Abstract
Introduction:
Polyvictimization is a significant public health issue. Sexual and gender minority youth are important to include in polyvictimization research because they report higher rates of victimization than non-sexual and gender minority youth. This study examines whether polyvictimization attenuates the associations between individual types of victimization and depressed mood and substance use across gender and sexual identities.
Methods:
Data were collected cross-sectionally from 3,838 youth aged 14-15 years. Youth were recruited through social media between October 2018 and August 2019 across the U.S. Sexual and gender minority youth were over-sampled. Depressed mood and substance use were dependent variables.
Results:
Transgender boys were the most likely to be polyvictims (25%). Transgender girls (14.2%) and cisgender sexual minority girls (13.4%) also reported high rates. Cisgender heterosexual boys were the least likely to be classified as polyvictims (4.7%). When adjusting for polyvictimization, existing relationships between individual types of victimization (e.g., theft) and depressed mood became nonsignificant in most cases. Of exception, witnessing violence and peer victimization remained significant predictors of the odds of depressed mood. Most associations between individual types of victimization and substance use became nonsignificant after considering polyvictimization, with the exception of cisgender heterosexual boys and girls, for whom many remained significant but attenuated (e.g., emotional interpersonal violence).
Conclusions:
Sexual and gender minority youth experience a disproportionate number of victimizations across multiple domains. A comprehensive assessment of victimization exposure may be important when considering prevention and intervention approaches for depressed mood and substance use.
Keywords: polyvictimization, sexual and gender identity, depressed mood, substance use, cumulative victimization
Introduction
Polyvictimization, or exposure to multiple different forms of violence across different domains (e.g., peer, child maltreatment), is a significant public health issue affecting approximately 1 in 10 youth nationally.1 Adolescence is a critical period of development and impacts that occur at this age have the potential to persist into and through adulthood.2 Previous research has found that total exposure to violence more strongly predicts mental health and somatic outcomes than any single measure of victimization.3,4 At the same time, certain types of victimization, including lifetime physical assault, maltreatment, peer and sibling victimization, and exposure to family violence, have been noted to have independent effects on trauma symptomatology, above that of polyvictimization.1
Sexual and gender minority (SGM) youth - those who identify as gay, lesbian, bisexual or another non-heterosexual sexual identity; and those who identify as transgender, non-binary, or another non-cisgender gender identity - are a particularly important group to include in polyvictimization research because they report higher rates of lifetime victimization than non-SGM youth.4–9 Although there is a growing body of research exploring specific victimization experiences among SGM youth, there is a paucity of research on polyvictimization. Extant research finds that SGM youth experience higher rates of lifetime polyvictimization than their non-SGM peers,4,10,11 and that SGM youth report higher levels of depression and substance use than to their non-SGM peers.6,7,12–18 Few studies have explored the relationship between polyvictimization and mental health across different sexual and gender identity. This is a significant gap given the documented research on the impact of polyvictimization on mental health, 3,19 including substance use19,20 in general population samples of youth.
This study will address these noted gaps by: 1) examining different types of victimization and polyvictimization status across gender and sexual identities, and 2) testing the hypothesis that polyvictimization attenuates associations between individual types of victimization and depressed mood and substance use across gender and sexual identities.
Methods
Study Population
Growing up with Media is a national, longitudinal survey of youth designed to study the etiology of sexual violence.21,22 In addition to the original cohort recruited in 2006, a new cohort of 4,404 youth, aged 14-16 years, was recruited between June 2018, and March, 2020. The response rate, using AAPOR Response Rate 4, was 7.5%. 23 Greater detail about the methodology can be found elsewhere.24
The analytic sample for this study, 3,838 youth aged 14-15 years, varies from those of some previous articles because it does not include those aged 16 years. This is because there were too few to apply weights to and thus were excluded from the current analyses.
Participants were recruited through advertisements on Facebook and Instagram. Racial and ethnic diversity was attained by targeting interests such as Latin culture. To support subsample analyses, an oversample of SGM youth was achieved by targeting ads to youth who reported same-sex attraction. Sample characteristics are provided in Table 1.
Table 1.
