Abstract
Objectives. We examined the relationships among sexual minority status, sex, and mental health and suicidality, in a racially/ethnically diverse sample of adolescents.
Methods. Using pooled data from 2005 and 2007 Youth Risk Behavior Surveys within 14 jurisdictions, we used hierarchical linear modeling to examine 6 mental health outcomes across 6 racial/ethnic groups, intersecting with sexual minority status and sex. Based on an omnibus measure of sexual minority status, there were 6245 sexual minority adolescents in the current study. The total sample was n = 72 691.
Results. Compared with heterosexual peers, sexual minorities reported higher odds of feeling sad; suicidal ideation, planning and attempts; suicide attempt treated by a doctor or nurse, and self-harm. Among sexual minorities, compared with White youths, Asian and Black youths had lower odds of many outcomes, whereas American Native/Pacific Islander, Latino, and Multiracial youths had higher odds.
Conclusions. Although in general, sexual minority youths were at heightened risk for suicidal outcomes, risk varied based on sex and on race/ethnicity. More research is needed to better understand the manner in which sex and race/ethnicity intersect among sexual minorities to influence risk and protective factors, and ultimately, mental health outcomes.
Over the past 20 years, research has documented elevated suicidality1—defined as behavior related to contemplating, attempting, or completing suicide2—among sexual minority youths (an umbrella term, generally including those who identify as lesbian, gay, bisexual, or transgender [LGBT]; engage in same-sex sexual behavior; or have same-sex attractions). This research has consistently demonstrated substantial sexual orientation disparities in suicidality, with sexual minority youths having higher prevalence of suicidality than their heterosexual peers.3–6 A recent review of the literature indicated that sexual minority youths are at least twice as likely as heterosexual youths to contemplate suicide, and 2 to 7 times as likely to attempt suicide.7 A meta-analysis found that 28% of sexual minority youths had a history of suicidality, compared with 12% of their heterosexual peers.8
Despite the development of knowledge about suicidality among sexual minority youths, little is known about suicidality in sexual minority youths of color. To the extent that existing researchers have explored racial/ethnic differences, analyses have rarely gone beyond dichotomous (White vs “youths of color”) or trichotomous (White vs Black vs Latino) comparisons. As a result, there exists scant literature exploring the full spectrum of racial/ethnic differences in suicidality among sexual minority youths.
The literature on suicide in the general adolescent population demonstrates racial/ethnic differences in suicide ideation and attempts. For example, prevalence of suicide among Native American and Alaska Native youths is twice that of other youths,9 and Latino youths are more likely than either Black or White youths to have considered and attempted suicide.10 Differences are further moderated by participants’ gender: girls are more likely to consider suicide and attempt suicide than boys,10 although boys are more likely to complete suicide.11
Studies that have considered racial/ethnic differences in suicidality among sexual minority youths have found differences, though the patterns have been inconsistent. A study based on Youth Risk Behavior Survey (YRBS) data from Massachusetts, reported that among self-identified lesbian, gay, or bisexual (LGB) youths3 Latinos were significantly more likely than Whites to report past-year suicide attempt. Another study found that same-sex–attracted Black and White youths were more likely than their other-sex–attracted peers to report suicidal ideation, whereas same-sex–attracted Latino and Asian/Pacific Islander youths did not differ from other-sex–attracted peers.12 In a nonprobability sample of urban LGBT youths,13 Black and White youths were more likely to report suicidal ideation than Latinos; however, Latinos reported the highest frequency of suicide attempts. A study of New York City adults found that Latino and Black LGB participants were more likely to report serious suicide attempts than were White LGB participants, with most reported attempts occurring during adolescence and young adulthood.14 These conflicting results suggest that there are important differences in suicidality at the intersections of sexual minority status and race/ethnicity, yet further study requires data of sufficient scale and scope to enable analyses of low-prevalence behaviors across small subgroups of youths.
To address the need for information about suicidality among racially/ethnically diverse sexual minority populations, we assess suicidality patterns among youths based on sexual orientation, race/ethnicity, and sex. With this, we respond to calls for public health to utilize minority stress and intersectional frameworks as potential lenses through which to understand health and health disparities among sexual minority populations.15,16 Rather than treating social identities as separate and discrete phenomena, our inquiry allows that co-occurring minority identities operate together. An intersectional approach suggests that sexual identity–race–sex intersections are informed by unique cultural, historical, social, and political factors that differentially influence life experiences, including discrimination based on such identities.17–19 In turn, minority stress theory posits that discriminatory experiences predispose populations to stress and adverse mental health outcomes, including suicidality.20
The focus on health differences among sexual minority youths across race/ethnicity and sex is vital to creating effective health interventions and programs. Such a focus is particularly relevant within the context of youth suicide, as risk and protective factors associated with suicidality vary across both racial/ethnic and sexual minority groups, and there is a need to better integrate these bodies of research.21
METHODS
We conducted analyses using pooled 2005 and 2007 YRBS data from several jurisdictions that included 1 or more measures of sexual orientation. The general approach to pooling the data and analyzing the pooled dataset, along with the sexual orientation items and characteristics of the sample by jurisdiction, are described in detail elsewhere in this issue.22 The current study analyzed data from the 14 jurisdictions that measured sexual orientation identification, sex of sexual partners, or sex of sexual attraction, including Boston, Massachusetts; Chicago, Illinois; Connecticut; Delaware; Hawaii; Maine; Massachusetts; New York City, New York; San Diego, California; San Francisco, California; Vermont; Rhode Island; Wisconsin; and Milwaukee, Wisconsin. Because the outcome variables were not assessed in all jurisdictions, the total sample size varies for each model depending on which outcome is examined. Respondents who did not answer the sexual orientation questions were excluded from analysis. Nonresponse to questions about sexual identity, behavior, and attraction was 3.2%, 3.9%, and 1.8% respectively.22
Also excluded were participants who were 12 years old or in seventh grade because of the very small number of such participants and concerns about the quality of the data among this group. The unweighted final sample size was n = 73 154.
