Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2007 Dec 13.
Published in final edited form as: Sex Res Social Policy. 2006 Mar;3(1):13–21. doi: 10.1525/srsp.2006.3.1.13

Resilience Among Young Men Who Have Sex With Men in New York City

Marya Viorst Gwadz 1, Michael C Clatts 2, Huso Yi 3, Noelle R Leonard 4, Lloyd Goldsamt 5, Steve Lankenau 6,✉
PMCID: PMC2136414  NIHMSID: NIHMS21976  PMID: 18079993

Abstract

This article describes a study of resilience among young men who have sex with men (YMSM). Resilience is defined as positive adaptation in the context of hardship. Using targeted sampling to capture the diversity and range of this hidden population, we recruited 569 YMSM ages 17–28 years old and examined a subset of 134 YMSM who had experienced severe childhood adversity, as indicated by placement in foster care. Most of the YMSM in this subset were from racial or ethnic minority backgrounds and fewer than half identified as gay or homosexual (46.3 percent). More than half (58.3 percent) exhibited positive outcomes on four of seven indicators of adaptive functioning. YMSM who identified as either bisexual or heterosexual exhibited lower rates of resilience. Structural- as well as individual-level factors appear to be implicated in resilience among YMSM. Findings underscore the importance of fostering stable sexual identity as a means of building resilience.

Keywords: sexual identity, homosexual, bisexual, transgender, adaptation, foster care

Young men who have sex with men (YMSM) are at greater risk for poor health and mental health outcomes in comparison to their peers in the general population. A high prevalence of substance abuse, HIV infection, and mental health problems has been routinely documented in this population (Centers for Disease Control and Prevention [CDC], 2001; O’Donnell, Agronick, San Doval, Duran, Myint-U, & Stueve, 2002; Thiede et al., 2003; Valleroy et al., 2000). Among YMSM, those who do not identify as gay or homosexual and those who consider themselves transgender appear to be at greatest jeopardy for adverse outcomes (Agronick et al., 2004; Cooper, 1999; Hughes & Eliason, 2002). Overall, YMSM are a hidden and hard-to-reach population, mainly because of stigma and often because of their involvement in illegal activities (MacKellar, Valleroy, Karon, Lemp, & Janssen, 1996; Muhib et al., 2001). In particular, YMSM who are involved in the street economy, have substance abuse problems, are from low socioeconomic status backgrounds, do not identify as gay, are younger, and/or are from racial/ethnic minority backgrounds may be more difficult to access for treatment and research than their peers (Barendregt, van der Poel, & van de Mheen, 2005; Muhib et al., 2001; Van Ness, Davis, & Johnson, 2004) and therefore may be viewed as hidden subpopulations. As a result, these young men are understudied and underrepresented in clinical settings since few targeted outreach efforts have focused on them (Clatts, Davis, & Atillasoy, 1995).

Resilience in Populations at Risk

While multiple unfavorable outcomes exhibited by YMSM have been well documented in the literature, there is also growing interest in better understanding resilience in populations at risk (Anderson, 1998; Russell, 2005; Savin-Williams, 1990). Resilience is conceptualized as a process of positive adaptation in the context of significant adversity (Luthar, Cicchetti, & Becker, 2000). Resilience is typically inferred by examining the functioning of individuals who have experienced hardship in their early upbringings in various domains later in life. Those who exhibit both early hardship and later positive outcomes are considered resilient (Masten, 2001; Rutter, 1985). For instance, McGloin and Widom (2001) studied adults with past substantiated cases of childhood sexual abuse or neglect. About a quarter (22 percent) were considered resilient by meeting positive outcomes in a minimum of six of the following eight domains: employment, stable housing, education, social activity, mental health, substance abuse, criminal justice system involvement, and violence. While such an examination of the presence or absence of resilience has great utility, resilience may, in fact, be multidimensional in nature. That is, within an individual, adaptive functioning may vary across domains and in different contexts (Freitas & Downey, 1998; Luthar, Doernberger, & Zigler, 1993).

Aims

The primary aim of this article is to describe levels of positive adaptation among YMSM and to explore the demographic correlates of positive outcomes in order to identify subgroups that are either thriving or exhibiting particular deficits. In order to study resilience, a criterion for early childhood adversity must first be selected. One could argue that nearly all YMSM experience at least some level of childhood hardship, given that they often must manage a stigmatized identity and behavior. However, in keeping with the predominant use of the construct of resilience, we applied a stringent criterion of prior history of childhood adversity. Specifically, we selected a sub-sample that had a history of placement in foster care from a larger cohort of YMSM. Foster care placement is generally an outcome of child maltreatment or neglect and is a robust indicator of serious childhood adversity. Further, it is strongly associated with later poor outcomes (Rosenfeld et al., 1997; Rutter, 2000). Consistent with McGloin and Widom (2001), we examined the extent of adaptive functioning in our sample across a range of indices. However, in contrast to McGloin and Widom, we did not examine resilience as a dichotomous state. Instead, we studied the degree to which YMSM exhibited adaptive functioning across a range of domains in relationship to demographic characteristics associated with these outcomes in this population.