Polyvictimization Among Different Subgroups of Youth
| Characteristic | All youth (N=3,838) n (%) |
Not poly-victim (n=3,369) n (%) |
Poly-victim (18+) (n=469) n (%) |
P value |
|---|---|---|---|---|
| Gender and sexual identity | ||||
| Cisgender heterosexual boys | 621 (38.8) | 593 (95.3) | 28 (4.7) | <0.001 |
| Cisgender sexual minority boys | 755 (8.7) | 682 (90.6) | 73 (9.4) | |
| Cisgender heterosexual girls | 782 (34.2) | 713 (91.7) | 69 (8.3) | |
| Cisgender sexual minority girls | 792 (10.0) | 684 (86.6) | 108 (13.4) | |
| Transgender girls a | 101 (1.5) | 89 (85.8) | 12 (14.2) | |
| Transgender boys a | 705 (6.9) | 531 (74.5) | 174 (25.5) | |
| Race/ethnicity | ||||
| White | 2,578 (66.7) | 2,241 (90.3) | 337 (9.7) | 0.03 |
| African American | 206 (6.2) | 196 (94.4) | 10 (5.1) | |
| Mixed racial background | 526 (12.3) | 442 (89.9) | 84 (10.1) | |
| Other race | 419 (11.7) | 385 (95.1) | 34 (4.9) | |
| Decline to answer | 109 (3.2) | 105 (95.3) | 4 (4.7) | |
| Hispanic ethnicity | ||||
| No | 3,107 (79.5) | 2,630 (90.8) | 387 (9.2) | |
| Yes | 821 (20.5) | 739 (93.0) | 82 (7.0) | 0.11 |
| Family income | ||||
| Higher than average family | 721 (21.9) | 662 (94.3) | 59 (5.7) | <0.001 |
| Similar to average family | 2,094 (55.4) | 1,860 (91.5) | 234 (8.5) | |
| Lower than average family | 894 (19.1) | 727 (85.9) | 167 (14.1) | |
| Decline to answer | 129 (3.5) | 120 (96.9) | 9 (3.1) | |
| Youth age | ||||
| 14 years | 1,446 (47.3) | 1,271 (91.3) | 175 (8.7) | 0.88 |
| 15 years | 2,392 (52.7) | 2,098 (91.1) | 294 (8.8) | |
| Psychosocial characteristics | ||||
| Depressed mood | <0.001 | |||
| No | 1,112 (35.2) | 1,078 (97.4 | 34 (2.6) | |
| Yes | 2,726 (64.8) | 2,291 (87.9) | 435 (12.1) | |
| Any substance use (past year) | <0.001 | |||
| No | 1,991 (53.6) | 1,873 (95.6) | 118 (4.4) | |
| Yes | 1,847 (46.4) | 1,496 (86.2) | 351 (13.8) | |
| Mean (SE) number of types | 1.07 (.03) | 0.96 (.03) | 2.22 (.14) | <0.001 |
Note. Boldface indicates statistical significance (p<0.05).
Rows add to 100%. Data on sexual and gender identity are missing for 82 participants.
Includes youth who identify their gender as transgender, gender queer, non-binary, pangender, questioning, unsure, and some other identity.
Survey aims were not mentioned in social media ads to reduce self-selection bias on the basis of interest in a particular topic. Multiple steps were taken to ensure the authenticity of the sample, including comparing the age entered in the screener with their date of birth, looking up IP (Internet Protocol) addresses, and checking multiple variables for patterns of responses. Duplicates were identified by comparing phone numbers and e-mail addresses across participants. A waiver of caregiver permission was granted to protect youth from having to disclose to their parents their sexual and gender identity. Appropriate mechanisms were in place to protect youth, including follow-up by a clinician to provide localized referrals to mental health supports if warranted.
Participants were given a $15 Amazon gift card for completing the survey. Ineligible youth were directed to a web page that included links to general resources for youth. The protocol was reviewed and approved by Pearl IRB and Advarra IRB.
Measures
Some victimization measures referred to the past year, and others referred to lifetime. Past-year variables may have resulted in an undercounting of lifetime events. Nonetheless, the diverse range of types of victimization is unique to this study. We chose to include all experiences to maximize the number of different types of victimization experiences included in the analyses. Response options for all measures were yes/no.
Eight different types of past-year sexual harassment were queried.25,26 To approximate a ‘hostile environment,’ those who endorsed at least 1 type were asked whether the place where it happened now felt scary, unfriendly or uncomfortable. Those who agreed were coded as experiencing any sexual harassment.
Lifetime sexual victimization included 4 items: sexual assault, coercive sex, attempted rape and rape.27
Past-year theft was indicated if “Someone stole something from me.” Past-year physical assault was queried with 2 items: “Another person or group attacked me” and “Someone pulled a knife or gun on me.”