MEASURES
All measures, including demographic characteristics and sexual orientation, were assessed via self-report. The measurement and pooling of sexual orientation and race/ethnicity questions are described elsewhere.22
Sexual Orientation
A binary variable, constructed from self-reported sexual identity, behaviors, and attractions, indicated whether participants were classifiable as sexual minority versus sexual majority. Participants who reported a nonheterosexual identity, any same-sex behavior, or any same-sex attractions were labeled as sexual minority; all others were coded sexual majority. The use of the omnibus “sexual minority status” variable allowed for the broadest conceptualization of sexual orientation and ensured sufficient power to detect effects among many different racial groups. Given the developmental stage of participants, during which each of these dimensions of sexuality emerges, this broad definition of sexual minority status is warranted.21 See Mustanski et al.22 for a fuller discussion of how this variable was created.
Race/Ethnicity
We constructed a nominal race/ethnicity variable by collapsing 8 racial/ethnic groups into 6 groups to assure adequate cell sizes. The 6 subgroups were: (1) American Native/Pacific Islander (AN/PI, henceforth), which combined American Indian/Alaska Native with Native Hawaiians/Pacific Islanders; (2) Asian; (3) Black; (4) White; (5) Hispanic/Latino; and (6) Multiracial, which combined 2 existing multiracial categories (multiracial Hispanic/Latino and multiracial Non-Hispanic/Latino) into a single group. White participants were used as the reference group for these analyses.
Sex
We asked the participants, “What is your sex?” The response options were coded as 0 = male and 1 = female. In stratified analyses, male participants were used as the reference group. No jurisdiction included transgender as an option.
Outcome Variables
Feel sad.
We asked the participants, “During the past 12 months, did you ever feel so sad or hopeless almost every day for 2 weeks or more in a row that you stopped doing some usual activities?” (Yes = 1, No = 0).
Suicide ideation.
We asked the participants, “During the past 12 months, did you ever seriously consider attempting suicide?” (Yes = 1, No = 0). Because Vermont did not assess suicidal thoughts, data from Vermont was excluded from models of this outcome.
Suicide plan.
We asked the participants, “During the past 12 months, did you make a plan about how you would attempt suicide?” (Yes = 1, No = 0).
Suicide attempts.
We assessed this variable by asking, “During the past 12 months, how many times did you actually attempt suicide?” The response options ranged from 0 times to 6 or more times. Responses were collapsed and dichotomized as “any” or “none.”
Suicide attempt treated by a doctor or nurse.
We asked the participants, “If you attempted suicide during the past 12 months, did any attempt result in an injury, poisoning, or overdose that had to be treated by a doctor or nurse?” The response options were (1) did not attempt suicide during the past 12 months, (2) yes, and (3) no. In the multivariate models, only those who reported a suicide attempt (answered 2 or 3 to the question) where included. Because Connecticut did not assess suicide attempt treated by a doctor or nurse, data from Connecticut were excluded from models of this outcome.
Self-Harm.
In most jurisdictions that included this item, we asked the participants, “During the past 12 months, how many times did you do something to purposely hurt yourself without wanting to die, such as cutting or burning yourself on purpose?” The responses ranged from 0 times to 12 or more times. Variations to question and response wording occurred in Delaware in 2005 and 2007: “During the past 12 months, have you done something to purposely hurt yourself without wanting to die, such as cutting, scraping, or burning yourself on purpose?” (Yes/No). In 2005, Massachusetts and Boston asked, “During the past 12 months, how many times did you hurt or injure yourself on purpose without wanting to die?” with responses ranging from 0 times to 20 or more times. Finally, the Boston 2007 YRBS asked, “During the past 12 months, how many times did you do something to purposely hurt yourself without wanting to die, such as cutting or burning, or bruising yourself on purpose?” Responses were collapsed and dichotomized as “any” or “none”.
Age.
We used a continuous age variable as a control in all multivariate models. Participants included in the analyses were 13 to 18 years old.
Data Analysis
Descriptive analyses were conducted using SPSS version 21.0 (IBM, Somers, NY) to examine the distribution and associations of the study variables. Given the complex sampling design of the YRBS administration, the complex samples module of SPSS 21.0 was utilized for all analyses within SPSS. Final models were fit using the multilevel software HLM version 7 (Scientific Software International, Lincolnwood, IL) to account for jurisdictional clustering of the data. HLM analyses accounted for the complex sampling design of the YRBS by adjusting the relative weights and altering the effective sample size using design effects (DEFTs) calculated for each jurisdiction. The approach to calculating design effects and accounting for the clustering of the data are described in detail elsewhere.22 Full-information maximum likelihood estimation was used. We identified significant effects as having associated P values of less than 0.05.
