Suicidal ideation and attempts among sexual minority youths receiving social services.
In general, numerous risk factors for youth suicide have been established. These risks include depression, substance abuse, conduct and other disruptive disorders, poor interpersonal problem-solving skills, family history of suicidal behavior and psychopathology, poor parent-child relationships, physical and sexual abuse, school and work problems, media exposure to suicidal behavior, and lower levels of religiosity and family cohesion (Gould, Greenberg, Velting, & Shaffer, 2003). Homelessness among adolescents has also been linked to elevated rates of suicidal behavior, with 30 percent of runaway youths reporting a suicide attempt in one study (Stiffman, 1989). Some of these risk factors appear to be more common among sexual minority youths than other youths, especially depression (Russell & Joyner, 2001; Safren & Heimberg, 1999), substance abuse (Russell & Joyner, 2001), and hopelessness (Safren & Heimberg, 1999).
Even controlling for these general risk factors, however, sexual orientation remained an independent predictor of suicidality in one large, national study (Russell &Joyner, 2001).This may be due partially to the fact that some risk factors for suicidality accrue uniquely to gay and lesbian youths as a result of environmental stressors such as societal oppression and discrimination (McDaniel et al., 2001). These risk factors include victimization based on sexual identity (D'augelli et al., 2005; Savin-Williams, 1994), rejection by peers or family because of sexual orientation (D'augelli et al., 2005; Hershberger, Pilkington, & D'augelli, 1997), internalized homophobia, fear of rejection and ridicule, and actual violence (Bontempo & D'augelli, 2002). For these youths, schools are often the site of their harassment and victimization (Bontempo & D'augelli, 2002), with students, teachers, and administrators often marginalizing them (Dennis & Harlow, 1986), thereby creating a "toxic environment" for the youths (van Wormer & McKinney, 2003).
Recent studies drawing from community-based representative samples have helped to establish the legitimacy of sexual minority youths' heightened needs for attention from the research and clinical communities and are invaluable from epidemiologic, methodological, and policy perspectives. However, because the majority of youths in the general population-even those facing a mental health crisis such as a suicide attempt--do not use professional services (Busch & Horwitz, 2004; Kataoka, Zhang, & Wells, 2002), research using random samples largely conveys information about youths who may never encounter a social services professional. As such, research focusing specifically on those who do access social services is still needed to inform the day-today work of social workers in the development of interventions, policy, and further research.
To help expand the knowledge base about sexual minority youths who use social services, the current study draws specifically from self-identified gay, lesbian, bisexual, questioning, and transgender adolescents who sought services at an urban social services agency. A general and pragmatic question underlay the research and interpretation of results: How can knowledge about risk and protective factors for suicidal behavior among these youths help inform the practical day-to-day work of social workers who are in a position to help?
Sample Recruitment and Characteristics
Survey respondents were 142 youths and young adults (14 to 21 years of age) who received services at Rainbow Alley, a Denver-area program of the Gay, Lesbian, Bisexual, and Transgender Community Center of Colorado. The program provides support, education, advocacy, and social activities for sexual minority youths and their allies. Surveys were administered during June 2004 and June 2005 as part of the program's annual evaluation process that has historically taken place in June. Youths completed the instruments in private. Of the final sample, 78 respondents (54.93 percent) completed the survey in 2004 and the remaining 64 respondents completed it in 2005. No significant differences in variables were found between the two years.
Participation was requested of all center participants during a three-week period, and data were collected until at least 100 surveys were completed for each of the two years, resulting in 200 participants. Data on response rates were not collected, but facility staff reported that virtually every person asked agreed to participate in the survey. Examination of agency records indicated that eight participants had participated in both years' surveys. Demographic matching was used to eliminate the earlier of the two duplicate records for each of these participants.
After removal of duplicate records, the sample consisted of 182 participants. Because of our interest in victimization at school and the presence of gay-straight alliances (GSAs), an additional 31 participants were dropped who were not currently attending school. In the context of this article, we use the term school to encompass junior high-middle school, high school, and college. Also, because of missing data on one or more of the variables of interest, an additional nine (5.9 percent) records were excluded from the sample, resulting in a final sample size of 142. All variables except the presence of a GSA and victimization at school were examined to determine if significant differences emerged between participants who were dropped from the sample and those who remained in the sample. No significant differences emerged.
