96 | 0.73 | 0.006 | 0.58 | 0.92 |
| Race/ethnicity (white=referent) | ||||||||||||||||
| Latinx/Hispanic | 0.70 | 0.325 | 0.35 | 1.42 | −0.03 | 0.527 | −1.23 | 0.63 | 0.50 | 0.032 | 0.26 | 0.94 | 0.98 | 0.953 | 0.52 | 1.85 |
| Multi-racial/multi-ethnic | 1.47 | 0.432 | 0.56 | 3.87 | 0.00 | 0.984 | −1.12 | 1.14 | 1.19 | 0.661 | 0.55 | 2.57 | 1.14 | 0.732 | 0.54 | 2.41 |
| Black | 0.91 | 0.844 | 0.34 | 2.44 | −0.03 | 0.497 | −1.59 | 0.77 | 0.80 | 0.576 | 0.36 | 1.76 | 1.77 | 0.137 | 0.83 | 3.77 |
| Asian | 0.15 | 0.004 | 0.04 | 0.54 | −0.07 | 0.183 | −3.27 | 0.63 | 0.32 | 0.072 | 0.09 | 1.11 | 0.68 | 0.558 | 0.19 | 2.48 |
| Native American/Alaskan Native | 0.32 | 0.026 | 0.11 | 0.87 | 0.05 | 0.336 | −0.82 | 2.39 | 0.30 | 0.021 | 0.11 | 0.83 | 0.28 | 0.060 | 0.08 | 1.05 |
| Sexual Orientation (gay/lesbian=referent) | ||||||||||||||||
| Bisexual/pansexual | 0.71 | 0.229 | 0.40 | 1.24 | 0.02 | 0.670 | −0.55 | 0.86 | 0.97 | 0.915 | 0.60 | 1.58 | 0.84 | 0.481 | 0.52 | 1.36 |
| Another sexual orientation | 0.81 | 0.655 | 0.33 | 2.01 | 0.08 | 0.120 | −0.24 | 2.11 | 1.56 | 0.287 | 0.69 | 3.52 | 2.01 | 0.074 | 0.94 | 4.33 |
| Assigned female at birth | 0.47 | 0.006 | 0.27 | 0.81 | −0.14 | 0.006 | −1.75 | −0.30 | 0.63 | 0.060 | 0.39 | 1.02 | 0.76 | 0.288 | 0.47 | 1.25 |
| Experienced parental rejection | 1.59 | 0.075 | 0.95 | 2.64 | 0.14 | 0.006 | 0.27 | 1.57 | 1.86 | 0.006 | 1.19 | 2.91 | 1.58 | 0.045 | 1.01 | 2.47 |
| U.S. Region (West=referent) | ||||||||||||||||
| Southwest | 1.37 | 0.475 | 0.58 | 3.22 | 0.03 | 0.656 | −0.84 | 1.33 | 0.63 | 0.278 | 0.33 | 1.38 | 1.02 | 0.966 | 0.48 | 2.17 |
| Midwest | 1.38 | 0.421 | 0.63 | 3.05 | 0.01 | 0.915 | −0.96 | 1.08 | 1.28 | 0.481 | 0.64 | 2.58 | 1.03 | 0.926 | 0.51 | 2.11 |
| Southeast | 1.49 | 0.301 | 0.70 | 3.16 | 0.12 | 0.055 | −0.02 | 1.88 | 1.19 | 0.597 | 0.62 | 2.29 | 1.55 | 0.188 | 0.81 | 2.97 |
| Northeast | 0.73 | 0.408 | 0.35 | 1.53 | 0.08 | 0.216 | −0.37 | 1.65 | 1.27 | 0.501 | 0.63 | 2.55 | 1.59 | 0.195 | 0.79 | 3.18 |
| Lives in urban area | 1.17 | 0.610 | 0.64 | 2.14 | 0.04 | 0.428 | −0.46 | 1.09 | 1.14 | 0.627 | 0.67 | 1.94 | 0.93 | 0.795 | 0.55 | 1.59 |
| Received free or reduced price lunch | 1.64 | 0.064 | 0.97 | 2.76 | 0.05 | 0.315 | −0.33 | 1.02 | 1.48 | 0.095 | 0.93 | 2.35 | 1.37 | 0.189 | 0.86 | 2.19 |
Logistic regression
Linear regression
Other correlates of moderate to severe anxiety in the models restricted to those experiencing homelessness include race/ethnicity (Asian: OR=0.15; p=0.004; Native American/Alaska Native: OR=0.32; p=0.026) and being assigned female at birth (OR=0.47; p=0.006). Depression symptoms were less likely among those assigned female at birth (β= −0.14; p=0.006) and more likely among those experiencing parental rejection (β=0.14; p=0.006). Suicidal ideation and attempt were associated with age (ideation: OR=0.75; p=0.21; attempt: OR=0.73; p=0.006), and ideation was less likely among Latinx/Hispanic (OR=0.50; p=0.032) and Native American/Alaska Native (OR=0.30; p=0.021) participants, and more likely among those with parental rejection experiences (OR=1.86; p=0.006).
