Learn more: PMC Disclaimer | PMC Copyright Notice
Challenges Faced by Homeless Sexual Minorities: Comparison of Gay, Lesbian, Bisexual, and Transgender Homeless Adolescents With Their Heterosexual Counterparts
Abstract
Objectives. The goal of this study was to identify differences between gay, lesbian, bisexual, and transgender (GLBT) homeless youths and their heterosexual counterparts in terms of physical and mental health difficulties.
Methods. A sample of 84 GLBT adolescents was matched in regard to age and self-reported gender with 84 heterosexual adolescents. The 2 samples were compared on a variety of psychosocial variables.
Results. GLBT adolescents left home more frequently, were victimized more often, used highly addictive substances more frequently, had higher rates of psychopathology, and had more sexual partners than heterosexual adolescents.
Conclusions. Homeless youths who identify themselves as members of sexual minority groups are at increased risk for negative outcomes. Recommendations for treatment programs and implications for public health are discussed.
Homeless youths represent a diverse population that reaches the streets for a variety of reasons and whose numbers have grown in recent decades. Gay, lesbian, bisexual, and transgender (GLBT) homeless youths face the obstacle of survival on the streets as well as the stigma of sexual minority group membership.
It is difficult to estimate the proportion of GLBT youths in the street population.1–3 The National Network of Runaway and Youth Services4 estimates that about 6% of homeless adolescents are gay or lesbian. However, the few studies assessing sexual orientations of homeless adolescents have revealed rates ranging from 11% to 35%.5,6
Among adolescents in general, GLBT youths are more vulnerable to health and psychological problems than are heterosexual youths. Many are victims of parental physical abuse, are substance abusers, and have both mental and general physical health problems.7–10 These problems may be amplified for GLBT youths who become homeless.
Homeless youths are vulnerable to victimization, including robbery, rape, and assault.11–15 Also, homelessness often leads to initiation or escalation of substance use.16–24 High rates of externalizing and internalizing problems, including psychosis, have been found among this population.25–32 Moreover, high rates of risky sexual behavior, including prostitution and survival sex (sex in exchange for money, drugs, or shelter), place these young people at risk for victimization and sexually transmitted diseases.33–36
The current study was the first of its kind to examine psychosocial outcomes for GLBT homeless youths. The objectives of the study were to identify risks faced by GLBT homeless youths and to determine whether these risks transcend those of their heterosexual counterparts. Our hypothesis was that GLBT homeless youths would have more negative indicators of psychological and physical well-being than do heterosexual homeless adolescents.
METHOD
Participants
The sample consisted of 375 adolescents aged 13 to 21 years (mean: 17.14 years); data were collected between 1995 and 1998. Youths were recruited for the Seattle Homeless Adolescent Research and Education project at street locations or social service agencies in the Seattle metropolitan area. Youths were eligible to participate if they spoke English, had not lived in the residence of a primary caretaker for at least 1 week, and had no stable home in which to live.
Trained outreach workers informed youths of their right to refuse participation, skip individual questions, or stop participation at any time. Informed consent was obtained from all participants, and youths were offered $25 for taking part. The overall response rate was 95%. The sampling methods were described in full in an earlier report.37
The majority of study participants identified themselves as White (52.5%). Other participants self-identified as American Indian or Alaska Native (18.9%), African American (17.6%), Hispanic/Latino (7.2%), or Asian/Pacific Islander (2.7%). About half of the sample was female (45.1%).
Self-reports of sexual orientation (heterosexual, gay, lesbian, bisexual, transsexual, or transgender) were used to classify the majority of participants. In a few cases, youths did not select one of the choices (n = 9) or selected multiple choices (n = 12). Questions about same- and opposite-sex attraction were used to classify these individuals as members of sexual minorities (n = 6) or as heterosexuals (n = 15). Individuals who both rated their same-sex attraction higher than the midpoint on a 7-point Likert scale and reported same-sex sexual behavior were classified as sexual minorities.
The majority of the 84 sexual minority youths identified themselves as bisexual (n = 71). Only 4 female and 8 male youths self-identified as exclusively lesbian or gay, and 1 youth self-identified as transgender. GLBT youths were matched in terms of age and gender with an equal number of self-identified heterosexual participants; thus, the study included a total of 168 participants.
Measures
Private, face-to-face structured interviews (1.5 to 2 hours in duration) were conducted with each youth. Participants were asked about their reasons for leaving home, street victimization, alcohol and drug use, and sexual behaviors. In all, 21 reasons for leaving home were provided, and youths were asked to select any reason that influenced their decision to leave. Youths were also asked whether they had encountered any physical or sexual victimization after they had left home, and if so, how often it had occurred (e.g., how often they had been assaulted or forced to have sex).