RESULTS
The demographic characteristics of the total sample, as well as the racial/ethnic subsamples, are presented in Table 1. Based on our omnibus definition, 8.4% of the respondents were sexual minorities. There was some variability across racial/ethnic groups in the proportion of youths who were sexual minority, with proportions ranging from 7.4% among Asian youths to 14.0% among AN/PI youths.
TABLE 1—
Sample Characteristics by Race/Ethnicity: United States, 2005 and 2007 Youth Risk Behavior Surveys
| Variable | Total Samplea (n = 72 691), No. (%) | American Native/Pacific Islanderb (n = 2057), No. (%) | Asian (n = 7028), No. (%) | Black (n = 11 274), No. (%) | White (n = 33 028), No. (%) | Hispanic/Latino (n = 9626), No. (%) | Multiracial (n = 7905), No. (%) |
| Age, y | |||||||
| 13 | 2183 (0.5) | 63 (1.5) | 59 (0.6) | 50 (0.1) | 1751 (0.7) | 46 (0.2) | 164 (0.6) |
| 14 | 10 495 (12.6) | 399 (20.7) | 1025 (18.4) | 1182 (11.7) | 5260 (11.1) | 1157 (13.3) | 1250 (15.3) |
| 15 | 17 656 (26.2) | 532 (29.6) | 1635 (25.6) | 2677 (26.9) | 7932 (24.8) | 2384 (28.0) | 2127 (30.1) |
| 16 | 18 384 (25.6) | 517 (25.3) | 1739 (23.7) | 3051 (25.6) | 7929 (25.9) | 2673 (26.0) | 2079 (25.6) |
| 17 | 16 127 (22.7) | 371 (16.4) | 1674 (19.7) | 2841 (22.7) | 7067 (24.2) | 2231 (21.8) | 1583 (19.1) |
| 18 | 7550 (12.4) | 174 (6.6) | 890 (12.1) | 1446 (13.0) | 3066 (13.3) | 1119 (10.7) | 683 (9.3) |
| Sex | |||||||
| Male | 35 501 (50.3) | 1131 (55.1) | 3596 (54.2) | 5416 (48.0) | 16 168 (50.9) | 4588 (48.9) | 3749 (48.4) |
| Female | 36 678 (49.7) | 912 (44.9) | 3404 (45.8) | 5813 (52.0) | 16 730 (49.1) | 5017 (51.1) | 4102 (51.6) |
| Sexual orientation status | |||||||
| Sexual majorityc | 66 446 (91.6) | 1781 (86.0) | 6462 (92.6) | 10 255 (91.1) | 30 657 (92.5) | 8792 (91.7) | 6930 (87.5) |
| Sexual minorityd | 6245 (8.4) | 276 (14.0) | 566 (7.4) | 1019 (8.9) | 2371 (7.5) | 834 (8.3) | 975 (12.5) |
Note. The sample size was n = 72 691.
Percentages are based on weighted data; reported numbers reflect unweighted data. Total and percentages may differ because of missing, excluded cases, and weighting.
Alaska Native/Native Hawaiian/Other Pacific Islander/American Indian.
Respondents who identified as heterosexual, and reported no same-sex behavior or same-sex attraction.
Respondents who reported either a lesbian, gay, bisexual, or unsure identity; any same-sex attraction; or any same-sex behavior.
Table 2 presents the prevalence of all outcome variables among the 6 racial/ethnic groups, based on sexual minority status and sex. Table 3 presents the odds ratios and 95% confidence intervals for the differences in the 6 outcome variables among groups defined by race/ethnicity, sex, and sexual minority status for the entire sample. Compared with sexual majority youths, sexual minorities had a higher prevalence of each outcome—often 3-fold the odds. For example, suicidal ideation (OR = 3.2), making a suicide plan (OR = 3.2), any suicide attempt (OR = 3.8), and any self-harm (OR = 3.2) were all significantly higher among sexual minority youths.
TABLE 2—
Prevalence of Mental Health and Suicidality Outcomes by Race/ethnicity, Sexual Minority Status and Sex: United States, 2005 and 2007 Youth Risk Behavior Surveys
| Feel Sad (n = 71 776) |
Suicide Ideation (n = 54 595) |
Suicide Plan (n = 71 972) |
Suicide Attempts (n = 63 456) |
Suicide Attempt Treated by Doctor or Nurse (n = 61 004) |
Self-Harm (n = 17 679) |
|||||||
| Variable | Sexual Minority, % | Sexual Majority, % | Sexual Minority, % | Sexual Majority, % | Sexual Minority, % | Sexual Majority, % | Sexual Minority, % | Sexual Majority, % | Sexual Minority, % | Sexual Majority, % | Sexual Minority, % | Sexual Majority, % |