Most measures used in this study were modeled after questions in the National Youth Risk Behavior Surveillance Survey (Centers for Disease Control and Prevention, 2003). Based on existing scholarship regarding suicidal ideation and attempts, several variables were used as controls: hopelessness, alcohol use, and methamphetamine use. Model variables included sociodemographic variables (gender and race), family variables (homelessness and familial abuse), and school variables (in-school victimization and presence of GSAs).
Youths were asked whether they had felt so sad or hopeless for two weeks or more in a row that they had stopped doing some usual activities. Hopelessness is judged to be a more accurate predictor of suicide than is depression (Beck, Steer, Kovacs, & Garrison, 1985). Respondents were asked to indicate the number of days they had ever had at least one alcoholic drink. From this, a dichotomized variable indicated whether they had used alcohol at least once. Various derivations of the variable were examined for their performance in the models, but the dichotomized variable was as informative as other options. Similarly, a dichotomous variable assessed methamphetamine use. Data collected about other types of drug use (for example, opiates) were not included in the models because of collinearity.
As both gender and race differences in adolescent suicidality have been documented, both were examined as potential sociodemographic predictors. Participants were asked to identify their gender and were given five potential responses: female, male, transgender male, transgender female, and other. Gender was recoded into two dichotomous variables, male and transgender, with female used as the reference category. Only male was retained in the reported models as differences did not emerge for transgender individuals. Although respondents indicated a wide array of responses for race and ethnicity, final models included only a dichotomous variable for African Americans (with white and other races as the reference group) as preliminary analyses indicated no significant differences among other racial groups and because existing literature suggested that African Americans have decreased risk of suicidality compared with other racial groups in the United States that were available in our sample.
To capture verbal and physical abuse by family members, one question assessed verbal harassment and another assessed physical harassment or attack in the past 12 months because of sexual orientation. The two variables were combined into a single dichotomous variable indicating whether any type of abuse had occurred. To assess homelessness, survey participants were asked to indicate the number of times they had slept in a homeless shelter, outside, or on someone else's couch because they had nowhere else to stay. From this a dichotomous variable was constructed to indicate whether the respondent had experienced a spell of homelessness during the past year.
Respondents were asked if they had experienced harassment at school because of their sexual orientation. To test the potential role of GSAs as a protective factor, respondents were asked if their school had a GSA. Responses were "yes," "no" and "I don't know." Participants who said that they did not know (n = 11,7.75 percent) were grouped with students who reported no GSA, based on the assumption that their school did not have an alliance or, if their school did, the youths did not perceive it as a resource because they were unaware of its existence.
Although most youths who think about or attempt suicide do not go on to die by suicide, researchers frequently use suicidal ideation as a proxy for suicide risk (Hawton et al., 1998). Suicidal ideation precedes nearly all suicides, and the largest risk factor for completed suicide is a prior suicide attempt (Harris & Barraclough, 1997). As such, two dependent variables were examined. First, youths were asked the number of times in the past 12 months they had attempted suicide. A dichotomous variable was created for at least one attempt. To capture a broader risk pool, a second dependent variable combined the suicide attempt variable with a question that asked if the respondent had seriously considered attempting suicide in the past 12 months.
To test the hypothesis that variation in likelihood of suicidality and suicide attempts can be explained by demographics, family factors, or school factors beyond that explained by the control variables, five multivariate logistic regression analyses were conducted. Appropriate for binary dependent variables, logistic models result in odds ratios.
For each dependent variable, the first models contained only control variables, followed by three models in which the control variables were combined with either the sociodemographic variables, the family factor variables, or the school factor variables. The final models combined the control variables with the variables that demonstrated at least marginal significance (p < .10) in prior models.
Demographic characteristics and descriptive statistics are reported in Table 1. As the samples consisted of young adults who were both high school and college age, we examined whether controlling for school status influenced any of the results. Only one difference emerged in the models; that difference is reported in the discussion of the models.