In the substance use section, participants were asked how often they had used all major forms of drugs and alcohol in the previous 6 months (ratings were made on a 7-point scale ranging from not at all [0] to every day [6]). Sexual behavior questions focused on areas such as number and gender of sexual partners and frequency of safe-sex practices. Depressive symptoms were measured with the Center for Epidemiologic Studies Depression Scale (CES-D),38,39 and other behavior problems were assessed with Achenbach's Youth Self-Report (YSR).40 A full description of measures is available from the authors.
Data Analyses
Given our primary goal of determining how GLBT youths differed from heterosexual youths in several domains, we had no a priori multivariate predictions about the relationships among variables. Therefore, we conducted primarily t tests (for continuous variables) and, in some cases, χ2 tests (for dichotomous variables).
RESULTS
Pathways to Homelessness
By and large, GLBT youths left home for reasons similar to those of their heterosexual counterparts. However, GLBT youths left home more often than did heterosexual youths (means of 12.38 times and 6.69 times, respectively; t160 = 1.91, P = .058). The most common reasons reported by youths for leaving home were family conflict (59.9%), desire for freedom (51.5%), and difficulties with a family member (48.5%). GLBT youths were more likely to leave as a result of physical abuse in the home (χ21 = 3.6, P = .044), and there was a trend toward more GLBT youths leaving as a result of alcohol use in the home (χ21 = 3.2, P = .055). Twelve (14.3%) GLBT youths indicated that they had left home because of conflicts with their parents over their sexual orientation.
Victimization
GLBT youths experienced higher levels of physical victimization than their heterosexual counterparts (Table 1 ▶). When analyses were conducted separately by gender, this effect was significant for male youths during the preceding 3 months and for female youths since the onset of their homelessness. Also, GLBT youths had more often been sexually victimized since the time they had first become homeless. When examined by gender, this effect was found among male youths but not among female youths. In addition, GLBT youths reported an average of 7.4 more perpetrators of sexual victimization than did heterosexual youths.
TABLE 1—
Physical and Sexual Victimization of Sexual Minority and Heterosexual Homeless Adolescents: Seattle, Wash, 1995–1998
| Measure | Sexual Minorities, Mean (SD) | Heterosexuals, Mean (SD) | t (df) | P |
| Physical victimization | ||||
| Since homeless | 1.88 (0.70) | 1.66 (0.72) | 1.98 (165) | .05 |
| Male | 2.10 (0.72) | 2.00 (0.77) | 0.58 (73) | .56 |
| Female | 1.70 (0.64) | 1.39 (0.54) | 2.49 (90) | .02 |
| During past 3 months | 1.45 (0.50) | 1.25 (0.44) | 2.67 (166) | .008 |
| Male | 1.67 (0.58) | 1.39 (0.55) | 2.21 (74) | .03 |
| Female | 1.26 (0.32) | 1.14 (0.29) | 1.87 (90) | .07 |
| Sexual victimization | ||||
| Since homeless | 1.52 (0.75) | 1.27 (0.52) | 2.46 (161) | .02 |
| Male | 1.36 (0.68) | 1.05 (0.16) | 2.65 (71) | .01 |
| Female | 1.65 (0.78) | 1.46 (0.63) | 1.31 (88) | .19 |
| During past 3 months | 1.45 (0.82) | 1.24 (0.56) | 1.80 (138) | .07 |
| Male | 1.50 (0.99) | 1.09 (0.37) | 2.25 (65) | .03 |
| Female | 1.40 (0.63) | 1.37 (0.66) | 0.21 (71) | .83 |
| No. of perpetrators | 8.61 (24.06) | 1.24 (3.76) | 2.69 (156) | .009 |
| Male | 6.74 (22.95) | 0.17 (0.45) | 1.70 (69) | .01 |
| Female | 10.09 (25.07) | 2.14 (4.93) | 2.06 (85) | .05 |
Note. For all measures except number of perpetrators, possible responses ranged from 1 (never) to 4 (many times).
Substance Abuse
For each substance category except marijuana, sexual minority youths had used the substance more frequently in the previous 6 months than had heterosexuals (Figure 1 ▶). Differences were significant for 3 types of drugs: cocaine or crack (respective means of 0.81 and 0.33 on the 7-point scale; t166 = 2.75, P = .007), crack or crack mixed with amphetamines (means of 1.02 and 0.46, respectively; t166 = 2.43, P = .016), and speed or crystal methamphetamines (means of 1.56 and 0.74; t166 = 2.88, P = .005). Sexual minority youths also used more types of drugs than did heterosexuals (means of 7.88 and 6.51, respectively; t166 = 2.39, P = .018).