| Total sample | 48.1 | 24.5 | 32.2 | 11.7 | 27.4 | 9.7 | 22.8 | 6.6 | 8.3 | 2.0 | 39.1 | 14.2 |
| Sex | ||||||||||||
| Male | 37.6 | 18.1 | 25.4 | 8.6 | 22.4 | 8.0 | 20.9 | 5.4 | 9.1 | 2.0 | 30.9 | 10.1 |
| Female | 54.7 | 31.2 | 36.4 | 14.9 | 30.5 | 11.4 | 23.9 | 7.7 | 7.8 | 2.1 | 44.6 | 18.4 |
| Race/ethnicity | ||||||||||||
| Alaska Native/Pacific Islander | 53.8 | 33.4 | 33.6 | 18.1 | 36.5 | 15.6 | 32.2 | 14.5 | 12.8 | 4.4 | 59.9 | 14.8 |
| Asian | 42.9 | 24.5 | 25.2 | 13.2 | 22.4 | 11.5 | 21.1 | 6.4 | 8.7 | 1.8 | 31.3 | 13.7 |
| Black | 38.7 | 26.4 | 26.0 | 11.3 | 22.1 | 9.8 | 20.7 | 7.8 | 5.3 | 2.7 | 18.3 | 8.2 |
| White | 48.0 | 20.5 | 33.9 | 10.9 | 27.5 | 8.7 | 21.1 | 4.9 | 8.8 | 1.4 | 42.5 | 15.4 |
| Hispanic/Latino | 55.2 | 32.3 | 34.5 | 11.8 | 25.9 | 9.7 | 26.9 | 9.0 | 7.4 | 2.5 | 35.0 | 11.7 |
| Multiracial | 59.4 | 31.2 | 37.3 | 14.9 | 34.6 | 13.1 | 26.9 | 9.8 | 8.2 | 3.2 | 41.7 | 16.3 |
| Females | ||||||||||||
| Alaskan Native/Pacific Islander | 65.5 | 43.4 | 42.9 | 25.2 | 36.1 | 20.0 | 30.8 | 17.9 | 14.5 | 3.1 | 64.0 | 19.7 |
| Asian | 51.6 | 29.0 | 28.6 | 16.5 | 27.1 | 14.3 | 23.6 | 8.7 | 5.9 | 1.9 | 42.3 | 18.5 |
| Black | 45.1 | 33.2 | 29.5 | 14.1 | 25.0 | 11.0 | 20.1 | 7.9 | 4.5 | 2.5 | 11.7 | 10.1 |
| White | 55.5 | 26.6 | 38.4 | 13.8 | 31.6 | 10.2 | 22.6 | 5.9 | 9.3 | 1.6 | 50.4 | 20.4 |
| Hispanic/Latino | 56.4 | 40.3 | 38.5 | 14.9 | 29.4 | 11.4 | 30.4 | 10.8 | 7.3 | 2.6 | 36.0 | 13.7 |
| Multiracial | 63.3 | 40.0 | 38.2 | 20.1 | 35.0 | 16.7 | 27.4 | 11.5 | 6.2 | 3.2 | 39.0 | 18.8 |
| Males | ||||||||||||
| Alaska Native/Pacific Islander | 42.6 | 25.3 | 25.8 | 12.4 | 36.9 | 12.0 | 32.7 | 11.7 | 10.9 | 5.4 | 55.1 | 12.6 |
| Asian | 34.8 | 20.7 | 22.1 | 10.4 | 18.0 | 9.2 | 19.0 | 4.2 | 11.4 | 1.7 | 17.7 | 9.7 |
| Black | 28.9 | 18.9 | 20.7 | 8.1 | 17.6 | 8.2 | 21.7 | 7.7 | 6.9 | 2.9 | 23.3 | 6.3 |
| White | 34.8 | 14.9 | 26.2 | 8.1 | 20.7 | 7.3 | 18.5 | 5.9 | 8.1 | 1.2 | 28.6 | 10.7 |
| Hispanic/Latino | 53.6 | 24.1 | 27.6 | 8.6 | 19.9 | 8.0 | 20.2 | 7.0 | 7.5 | 2.3 | 34.9 | 9.7 |
| Multiracial | 51.3 | 22.7 | 34.2 | 9.8 | 32.9 | 9.7 | 25.0 | 7.8 | 12.2 | 2.8 | 46.0 | 12.6 |
Note. Percentages are based on weighted data; reported numbers reflect unweighted data.
TABLE 3—
Main Effects for Sex, Sexual Minority Status, and Race/Ethnicity on Mental Health and Suicidality Outcomes: United States, 2005 and 2007 Youth Risk Behavior Surveys
| Variable | Feel Sad, OR (95% CI) | Suicide Ideation, OR (95% CI) | Suicide Plan, OR (95% CI) | Suicide Attempts, OR (95% CI) | Suicide Attempt Treated by Doctor or Nurse,a OR (95% CI) | Self-Harm, OR (95% CI) |
| Female sex | 2.03 (1.92, 2.14) | 1.82 (1.70, 1.96) | 1.48 (1.38, 1.59) | 1.44 (1.31, 1.57) | 0.49 (0.40, 0.61) | 2.07 (1.86, 2.30) |
| Sexual minority | 2.42 (2.22, 2.64) | 3.20 (2.89,3.54) | 3.21 (2.91, 3.55) | 3.85 (3.42, 4.33) | 1.32 (1.05, 1.66) | 3.25 (2.79, 3.77) |
| Race/ethnicity (Ref: White) | ||||||
| Alaska Native/Pacific Islander | 1.61 (1.36, 1.90) | 1.21 (0.97,1.51) | 1.52 (1.22, 1.88) | 2.37 (1.83, 3.07) | 1.20 (0.73, 1.97) | 1.58 (1.14, 2.15) |
| Asian | 1.07 (0.97, 1.18) | 1.08 (0.89,1.14) | 1.12 (0.99, 1.27) | 1.13 (0.95, 1.34) | 1.23 (0.86, 1.78) | 0.85 (0.69, 1.06) |
| Black | 1.08 (1.00, 1.18) | 0.88 (0.80, 0.99) | 0.96 (0.86, 1.08) | 1.37 (1.20, 1.57) | 1.24 (0.92, 1.68) | 0.76 (0.59, 0.98) |
| Hispanic/Latino | 1.58 (1.45, 1.72) | 1.01 (0.90. 1.14) | 1.03 (0.91, 1.16) | 1.71 (1.48, 1.97) | 0.98 (0.72, 1.34) | 0.93 (0.76, 1.15) |
| Multiracial | 1.56 (1.43, 1.69) | 1.28 (1.15, 1.43) | 1.42 (1.27, 1.58) | 1.77 (1.54, 2.03) | 1.16 (0.89, 1.53) | 1.17 (0.98, 1.39) |
Note. All models controlled for age.