Methods

Procedures

Targeted sampling is a technique for recruiting a sample from a hidden population, including YMSM, for whom traditional sample survey methods such as random sampling often yield low numbers of eligible participants (MacKellar et al., 1996; Muhib et al., 2001). As described by Muhib and colleagues, targeted sampling entails identifying days and times when the target population gathers at specific venues, constructing a sampling frame of venues, randomly selecting and visiting venues, and recruiting members of the target population from venues. The resulting sample is therefore reflective of the diversity of the population (Watters & Biernacki, 1989). The aim of targeted sampling is not, however, to recruit a representative sample; the method is constructed to capture a population’s variability and range of characteristics. As such, it yields comprehensive and complex descriptions of the population, including capturing subgroups that would likely be missed by other sampling methods.

We followed this approach for the present study. In 2 years of ethnographic fieldwork, we mapped 75 diverse potential venues where YMSM were known to congregate in various parts of New York City (Lankenau, Clatts, Welle, Goldsamt, & Gwadz, 2005). Venues were selected to reach YMSM with a wide range of behavioral characteristics, including those who were both homeless and stably housed, street-involved (e.g., participating in drug dealing and sex work) and employed in the formal economy, and from a range of socioeconomic and ethnic/racial backgrounds. Venues included bars, clubs, cafes, parks, street hangouts, indoor and outdoor sites where sex work took place (e.g., bus stations, outdoor strolls), and community-based organizations. Recruitment procedures are described in greater detail elsewhere (Gwadz, Clatts, Leonard, & Goldsamt, 2004), and we provide a summary of these methods below.

Venues were first combined into four primary categories, based on the demographic and behavioral similarities of the subpopulations found there: (1) bars, clubs, and cafes, (2) street hangouts and parks, (3) sites where sex work took place, and (4) community-based organizations. Within each category, the specific venues from the ethnographic phase were ranked according to the number and proportion of YMSM found. About 75 percent of recruitment visits were scheduled for bars/clubs/cafes and street hangouts/parks, the predominant venue types, and the remainder from sex work venues and community-based organizations. Prior to beginning sampling each month, venues were randomly selected and visited by staff researchers. Young men were screened for eligibility within the venues. Inclusion criteria included being male, age 17 to 28 years, and having had sexual contact with a man in the past 6 months. In order to mask screening criteria, screening questions included information about residence, health, recent sexual behavior with both male and female partners, and substance use. Approximately 80 percent of those approached participated in the screening interview. Almost all (>80 percent) were eligible, and 80 percent of those eligible participated in an hour-long structured interview, generally on the same day. From 43 venues we recruited 569 YMSM; 32 percent of the participants were recruited from bars/clubs/cafes; 42 percent from outdoor street hangouts/parks; 16 percent from sex work venues; and 10 percent from community-based agencies. The study procedures, including a plan to enroll 17-year-olds without parental consent, were approved by the Joint Institutional Review Board of the National Development and Research Institutes, Inc.

Measures

Outcome Variable: Adaptive Functioning

Modeled on McGloin and Widom (2001), our measure of adaptive functioning was comprised of seven indicators of positive behaviors and outcomes. Each of the following was coded as present (1) or absent (0): (1) employment in the formal economy; (2) stable housing (that is, not residing in a shelter, on the street, or in a marginal or temporary situation such as a single-room-occupancy hotel); (3) in school or having graduated from high school, including receiving a General Educational Development test certificate (GED); (4) absence of criminal justice system involvement, including being held by police or incarceration; (5) positive mental health, as indicated by a subclinical score (less than 7) on a standard measure, the Center for Epidemiological Studies-Depression Scale (CES-D; Huba et al., 1995; Melchior, Huba, Brown, & Reback, 1993); (6) social support, represented by a score at or above the sample median (median = 3) on a 4-item scale of the presence or absence of social support (range 0 to 4); and (7) no or only casual hard drug use in the past month, that is, use of crack, cocaine, heroin, methamphetamine, MDMA/Ecstasy, and/or ketamine restricted to less than 3 days in the past month. The resulting quantitative variable ranged from 0 to 7.

Independent Variables

Demographic and background characteristics

We assessed age, race/ethnicity, sexual orientation, trans-gender status, living situation, age of first foster care placement (range 0 to 18), and sources of financial support in the formal, informal, and street economies. The Hollingshead Four Factor Index (Hollingshead, 1975) was used to calculate family socioeconomic status.

Life events

Life events were assessed using a modified version of the Psychiatric Epidemiological Research Interview Life Events Scale (Dohrenwend, Krasnoff, Askenasy, & Dohrenwend, 1978; Huba et al., 1996), assessing the lifetime prevalence of 14 family events. The scale was reliable in this sample (Cronbach’s α = .70).

Childhood victimization events

Childhood victimization events, characterized as physical and sexual abuse and neglect, were assessed with a 9-item scale (Widom & Morris, 1997; Widom, Weiler, & Cottler, 1999). Reliability for the scale was high in this sample (Cronbach’s α = .80) and items were summed to create a quantitative variable ranging from 0 to 9, consistent with other research that indicates that types of abuse are more similar than they are different (Mullen, Martin, Anderson, Romans, & Herbison, 1996). The measure has good discriminant validity when compared to court-documented victimization experiences (Widom & Shepard, 1996).