Model 1--Control Variables
Three control variables (hopelessness, alcohol use, and methamphetamine use) were examined in the first model. Youths who reported being hopeless were, relative to those who did not report being hopeless, more than nine times as likely to report suicidality and almost five times as likely to report a suicide attempt. Not surprising, these findings mirror the documented relationship between hopelessness and suicidality in both sexual minority youths and the general youth populations. (See Tables 2 and 3 for regression model coefficients and standard errors.)
Alcohol use was not a significant predictor of either suicidality or a suicide attempt. This is in contrast to much literature that suggests alcohol use has a significant relationship with suicide (see Gould et al., 2003). Given that the overwhelming majority of youths in the sample had used alcohol at some point in their life (86.19 percent) and in the past 30 days (71.27 percent), the lack of variability may be partially responsible for failure to find significance.
Methamphetamine use significantly predicted suicidality, but not exclusively suicide attempts. Youths who reported methamphetamine use were 3.5 times as likely to have thought about or attempted suicide. These findings are similar to recent studies documenting a relationship between methamphetamine use and suicidality (Callor et al., 2005; Zweben et al., 2004).
Although the literature on the sociodemographics of gender and race suggests that young women are more likely than young men to experience suicidality or to attempt suicide (Lewinsohn, Rohde, Seeley, & Baldwin, 2001; Wunderlich, Bronisch, Wittchen, & Carter, 2001), this finding has not been consistently supported in the literature on sexual minority youths (D'augelli, 2002). Similarly, we found male respondents were no more or less likely than female respondents to report suicidality or a suicide attempt.
Racial differences in suicidality have been documented indicating that African Americans were less likely to experience suicidality than were white and other races (Garlow, Purselle, & Heninger, 2005). However, we found no significant differences in likelihood of suicidality or reported attempts between African Americans and other races, suggesting that African American sexual minority youths may be at a similar risk as those of other races.
Model 3--Family Factors
As with most of the existing literature on familial abuse and spells of homelessness, youths who experienced a spell of homelessness during the past 12 months were more likely to report suicidality. They were more than four times as likely to report suicidality, and almost three times as likely to report an attempt. Verbal or physical abuse by a family member was not a significant predictor. However, by limiting suicidality to include only serious attempts, recent work by D'augelli et al. (2005) found that parental psychological abuse or disapproval does have a significant role to play. Our inability to differentiate level of severity of attempt in this study may have obscured this important relationship. Finally, even though homelessness and prior abuse were weakly correlated in this study (r = .26, p < .01), family abuse remained statistically nonsignificant even after the removal of the homelessness variable.
Model 4--School Factors
In-school victimization was a significant predictor of suicidality, but not of the smaller risk pool of suicide attempts alone. Those who reported victimization were 2.76 times as likely to report suicidallty as those who did not. Finally, those who went to schools in which there were GSAs were significantly less likely both to experience suicidality and to report suicide attempts. Students at schools with GSAs were only two-thirds as likely as students at non-GSA schools to report suicidality, and about one-third as likely to report an attempt. The influence of having at least one adult ally in the school was related to a decreased likelihood of suicidality, but the results were not statistically significant (not shown). Although the adult ally and GSAs were significantly correlated (r = .33, p < .001), the GSAs remained significant even when including the adult ally variable in the models (not shown).
MODEL 5--FULL MODEL
In Table 2, model 5, the final multivariate model predicting suicidality is shown. Hopelessness, methamphetamine use, and homelessness were all associated with increased odds of considering suicide. In-school victimization was marginally significant in predicting an increased likelihood, and the presence of a school GSA was marginally significant in predicting a decreased likelihood of suicidality. The presence of a GSA was the only factor tested that was associated with lower odds of suicidality, suggesting that GSAs--or their perceived influence on school culture--may serve a protective function for sexual minority youths. In other analyses, both grades and the presence of an adult ally were examined as potential protective factors, but neither was significant in any models examined. When we reran the full model including the dummy variable based on whether the respondent was high school age or younger (18 years or under) or college age (19 years or older), we found that the marginal significance of in-school victimization became nonsignificant in the model. Further analyses of this finding suggested that in-school victimization is significantly more influential for youths and young adults who are high school age or younger than it is for college-age sexual minority youths. Given the flexibility in choosing colleges and the increased maturity level of college students, this finding is not particularly surprising.