Only includes those who reported a suicide attempt.
There were significant differences among race/ethnicity groups, though the direction of results was mixed. For example, compared with White youths, Black respondents had significantly lower prevalence of 1-year suicidal ideation and self-harm (OR = 0.9 and 0.8, respectively), yet they had a significantly higher prevalence of suicide attempts (OR = 1.4). Multiracial youths were at significantly higher odds than Whites on 4 of the 6 study outcomes, whereas Asian youths did not differ significantly from Whites on any of the outcomes.
Female participants differed significantly from male particpants on all outcomes. Girls had higher prevalence of all outcomes, except that they were half as likely as boys to report being treated by a doctor or nurse as a result of a suicide attempt (OR = 0.49).
Table 4 shows results for sexual minority youths by race/ethnicity and sex. Sex differences generally mirrored those found in the total sample. That is, sexual minority females had higher prevalence on all outcomes, with the exception of lower prevalence of past year treatment by a doctor or nurse after a suicide attempt. However, there was no significant difference by sex for suicide attempts in the past year.
TABLE 4—
Mental Health and Suicidality Outcomes by Race/Ethnicity and Sex, Sexual Minorities Only: United States, 2005 and 2007 Youth Risk Behavior Surveys
| Variable | Feel Sad, OR (95% CI) | Suicide Ideation, OR (95% CI) | Suicide Plan, OR (95% CI) | Suicide Attempts, OR (95% CI) | Suicide Attempt Treated by Doctor or Nurse,a OR (95% CI) | Self-Harm, OR (95% CI) |
| Female sex | 1.84 (1.57, 2.15) | 1.60 (1.35, 1.91) | 1.46 (1.22, 1.73) | 1.16 (0.95, 1.42) | 0.47 (0.32, 0.69) | 1.78 (1.21, 2.54) |
| Race/ethnicity (Ref: White) | ||||||
| Alaskan/Pacific Islander | 1.49 (1.02, 2.18) | 0.81 (0.54, 1.22) | 1.48 (1.01, 2.18) | 1.66 (1.06, 2.60) | 0.75 (0.33, 1.68) | 1.93 (0.48, 7.87) |
| Asian | 0.86 (0.67, 1.10) | 0.64 (0.48, 0.85) | 0.71 (0.53, 0.95) | 0.90 (0.65, 1.26) | 1.10 (0.59, 2.07) | 0.39 (0.23, 0.67) |
| Black | 0.70 (0.57, 0.85) | 0.69 (0.55, 0.86) | 0.74 (0.60, 0.93) | 0.89 (0.69, 1.16) | 0.74 (0.44, 1.24) | 0.45 (0.23, 0.87) |
| Latino | 1.27 (1.01, 1.60) | 1.01(0.79, 1.29) | 0.92 (0.71, 1.18) | 1.50 (1.14, 1.97) | 0.52 (0.30, 0.89) | 0.70 (0.35, 1.42) |
| Multiracial | 1.35 (1.04, 1.73) | 0.98 (0.75, 1.28) | 1.24 (0.95, 1.61) | 1.17 (0.86, 1.60) | 0.88 (0.49, 1.59) | 1.04 (0.59, 1.84) |
| Females | ||||||
| Alaskan/Pacific Islander | 1.92 (0.82, 4.47) | 0.80 (0.36, 1.79) | 0.99 (0.45, 2.20) | 1.32 (0.51, 3.43) | 2.03 (0.71, 5.79) | 2.01 (0.64, 6.30) |
| Asian | 0.85 (0.57, 1.27) | 0.54 (0.36, 0.82) | 0.70 (0.46, 1.10) | 0.83 (0.51, 1.42) | 0.70, (0.24, 2.10) | 0.41 (0.24, 0.73) |
| Black | 0.62 (0.46, 0.84) | 0.66 (0.48, 0.92) | 0.72 (0.52, 1.01) | 0.74 (0.49, 1.14) | 0.88 (0.40, 1.94) | 0.17 (0.06, 0.47) |
| Latino | 1.04 (0.72, 1.39) | 1.07 (0.77, 1.47) | 1.01 (0.72, 1.41) | 1.84 (1.26, 2.69) | 0.72 (0.34, 1.53) | 0.57 (0.31, 1.10) |
| Multiracial | 1.26 (0.92, 1.72) | 0.85 (0.62, 1.18) | 1.11 (0.81, 1.54) | 1.09 (0.74, 1.62) | 0.87 (0.46, 1.63) | 0.66 (0.42, 1.10) |
| Males | ||||||
| Alaskan/Pacific Islander | 1.34 (0.60, 2.99) | 0.77 (0.29, 1.99) | 2.23 (0.99, 4.98) | 2.29 (0.93, 5.64) | 0.46 (0.05, 4.09) | 2.01 (0.77, 5.21) |
| Asian | 1.04 (0.68, 1.59) | 0.76 (0.45, 1.27) | 0.76 (0.47, 1.22) | 0.99 (0.54, 1.67) | 2.97 (0.67, 13.2) | 0.33 (0.14, 0.77) |
| Black | 0.82 (0.54, 1.24) | 0.76 (0.47, 1.23) | 0.82 (0.51, 1.33) | 1.28 (0.75, 2.17) | 1.13 (0.37, 3.44) | 0.91 (0.42, 2.02) |
| Latino | 1.98 (1.32, 2.97) | 0.97 (0.60, 1.55) | 0.83 (0.52, 1.35) | 1.13 (0.65, 1.96) | 0.51 (0.11, 2.31) | 1.17 (0.44, 3.09) |
| Multiracial | 1.59 (1.10, 2.31) | 1.23 (0.80, 1.90) | 1.43 (0.97, 2.13) | 1.07 (0.66, 1.74) | 1.75 (0.66, 4.67) | 1.52 (0.74, 3.13) |
Note. All models controlled for age.
