Predictors of Housing Trajectories Among Young Adults Experiencing Homelessness in Los Angeles

Homelessness has adverse consequences on the health and development of young people, and young adults who experience homelessness are a large and understudied population who are often marginalized. Nearly 3.5 million young adults (ages 18–24 years) in the USA experience homelessness annually,¹ with over 27,000 young adults in the USA experiencing homelessness on a given night.² Experiencing homelessness during young adulthood is associated with both immediate and long-term negative health and psychological outcomes, including behavioral health problems (e.g., posttraumatic stress disorder, depression), physical health complications, and heavy use of alcohol, cannabis, and illicit drugs.^3‐7 These physical and behavioral health problems serve as both determinants and consequences of other challenges faced by young people who experience homelessness, including physical and sexual violence, risky sexual behaviors, and legal difficulties.^8‐11 Young adults who experience homelessness are also more likely to experience early mortality compared to those who are more stably housed.¹²

Understanding factors associated with both stable and unstable housing among young adults experiencing homelessness is a necessary first step towards designing programs and policies to reduce the negative physical and mental health effects. The experience of homelessness is a heterogenous one,¹³ making longitudinal studies outlining the course of housing and its determinants crucial in providing a more in-depth understanding of these issues. Existent longitudinal studies have examined housing trajectories of young people experiencing homelessness. For example, Milburn and colleagues¹⁴ examined 183 adolescents experiencing homelessness in Los Angeles and found that social and familial support (i.e., engagement with prosocial peers, maternal support) and attendance in school predicted stable housing outcomes over 2 years, whereas exposure to family violence and reliance on use of shelter services predicted less stable housing. Braciszewski and colleagues¹⁵ found in a 7-year longitudinal study of 243 adolescents that previous homelessness, racial/ethnic minority status, and neighborhood income were related to more difficulty securing stable housing. In another study of 359 Canadian young adults experiencing homelessness, Roy and colleagues¹⁶ found that factors related to social integration (e.g., having a high school degree, seeking psychological services) were associated with housing stability over 90 days, whereas factors related to street entrenchment (e.g., injection drug use, informal income sources such as from selling drugs or panhandling) were associated with housing instability. Other studies assessing mixed samples of both adolescents and young adults in the USA have found that risk factors for housing instability included illicit drug use and engagement in high-risk sexual behaviors, whereas strong social support networks, younger age, less time homeless, having the ability to return home if needed (e.g., not forced out of their home by parents), and receipt of behavioral health services were all protective.^17‐20

Though the available studies help with our understanding of trajectories of homelessness among young people, most focus on adolescents or have samples comprised of both adolescents and young adults. Young adults are in a crucial developmental period, where they start experiencing more autonomy from parents and begin developing their own identity, including establishing financial independence and fostering intimate relationships. It is also a time, after the age of 18, when there are less protections from local, state, and federal government, making options for housing and educational opportunities more limited for young adults experiencing homelessness. Experiencing homelessness during this age period can have significant implications for future outcomes,^21‐23 and thus, an increased understanding of housing trajectories of young adults experiencing homelessness in the USA may aid in the development of evidence-based public health programs and policies designed to serve this age group. Moreover, though research has established that mental, physical, and social health outcomes are affected by homelessness, it is less clear how these factors are associated with prospective housing trajectories among young adults. Moreover, young adults’ experiences of homelessness are often heterogeneous,¹³ with some young adults spending periods of time in their own home, interspersed with temporary housing accommodations with family, friends, or strangers (“couch surfing”); in shelters; or in highly unstable and dangerous unsheltered settings. Thus, in addition to examining stable housing as an optimal outcome, studies need to examine predictors of unsheltered housing trajectories, which encompass living in outdoor locations, vehicles, or abandoned buildings. Such experiences can be particularly detrimental for young adults, even if experienced temporarily.^{24, 25}

The present study addresses gaps in the literature by examining predictors of trajectories of housing stability and unsheltered housing, grouped into four key areas pertinent to young adults experiencing homelessness: (1) demographics: age, birth sex, sexual and gender minority identification, race and ethnicity, and age of first homeless experience; (2) substance use behaviors: heavy drinking days, cannabis use days, and days of illicit drug use; (3) health and social functioning; and (4) use of housing services. Establishing the association between these factors (measured at a single point-in-time) and prospective housing trajectories over 2 years is crucial to advancing efforts to support the housing needs of these young adults.

