Suicidality among Gay, Lesbian and Bisexual Youth: The Role of Protective Factors
Article Outline
Abstract
Purpose
Many reports have indicated that gay, lesbian and bisexual (GLB) youth are particularly vulnerable to poor outcomes, including suicide. Certain protective factors are likely to reduce this risk. The present study examines four protective factors (family connectedness, teacher caring, other adult caring, and school safety) and their association with suicidal ideation and attempts among adolescents with same-gender experience.
Methods
Data come from the 2004 Minnesota Student Survey of 9th and 12th grade students; 21,927 sexually active youth were grouped according to the gender of their sex partner(s) into GLB and non-GLB groups. Four protective factors and suicidal ideation and attempts were compared across groups. Logistic regression was used to examine the influence of protective factors on suicide, and predicted probabilities of suicidal ideation and attempts were estimated using general linear modeling.
Results
There were 2,255 respondents who reported same-gender experience. Over half of GLB students had thought about suicide and 37.4% reported a suicide attempt. GLB youth reported significantly lower levels of each protective factor than their non-GLB peers. Family connectedness, adult caring, and school safety were significantly protective against suicidal ideation and attempts. Risk associated with a GLB sexual orientation is largely mediated through protective factors.
Conclusions
Sexual orientation alone accounts for only a small portion of variability in suicidal ideation and attempts. If protective factors were enhanced among GLB youth, suicide in this population is expected to be considerably lower. Protective factors examined here are amenable to change and should be targeted in interventions.
Keywords: Sexual orientation, Suicide, Protective factors, Resiliency
Beginning in the 1980s, various reports have indicated an association between a gay, lesbian, or bisexual (GLB) sexual orientation and an increased suicide risk [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12]. The 1989 Report of the Secretary’s Task Force on Youth Suicide stated that GLB youth were two to three times more likely to attempt suicide than other young people and might comprise up to 30% of completed youth suicides annually [3]. Despite the methodological concerns and controversy surrounding the studies on which these findings were based [13], [14], more recent research with population-based samples and appropriate comparison groups has nonetheless corroborated these earlier findings [6], [7], [8], [9], [11], [12]. One such study reports that GLB youth are 50% more likely to have seriously considered suicide in the past 12 months, are twice as likely to attempt suicide at least once, and eight times as likely to report four or more attempts than their heterosexual peers [6]. One study reported that almost half (48%) of GLB youth who had thought about suicide reported that these thoughts were at least somewhat related to their sexual orientation [15].
This association can be explained in part by the greater prevalence of known risk factors in GLB adolescents, including depression, substance use and violence victimization [6], [7], [10], [11], [16], [17]. Other stressors theorized to account for this relationship include family conflict, ostracism at school, and broader stigmatizing socio-cultural factors such as homophobia [15], [16], [18], [19], [20]. For example, among GLB suicide attempters, almost half reported that their fathers were intolerant or rejecting of their sexual orientation, compared with approximately one-quarter among nonattempters [15]. Fear of harassment and mistreatment is widespread among GLB youth and may be linked to suicidal ideation [16], [21], and discrimination and stigmatization have been labeled the “primary culprit” in the epidemic of GLB youth suicide [2], [4], [20]. Furthermore, research suggests that certain risk factors, such as victimization at school, may be more potent for GLB youth than their heterosexual peers [17].
Despite these powerful negative forces, GLB youth are not a homogenous “at risk” group. Many have multiple “resiliency factors”—personal traits or characteristics of their social environment that protect young people from harm [22], [23], [24]–and indeed the majority of GLB adolescents grow up to lead happy, healthy, productive lives [20], [25]. Certain personal, family and community protective factors have been shown to affect the likelihood of negative outcomes among GLB youth and other vulnerable populations [24], [26], [27], [28], [29]. Identifying modifiable protective factors at multiple levels is critical to clinical assessment, intervention development and an ultimate reduction in suicide risk among GLB young people [25], [30], [31]. Indeed, the Surgeon General’s Call to Action to Prevent Suicide points out that “Risk and protective factors [for suicide] encompass genetic, neurobiological, psychological, social and cultural characteristics of individuals and groups and environmental factors such as easy access to firearms … Expanding the base of scientific evidence will help in the development of more effective interventions for these harmful behaviors” ([32], p. 8).