Discussion
Several key findings emerge from the current study. We found that sexual minority adolescents reported high rates of homelessness. In our sample, 21% of SMA reported homelessness experiences and 92% of those with homelessness experience reported couch-surfing. The prevalence of couch-surfing among SMA in our study highlights their high level of engagement with informal supports when faced with a housing loss and corroborates previous research indicating SMA/Y experiencing homelessness are more likely to couch-surf compared to their heterosexual peers (Petry et al., 2022). Moreover, this finding calls further attention to complications surrounding definitions of homelessness that differ across the federal agencies tasked with addressing youth homelessness. Definitions that do not account for the varied forms and experiences of youth homelessness lead to the legislative invisibility of SMA/Y and contribute further to their maringalization by the mainstream homeless services system. Inclusive definitions that recognize the unique vulnerability of couch-surfers can bolster the visibility of minoritized youth and pave the way for a more equitable system of care.
Several correlates of homelessness among SMA were also identified. Youth who identified as Asian were less likely to report couch-surfing, whereas older age and living in the Midwest was associated with an increased likelihood of couch-surfing. However, the latter two were not correlates of other types of homelessness. It is possible that a lower concentration of services — especially youth-specific services — in small/midsize cities (Ferguson et al. 2012; Bowen et al. 2017), leads to a greater reliance on personal networks in order to meet needs, resulting in a greater likelihood of couch-surfing. It is also possible that, even in the presence of youth-specific services, SMA may prefer to engage their personal networks to meet their housing needs due to a desire for identity-affirming resources, more positive experiences with service providers, and a sense of personal agency (Samuels et al., 2018).
Findings from this study provide additional support for existing research that sexual minority young people who experience parental rejection are more likely to experience homelessness (Gattis 2013; Katz-Wise, Rosario, and Tsappis 2016; Pearson, Thrane, and Wilkinson 2017). Although couch-surfers were more likely to report parental rejection and low SES compared to SMA with no homelessness experience, they indicated less parental rejection, less economic adversity, and were more likely to be located in the Midwest relative to the West when compared to SMA with other homelessness experiences. These findings underscore the unique circumstances of couch-surfing SMA and contextualize these experiences within broader experiences of homelessness among SMA and can help inform national responses to youth homelessness. Parental rejection and economic adversity may be less acute issues for couch-surfing SMA than for their peers experiencing other forms of homelessness, but they are still important contributing factors to their housing insecurity. Homelessness prevention and intervention programs should aim to support the personal networks of SMA while also investing in family reunification efforts that help parents decrease the risks for and increase the well-being of SMA. Providing time-limited housing assistance to enable SMA to reside with extended family members or other trusted persons within their network while service providers assist in negotiating family reunification efforts can help ensure safety and stability for SMA. Future research should investigate how parental rejection and economic adversity contribute to varying experiences of homelessness among SMA, how their personal support networks might differ, and how these factors influence their housing.