Similar findings emerged for predicting suicide attempts only (see Table 3, model 5). Hopelessness and homelessness were the two factors that predicted greater odds of reporting a suicide attempt. As with suicidality, the presence of a GSA was marginally significantly related to a decrease in likelihood of reporting suicide attempts.
In line with earlier findings, our results indicate that sexual minority youths who seek social services are at a higher risk of suicidality than has been found in epidemiologic studies of the general youth population. Similarly, factors such as hopelessness, homelessness, in-school victimization, and methamphetamine use are associated with increased risks for these youths. Unlike research in the general youth population, African Americans and male sexual minority youths using social services had similar rates of suicidal behavior compared with white and female youths in the sample.
One unique association significant in model 3 examining school variables (but only marginally significant in the full models) is the association between the presence of GSAs and a decreased risk of both suicidality and suicide attempts. GSAs are gaining popularity as a means to improve school climate for sexual minority students and to reduce their psychological stress (Miceli, 2005). Whether the relationship found between the presence of a GSA and decreased risk of suicidality and suicide attempts is because of an actual improvement in school climate or because of sexual minority youths' perceptions of having resources available cannot be determined from the current data and warrants further exploration. The data suggest, however, that institutional support in the form of approved student clubs may be an effective strategy in supporting sexual minority youths. Although neither having adult allies in school nor having good grades reached a level of significance in the models, this finding may simply be an artifact of the sample size.
The focus on self-identified sexual minority youths who encounter community-based social services precludes generalizing the findings to the entire population of sexual minority youths. Given that early disclosure of a sexual minority identity is associated with increased mental health problems (Orenstein, 2001; Savin-Williams & Ream, 2003), this article may paint a grimmer picture than if the sample included same-age, same-sex-attracted youths who have not yet identified themselves as nonheterosexuaL At the same time, although not necessarily a reflection of all sexual minority youths, the sample biases reflect the reality of applied social work practice. Compared with youths randomly sampled in epidemiological surveys, it is likely that these sexual minority youths more accurately reflect those seen in agency settings.
The current research on adolescents in one agency setting also precludes a thorough examination of high-functioning, well-adjusted sexual minority adolescents who identify no need for help. As such, caution must be taken not to negatively stereotype all sexual minority youths as problematic. As Savin-Williams (2005) pointed out, the majority of sexual minority youths are resilient, adaptive youths who--like their heterosexual counterparts--negotiate the struggles of adolescence successfully to become contributing members of society. Because the youths in the study received services from an agency that serves the sexual minority community, adolescents seeking assistance from agencies that do not have this focus may differ on issues such as degree of openness about their sexual orientation, comfortableness with their sexual identity, or in other similar ways. Similarly, because the sample is drawn from clients served by a single agency, the results are not necessarily generalizable to other agencies serving the same population.
The timing of the annual survey in June, after some of the Denver-area schools and colleges have ended for the year, may have had some affect on who was included in the sample. However, Rainbow Alley provides services year round and experiences both positive and negative fluctuations in the number of clients served in summer, depending on the year. It should be remembered that the sample consisted only of youths who were currently in school, and the findings should not be generalized to youths who are no longer attending school.
Recent findings suggest the importance of differentiating whether or not suicidality is related directly to sexual orientation and assessing the seriousness of suicide attempts for sexual minority youths (D'augelli et al., 2005). Neither of these variables was available in our study, and although these aspects are important from a theoretical and epidemiological perspective, treating all suicidality as if it were potentially lethal regardless of whether it is directly related to sexual orientation is probably the most prudent route of action for social work practitioners. Finally, the use of single-item measures to capture complex constructs such as suicidal thoughts, suicide attempts, and substance abuse, although common in social science research, may obscure a more nuanced understanding of the relationships examined.