The present study, therefore, examines four potential protective factors repeatedly identified in research literature guided by a resiliency paradigm (family connectedness, teacher caring, other adult caring and school safety) and their association with suicidal ideation and attempts, hypothesizing that those characterized by higher levels of each factor will be less likely to report a history of these suicide behaviors. In addition, we examine the predicted probabilities of suicidal ideation and attempts among students with opposite- and same-gender experience, controlling for levels of each protective factor [23], [33].
Methods
The Minnesota Student Survey
Data come from the 2004 Minnesota Student Survey (MSS), a statewide survey of the health, safety and academic issues of 6th, 9th, and 12th grade students. The MSS has been conducted every three years since 1989, and includes measures of a wide variety of health-related attitudes and behaviors, including school climate, violent behavior, substance use, and connections with school, family and community. The secondary-level survey, given to students in grades 9 and 12, also includes measures of sexual activity. Data collection was undertaken in schools during class time. In most school districts, passive parental consent was used; a small number of districts required active parent consent. The University of Minnesota’s Institutional Review Board approved this secondary data analysis project using the MSS.
Eighty-eight percent of school districts in Minnesota participated. Student participation rates within school districts varied widely. Statewide, 75% of 9th graders and 55% of 12th graders participated. A small percentage of surveys (approximately 3%) was eliminated from analysis because gender was missing, responses were highly inconsistent, or there was a pattern of likely exaggeration. The final sample consisted of 83,731 students (41,044 males and 42,687 females). The racial/ethnic breakdown of participating 9th and 12th graders was: 80.5% white, 4.0% African American, 2.5% Hispanic or Latino, 4.8% Asian/Pacific Islander, 1.1% American Indian, 7.1% mixed race (checked more than one of the above) or “I don’t know.” Participants reporting no sexual partners (n = 51,901) were excluded from analysis, leaving an analytic sample of 21,927 sexually active youth.
Measures
The survey items “During the last 12 months, with how many different [male/female] partners have you had sexual intercourse” were crossed with gender of the respondent to identify participants who had a same-gender sexual experience. All those reporting one or more partners of their same gender were included in the GLB group. Participants with both male and female partners were grouped with those reporting only same-gender partners, as no significant differences were found between these groups on either suicide variable. Dependent variables were measured with the items “Have you ever thought about killing yourself” and “Have you ever tried to kill yourself.” Response options for each were “No,” “Yes, during the last year” and “Yes, more than a year ago.” The “Yes” options were combined for each to assess past suicidal ideation and attempts.
Four protective factors were measured using multiple survey items. Family connectedness included seven items: “Can you talk to your [father/mother] about problems you are having?”; “How much do you feel… a) your parents care about you?; b) your family cares about your feelings?; c) your family understands you?; d) your family has lots of fun together?; e) your family respects your privacy?” (Cronbach alpha = .87). Teacher caring included three items: “How much do you feel… teachers or other adults at school care about you?”; “How many of your teachers… a) are interested in you as a person?; b) show respect for the students?” (Cronbach alpha = .77). Other adult caring included three items: “How much do you feel … a) church or spiritual leaders care about you?; b) other adults in your community care about you?; c) other adult relatives care about you?” (Cronbach alpha = .67). School safety included: “I feel safe going to and from school,” “I feel safe at school,” and “Bathrooms in this school are a safe place to be” (Cronbach alpha = .84). For each factor, individual items were re-coded so all were in a uniform direction. The mean of the included items constituted the score for that factor. Each protective factor was then dichotomized for use in logistic regression to compare those reporting the highest quartile of each factor to those in the lowest three quartiles.
Participants were asked to describe themselves by checking one or more of six racial/ethnic groups, and race was included as a covariate in analysis. A dichotomous family structure variable was also included in analysis; all respondents who lived with two biological or adoptive parents were compared with those in other family-living situations.