Confirming existing research (Bruce et al. 2014; Rhoades et al. 2018; Walls, Potter, and Leeuwen 2009), homelessness experiences among SMA in this study were found to be associated with significantly higher rates of mental health disorder symptoms, in particular past year suicide attempt, which is reported at a rate 3–5 times higher among SMA with homelessness experiences in this study. Overall, both couch-surfing adolescents and those experiencing multiple types of homelessness reported more anxiety, depression, and suicide attempts than their peers who had not experienced homelessness. These findings suggest that couch-surfing, while less visible and often perceived as less serious than other forms of homelessness, is nevertheless associated with many of the same mental health disparities identified among youth with more traditional homelessness experiences. Previous research in Australia showed couch-surfers to be more likely to have a diagnosed mental health concern (Hail-Jares, Vichta-Ohlsen, and Nash 2020), perhaps indicating couch-surfers’ ability to access mental health services relative to youth experiencing other forms of homelessness. However, this is invariably dependent upon the services infrastructure of a given location and warrants further investigation—especially given the barriers to accessing homeless services cited by couch-surfing youth in the U.S. (McLoughlin 2013). Regardless, this study’s findings highlight the vulnerability of couch-surfers and the risk of serious mental health concerns faced by couch-surfing SMA relative to their housed peers. Additional research is needed to identify best practices for identifying couch-surfing youth so that schools and youth-serving organizations can help intervene and connect these young people with the appropriate supports, including mental health resources. Health care professionals should also assess for housing status and stability and support the mental health of youth who may be experiencing homelessness. In addition to screening practices, policies allowing unaccompanied youth experiencing homelessness—including couch-surfers—to independently access and consent to mental health care are paramount to ensuring the safety and well-being of SMA. Future research should determine the most effective strategies for mental health outreach to SMA in various living situations and to explore how different service delivery models might improve mental health care access for SMA experiencing homelessness.
While both couch-surfing only and multiple forms of homelessness were each associated with increased odds of anxiety, the slightly larger effect size observed among couch-surfers may reflect the aforementioned psychological burdens of couch-surfing, including the relative uncertainty of their housing situation (McLoughlin 2013). Anxiety may be less acute for youth experiencing other forms of homelessness, in part due to the degree of self-reliance required to survive on the streets.
Homeless services are often considered as the last safety net for this vulnerable population; however, SMA/Y often to not have access to LGBTQ+ culturally competent shelters or services, and as such, these services can perpetuate biased and prejudiced environments that may exacerbate negative mental health consequences (Prock and Kennedy 2020). Additionally, many SMA/Y experiencing homelessness use emergency shelters and services which do not have the resources to systematically address trauma that is unfortunately so commonplace for this population (Ream and Forge 2014).
Given the complex nature of familial relationships and mental health disparities among SMA, homelessness services and providers should understand and be responsive to the unique needs of SMA, including recognizing that family reunification may not be possible or recommended in all cases. Homelessness service providers should conduct anti-discrimination training, build cultural competency, and provide LGBTQ-affirming services and legal protections to SMA youth experiencing homelessness (Dolamore and Naylor 2018; Ferguson and Maccio 2015). Additionally, agencies and services providers working on reducing mental health disparities among SMA need to be cognizant of additional stressors associated with homelessness experiences. There is therefore a need for also changing the climate in a range of human services (including child welfare systems and schools) where discriminatory services for SMA may further exacerbate their disproportionate burden of negative mental health outcomes.
Prior research into couch-surfing suggests that experiences vary and while not all couch-surfing experiences necessarily meet a threshold necessitating homelessness services, many do (Curry et al. 2017). Furthering these findings, the striking associations found in this paper between exclusive couch-surfing experiences and steeply increased rates of anxiety, depression, and particularly suicidal ideation and attempt, suggest that SMA who report couch-surfing are likely in need of a wide of supportive services and should not be excluded from services targeted to youth experiencing homelessness.