Only includes those who reported a suicide attempt.
Sexual minorities differed across race/ethnicity on a number of outcomes. Compared with White sexual minority youths, Asian and Black youths were at significantly lower odds of suicidal ideation, suicide planning, and self-harm. AN/PI, Latino and Multiracial youths, however, were at significantly higher odds of feeling sad when compared with White youths, and AN/PI and Latino youths were also at significantly higher odds of a suicide attempt in the past year. This pattern differs from findings among the total sample: Asian youths did not significantly differ from Whites, and Multiracial youths fared significantly worse than White youths on 4 of the 6 outcomes, as opposed to just the single outcome (feeling sad) among sexual minority youths.
When analyses were stratified by sex, few significant differences emerged. Compared with White sexual minority females, Asian and Black sexual minority females had lower prevalence of 1-year suicidal ideation and self-harm. Sexual minority Latinas were the only group with significantly higher prevalence of 1-year suicide attempt compared with White sexual minority females. Among sexual minority males, Latinos had twice the odds of 1-year prevalence of feeling sad (OR = 1.98) and Asian sexual minority males had one third the odds of self-harm (OR = 0.33) compared with Whites.
DISCUSSION
We assessed mental health and suicidality outcomes across groups based on sexual minority status, race/ethnicity, and sex. Results add to existing evidence to support findings that sexual minority youths are at higher risk for suicidality compared with heterosexual youths. Results also reveal important nuances that are often missing from extant literature. For example, Asian and Black sexual minority youths tended to fare better than White sexual minority youths on a number of outcomes. However, when we stratified results by sex, this protective effect held only for Asian and Black females, with the exception of Asian sexual minority males being less likely to report self-harming behavior than White sexual minority males. Although female sexual minority participants were at higher odds for most outcomes than their male counterparts, there was no significant sex difference among sexual minority youths in reports of suicide attempt in the prior year.
These patterns accentuate the complexity of multiple, intersecting identities and their interaction with health, health behaviors, and health outcomes. Intersectionality has been suggested as an important conceptual framework through which to understand sexual minority health.15,16 These results affirm that the consequences of possessing multiple marginalized identities are not simply additive (i.e., that “more” marginalization necessarily leads to more negative health outcomes). Rather, it appears that for some health outcomes and behaviors, in particular self-harm, the intersection of minority identities conferred a protective effect; this was the case for sexual minority Asian and Black females compared with their White counterparts.
Differences between groups in the current study may stem, in part, from varying cultural conceptions regarding the acceptability of self-injurious behaviors, including suicide. In their comprehensive synthesis of the suicide literature across cultural minority groups, including LGBT populations, Chu et al.21 specifically address this issue. They note, for example, that cultural sanctions around suicide, or messages of disapproval or acceptability of suicide, vary across racial and ethnic groups, pointing to a number of studies in which Blacks are more likely than their White counterparts to view suicide negatively and as immoral.21 Our findings support the inter- and intrapersonal aspects of identity and the interdependence of these aspects.20 For some sexual and racial/ethnic minority youths, their racial/ethnic identity may be a prominent aspect of their self-schema. The cultural values and norms associated with their race/ethnicity may serve as a protective factor or buffer against sexual minority stigma, which may include strong sanctions against any self-harming behaviors.
Results of the study point to the limitations of using categories such as “youths of color” when conducting research because salient differences and distinctions among racial/ethnic minorities can be blurred and nullified. This, in turn, has consequences for how we design mental health policies and interventions, for both sexual minority youths and for youths in general. An intersectional lens illuminates the value of tailoring policies and interventions so that they address the unique and particular needs of specific groups.
An additional notable finding was that the AN/PI group fared worse than other racial/ethnic groups on many outcomes, irrespective of sexual minority status or sex. This is consistent with previous work among Native American and Alaska Native groups specifically9 as well as a study of Asian/Pacific Islander adolescents in Guam.23 Although significant differences were not found among sexual minorities when stratified by sex, on the whole, AN/PI youths reported the highest odds of past-year suicide attempt. Comparability of the findings to other studies is highly provisional given the manner in which the AN/PI category was constructed in the current study. Nevertheless, results are a cause for concern, and highlight the need for more focused studies on Native American sexual minority youths and for tailored intervention and prevention programs for this population.