Data Analysis
Bivariate analyses included chi-square tests of association between sexual orientation and each suicide measure and protective factor. Bivariate associations among sociodemographic factors and orientation, suicide behaviors, and protective factors were also examined using chi-square tests. Multiple logistic regression was used to examine the influence of sexual orientation and four protective factors (dichotomized) on suicidal ideation and suicide attempts. Because grade level (9th/12th), race and family structure were significantly associated with all key independent and dependent variables, they were included in logistic regression models to avoid confounding by these characteristics. General linear modeling was also used to calculate predicted probabilities of each dependent variable for all participants, controlling for four protective factors (used as continuous variables). A significance level of .01 was selected to reduce the Type I error rate, due to the very large size of the sample. All analyses were stratified by gender.
Results
Characteristics of the Sample
Overall, 2255 respondents (10.3% of sexually active participants) reported same-gender sexual experience. Among the 1452 GLB males, 1214 (83.6%) reported both male and female sexual partners and 238 (14.4%) reported only male partners. Among 803 GLB females, 694 (84.4%) reported partners of both genders, whereas 109 (13.6%) had only female partners. Of all GLB youth, 55.9% were in 9th grade, 67.2% were white, and half (49.9%) lived in a two-parent family. Over half (56.5%) of GLB students had thought about suicide and 37.4% reported a previous suicide attempt.
The distributions of sexual orientation in each demographic group are shown in Table 1. GLB youth were significantly more likely to be in 9th grade (53.5% of males, 60.2% of females) compared with non-GLB youth (35.7% of males, χ2 = 168.6, p < .001; 30.6% of females, χ2 = 296.4, p < .001). GLB youth were also less likely to be white than their non-GLB peers. GLB females were less likely to live in intact original families (42.1%) compared with non-GLB females (54.5%, χ2 = 45.9, p < .001).
Table 1. Characteristics of the sample (n, %)
| Total | Males | Females | |||||
|---|---|---|---|---|---|---|---|
| n = 21,927 | GLB n = 1452 (13.6) | non-GLB n = 9220 (86.4) | χ2 (p) | GLB n = 803 (7.1) | non-GLB n = 10,452 (92.9) | χ2 (p) | |
| Grade | 168.6 (<.001) | 296.4(<.001) | |||||
| 9th | 7748(35.3) | 777(53.5) | 3292(35.7) | 483(60.2) | 3196(30.6) | ||
| 12th | 14,179(65.7) | 675(46.5) | 5928(64.3) | 320(39.9) | 7256(69.4) | ||
| Race | 129.0(<.001) | 120.8(<.001) | |||||
| White | 17,364(79.5) | 970(67.0) | 7282(79.3) | 539(67.5) | 8573(82.3) | ||
| Black/African American | 931(4.3) | 90(6.2) | 453(4.9) | 43(5.4) | 345(3.3) | ||
| Hispanic | 678(3.1) | 81(5.6) | 289(3.2) | 35(4.4) | 273(2.6) | ||
| Asian | 729(3.3) | 71(4.9) | 303(3.3) | 31(3.9) | 324(3.1) | ||
| Native American | 332(1.5) | 25(1.7) | 150(1.6) | 21(2.6) | 136(1.3) | ||
| Mixed/other | 1815(8.3) | 211(14.6) | 702(7.7) | 130(16.3) | 772(7.4) | ||
| Original family (yes) | 11,832(54.3) | 776(54.2) | 5039(55.2) | .46(NS) | 336(42.1) | 5681(54.5) | 45.9(<.001) |
As expected, GLB students were significantly more likely to report thinking about and attempting suicide than non-GLB students, for both males and females (Table 2). For example, over half (52.4%) of GLB girls reported a suicide attempt, compared with 24.8% of non-GLB girls (χ2 = 283.0, p < .001).