Although many of the youths lived in less than ideal environments--spells of homelessness, drug and alcohol use, in-school victimization, or familial abuse--the majority demonstrated remarkable resilience. Almost 85 percent of those in school indicated that they were getting average or above average grades, with a substantial proportion (49.64 percent) reporting mostly As and Bs. A majority reported increased motivation for school and improvement in their relationships with adults as a result of their involvement in Rainbow Alley programming. Most had identified potential available avenues of support in their schools, including adults who were allies or the presence of a GSA.
Implications for Social Work Practice
Our results suggest a number of interventions to help social workers support sexual minority youths. Screening and monitoring for hopelessness should be a critical, ongoing component of work with these youths, as should exploring drug and alcohol usage, especially use of methamphetamine. Because of the frequency of familial abuse, the appropriateness of interventions with families should be assessed. Many parents go through a coming out process of their own and may need resources such as support groups or counseling as they do so. However, because parents may not be aware of their adolescent's same-sex orientation, social workers must make safety the first priority, including identifying safe ways to contact the adolescent without inadvertently outing him or her. Outing youths to their parents may result in homelessness for them or put them at greater risk of abuse.
Because of the increased risk of homelessness and the strength of the relationship between homelessness and suicidality, relationships with services providers who support homeless youths should be established. Social workers should watch for indicators of homelessness, such as youths sleeping over at friends most nights, as some youths may not consider themselves homeless. If at all possible, work should be done with the youths and their families to prevent homelessness. However, given that familial rejection may lead to homelessness when these youths come out to family members, social workers may want to explore the potential ramifications of coming out to family members, helping youths develop contingency plans should they decide to do so. This is not to imply that social workers should discourage youths from coming out to their families, but rather that social workers have a role to play in helping the youths think through their particular life context as part of the discernment process in deciding to whom and at what point in time they disclose their sexual orientation.
Social workers should ask about verbal and physical harassment at school, given the frequency of victimization at school. Schools have not historically protected sexual minority youths from bullying and victimization. Advocating for antibullying policies and training for teachers, staff, and students would help support sexual minority youths. Social workers can also play a significant role in supporting the development of GSAs by acting as club advisers, advocating with and educating school administrators, and publicizing the groups' availability.
Although not emerging from the findings of this study, a number of additional suggestions for practice have been offered by other scholars that bear repeating. First, one should not convey the assumption that all clients are opposite-sex oriented (Appleby & Anastas, 1998). Using gender-neutral language when inquiring about intimate relationships and not assuming that all sexually active females need birth control are two such examples. Second, social workers should recognize that identity, behavior, and attraction may not align neatly for some adolescents (Savin-Williams, 2005). Third, sexuality may be experienced as fluid for adolescents, perhaps more so now than for previous generations (Diamond, 2000; Savin-Williams, 2005). Thus, an adolescent may identify as same-sex oriented at one point in time and opposite-sex oriented at a later point. Fourth, the act of claiming a nonheterosexual sexual identity does not imply that the adolescent is necessarily sexually active as some adolescents are aware of their same-sex attraction before engaging in sexual activity (Savin-Williams, 2005). Fifth, assumptions should not be made that problems are associated with sexual orientation or heterosexist culture, nor should the converse assumption that sexual orientation or everyday heterosexism are unrelated to problems be made (Appleby & Anastas, 1998; Hunter, Shannon, Knox, & Martin, 1998). The role of the influence of such cultural variables on these issues must be explicitly assessed (Rodriguez &Walls, 2000). Finally, practitioners should openly demonstrate being an ally to lesbian, gay, bisexual, or transgender clients (Van Den Bergh & Crisp, 2004) by participating in a safe zone program, displaying posters that indicate an accepting stance, or having books about lesbian and gay issues visible to young people.
Although many sexual minority youths who use social services experience numerous life stressors, it is critically important that social workers not only see these environmental challenges in their lives, but also acknowledge and support the adaptability and resilience of these youths, including the strength inherent in seeking support.