Mental Health
There was a trend for GLBT youths to report higher levels of depressive symptoms (as measured with the CES-D) than did heterosexual youths (t147 = 1.43, P = .09). Also, YSR subscale scores indicated that GLBT youths had higher rates of psychopathology (Figure 2 ▶). Independent-samples t tests revealed significant differences in the following domains: withdrawn behavior (t142 = 2.28, P = .024), somatic complaints (t141 = 2.07, P = .041), social problems (t150 = 2.03, P = .044), delinquency (t142 = 2.07, P = .040), aggression (t149 = 2.45, P = .016), internalizing behavior (t146 = 2.38, P = .018), and externalizing behavior (t148 = 2.71, P = .007). The total YSR mean difference was also significant; overall, GLBT adolescents reported higher levels of symptomatology (t149 = 3.24, P = .001).
Sexual Behavior
Most adolescents (94%) reported engaging in voluntary sex at least once; however, GLBT youths reported a higher number of lifetime sexual partners than did heterosexual youths (means of 24.19 and 12.49, respectively; t143 = 2.84, P = .005). GLBT youths were significantly younger than heterosexuals in regard to age at first voluntary intercourse (means of 13.00 and 13.83 years, respectively; t154 = 2.11, P = .037). Also, GLBT youths reported high rates of unprotected intercourse, endorsing a mean of 2.91 for this item (approximately “half of the time”), in comparison with a mean of 2.51 among heterosexual adolescents (t115 = 1.08, NS). Although this difference was not statistically significant, more than twice as many GLBT youths as heterosexual youths reported that they neglected to use protection during sex “all of the time.”
DISCUSSION
The present results indicate negative outcomes in multiple domains for GLBT homeless adolescents. These outcomes include more-frequent departures from home, greater vulnerability to physical and sexual victimization, higher rates of addictive substance use, more psychopathology, and riskier sexual behavior in comparison with homeless heterosexual adolescents.
The coping model used by MacLean et al.21 to explain substance use among homeless adolescents may be extended to these findings. GLBT homeless adolescents experience not only the vulnerabilities, daily difficulties, and survival challenges of living on the street but also the discrimination faced by GLBT youth in general. In coping with these stressors, they may use more substances and use them more frequently than do heterosexual youths. Other psychological problems may be direct effects of this discrimination or may be intensified by coping-focused substance abuse.
GLBT adolescents face great challenges as they work to come to terms with their sexual orientation. Those living in homes forge this identity within the context of their family and school or peer networks, facing risks of isolation, rejection, and sometimes victimization by others. Their homeless counterparts, however, frequently have no family members available, no school environment to support them, and transient or insufficient peer networks.
Primary interventions directed toward these youths should focus on preventing initial and recurrent episodes of homelessness by providing therapeutic services to assist families in dealing with adolescents' sexual identity, thus improving the home environment and reducing the likelihood that GLBT adolescents will leave. Secondary interventions that could benefit GLBT homeless adolescents would involve recognition of the particular difficulties faced by this group and provision of services sensitive to issues of sexual orientation.
Community public health programs must recognize the increased risks faced by homeless GLBT youths and the need for assisting this population. First, it is important to ask adolescents about their sexual orientation, both to identify their need for services and to indicate that sexual orientation is an acceptable topic of discussion. Second, intensified services similar to those described by Tenner et al.6 could be tailored to homeless GLBT youths to reduce risks of HIV infection. Third, programs should acknowledge and address the contribution of homophobia to the etiology and maintenance of substance abuse problems. Finally, programs that encourage acceptance of sexual minorities among street youths and in shelters may reduce the risk of GLBT adolescents being victimized in these locales. Further studies may increase our understanding of this often-overlooked population.
Our study involved some limitations that should be noted. All of the data were obtained through self-reports, and most domains were assessed via single measurements. Our sexual minority group was largely bisexual, and results may not generalize to homeless adolescents who self-identify as exclusively lesbian, gay, or transgender. However, sexual orientation is a dynamic variable subject to developmental changes,3 and future studies may elucidate whether and how outcomes change as homeless adolescents redefine their sexual identities. Finally, our analyses were limited to bivariate tests. Given our interest in exploring differences between sexual minority and heterosexual homeless adolescents in a number of different domains, our approach was appropriate; in the future, however, we will use multivariate modeling techniques in an attempt to improve our understanding of the challenges faced by GLBT youths in the context of homelessness.
Acknowledgments
This study was supported by a research grant (AA10253-03) from the National Institute on Alcohol Abuse and Alcoholism.
We thank Brynn Malone for her contributions to the editing of the manuscript and Kevin Wruck and Yvette Lohr for their conscientious data collection efforts.
Notes
Peer Reviewed
B. N. Cochran conceptualized the project, conducted data analyses, and contributed to the literature review and the writing of the manuscript. A. J. Stewart contributed to the study design, to the literature review and data analysis, and to the writing and review of the manuscript. J. A. Ginzler assisted with the substance abuse outcome data analysis and the literature review and linked findings to the risk amplification model. A. M. Cauce contributed to study design and implementation and to the writing and review of the manuscript.