Table 2. Suicide behaviors and protective factors (dichotomized) in the sample (n, %)
| Males | Females | |||||
|---|---|---|---|---|---|---|
| GLB (n = 1452) | non-GLB (n = 9220) | χ2 (p) | GLB (n = 803) | non-GLB (n = 10,452) | χ2 (p) | |
| Behaviors | ||||||
| Suicidal ideation | 656(47.3) | 3127(34.7) | 82.5(<.001) | 567(72.9) | 5484(53.0) | 114.8(<.001) |
| Suicide attempt | 402(29.0) | 1135(12.6) | 255.4(<.001) | 409(52.4) | 2559(24.8) | 283.0(<.001) |
| Protective factors (high) | ||||||
| Family connectedness | 355(24.5) | 2827(30.7) | 22.6(<.001) | 125(15.6) | 2957(28.3) | 60.5(<.001) |
| Teacher caring | 417(28.8) | 3460(37.6) | 41.4(<.001) | 189(23.6) | 4200(40.2) | 85.8(<.001) |
| Other adult caring | 320(23.1) | 2314(25.7) | 4.4(.035) | 133(17.1) | 2960(28.5) | 47.3(<.001) |
| Safe school | 295(20.5) | 2815(30.7) | 61.8(<.001) | 110(13.8) | 2892(27.7) | 73.2(<.001) |
Protective Factors
GLB males and females were less likely than non-GLB youth to be in the top quartile of each protective factor, as shown in Table 2. For example, 15.6% of GLB females reported high family connectedness compared with 28.3% of non-GLB females (χ2 = 60.5, p < .001). Similarly, there was a difference of 10 percentage points between GLB males reporting a high degree of safety in school (20.5%) and non-GLB males reporting safety in school (30.7%, χ2 = 61.8, p < .001).
As seen in previous literature, GLB students were significantly more likely to report thinking about and attempting suicide than non-GLB students, for both males and females (Table 3). Logistic regression models indicated that youth with a GLB orientation had odds of suicidal ideation and attempts that were 1.60–2.63 times the odds for non-GLB youth, controlling for grade level, race and family structure. When the four protective factors were added to each model, the odds ratios (OR) for GLB orientation were attenuated to 1.35–2.16, but remained statistically significant. As hypothesized, family connectedness, other adult caring and school safety were all significant protective factors against suicide behaviors. In particular, youth with high family connectedness scores had odds of suicidal ideation that were approximately half the odds for youth with lower family connectedness scores (OR = .52 for males; OR=.53 for females). In all models, family connectedness accounted for a much greater amount of the variance in suicide behaviors than sexual orientation or any other protective factor (χ2 = 267.0–452.5). Teacher caring was not a significant protective factor in the context of other variables in these models.
Table 3. Odds ratios (99% CIs) for suicidal ideation and attempts
| Suicidal ideation | Suicide attempts | |||||
|---|---|---|---|---|---|---|
| OR | 99% CI | χ2 (p) | OR | 99% CI | χ2 (p) | |
| Males | ||||||
| GLB sexual orientationa | 1.60 | 1.37,1.86 | 61.9(<.001) | 2.49 | 2.08,2.98 | 173.2(<.001) |
| GLB sexual orientationb | 1.35 | 1.14,1.59 | 20.9(<.001) | 2.06 | 1.69,2.50 | 89.1(<.001) |
| Family connectedness | .52 | .48,.56 | 452.5(<.001) | .51 | .46,.57 | 268.5(<.001) |
| Teacher caring | 1.00 | .92,1.07 | .03(NS) | .99 | .90,1.10 | .02(NS) |
| Other adult caring | .92 | .85,.99 | 7.5(.006) | .89 | .80,.99 | 8.5(.004) |
| Safe school | .83 | .76,.91 | 27.8(<.001) | .70 | .62,.78 | 65.9(<.001) |
| Females | ||||||
| GLB sexual orientationa | 1.92 | 1.54,2.40 | 56.7(<.001) | 2.63 | 2.15,3.23 | 151.6(<.001) |
| GLB sexual orientationb | 1.54 | 1.21,1.95 | 21.4(<.001) | 2.16 | 1.74,2.68 | 83.4(<.001) |
| Family connectedness | .53 | .50,.57 | 441.7(<.001) | .60 | .55,.65 | 267.0(<.001) |
| Teacher caring | .97 | .89,1.04 | 1.4(NS) | 1.01 | .93,1.11 | .17(NS) |
| Other adult caring | .82 | .76,.88 | 47.6(<.001) | .81 | .74,.88 | 45.8(<.001) |
| Safe school | .83 | .75,.91 | 23.8(<.001) | .73 | .66,.81 | 55.1(<.001) |
aAdjusted for grade level, race and family structure. |
bAdjusted for grade level, race, family structure and four protective factors. |
Predicted Probabilities of Suicidal Ideation and Attempts