Original manuscript received January 30, 2006
Final revision received November 15, 2006
Accepted January 25, 2007
Appleby, G.A., & Anastas, J.W. (1998). Not just a passing phase: Social work with gay, lesbian and bisexual people. New York: Columbia University Press.
Beck, A.T., Steer, R.A., Kovacs, M., & Garrison, B. (1985). Hopelessness and eventual suicide: A 10year prospective study of patients hospitalized with suicidal ideation. American Journal of Psychiatry, 142, 559-563.
Bontempo, D. E., & D'augelli, A. R. (2002). Effects of at-school victimization and sexual orientation on lesbian, gay, or bisexual youths' health risk behavior. Journal of Adolescent Health, 30, 364-374.
Busch, S. H., & Horwitz, S. M. (2004).Access to mental health services: Are uninsured children falling behind? Mental Health Services Research, 6, 109-116.
Callor, W. B., Petersen, E., Gray, D., Grey, T., Lamoreaux, T., & Bennett, P.J. (2005). Preliminary findings of noncompliance with psychotropic medication and prevalence of methamphetamine intoxication associated with suicide completion. Crisis: Journal of Crisis Intervention And Suicide Prevention, 26, 78-84.
Centers for Disease Control and Prevention. (2003). 2003 National school-based Youth Risk Behavior Survey public-use data documentation. Retrieved January 13, 2006, ftp://ftp.cdc.gov/pub/data/yrbs/2003/yrbs2003codebook.txt
D'augelli, A. R. (2002). Mental health problems among lesbian, gay, and bisexual youths ages 14 to 21. Clinical Child Psychology and Psychiatry, 7, 433-456.
D'augelli, A. R., Grossman, A., Salter, N. P., Vasey, J.J., Starks, M.T., & Sinclair, K. O. (2005). Predicting the suicide attempts of lesbian, gay, and bisexual youth. Suicide and Life-Threatening Behavior, 36, 646-660.
Dennis, D. L., & Harlow, R. E. (1986). Gay youth and the right to education. Yale Law Journal and Policy Review, 4, 447-478.
Diamond, L. M. (2000). Sexual identity, attractions, and behavior among young sexual-minority women over a two-year period. Developmental Psychology, 36, 241-250.
Garlow, S.J., Purselle, D., & Heninger, M. (2005). Ethnic differences in Patterns of suicide across the life cycle. American Journal of Psychiatry, 162, 319-323.
Garofalo, R., Wolf, R., Cameron, M. S., Kessel, S., Palfrey, J., & Durant, R. H. (1998).The association between health risk behaviors and sexual orientation among a school-based sample of adolescents. Pediatrics, 101, 895-902.
Gould, M. S., Greenberg, T., Velting, D. M., & Shaffer, D. (2003).Youth suicide risk and preventive interventions: A review of the past 10 years. Journal Of The American Academy of Child &Adolescent Psychiatry, 42, 386-405.
Harris, E. C., & Barraclough, B. (1997). Suicide as an outcome for mental disorders: A meta-analysis. British Journal of Psychiatry, 170, 205-228.
Hawton, K., Arensman, E., Townsend, E., Bremner, S, Feldman, E., Goldney, R., Gunnell, D., Hazell, P., van Heeringen, K., House, A, Owens, D., Sakinofsky, I., & Triiskman-Bendz, L. (1998). Deliberate self harm: systematic review of efficacy of psychosocial and pharmacological treatments in preventing repetition. British Journal of Psychiatry, 317, 441-447.
Hershberger, S. L., Pilkington, N.W., & D'augelli,A. R. (1997). Predictors of suicide attempts among gay, lesbian, and bisexual youth.Journal of Adolescent Research, 12, 477-497.
Hunter, S., Shannon, C., Knox, J., & Martin, J. I. (1998). Lesbian, gay, bisexual youths and adults: Knowledge for human services. Thousand Oaks, CA: Sage Publications.
Kataoka, S. H., Zhang, L., & Wells, K. B. (2002). Unmet need for mental health care among U.S. children: Variation by ethnicity and insurance status. American Journal of Psychiatry, 159, 1548-1555.