The only other racial/ethnic differences among sexual minorities were among Latino youths: sexual minority Latinas were significantly more likely than their White counterparts to report a past-year suicide attempt, and Latinos were significantly more likely than White male sexual minority youths to report feeling sad. Overall, these findings are in concert with some previous studies,3,12 which found that Latino sexual minority youths were more likely to report suicide attempts than their White counterparts, but are inconsistent with findings from other studies that found Black sexual minority youths13 and adults14 were also at heightened risk for ideation and attempt. Such differences across studies are likely a function of variations in measurement of both sexual orientation and “suicidality,” as well as how samples were obtained. For example, in the study by Mustanski et al.,13 they assessed suicidal ideation and attempt in a structured psychiatric interview, whereas in the current study, those aspects were assessed via a single item in a paper-and-pencil questionnaire.
Because all data reported in the current study are based on self-report, there is the possibility that findings underrepresent the prevalence of sexual minority youths and of suicidal behaviors. Although we were unable to estimate the extent of biased reporting, the questions used to assess health-risk behaviors in the YRBS have been shown to have good test-retest reliability.24 Several other limitations should also be considered. YRBS data are not representative of sexual minority students in other jurisdictions. This is particularly important because regions with more liberal policies may be more likely to include sexual orientation questions in their YRBS studies than regions with less liberal attitudes. Regional attitudes and policies toward sexual minorities are also an important factor in determining the health of sexual minorities, including suicidality.25 Therefore, disparities between sexual minority and sexual majority youths in this study are likely to underrepresent the true disparities across the United States.26 Furthermore, YRBS data are collected from youths who attend public school. It is plausible that sexual minority youths or those at highest risk for suicidal behaviors are more likely to drop out of school or attend private schools. The etiology of suicidality is complex and multifactorial. The current study did not assess many factors that may contribute to suicidality, such as violence and victimization; nor did it assess protective factors, such as social support, that may buffer suicidality risks.27,28
A final limitation is the use of an omnibus sexual orientation variable. This approach obscures variation between different dimensions of sexual orientation (e.g., same-sex behavior vs sexual minority identity), as well as variation among sexual minority groups (e.g., bisexual vs lesbian/gay identity), vis-à-vis both race/ethnicity and mental health and suicidality. However, our primary goal was to present a broad analysis of racial/ethnic differences based on sexual minority status. Minority stress theory posits that sexual minority status, regardless of how it is defined (attraction, behavior, identity), is stigmatized, and therefore, stressors resulting from this stigma may be associated with any of these dimensions. Future studies should address the complexities of intersections of race/ethnicity with multiple dimensions of sexuality in adolescence.
A general omnibus measure allowed for inclusion of the maximum number of participants and jurisdictions, which assured sufficient statistical power and variability. We note, however, the irony of using such a measure of sexual orientation, while simultaneously making the case against a similar measure of race/ethnicity (i.e., “youths of color”). This trade-off highlights some of the inherent conundrums and complications of attempting to do a truly “intersectional” analysis with secondary, quantitative data,15 and begs the question of whether such work can effectively be accomplished with quantitative data—or at least quantitative data alone.
In conclusion, the current study supports previous findings showing that sexual minority youths are at substantially higher risk for suicidality. Beyond this, our findings indicate that among sexual minority youths, sex and race/ethnicity interact to influence health and health behaviors in complex and sometimes unpredicted ways. For example, although some racial/ethnic sexual minority youths appear to have higher risk for suicidality than White sexual minority youths, others (e.g., Black and Asian youths), do not show disparate risk and, in fact, are in some cases at lower risk than Whites. Additional research is necessary to better understand patterns of risk, as well as how risk and resilience function based on various combinations of intersecting identities.
Acknowledgments
This project was supported by a grant from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (award number R21HD051178) and by the IMPACT LGBT Health and Development Program at Northwestern University. Award Assistance from the Centers for Disease Control and Prevention (CDC) Division of Adolescent and School Health and the work of the state and local health and education departments who conduct the Youth Risk Behavior Survey made the project possible. Preparation of this article was also supported, in part, by the National Institute on Alcohol Abuse and Alcoholism, 2R01 AA013328 (T. H.).
The authors would like to thank Aimee Van Wagenen for coordination and management of the pooled data set.
Note. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, the CDC, or any of agencies involved in collecting the data.
Human Participant Protection
Human participant protection was not required because data were acquired from secondary, de-identified sources.