Data Analyses

We examined characteristics of variables using measures of central tendency and variability. In tests of statistical significance, we used t-tests for continuous variables and chi-square tests for categorical variables. We examined correlations among independent variables and in a multivariate model to ensure we did not have high levels of multicollinearity (data available from first author). Categorical variables were recoded as dummy variables; the reference group for sexual identity variables was gay/homosexual, and the reference group for race/ethnicity variables was White/Caucasian. We performed two ordinary least squares (OLS) regression models, entering independent variables in a single block first for the cohort with higher childhood adversity and, for the purpose of comparison, for the lower childhood adversity cohort. We included five demographic and background characteristics relevant for YMSM: age, socioeconomic status, race/ethnicity, sexual orientation, and transgender status. Analyses were conducted in SPSS 10.0 for Windows (SPSS Inc., Chicago, IL, U.S.).

Results

Participants

The present article focuses principally on the sub-sample of YMSM who had demonstrably higher rates of childhood adversity, that is, those who had been placed in foster care in the past (N = 134). In Table 1, we provide a comparison of the demographic and background characteristics of those with lower (N = 435) and higher childhood adversity. Compared to their peers with lower childhood adversity, YMSM with higher childhood adversity were slightly younger and originated from lower socioeconomic status family backgrounds. They were more likely to be Latino/Hispanic, to identify as heterosexual or bisexual, to be homeless, and to be involved in the street economy. They were less likely to be White and to identify as gay or homosexual. They had first been placed in foster care at an average age of 10 years (SD = 4.81). Because childhood adversity is not limited to those who had been placed in foster care, we also examined rates of two other forms of childhood maltreatment. Those in the higher-adversity subgroup had experienced a greater number of different types of childhood maltreatment (Widom & Morris, 1997) and serious adverse life events (Huba et al., 1996) compared to those in the lower-adversity group. Thus, although childhood adversity was not entirely absent among the lower-adversity group, these findings provide support for the utility of foster care placement as an indicator of serious childhood hardship.

Table 1.

Comparison of Demographic and Background Characteristics Between Lower and Higher Childhood Adversity History Groups

Characteristics Lower (N = 435) % or M (SD) Higher (N = 134) % or M (SD)
Age 21.8 (2.9) 21.1 (2.9) *
 17–19 years 25.0 30.8
 20–22 years 33.3 41.0
 23–25 years 31.1 14.9
 26–28 years 10.4 12.6
Socioeconomic Status Index 37.6 (16.4) 32.1 (17.6) **
 Level V (highest level) 18.1 33.6
 Level IV 19.5 18.5
 Level III 12.3 9.2
 Level II 30.4 23.5
 Level I (lowest level) 19.7 15.1
Race/ethnicity
 White 31.0 12.7***
 African American/Black 21.1 27.6
 Latino/Hispanic 37.7 49.3*
 Other 10.1 10.4
Sexual identity
 Gay, homosexual 68.7 46.3***
 Bisexual 20.7 33.6**
 Heterosexual 5.7 11.9*
 Other or none-identified 4.8 8.2
Transgendera 8.1 11.2
Age first in foster care/group home — 9.93 (4.81)
 0–3 years — 10.9
 4–6 years — 18.0
 7–9 years — 11.7
 10–12 years — 21.1
 13–15 years — 28.9
 16–18 years — 9.4
Childhood maltreatment (0–9) 1.74 (1.71) 4.10 (2.69) ***
Serious adverse life events (0–13) 3.53 (2.35) 6.24 (3.10) ***
Ever homeless 17.2 53.7***
Currently homeless 6.9 25.4***
Street economy involvement— current
 Drug economy 5.1 15.7***
 Panhandling 1.8 9.7***
 Stealing, mugging, robbing 0.7 7.5***
 Sex work 14.0 47.0***
Open in a new tab
a

YMSM reported sexual identity and transgender status separately.

*

p < .05.

**

p < .01.

***

p < .001.

Indicators of Adaptive Functioning

A lower proportion of YMSM with higher childhood adversity had positive outcomes on all seven indices compared to those with lower childhood adversity (see Table 2). Among those with higher rates of childhood adversity, about half (58.3 percent) exhibited four or more indicators of adaptive functioning and more than a third (39.6 percent) showed five or more indicators (see Table 3). For example, one such pattern would entail an individual with casual hard drug use, stably housed, who had graduated from high school, with at least three people he could call on for social support, but who was not employed, had significant rates of depressive symptoms, and had been involved with the criminal justice system (i.e., 4 out of 7 indicators). Among those with lower rates of childhood adversity, 86.9 percent reported four or more indicators, and 71.5 percent exhibited five or more indicators.

Table 2.

Comparison of Indicators of Adaptive Functioning Between Lower and Higher Childhood Adversity History Groups

Indicators of Adaptive Functioning Childhood Adversity History
Lower (N = 435) Higher (N = 134)
% %
Employed 73.3 55.2***
Stably housed 93.1 74.6***
In school or graduated high school 87.8 68.7***
No criminal justice system involvement 63.4 18.7***
Positive mental health 54.5 35.8***
Social support 72.0 59.7**
No/casual hard drug use 79.3 70.1*
Adaptive functioning score M (SD) 5.23 (1.57) 3.83 (1.71)***
Open in a new tab
*

p < .05.