Lewinsohn, P. M., Rohde, E, Seeley, J. R., & Baldwin, C. L. (2001). Gender differences in suicide attempts from adolescence to young adulthood. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 427-434.
McDaniel, J. S., Purcell, D., & D'augelli, A. R. (2001).The relationship between sexual orientation and risk for suicide: Research findings and future directions for research and prevention. Suicide and Life-Threatening Behavior, 3/(Suppl.), 84-105.
Miceli, M. (2005). Standing out, standing together: The social and political impact of gay-straight alliances. New York: Routledge.
Orenstein, A. (2001). Substance use among gay and lesbian adolescents. Journal of Homosexuality, 41, 1-15.
Remafedi, G., French, S., Story, M., Resnick, M., & Bhim, B. (1998).The relationship between suicide risk and sexual orientation: Results of a population-based study. American Journal of Public Health, 88, 57-60.
Rodriguez, R. R., &Walls, N. E. (2000). Culturally educated questioning: Toward a skills-based approach in multicultural counselor training. Journal of Applied and Preventive Psychology, 9, 89-99.
Russell, S.T. (2003). Sexual minority youth and suicide risk. American Behavioral Scientist, 46, 1241-1257.
Russell, S.T., & Joyner, K. (2001). Adolescent sexual orientation and suicide risk: Evidence from a national study. American Journal of Public Health, 91, 1276-1281.
Safren, S.A., & Heimberg, R. G. (1999). Depression, hopelessness, suicidality and related factors in sexual minority and heterosexual youth. Journal of Consulting and Clinical Psychology, 67, 859-866.
Savin-Williams, R. C. (1994).Verbal and physical abuse as stressors in the lives of lesbian, gay male, and bisexual youths: associations with school problems, running away, substance use, prostitution, and suicide. Journal of Consulting and Clinical Psychology, 62, 261-269.
Savin-Williams, R. C. (2005). The new gay teenager. Cambridge, MA: Harvard University Press.
Savin-Williams, R. C., & Ream, G. L. (2003). Suicide attempts among sexual minority male youth. Journal of Clinical Child and Adolescent Psychology, 32, 509-522.
Stiffman, A. R. (1989). Suicide attempts in runaway youths.
Suicide and Life Threatening Behavior, 19, 147-159.
Van Den Bergh, N., & Crisp, K. (2004). Defining culturally competent practice with sexual minorities: Implications for social work education and practice. Journal of Sodal Work Education, 40, 221-238.
van Wormer, K., & McKinney, R. (2003).What schools can do to help gay/lesbian/bisexual youth: A harm reduction approach. Adolescence, 38, 409-420.
Wunderlich, U., Bronisch, T., Wittchen, H.-U., & Carter, R. (2001). Gender differences in adolescents and young adults with suicidal behaviour. Acta Psychiatrica Scandinavica, 104, 332-339.
Zweben, J. E., Cohen, J. B., Christian, D., Galloway, G. E, Salinardi, M., Parent, D., & Iguchi, M. (2004). Psychiatric symptoms in methamphetamine users. American Journal on Addictions, 13, 181-190.
N. Eugene Walls, PhD, is assistant professor, Graduate School of Social Work, University of Denver, 2148 S. High Street, Denver, CO 80208; e-mail: email@example.com. Stacey Freedenthal, PhD, LCSW, is assistant professor, Graduate School of Sodal Work, University of Denver. Hope Wisneski, MSW, LCSW, is deputy executive director, the Gay, Lesbian, Bisexual, and Transgender Community Center of Colorado, Denver.