The current study examined 2-year housing trajectories in a sample of 271 young adults initially experiencing homelessness in Los Angeles, focused on housing stability (i.e., living in one’s own home every day) and being unsheltered (i.e., spending at least one night on the street, in a vehicle, or in an abandoned building) in the past 3 months. Regarding housing stability, several factors associated with trajectories of young adults report of living in their own home every day for the past 3 months. After accounting for other factors significant in bivariate models, both female birth sex and better quality of one’s friendships were associated with a greater likelihood of being stably housed at baseline, and reporting multi-racial/other race/ethnic identification (compared to White) and better friendship quality were associated with a less pronounced increase in the probability of being stably housed over time. Findings suggested that targeted outreach efforts to provide stable housing interventions for male young adults and those identifying with multiple racial/ethnic identities may be necessary. As in prior work, social support (examined in this study as self-reported quality of friendships) was associated with stable housing cross-sectionally at baseline.^{17, 19, 20} Strong peer relationships have been associated with increased use of drop-in centers and higher-level services.^{9, 32, 33} and also reduce the mental and physical health consequences of homelessness as young people have friends to rely on for a place to stay or to help them in times of mental or physical health needs.^{34, 35} To our surprise, however, greater social support at baseline was associated with a lower probability of stable housing over time. It is possible that stronger friend networks may help address young adults’ housing needs for a brief period of time (i.e., “couch surfing”), but staying with friends is inherently an unstable long-term housing situation. It is also possible that young adults experiencing homelessness develop strong relationships within “street families,”³⁶ which can mimic traditional family roles and be difficult to give up if transitioning to a more stable housing setting. Entrenchment in street life has also been associated with less housing stability in other work.¹⁶ Thus, to encourage obtaining more stable housing, interventions that target young adults more fully engrained in street life may be necessary.

Regarding trajectories of being unsheltered for at least one night in the past 3 months, several factors were significant at the bivariate level (reporting sexual and gender minority identity, being in school, being employed, cannabis and illicit drug use, and friendship quality). However, when all variables were included in the model, only being in school and being employed were associated with a lower probability of being unsheltered at baseline, and being employed was associated with a less steep decrease in the probability of being unsheltered over time. This is contrary to other work, which showed that engagement in school and work may be important protective factors against being unsheltered, a particularly high-risk living situation.^{1, 37, 38} It should be noted that we did not assess forms of employment, so it is unclear if reported employment was formal/steady jobs with a paycheck or more “off-the-books” types of employment. Frequency of illicit drug use was associated with a greater probability of being unsheltered at baseline, consistent with prior research showing that drugs besides alcohol and cannabis are typically associated with greater risk of being unsheltered among young adults experiencing homelessness.^{18, 39} Interestingly, greater illicit drug use at baseline was associated with a more pronounced decrease in the probability of being unsheltered over time, potentially indicating that young adults reporting more problematic substance use could have been receiving targeted housing services (shelter accommodations or permanent supportive housing) or been prioritized for receiving alcohol and other drug services due to severity.

Results should be considered with several limitations in mind. First, our sample was limited to young adults in the Los Angeles area, and although the sample was diverse and reflected the broader population of homeless young adults in Los Angeles, findings may not be generalizable to populations outside of Los Angeles. In addition, all data, including reports of pregnancy and illicit drug use, were based on self-report, which has potential for respondent bias. Sexual and gender minority participants were included as one group for analyses, and small sample sizes within each specific identity (e.g., transgender or gender non-binary participants, lesbian cisgender women or gay men) did not allow for meaningful examination of trajectories by these unique groups. Lastly, though relevant literature was reviewed to inform selection of factors associated with prospective housing trajectories in our analyses, there are other factors that may affect housing which were not included, such as exposure to family violence and neighborhood/environmental level factors. More research is needed on both risk and protective factors that may contribute to stable housing to provide a better understanding of the services needed to alleviate the burden of homelessness among young people.

In conclusion, this study fills an important gap in the literature on young adults experiencing homelessness. Though studies exist, there are few longitudinal studies of young adult housing outcomes. It is important to continue to understand factors associated with both stable and unstable housing among young adults experiencing homeless in the effort to develop programs and policies to reduce homelessness and its negative physical and mental health effects on young people.

Though it was not surprising that use of housing services was negatively associated with stable housing at baseline (those with stable housing tended to use housing services less because they did not need the services), use of housing services at baseline was not associated with trajectories of housing stability or being unsheltered over time. Housing services are often not a primary reason for young adults’ use of drop-in center services,³² because these centers tend to focus on providing basic needs services (food, showers); this is also shown in our drop-in center sample here, with participants using housing services on an average of 3 to 5 days in the past 3 months at baseline. Helping engage youth in drop-in center services is a necessary first step towards connecting young adults to services that address their housing and health needs. In addition, none of the physical or mental health factors (e.g., depression, general health) was associated with either stable housing or unsheltered housing trajectories. It is possible that because these young adults were recruited from drop-in centers, they were more likely to have their health needs met (e.g., physical health was rated an average of “very good”), and perhaps those with depression (30% met criteria in our sample) were already receiving some services at the drop-in for symptoms.^{40, 41} Still, poor mental and physical health are clearly linked to homelessness,⁴² and continued outreach, assessment, and intervention are necessary to meet young adults’ needs.