**

p < .01.

***

p < .001.

Table 3.

Comparison of Frequency Distribution of Adaptive Functioning Scores Between Lower and Higher Childhood Adversity Groups

Score Childhood Adversity History
Lower (N = 435) Higher (N = 134)
% %
0 0.7 4.5
1 2.3 6.7
2 3.2 8.2
3 6.9 22.4
4 15.6 18.7
5 20.9 22.4
6 26.0 14.2
7 24.4 3.0
Open in a new tab

Multivariate Models

YMSM with higher childhood adversity histories

Next, we conducted analysis using OLS hierarchical regression. For the groups with higher adversity, we entered demographic and background characteristics in a single block. The dependent variable was adaptive functioning (range 0 to 7). As shown in Table 4, these characteristics accounted for approximately 17 percent of the variance in adaptive functioning. Those who identified as either bisexual or heterosexual had lower rates of adaptive functioning. Those who were transgender tended to have lower rates of adaptive functioning and those in the “other” racial/ethnic category (e.g., White, Asian, or biracial/ethnic) tended to have higher rates of adaptive functioning.

Table 4.

Hierarchical Linear Regression Model for Cohort With Higher Childhood Adversity

B SE β
Age −0.01 0.05 −0.10
Socioeconomic status 0.001 0.008 0.12
Race/ethnicity
Black 0.57 0.53 0.15
Hispanic 0.48 0.49 0.14
Other 1.20 0.65 0.22†
Sexual self-identity
Bisexual −0.83 0.33 −0.23*
Heterosexual −1.31 0.46 −0.25**
Other −0.13 0.56 −0.02
Transgender −0.84 0.47 −0.16†
Open in a new tab

Note. B = unstandardized coefficient; SE = standard error; total R2 = .17.

†

p ≤.10.

*

p ≤ .05.

**

p ≤ .01.

YMSM with lower childhood adversity histories

We also conducted a corresponding analysis with the sub-sample of YMSM with lower childhood adversity for the purposes of comparison (data not shown). Results were similar to those found with the higher-adversity cohort in that bisexual identity (β = − .15, p ≤ .01), heterosexual identity (β = −.16, p ≤ .001), and transgender status (β = −.16, p ≤ .001) were negatively associated with adaptive functioning. In addition, among this subgroup SES was positively associated (β = .17, p ≤ .001), and being Hispanic (β = −.13, p ≤ .05) was negatively associated, with adaptive functioning. The model accounted for 16 percent of the variance in adaptive functioning. For both models, variables did not exhibit high levels of multi-collinearity at the bivariate or multivariate levels, and the observed residuals fell on, or very close to, the normal distribution, thus meeting the assumption of normality.

Discussion

In the present study we focused on a population of late adolescents and young adults at high risk for detrimental health and mental health outcomes, YMSM with substantial experiences of childhood adversity. Despite the emphasis in the literature on the risk factors YMSM experience and the deficits they evidence, we found that those with significant historical risk factors exhibited a combination of risk behavior and resilience to a substantial degree. Further, this study demonstrated the utility in examining the extent to which individuals exhibited positive outcomes, rather than in simply characterizing resilience as present or absent. While we know of no gold standard for the level at which an individual would be considered to be resilient for these data, we did find that almost all YMSM exhibited at least some level of adaptive functioning. For example, over half the group with higher levels of childhood adversity demonstrated positive outcomes on 4 or more indicators of adaptive functioning and more than a third on 5 or more indicators. Thus, although very high levels of resilience may not have been the norm in this group, indicators of adaptive functioning were at least not scarce or infrequent. This is consistent with Masten’s (2001) contention that resilience, rather than being rare, is actually a commonplace phenomenon. Yet we do not wish to dismiss the impact of childhood hardship on development. We found that YMSM who experienced serious childhood adversity were significantly less likely to exhibit positive outcomes when compared to their peers who did not. Thus these earlier negative experiences appear to contribute to detrimental outcomes among YMSM, even into late adolescence and adulthood.

When compared to their heterosexual peers in the general population, gay-identified youth exhibited substantial difficulties in numerous areas of functioning (Lock & Steiner, 1999; Sullivan & Wodarski, 2002). Yet in the present article, YMSM who identified as heterosexual or bisexual exhibited lower levels of adaptive functioning compared to their gay-identified peers. Heterosexually identified young men may have sex with men for a variety of reasons. In most, or at least some, cases, this sexual activity takes place for pay, typically as an economic survival strategy, and is an oftentimes stressful or traumatic and occasionally risky activity (Farley, Baral, Kiremire, & Sezgin, 1998; Rietmeijer, Wolitski, Fishbein, Corby, & Cohn, 1998; Valera, Sawyer, & Schiraldi, 2001). Although youth who identify as bisexual have received less attention in the literature than gay-identified individuals, these findings suggest that they may be similar to heterosexual YMSM in some respects. Further, our findings are consistent with other research studies that have found bisexuals to be at heightened risk for sexual risk behavior and adverse outcomes, including participation in sex work (Agronick et al., 2004; Logan & Leukefeld, 2000; National Institute on Drug Abuse, 2000). We expect that factors other than participation in sex work are involved in lower rates of resilience in these two important subpopulations, and future research on these groups is warranted. Yet our findings highlight the potential burden and complexity of a sexual identity that may be at odds with one’s behavior, driven in some cases by economic concerns, and that may be in flux and causing internal conflict. In addition, these findings indicate the potential significance of a stable gay or homosexual identity. Although hard-won in a largely homophobic society, such an identity may serve as a protective factor, as it may guide behavior and provide access to an important supportive community (Cross & Epting, 2005; Dworkin & Yi, 2003; Halpin & Allen, 2004; Sullivan, 2003).