Limitations and Conclusions
As these data are cross-sectional, this study cannot make causal inferences about the relationship between exclusive couch-surfing, multiple forms of homelessness and mental health outcomes; future research should examine longitudinal relationships among these factors to better identify temporal action points for the most effective interventions. We also recognize that because this study focused on cisgender youth, it ignored the important disparities experienced by transgender and gender nonbinary adolescents, particularly trans and nonbinary youth of color (Shelton and Bond 2017). Further, experiences of homelessness and couch-surfing for young people can be complex (Curry et al., 2017) and there may be SMA in this study whose couch-surfing experiences do not meet a threshold that we would consider homelessness; future research may want to explore more nuanced ways of assessing housing instability among adolescents.
Despite these limitations, this manuscript highlights several potentially important touchpoints for intervention with a vulnerable population at high risk for negative mental health outcomes. Firstly, by identifying correlates of homelessness, this paper expands our knowledge of which SMA are more or less likely to experience any homelessness, multiple forms of homelessness, and exclusive couch-surfing. Secondly, these findings provide further evidence of the stark negative mental health disparities experienced by SMA with homelessness experiences. Finally, these findings underscore the need of exclusive couch-surfing youth to receive attention in research and intervention, as their mental health vulnerabilities are similar to their peers who have more “traditional” homelessness experiences.
Footnotes
The age groups included in specific research studies cited here vary, but in general we define adolescents as those ages 12–17 and use ‘youth’ to be inclusive of studies with populations up to age 24, many of which have respondent populations that span both adolescence and young adulthood (typically defined as 18–24 years of age).
References
- Bowen EA, Miller B, Barman-Adhikari A, Fallin K, and Zuchlewski D. 2017. “Emerging Adult Homelessness in Geographic Perspective: A View from the Rust Belt.” Children and Youth Services Review 73: 213–19. [Google Scholar]
- Bruce D, Stall R, Fata A, and Campbell RT 2014. “Modeling Minority Stress Effects on Homelessness and Health Disparities among Young Men Who Have Sex with Men.” Journal of Urban Health 91 (3): 568–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Castellanos HD. 2016. “The Role of Institutional Placement, Family Conflict, and Homosexuality in Homelessness Pathways among Latino LGBT Youth in New York City.” Journal of Homosexuality 63(5): 601–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cleary SD 2000. “Adolescent victimization and associated suicidal and violent behaviors.” Adolescence 35 (140): 671. [PubMed] [Google Scholar]
- Curry SR, Morton M, Matjasko JL, Dworsky A, Samuels GM, and Schlueter D. 2017. “Youth Homelessness and Vulnerability: How Does Couch-Surfing Fit?” American Journal of Community Psychology 60: 17–24. [DOI] [PubMed] [Google Scholar]
- Curry SR, VanMeeter M, Tully BA, Ault S, Garst N, Adam K, Nesmith A, and White J. 2020. “Beyond a Bed: Supportive Connections Forged between Youth Who Are Couch Hopping and Adult Hosts.” Child and Adolescent Social Work Journal 28: 12–26. [Google Scholar]
- Cutuli JJ, Treglia D, and Herbers JE 2020. “Adolescent Homelessness and Associated Features: Prevalence and Risk across Eight States.” Child Psychiatry & Human Development 51 (1): 48–58. [DOI] [PubMed] [Google Scholar]
- Dolamore, and Naylor LA 2018. “Providing Solutions to LGBT Homeless Youth: Lessons From Baltimore’s Youth Empowered Society.” Public Integrity 20 (6): 595–610. [Google Scholar]