Using general linear modeling controlling only for grade level, race and family structure, 45.4% of male and 62.8% of female GLB students are predicted to report suicidal ideation, compared with 34.3% of non-GLB male and 49.0% of non-GLB female students (Table 4). These models explain approximately 2% and 6% of the variability in suicidal ideation for males and females, respectively. When all four protective factors are added to these models, the overall model R2 increases to 12% for males and 14% for females, and the predicted probability of suicidal ideation falls to 39.7% for GLB males and 55.8% for GLB females. A similar pattern is evident for suicide attempts: after accounting for the role of protective factors, the predicted probability of a suicide attempt drops (e.g., from 45.4% to 39.6% among GLB females).
Table 4. Predicted probabilities of suicidal ideation and attempts among GLB and non-GLB youtha
| Suicidal ideation | Suicide attempts | |||
|---|---|---|---|---|
| Sexual orientation | Orientation + protective factors | Sexual orientation | Orientation + protective factors | |
| Males | R2 = .02 | R2 = .12 | R2 = .04 | R2 = .12 |
| GLB | 45.4 | 39.7 | 29.0 | 24.4 |
| Non-GLB | 34.3 | 33.3 | 14.7 | 13.7 |
| Females | R2 = .06 | R2 = .14 | R2 = .07 | R2 = .14 |
| GLB | 62.8 | 55.8 | 45.4 | 39.6 |
| Non-GLB | 49.0 | 48.4 | 23.7 | 23.1 |
aAdjusted for grade level, race and family structure. |
Discussion
Findings from the current study are in keeping with the body of evidence that GLB youth are at greater risk for suicidal ideation and attempts than non-GLB youth [6], [7], [8], [9], [11], [12]. It is important to note, however, that sexual orientation alone accounted for only a small portion of the variability in suicidal ideation and attempts. Four protective factors contributed greater explanatory power, and the largest proportion of the variance was unexplained by variables used here, indicating that many other factors influence suicide risk above and beyond a GLB orientation. This finding is consistent with previous work demonstrating the important mediating effect of protective factors in suicidality among GLB youth. Safren and Heimberg [27], for example, found that sexual orientation contributed relatively little to the explanatory power of multiple regression models for suicide behaviors after accounting for additional personal and environmental risk and protective factors, such as satisfaction with social support and depression.
The important role of protective factors in the lives of general population and higher-risk young people has been documented [22], [29]. Previous studies have indicated that GLB youth suffer a variety of challenges related to their sexual orientation, including family conflict and rejection, and violence victimization [15], [17], [21]; it is therefore not surprising that lower levels of protective factors were reported among GLB participants in the current study. In spite of their lower levels, however, family connectedness, other adult caring and safe schools provided significant protection against suicidal ideation and attempts for GLB participants. One previous study examining the role of protective factors did not find these significant associations [26]. That study, however, relied on a global measure of satisfaction with social support, rather than distinct domains of support, and included a smaller sample (n = 90) of youth who self-identified as GLB. These methodologic differences, as well as demographic differences between that sample and the current study, may account for the difference in findings.
Predicted probabilities for suicide behaviors in this analysis point to the potential for obviating suicide risk among GLB youth. If all GLB youth reported protective factors at the level of non-GLB youth, suicidal ideation and attempts are expected to be considerably lower. For example, the predicted probability of a suicide attempt among GLB girls was 5.8 percentage points lower after introducing protective factors into the model. If this difference is extrapolated to 2% (the percentage of females reporting GLB orientation in the full MSS sample) of all 15–19-year-old girls in the United States (roughly 10 million [34]), one would expect approximately 12,000 fewer suicide attempts among females in this age group across the country.