The present study raised an additional issue that deserves further attention. We found that YMSM are placed in foster care at relatively late ages: on average, at the age of 10 years. This raises questions about whether YMSM are placed out-of-home at least partly in response to their emerging sexual attractions to and/or sexual activities with males as they approach puberty. Herdt and McClintock (2000) speculated that age 10 is critical in the development of attraction and sexuality, perhaps even a human universal. If YMSM are indeed being placed out-of-home in response to their developing awareness of sexual attractions to other males, a better understanding of the effect of such a placement on development, including on sexual identity development, is needed. Further, research efforts can endeavor to understand the potential role of homophobia in families and the larger culture as it contributes to youth being placed out-of-home and how to best intervene to strengthen families and prevent foster care placements for gay-identified youth. At present, little is known about the timing, duration, and factors underlying foster care placement for YMSM.

Although the principal focus in the present article was on YMSM with higher levels of childhood adversity—in keeping with the literature on resilience, which requires early hardship as a precondition for resilience—we also explored adaptive functioning among those who did not report a history of foster care placement. Interestingly, correlates of resilience were generally similar between the two subgroups, suggesting the importance of identity and factors that contribute to identity development, independent of levels of childhood adversity. In addition, among this lower-adversity subgroup, transgender individuals had lower rates of adaptive outcomes than their peers who do not identify as transgender (among the higher-adversity subgroup this tended to be the case). Transgender individuals are a vulnerable population, as they experience heightened stigma and additional barriers to care as a result of this identity (Hughes & Eliason, 2002; Mathy, Lehmann, & Kerr, 2003). Last, a Latino/Hispanic background was associated with lower rates of resilience, a finding similar to that in other recent research (Diaz, Ayala, Bein, Henne, & Marin, 2001; Frank & Lester, 2001). However, the cultural, socioeconomic, and other factors promoting risk or resilience in this group are not currently well understood.

Furthermore, among the lower-adversity cohort, socioeconomic status was positively associated with adaptive functioning, highlighting the relevance of structural and contextual factors in resilience. While an individual’s own behavior, skills, relationships, and choices no doubt play a significant role in whether he meets age-appropriate milestones and functions successfully, structural and institutional factors also affect behavior and outcomes. For example, we found in the present study that YMSM had commonly been in police custody or in jail. While YMSM can to a certain extent control whether they enter police custody, based on avoidance of illegal behavior, there are certainly also instances when they are targeted by police in the absence of misdeeds. Youth of color, in particular, may be disproportionately targeted by law enforcement officials, and those who are out-of-home and on the streets may come into more frequent contact with police as well (Parker, MacDonald, Alpert, Smith, & Piquero, 2004; Reid & Klee, 2000; Rosenbloom & Way, 2004). In New York City, there have been reports that gay-identified youth have been targeted in response to community pressure (Karp, 2002; Lee, 2002). Transgender individuals reported discrimination in the workplace, pressuring them to derive economic support from the street economy (Clements, Wilkinson, Kitano, & Marx, 1999). The degree to which resilience is dependent on structural factors, and how these can be addressed, merits future investigation.

We found utility in examining resilience among this group of young men at risk. Yet several limitations should be noted. First, we wish to reiterate that foster care placement is an indicator of severe childhood adversity. There is little question that serious child maltreatment and neglect can go undetected, raising the possibility that YMSM who are not placed in foster care actually experienced high levels of adversity. However, the utility of this criterion was supported by the fact that those with foster care placements had significantly higher rates of other forms of childhood maltreatment and stress. Second, these data are cross-sectional, and thus we are restricted in our ability to examine how these factors operate over time. Longitudinal studies are needed to better understand these developmental processes. Third, data were collected by self-report and were retrospective, raising the likelihood of recall and reporting biases. Last, although we used a targeted sampling plan to capture maximum diversity in this population, much of which is hidden, we cannot make definitive claims regarding the generalizability of these findings to the larger population, in part because the larger population is not defined.

In the present article we focus on an important and understudied population. An understanding of resilience among YMSM and the factors that promote healthy adaptation has great public health and clinical utility. Future studies with this population can seek to address resilience as well as risk and should also include and work to better understand the most vulnerable and hardest-to-reach YMSM, in particular transgender individuals and those who identify as heterosexual or bisexual.