References
- 1.Remafedi G, Farrow J, Deisher RW. Risk factors for attempted suicide in gay and bisexual youth. Pediatrics. 1991;87(6):869–875. [PubMed] [Google Scholar]
- 2.O’Carroll PW, Berman AL, Maris RW, Moscicki EK, Tanney BL, Silverman MM. Beyond the Tower of Babel: a nomenclature for suicidology. Suicide Life Threat Behav. 1996;26(3):237–252. [PubMed] [Google Scholar]
- 3.Garofalo R, Wolf RC, Wissow LS, Woods ER, Goodman E. Sexual orientation and risk of suicide attempts among a representative sample of youth. Arch Pediatr Adolesc Med. 1999;153:487–493. doi: 10.1001/archpedi.153.5.487. [DOI] [PubMed] [Google Scholar]
- 4.Russell ST, Toomey RB. Men's sexual orientation and suicide: evidence for adolescent- specific risk. Soc Sci Med. 2012;74(4):523–529. doi: 10.1016/j.socscimed.2010.07.038. [DOI] [PubMed] [Google Scholar]
- 5.Russell ST, Joyner K. Sexual minority youth and suicide risk. Am Behav Sci. 2003;9:498–508. [Google Scholar]
- 6.Silenzio VM, Pena PR, Duberstein JC, Cerel J, Knox KL. Sexual orientation and risk factors for suicidal ideation and suicide attempts among adolescents and young adults. Am J Public Health. 2007;97:2017–2019. doi: 10.2105/AJPH.2006.095943. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Haas AP, Eliason M, Mays V et al. Suicide and suicide risk in lesbian, gay, bisexual, and transgender populations: review and recommendations. J Homosex. 2011;58:10–51. doi: 10.1080/00918369.2011.534038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Marshal MP, Friedman MS, Stall R et al. Sexual orientation and adolescent substance use: a meta-analysis and methodological review. Addiction. 2008;103:546–556. doi: 10.1111/j.1360-0443.2008.02149.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Borowsky IW, Resnick MD, Ireland M, Blum RW. Suicide attempts among American Indian and Alaska Native Youth. Arch Pediatr Adolesc Med. 1999;153:573–580. doi: 10.1001/archpedi.153.6.573. [DOI] [PubMed] [Google Scholar]
- 10.Centers for Disease Control and Prevention. Youth Risk Behavior Surveillance — United States, 2011. MMWR Surveill Summ. 2012;61(4):1–162. [PubMed] [Google Scholar]
- 11.Spirito A, Esposito-Smythers C. Attempted and completed suicide in adolescence. Annu Rev Clin Psychol. 2006;2:237–266. doi: 10.1146/annurev.clinpsy.2.022305.095323. [DOI] [PubMed] [Google Scholar]
- 12.Consalacion TB, Russell ST, Sue S. Sex, race/ethnicity, and romantic attractions: multiple minority status adolescents and mental health. Cultur Divers Ethnic Minor Psychol. 2004;10:200–214. doi: 10.1037/1099-9809.10.3.200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mustanski BS, Garofalo R, Emerson EM. Mental health disorders, psychological distress, and suicidality in a diverse sample of lesbian, gay, bisexual, and transgender youths. Am J Public Health. 2010;100:2426–2432. doi: 10.2105/AJPH.2009.178319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.O’Donnell S, Meyer I, Schwartz S. Increased risk of suicide attempts among Black and Latino lesbians, gay men, and bisexuals. Am J Public Health. 2011;101(6):1055–1059. doi: 10.2105/AJPH.2010.300032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Bowleg L. The problem with the phrase women and minorities: intersectionality—an important theoretical framework for public health. Am J Public Health. 2012;102:1267–1273. doi: 10.2105/AJPH.2012.300750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Committee on Lesbian Gay Bisexual and Transgender Health Issues and Research Gaps and Opportunities, Institute of Medicine. The Health of Lesbian, Gay, Bisexual, and Transgender People: Building a Foundation for Better Understanding. Washington, DC: National Academies Press; 2011. [PubMed] [Google Scholar]
- 17.McCall L. The complexity of intersectionality. Signs (Chic) 2005;30:1771–1800. [Google Scholar]
- 18.Crenshaw K.Mapping the margins: Intersectionality, identity politics, and violence against women of color Stanford Law Rev 19911241–1299 [Google Scholar]
- 19.Davis K. Intersectionality as buzzword. Fem Theory. 2008;9(1):67–85. [Google Scholar]
- 20.Meyer IH. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychol Bull. 2003;129(5):674–697. doi: 10.1037/0033-2909.129.5.674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chu JP, Goldblum P, Floyd R, Bongar B. The cultural theory and model of suicide. App Prev Psychol. 2010;14(1):25–40. [Google Scholar]
- 22.Mustanski B, Van Wagenen A, Birkett M, Eyster S, Corliss H. Identifying sexual orientation health disparities in adolescents: analysis of pooled data from the Youth Risk Behavior Survey, 2005 and 2007. Am J Public Health. 2014;104(2):211–217. doi: 10.2105/AJPH.2013.301748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Pinhey TK, Millman SR. Asian/Pacific Islander adolescent sexual orientation and suicide risk in Guam. Am J Public Health. 2004;94:1204–1206. doi: 10.2105/ajph.94.7.1204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Brener ND, Kann L, McManus T, Kinchen SA, Sundberg EC, Ross JG.Reliability of the 1999 Youth Risk Behavior Survey questionnaire J Adolesc Health 2002336–342 [DOI] [PubMed] [Google Scholar]
- 25.Hatzenbuehler ML, Birkett M, Van Wagenen A, Meyer IH. Protective school climates and reduced risk for suicide ideation in sexual minority youths. Am J Public Health. 2014;104(2):279–286. doi: 10.2105/AJPH.2013.301508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hatzenbuehler ML. The social environment and suicide attempts in lesbian, gay, and bisexual youth. Pediatrics. 2011:127. doi: 10.1542/peds.2010-3020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Liu RT, Mustanski B. Suicidal ideation and self-harm in lesbian, gay, bisexual, and transgender youth. Am J Prev Med. 2012;42:221–228. doi: 10.1016/j.amepre.2011.10.023. [DOI] [PubMed] [Google Scholar]
- 28.Mustanski B, Liu RT. A longitudinal study of predictors of suicide attempts among lesbian, gay, bisexual, and transgender youth. Arch Sex Behav. 2013;42:437–448. doi: 10.1007/s10508-012-0013-9. [DOI] [PubMed] [Google Scholar]