- Durso LE, and Gates GJ. 2012. “Serving Our Youth: Findings from a National Survey of Services Providers Working with Lesbian, Gay, Bisexual and Transgender Youth Who Are Homeless or at Risk of Becoming Homeless.” 2012. http://williamsinstitute.law.ucla.edu/research/safe-schools-and-youth/serving-our-youth-july-2012/. [Google Scholar]
- Ferguson KM, Bender K, Thompson SJ, Maccio EM, and Pollio D. 2012. “Employment Status and Income Generation among Homeless Young Adults: Results from a Five-City, Mixed-Methods Study.” Youth & Society 44 (3): 385–407. [Google Scholar]
- Ferguson KM, Jun J, Bender K, Thompson S, Pollio D, and Louis. 2010. “A Comparison of Addiction and Transience among Street Youth.” Missouri. Community Mental Health Journal 46: 296–307. [DOI] [PubMed] [Google Scholar]
- Ferguson, and Maccio EM 2015. “Promising Programs for Lesbian, Gay, Bisexual, Transgender, and Queer/Questioning Runaway and Homeless Youth.” Journal of Social Service Research 41 (5): 659–83. [Google Scholar]
- Gattis MN 2013. “An Ecological Systems Comparison between Homeless Sexual Minority Youths and Homeless Heterosexual Youths.” Journal of Social Service Research 39 (1): 38–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haas AP EM, Mays VM, and Mathy RM. 2011. “Suicide and Suicide Risk in Lesbian, Gay, Bisexual, and Transgender Populations.” Review and Recommendations. Journal of Homosexuality 58 (1): 10–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hail-Jares K, Vichta-Ohlsen R, Butler T, and Dunne A. 2021. “Psychological Distress among Young People Who Are Couchsurfing: An Exploratory Analysis of Correlated Factors.” Journal of Social Distress and Homelessness, 1–5.
- Hail-Jares K, Vichta-Ohlsen R, and Nash C. 2020. “Safer inside? Comparing the Experiences and Risks Faced by Young People Who Couch-Surf and Sleep Rough.” Journal of Youth Studies.
- Katz-Wise SL, Rosario M, and Tsappis M. 2016. “LGBT youth and family acceptance.” Pediatric Clinics of North America 63 (6): 1011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kidd SA, Gaetz S, and O’Grady B. 2017. “The 2015 National Canadian Homeless Youth Survey: Mental Health and Addiction Findings.” The Canadian Journal of Psychiatry 62(7): 493–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Krause KD, Kapadia F, Ompad DC, D’Avanzo PA, Duncan DT, and Halkitis PN. 2016. “Early Life Psychosocial Stressors and Housing Instability among Young Sexual Minority Men: The P18 Cohort Study.” Journal of Urban Health 93(3): 511–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Marshal MP, Dermody SS, Cheong J, Burton CM, Friedman MS, Aranda F, and Hughes TL 2013. “Trajectories of Depressive Symptoms and Suicidality among Heterosexual and Sexual Minority Youth.” Journal of Youth and Adolescence 42 (8): 1243–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Marshal MP, Dietz LJ, Friedman MS, and a. 2011. “Suicidality and Depression Disparities between Sexual Minority and Heterosexual Youth: A Meta-Analytic Review.” Journal of Adolescent Health 49 (2): 115–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- McLoughlin PJ 2013. “Couch-Surfing on the Margins: The Reliance on Temporary Living Arrangements as a Form of Homelessness among School-Aged Home Leavers.” Journal of Youth Studies 16 (4): 521–45. [Google Scholar]
- Melchior LA, Huba G, Brown VB, and Reback CJ 1993. “A short depression index for women.” Educational and Psychological Measurement 53 (4): 1117–25. [Google Scholar]