Strengths and limitations
The current study has several strengths. First, the sample size is very large, which allowed for meaningful analysis of a group of youth who are often not well represented in survey research. The population-based sampling frame (i.e., schools) also enables us to generalize to a broader population of adolescents, which is often not possible in studies of stigmatized minorities. In addition, the response rate at the school level is fairly high (88%), so findings are likely to represent a good cross-section of youth in the state. However, student participation rates are somewhat lower. Differences between responders and nonresponders could not be assessed and may have biased the analysis. The measures used to determine sexual orientation groups are another limitation of the study. Saewyc et al [35] have suggested that a multi-dimensional assessment of sexual orientation (including attraction, behavior and self-labeling) may be needed to accurately identify GLB young people, and that sexual behavior may not be the best single item. The MSS, however, included only a behavioral assessment of sexual orientation. In particular, youth who self-identify as GLB may represent a different subset of the population than the group captured in this study. Results should therefore be interpreted cautiously. In addition, the student survey item asks about sexual behaviors only during the past year; those who had same-gender experiences more than one year prior would not be included in the GLB group. This misclassification may have biased results toward the null. Another limitation of the study is that additional risk and protective factors for suicide behaviors were not included in the analysis, thus, the large majority of variance in the dependent variables was not explained by factors examined here. Finally, this study uses data collected in a single mid-Western state with limited racial and ethnic diversity. Replication of these findings in largely minority and low income populations is needed.
Implications
The present study builds on existing knowledge of the importance of protective factors in the lives of adolescents, even in cases where young people are at increased risk of negative outcomes. These findings also pinpoint factors that may be particularly salient for GLB youth, including support from family and other adults, and a safe school environment. The reasons for the importance of these factors above others (e.g., teacher caring) are not clear. Future research should include qualitative methods to delve more deeply into the roles these and other protective factors play in the lives of GLB youth. Longitudinal research is also needed to determine temporality and begin to uncover any causal associations between protective factors and emotional well-being.
Nevertheless, these findings suggest that programs, policies and resources are needed to support GLB adolescents, as these youth remain at considerably increased risk for suicidal ideation and attempts. Family connectedness, support from other adults, and school safety are all characteristics that are amenable to change, and would be appropriate targets for interventions aimed at protecting young people from self-harm. Improving the ability of parents and other influential adults to connect with and support adolescents grappling with issues of sexual identity may be a critical component of mental health promotion and protection for these young people.
Acknowledgments
Notice of source of data: Minnesota Student Survey provided by public school students in Minnesota via local public school districts and managed by the Minnesota Student Survey Interagency Team, 2004. This work was supported by Cooperative Agreement Number 1-U48-DP-000063, Prevention Research Centers, from the Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention.