Acknowledgments

This study was supported by a grant from the National Institute on Drug Abuse (DA11596). We would like to thank the young men who participated in the study, as well as the project interviewers: Kyle Ayers, Carlos Cortez, Jorge Ignacio Cortiñas, Wil Edgar, Christian Kolarz, Kalil Vicioso, and Salvador Vidal-Ortiz. We would also like to express our appreciation to David Nish and the staff of the Streetwork Project/Safe Horizons, Carrie Steinman formerly of Safe Space, and Jennie Casciano formerly of the New Neutral Zone/GVYC for help with recruitment, and to Karla Gostnell, Maya Tharaken, Mindy Finkelstein, Dorinda Welle, Ph.D., three anonymous reviewers, and the managing editor of the journal. We also wish to acknowledge the members of the New York City Association on Homeless and Street-Involved Youth Organizations for their assistance with the project.

Contributor Information

Marya Viorst Gwadz, Institute for AIDS Research, National Development and Research Institutes, Inc. 71 West 23rd Street, 8th Floor, New York, NY 10010. E-mail: Gwadz@ndri.org.

Michael C. Clatts, Institute for International Research on Youth at Risk, National Development and Research Institutes, Inc. 71 West 23rd Street, 8th Floor, New York, NY 10010. E-mail: Clatts@ndri.org

Huso Yi, Institute for International Research on Youth at Risk, National Development and Research Institutes, Inc. 71 West 23rd Street, 8th Floor, New York, NY 10010. E-mail: Yi@ndri.org.

Noelle R. Leonard, Institute for AIDS Research, National Development and Research Institutes, Inc. 71 West 23rd Street, 8th Floor, New York, NY 10010. E-mail: Leonard@ndri.org

Lloyd Goldsamt, Institute for International Research on Youth at Risk, National Development and Research Institutes, Inc. 71 West 23rd Street, 8th Floor, New York, NY 10010. E-mail: Goldsamt@ndri.org.

Steve Lankenau, Keck School of Medicine, University of Southern California, Children’s Hospital Los Angeles, Division of Research on Children, Youth and Families, 6430 Sunset Boulevard, Suite 1500, Los Angeles, CA 90028. E-mail: SLankenau@chla.usc.edu.