- Morton MH, Dworsky A, Matjasko JL, Curry SR, Schlueter D, Chavez R, and Farrell AF. 2018. “Prevalence and Correlates of Youth Homelessness in the United States.” Journal of Adolescent Health 62(1): 14–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Moskowitz A, Stein JA, and Lightfoot M. 2013. “The Mediating Roles of Stress and Maladaptive Behaviors on Self-Harm and Suicide Attempts among Runaway and Homeless Youth.” Journal of Youth and Adolescence 42(7): 1015–27. [DOI] [PubMed] [Google Scholar]
- Noell JW, and Ochs LM. 2001. “Relationship of Sexual Orientation to Substance Use, Suicidal Ideation, Suicide Attempts, and Other Factors in a Population of Homeless Adolescents.” J Adolesc Health 29(1): 31–36. [DOI] [PubMed] [Google Scholar]
- Pearson J, Thrane L, and Wilkinson L. 2017. “Consequences of Runaway and Thrownaway Experiences for Sexual Minority Health during the Transition to Adulthood.” Journal of LGBT Youth 14 (2): 145–71. [Google Scholar]
- Prock KA, and Kennedy AC 2020. “Characteristics, Experiences, and Service Utilization Patterns of Homeless Youth in a Transitional Living Program: Differences by LGBQ Identity.” Children and Youth Services Review 116: 105176. [Google Scholar]
- Ream GL, and Forge NR 2014. “Homeless Lesbian, Gay, Bisexual, and Transgender Youth in New York City: Insights from the Field.” Child Welfare 93 (2): 7.26030986 [Google Scholar]
- Rhoades H, Rusow JA, Bond D, Lanteigne A, Fulginiti A, and Goldbach JT. 2018. “Homelessness, Mental Health and Suicidality among LGBTQ Youth Accessing Crisis Services.” Child Psychiatry & Human Development, 1–9. [DOI] [PubMed]
- Samuels GM, Cerven C, Curry SR, and Robinson SR 2018. “Nothing Is for Free...”: Youth Attitudes about Engaging Resources While Unstably Housed.” Cityscape 20 (3): 35–68. [Google Scholar]
- Schmitz RM, and Tyler KA. 2017. “The Complexity of Family Reactions to Identity among Homeless and College Lesbian, Gay, Bisexual, Transgender, and Queer Young Adults.” Archives of Sexual Behavior, 1–13. [DOI] [PubMed]
- Schrager SM, Goldbach JT, and Mamey MR 2018. “Development of the Sexual Minority Adolescent Stress Inventory.” Frontiers in Psychology 9 (319). [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shelton, and Bond L. 2017. “It Just Never Worked Out”: How Transgender and Gender Expansive Youth Understand Their Pathways into Homelessness.” Families in Society 98 (4): 284–91. [Google Scholar]
- Showden C, and Majc S. 2018. Youth Who Trade Sex in the United States. Philadelphia, PA: Temple University Press. [Google Scholar]
- Spitzer RL, Kroenke K, Williams JB, and Löwe B. 2006. “A Brief Measure for Assessing Generalized Anxiety Disorder: The GAD-7.” Archives of Internal Medicine 166 (10): 1092–97. [DOI] [PubMed] [Google Scholar]
- Suchting R, Businelle MS, Hwang SW, Padhye NS, Yang Y, and Santa Maria DM 2020. “Predicting Daily Sheltering Arrangements among Youth Experiencing Homelessness Using Diary Measurements Collected by Ecological Momentary Assessment.” International Journal of Environmental Research and Public Health 17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tyler KA, Johnson KA, and Melander LA 2010. “A Comparison of Homeless and High-Risk Young Adults: Are They One and the Same?” Vulnerable Children and Youth Studies 5 (2): 122–29. [Google Scholar]
- Tyler KA, Olson K, and Ray CM 2020. “Short Message Service Surveying with Homeless Youth: Findings from a 30-Day Study of Sleeping Arrangements and Well Being.” Youth & Society 52 (5): 850–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Van Leeuwen JM, S Boyle S Salomonsen-Sautel, and Baker DN. 2006. “Lesbian, Gay, and Bisexual Homeless Youth: An Eight-City Public Health Perspective.” Child Welfare 85(2): 151. [PubMed] [Google Scholar]
- Walls NE, Potter C, and Leeuwen J. 2009. “Where Risks and Protective Factors Operate Differently: Homeless Sexual Minority Youth and Suicide Attempts.” Child and Adolescent Social Work Journal 26 (3): 235–57. [Google Scholar]