Table 1: Demographic Characteristics of Sample Variable n (%) Gender Female 63 44.37 Male 73 51.41 Transgender 6 4.23 Sexual identity Gay 60 42.86 Lesbian 35 25.00 Bisexual 32 22.86 Questioning 9 6.43 Pansexual/Other 6 4.23 Race and ethnicity African American 12 8.45 Asian American 1 0.70 Bi/Multiracial 40 28.17 Latino 29 20.42 Native American 5 3.52 White 55 38.73 Age 14-15 19 13.38 1 G-17 71 50.00 18-19 46 32.40 20-21 6 4.22 Residence Parents 38 26.95 Relatives 10 7.09 Placement 2 1.42 On my own 6 4.26 Friends 7 4.96 Multiple 79 56.03 Afraid at school Never 61 42.96 Rarely 43 30.28 Sometimes 30 21.13 Most/all of the time 8 5.64 Harassed at school 75 52.82 Victimized by family 65 45.77 Alcohol use 119 83.80 Methamphetamine use 49 34.51 Felt hopeless 83 58.45 Suicidality 81 57.04 Suicide attempt 57 40.14 Table 2: Logistic Regression of Suicidality on Control Variables, Demographics, Family Factors, and School Factors Variable Model 1 Model 2 Model 3 Hopelessness 9.33 **** 10.58 **** 8.95 **** (3.8999) (4.6909) (4.0201) Alcohol use 2.32 2.17 2.66 (1.3142) (1.2993) (1.6517) Methamphetamine use 3.57 *** 3.62 *** 3.07 ** (1.6334) (1.6683) (1.4878) Male 1.42 (0.6220) African American 0.51 (0.3838) Homelessness 4.11 *** (1.8407) Familial abuse 1.33 (.6014) In-school victimization Gay-straight alliance N 142 142 142 Variable Model 4 Model 5 Hopelessness 8.42 **** 8.71 **** (3.7337) (4.0741) Alcohol use 1.73 2.27 (1.0378) (1.4180) Methamphetamine use 3.22 ** 2.98 ** (1.5263) (1.4693) Male African American Homelessness 3.76 *** (1.7124) Familial abuse In-school victimization 2.76 ** 2.37 * (1.2273) (1.0986) Gay-straight alliance 0.34 ** 0.38 * (0.1697) (0.1986) N 142 142 Note: Model 1 = control (hopelessness, alcohol use, methamphetamine use); model 2 = sociodemographics (gender, race); model 3 = family factors (familial abuse, spells of homelessness); model4 = school factors (in-school victimization, presence of gay-straight alliance); model 5=full (multivariate). Standard error of odd ratios in parentheses. * p < .10. ** p < .05. *** p < .01. **** p < .001. Table 3: Logistic Regression of Suicide Attempts on Control Variables, Demographics, Family Factors, and School Factors Variable Model 1 Model 2 Model 3 Hopelessness 4.95 **** 4.92 **** 4.17 **** (2.0097) (2.0476) (1.7669) Alcohol use 1.61 1.61 1.73 (0.8962) (0.9065) (0.9858) Methamphetamine use 1.85 1.85 1.47 (0.7152) (0.7151) (0.6043) Male 0.98 (0.3796) African American 1.03 (0.6998) Homelessness 2.82 ** (1.1595) Familial abuse 1.76 (0.7124) In-school victimization Gay-straight alliance N 142 142 142 Variable Model 4 Model 5 Hopelessness 4.55 **** 4.79 **** (1.9408) (2.0396) Alcohol use 1.31 1.70 (0.7477) (0.9759) Methamphetamine use 1.68 1.53 (0.6703) (0.6273) Male African American Homelessness 2.98 *** (1.21 i7) Familial abuse In-school victimization 1.71 (0.7039) Gay-straight alliance 0.41 ** 0.48 * (0.1686) (0.2038) N 142 142 Note: Model 1 = control (hopelessness, alcohol use, methamphetamine use); model 2 = sociodemographics (gender, race); model 3 = family factors (familial abuse, spells of homelessness); model 4- school factors (in-school victimization, presence of gay-straight alliance); model 5 = full (multivariate). Standard error of odd ratios in parentheses. * p <.10. ** p<.05. *** p<.01. **** p<.001.
|Printer friendly Cite/link Email Feedback|
|Author:||Walls, N. Eugene; Freedenthal, Stacey; Wisneski, Hope|
|Date:||Jan 1, 2008|
|Previous Article:||Development of a family intervention for Native Hawaiian women with cancer: a pilot study.|
|Next Article:||Moving toward culturally competent practice with Muslims: modifying cognitive therapy with Islamic tenets.|