References

  1. Agronick G, O’Donnell L, Stueve A, San Doval A, Duran R, Vargo S. Sexual behaviors and risks among bisexually-and gay-identified young Latino men. AIDS & Behavior. 2004;8(2):185–197. doi: 10.1023/B:AIBE.0000030249.11679.d0. [DOI] [PubMed] [Google Scholar]
  2. Anderson AL. Strengths of gay male youth: An untold story. Child and Adolescent Social Work Journal. 1998;15:55–71. [Google Scholar]
  3. Arnett JJ. Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist. 2000;55:469–480. [PubMed] [Google Scholar]
  4. Barendregt C, van der Poel A, van de Mheen D. Tracing selection effects in three non-probability samples. European Addiction Research. 2005;11(3):124–131. doi: 10.1159/000085547. [DOI] [PubMed] [Google Scholar]
  5. Center for Disease Control and Prevention. HIV incidence among young men who have sex with men—seven U.S. cities, 1994–2000. Morbidity and Mortality Weekly Report. 2001;50:440–444. [PubMed] [Google Scholar]
  6. Clatts MC, Davis WR, Atillasoy A. Hitting a moving target: The use of ethnographic methods in the development of sampling strategies for the evaluation of AIDS outreach programs for homeless youth in New York City. NIDA Research Monograph. 1995;157:117–135. [PubMed] [Google Scholar]
  7. Clements K, Wilkinson W, Kitano K, Marx R. HIV prevention and health service needs of the trans-gender community in San Francisco. International Journal of Transgenderism. 1999;3:1–2. n.p. Retrieved June 8, 2005, from http://www.symposion.com/ijt/hiv_risk/clements.htm.
  8. Cooper K. Practice with transgendered youth and their families. Journal of Gay and Lesbian Social Services: Issues in Practice, Policy and Research. 1999;10 (3/4):111–129. [Google Scholar]
  9. Cross M, Epting F. Self-obliteration, self-definition, self-integration: Claiming a homosexual identity. Journal of Constructivist Psychology. 2005;18(1):53–63. [Google Scholar]
  10. Diaz RM, Ayala G, Bein E, Henne J, Marin BV. The impact of homophobia, poverty, and racism on the mental health of gay and bisexual men: Findings from 3 US cities. American Journal of Public Health. 2001;91:927–932. doi: 10.2105/ajph.91.6.927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dohrenwend BS, Krasnoff L, Askenasy AR, Dohrenwend BP. Exemplification of a method for scaling life events: The PERI Life Events Scale. Journal of Health and Social Behavior. 1978;19:205–229. [PubMed] [Google Scholar]
  12. Dworkin SH, Yi H. LGBT identity, violence, and social justice: The psychological is political. International Journal for the Advancement of Counselling. 2003;25(4):269–279. [Google Scholar]
  13. Farley M, Baral I, Kiremire M, Sezgin U. Prostitution in five countries: Violence and post-traumatic stress disorder. Feminism and Psychology. 1998;8:405–426. [Google Scholar]
  14. Frank ML, Lester D. Risky behaviors in Hispanic youth. Hispanic Journal of Behavioral Sciences. 2001;23:498–504. [Google Scholar]
  15. Freitas AL, Downey G. Resilience: A dynamic perspective. International Journal of Behavioral Development. 1998;22:263–285. [Google Scholar]
  16. Gwadz M, Clatts MC, Leonard NL, Goldsamt L. Attachment style, childhood adversity, and behavioral risk among young men who have sex with men. Journal of Adolescent Health. 2004;34:402–413. doi: 10.1016/j.jadohealth.2003.08.006. [DOI] [PubMed] [Google Scholar]
  17. Halpin SA, Allen MW. Changes in psychosocial well-being during stages of gay identity development. Journal of Homosexuality. 2004;47(2):109–126. doi: 10.1300/J082v47n02_07. [DOI] [PubMed] [Google Scholar]
  18. Herdt G, McClintock M. The magical age of 10. Archives of Sexual Behavior. 2000;29:587–606. doi: 10.1023/a:1002006521067. [DOI] [PubMed] [Google Scholar]
  19. Hollingshead A. Four factor index of social status. Unpublished working paper, Yale University; New Haven, MA: 1975. [Google Scholar]
  20. Huba GJ, Melchior LA. Module 26B: CES-D8 Form (Interview) Culver City, CA: The Measurement Group; 1995. Staff of The Measurement Group, & HRSA/HAB’s SPNS Cooperative Agreement Steering Committee. [Google Scholar]
  21. Huba GJ, Melchior LA. Module 45: Life Stressors Form. Culver City, CA: The Measurement Group; 1996. Staff of The Measurement Group, & HRSA/HAB’s SPNS Cooperative Agreement Steering Committee. [Google Scholar]
  22. Hughes TL, Eliason M. Substance use and abuse in lesbian, gay, bisexual and transgender populations. Journal of Primary Prevention. 2002;22:263–298. [Google Scholar]
  23. Karp D. Fierce gay youth confront distressed West Village residents. Youth City News. 2002 April; Retrieved January 14, 2005, from http://www.jrn.columbia.edu/studentwork/children/2002/karpgay.asp.
  24. Lankenau S, Clatts M, Welle D, Goldsamt L, Gwadz M. Street careers: Homelessness, drug use, and hustling among young men who have sex with men. International Journal of Drug Policy. 2005;16:10–18. [Google Scholar]
  25. Lee D. Street fight. New York Times 2002 May 31; [Google Scholar]
  26. Lock J, Steiner H. Gay, lesbian, and bisexual youth risks for emotional, physical, and social problems: Results from a community-based survey. Journal of the American Academy of Child and Adolescent Psychiatry. 1999;38:297–304. doi: 10.1097/00004583-199903000-00017. [DOI] [PubMed] [Google Scholar]
  27. Logan TK, Leukefeld C. HIV risk behavior among bisexual and heterosexual drug users. Journal of Psychoactive Drugs. 2000;32:239–248. doi: 10.1080/02791072.2000.10400446. [DOI] [PubMed] [Google Scholar]
  28. Luthar SS, Cicchetti D, Becker B. The construct of resilience: A critical evaluation and guidelines for future work. Child Development. 2000;71:543–562. doi: 10.1111/1467-8624.00164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Luthar SS, Doernberger CH, Zigler E. Resilience is not a unidimensional construct: Insights from a prospective study of inner-city adolescents. Development Psychopathology. 1993;5:703–717. doi: 10.1017/S0954579400006246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. MacKellar D, Valleroy L, Karon J, Lemp G, Janssen R. The Young Men’s Survey: Methods for estimating HIV seroprevalence and risk factors among young men who have sex with men. Public Health Reports. 1996;111(Suppl 1):138–144. [PMC free article] [PubMed] [Google Scholar]