References
- . Suicide among homosexual adolescents. J Homosex. 1987;13(4):111–117
- . Mental health issues of gay and lesbian adolescents. J Adolesc Health Care. 1988;9:114–122
- . Gay male and lesbian youth suicide. Alcohol, Drug Abuse, and Mental Health Administration. Report of the Secretary’s Task Force on Youth Suicide, Volume 3: Prevention and Interventions in Youth Suicide (DHHS Pub No. (ADM)89-1623). Washington, DC: Superintendent of Documents, US Government Printing Office; 1989;
- . Verbal and physical abuse as stressors in the lives of lesbian, gay male, and bisexual youths: associations with school problems, running away, substance abuse, prostitution and suicide. J Consult Clin Psychol. 1994;62(2):261–269
- Gender differences in health and risk behaviors among bisexual and homosexual adolescents. J Adolesc Health. 1998;23:181–188
- . Correlates of same-sex sexual behavior in a random sample of Massachusetts high school students. Am J Public Health. 1998;88:262–266
- The association between health risk behaviors and sexual orientation among a school-based sample of adolescents. Pediatrics. 1998;101(5):895–902
- The relationship between suicide risk and sexual orientation: results of a population-based study. Am J Public Health. 1998;88(1):57–60
- Sexual orientation and risk of suicide attempts among a representative sample of youth. Arch Pediatr Adolesc Med. 1999;153(5):487–493
- . Suicidal acts and ideation in homosexual and bisexual young people: a study of prevalence and risk factors. Soc Psychiatry Psychiatr Epidemiol. 2000;35:494–499
- . Adolescent sexual orientation and suicide risk: evidence from a national study. Am J Public Health. 2001;91(8):1276–1281
- . Risk assessment of adolescents with same-sex relationships. J Adolesc Health. 2002;31(1):84–92
- . Suicide and sexual orientation: nearing the end of controversy?. Arch Gen Psychiatry. 1999;56:885–886
- . Sexual orientation and youth suicide. JAMA. 1999;282(13):1291–1292
- . Suicidality patterns and sexual orientation-related factors among lesbian, gay, and bisexual youths. Suicide Life Threat Behav. 2001;31(3):250–264
- . Elevated rates of suicidal behavior in gay, lesbian, and bisexual youth. Crisis. 2000;21(3):111–117
- . Effects of at-school victimization and sexual orientation on lesbian, gay, or bisexual youths’ health risk behavior. J Adolesc Health. 2002;30(5):364–374
- . Lesbian, gay, and bisexual youth and their families: disclosure of sexual orientation and its consequences. Am J Orthopsychiatry. 1998;68(3):361–371discussion 372–5
- . Practice parameter for the assessment and treatment of children and adolescents with suicidal behavior. J Am Acad Child Adolesc Psychiatry. 2001;40(suppl 7):24S–51S
- . Health care issues of gay and lesbian youth. Curr Opin Pediatr. 2001;13:298–302
- . Experiences of harassment, discrimination, and physical violence among young gay and bisexual men. Am J Public Health. 2004;94(7):1200–1203
- Protecting adolescents from harm (Findings from the National Longitudinal Study on Adolescent Health). JAMA. 1997;278(10):823–832
- . Adolescent suicide attempts: risks and protectors. Pediatrics. 2001;107(3):485–493
- . Life on the seesaw: a qualitative study of suicide resiliency factors for young gay men. J Homosex. 2003;45(1):1–22
- . A critique of research on sexual-minority youths. J Adolesc. 2001;24(1):5–13
- . Self-esteem and supportiveness as predictors of emotional distress in gay male and lesbian youth. J Homosex. 1998;35(2):25–39
- . Depression, hopelessness, suicidality, and related factors in sexual minority and heterosexual adolescents. J Consult Clin Psychol. 1999;67(6):859–866
- Preventing sexual risk behaviors among gay, lesbian, and bisexual adolescents: the benefits of gay-sensitive HIV instruction in schools. Am J Public Health. 2001;91(6):940–946
- . Violence risk and protective factors among youth held back in school. Ambul Pediatr. 2002;2(6):475–484
- . Developmental and adjustment issues of gay, lesbian, and bisexual adolescents: a review of the empirical literature. Clin Child Fam Psychol Rev. 1998;1(4):215–230
- . Suicide and gay/lesbian/bisexual youth: implications for clinicians. J Adolesc. 2001;24(1):39–49
- . The Surgeon General’s Call to Action to Prevent Suicide. Washington, DC: US Public Health Service; 1999;
- . Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. J Adolesc Health. 2004;35:424;e1–10
- U.S. Census Bureau. Table 1: Annual Estimates of the Population by Sex and Five-Year Age Groups for the United States: April 1, 2000 to July 1, 2004 (NC-EST2004-01). Population Division, U.S. Census Bureau: June 9, 2005.
- Measuring sexual orientation in adolescent health surveys: evaluation of eight school-based surveys. J Adolesc Health. 2004;35(4):345.e1–345.e15
PII: S1054-139X(06)00171-6
doi:10.1016/j.jadohealth.2006.04.024
© 2006 Society for Adolescent Medicine. Published by Elsevier Inc. All rights reserved.