  31. Masten AS. Ordinary magic: Resilience processes in development. American Psychologist. 2001;56:227–238. doi: 10.1037//0003-066x.56.3.227. [DOI] [PubMed] [Google Scholar]
  32. Mathy RM, Lehmann BA, Kerr DL. Bisexual and transgender identities in a nonclinical sample of North Americans: Suicidal intent, behavioral difficulties, and mental health treatment. Journal of Bisexuality. 2003;3(3–4):93–109. [Google Scholar]
  33. McGloin JM, Widom CS. Resilience among abused and neglected children grown up. Development & Psychopathology. 2001;13:1021–1038. doi: 10.1017/s095457940100414x. [DOI] [PubMed] [Google Scholar]
  34. Melchior LA, Huba GJ, Brown VB, Reback CJ. A short depression index for women. Educational & Psychological Measurement. 1993;53:1117–1125. [Google Scholar]
  35. Muhib FB, Lin LS, Stueve A, Miller RL, Ford WL, Johnson WD, Smith PJ the Community Intervention Trial for Youth Study Team. A venue-based method for sampling hard-to-reach populations. Public Health Reports. 2001;116(Suppl 1):216–222. doi: 10.1093/phr/116.S1.216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Mullen PE, Martin JL, Anderson JC, Romans SE, Herbison GP. The long-term impact of the physical, emotional, and sexual abuse of children: A community study. Child Abuse & Neglect. 1996;20(1):7–21. doi: 10.1016/0145-2134(95)00112-3. [DOI] [PubMed] [Google Scholar]
  37. National Institute on Drug Abuse. HIV risk behavior greater among bisexual than heterosexual drug users. National Institute on Drug Abuse Biennial Report to Congress for Years 1999–2000. 2000 Retrieved January 19, 2006, from http://www.drugabuse.gov/whgd/BiRep99-00.html.
  38. O’Donnell L, Agronick G, San Doval A, Duran R, Myint-U A, Stueve A. Ethnic and gay community attachments and sexual risk behaviors among urban Latino young men who have sex with men. AIDS Education & Prevention. 2002;14(6):457–471. doi: 10.1521/aeap.14.8.457.24109. [DOI] [PubMed] [Google Scholar]
  39. Parker KF, MacDonald JM, Alpert GP, Smith MR, Piquero AR. A contextual study of racial profiling: Assessing the theoretical rationale for the study of racial profiling at the local level. American Behavioral Scientist. 2004;47:943–962. [Google Scholar]
  40. Reid P, Klee H. Young homeless drug users: Ways of coping with harassment. Journal of Community & Applied Social Psychology. 2000;10(1):69–75. [Google Scholar]
  41. Rietmeijer CA, Wolitski RJ, Fishbein M, Corby NH, Cohn DL. Sex hustling, injection drug use, and non-gay identification by men who have sex with men. Associations with high-risk sexual behaviors and condom use. Sexually Transmitted Diseases. 1998;25:353–360. doi: 10.1097/00007435-199808000-00006. [DOI] [PubMed] [Google Scholar]
  42. Rosenbloom SR, Way N. Experiences of discrimination among African American, Asian American, and Latino adolescents in an urban high school. Youth & Society. 2004;35:420–451. [Google Scholar]
  43. Rosenfeld AA, Pilowsky DJ, Fine P, Thorpe M, Fein E, Simms MD, et al. Foster care: An update. Journal of the American Academy of Child & Adolescent Psychiatry. 1997;36:448–457. doi: 10.1097/00004583-199704000-00006. [DOI] [PubMed] [Google Scholar]
  44. Russell ST. Beyond risk: Resilience in the lives of sexual minority youth. Journal of Gay and Lesbian Issues in Education. 2005;2(3):5–18. [Google Scholar]
  45. Rutter M. Resilience in the face of adversity: Protective factors and resistance to psychiatric disorder. British Journal of Psychiatry. 1985;147:598–611. doi: 10.1192/bjp.147.6.598. [DOI] [PubMed] [Google Scholar]
  46. Rutter M. Children in substitute care: Some conceptual considerations and research implications. Children & Youth Services Review. 2000;22(9–10):685–703. [Google Scholar]
  47. Savin-Williams RC. Gay and lesbian youth: Expressions of identity. New York: Hemisphere; 1990. [Google Scholar]
  48. Sullivan M, Wodarski JS. Social alienation in gay youth. Journal of Human Behavior in the Social Environment. 2002;5(1):1–17. [Google Scholar]
  49. Sullivan MK. Homophobia, history, and homosexuality: Trends for sexual minorities. Journal of Human Behavior in the Social Environment. 2003;8(2–3):1–13. [Google Scholar]
  50. Thiede H, Valleroy LA, MacKellar DA, Celentano DD, Ford WL, Hagan H, et al. Regional patterns and correlates of substance use among young men who have sex with men in 7 US urban areas. American Journal of Public Health. 2003;93:1915–1921. doi: 10.2105/ajph.93.11.1915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Valera R, Sawyer RG, Schiraldi GR. Perceived health needs of inner-city street prostitutes: A preliminary study. American Journal of Health Behavior. 2001;25(1):50–59. doi: 10.5993/ajhb.25.1.6. [DOI] [PubMed] [Google Scholar]
  52. Valleroy LA, MacKellar DA, Karon JM, Rosen DH, McFarland W, Shehan DA, et al. HIV prevalence and associated risks in young men who have sex with men. Journal of American Medical Association. 2000;284:198–204. doi: 10.1001/jama.284.2.198. [DOI] [PubMed] [Google Scholar]
  53. Van Ness PH, Davis WR, Johnson BD. Socioeconomic marginality and health services utilization among Central Harlem substance users. Substance Use & Misuse. 2004;39(1):61–85. doi: 10.1081/ja-120027766. [DOI] [PubMed] [Google Scholar]
  54. Watters JK, Biernacki P. Targeted sampling: Options for the study of hidden populations. Social Problems. 1989;36:416–430. [Google Scholar]
  55. Widom CS, Morris S. Accuracy of adult recollections of childhood victimization: Part 2. Childhood sexual abuse. Psychological Assessment. 1997;9:34–46. [Google Scholar]
  56. Widom CS, Shepard RL. Accuracy of adult recollections of childhood victimization: Part 1. Childhood physical abuse. Psychological Assessment. 1996;8:412–421. [Google Scholar]
  57. Widom CS, Weiler BL, Cottler LB. Childhood victimization and drug abuse: A comparison of prospective and retrospective findings. Journal of Consulting and Clinical Psychology. 1999;67:867–880. doi: 10.1037//0022-006x.67.6.867. [DOI] [PubMed] [Google Scholar]

RESOURCES