Past-year generalized peer victimization was measured using 6 items.29 As an example, “How often have others around your age bullied you by: Hitting, kicking, pushing, or shoving you.”
Lifetime emotional interpersonal violence (IPV) was measured with 4 items.30,31 For example: “Would not let you spend time with other people or talk to someone you might be attracted to.” Both committed (i.e., “boyfriends or girlfriends you have had”) and noncommitted (i.e., “people you hooked up with or dated, but not as boyfriends/girlfriends”) relationships were queried.
Seven items measured lifetime physical IPV (e.g., scratched or slapped you).30,31 Experiences were qualified by the question, “Which, if any, of the following things have they ever done to you on purpose? (Only count it if they did it first. Do not count it if they did it in self-defense.)” Experiences were asked for both committed and non-committed relationships separately and combined for the current analyses.
Nine questions queried lifetime witnessing of violence: seeing or hearing perpetrated violence (2 items) (e.g., “Have you in real life seen someone get attacked or hit on purpose?”), witnessing IPV (3 items)32 (e.g., “Have you in real life seen or heard one of your parents get hit, slapped, punched, or beat up by your other parent, or by their boyfriend, girlfriend, or partner?”), and witnessing sexual violence (4 items) (e.g., “Have you ever seen or heard about someone you know in-person who said something sexual to someone when that person did not want to hear it?”)
Polyvictimization was operationalized in multiple ways. First, polyvictimization was included as a summary count measure of the total number of victimization items assessed (of a possible 41; range=0-41, mean =7.7, SD =6.4). Second, indicators of any experience of 8 domains of victimization were created: sexual harassment, theft, assault, peer victimization, sexual victimization, emotional IPV, physical IPV, and witnessing violence. Finally, a dichotomous indicator of polyvictimization was operationalized as those youth with total victimization scores 1 SD above the mean, roughly equivalent to the top 10%, as suggested by Finkelhor and colleagues.33 Depressed mood was measured by asking, “In the last month, was there ever a time of two weeks or longer when you were feeling unhappy, bummed out, depressed or down most of the day or nearly every day?”
Use of 9 different substances was queried:34 alcohol (31.1%), tobacco (11.6%), E-cigarettes (31.1%), marijuana (20.2%), inhalants (3.6%), prescription drug use without a doctor’s permission (6.7%), steroids without a doctor’s permission (0.6%), meth (0.8%), and club drugs (1.2%). A positive response to any of these items was coded as past-year substance use.
Youth reported their sex assigned at birth: male, female, or do not want to answer. Youth who endorsed a gender identity of female-to-male /transgender male/trans man; male-to-female /transgender female/trans woman; gender queer/non-binary/pangender or other (specify) were coded as gender minority. Youth who indicated their sexual identity was gay, lesbian, bisexual, questioning, queer, pansexual, asexual, other, or unsure were coded as sexual minority.
Youth were then assigned to 1 of 6 categories: 1) cisgender, heterosexual boys; 2) cisgender, sexual minority boys; 3) cisgender, heterosexual girls; 4) cisgender sexual minority girls, 5) transgender girls, and 6) transgender boys. Cisgender refers to people who identify their gender as the same as the sex they were assigned at birth. We use the term transgender to holistically refer to youth who identify their gender as transgender, gender queer, non-binary, pangender, questioning, unsure, and some other identity. Transgender youth were not further stratified by sexual identity because of sample size limitations.
Household income was queried with the question, “How would you describe your family’s income?” Youth had 3 options: lower than average, about average, and higher than average. Youth reported their race; multiple responses were possible (coded as White versus other). Ethnicity was measured as Hispanic versus other. Self-reported honesty was indicated by asking how honest respondents were when answering survey questions, on a scale of 1 (not at all honest) to 5 (extremely honest).
Statistical Analysis
Rates of “decline to answer” did not exceed 2.5% across variables. These responses were conservatively coded as 0.35
First, youth characteristics were examined by dichotomous polyvictimization status. Differences were measured using chi-square tests. Next, the prevalence rates of victimization types were examined by sexual and gender identity, again using chi-square tests. Pairwise comparisons between identities were conducted for the 8 domains of victimization. The relative odds of depressed mood and substance use were then quantified by gender and sexual identity, as well as victimization experiences.
Second, youth were stratified by sexual and gender identity, and the relative odds of depressed mood and substance use were each estimated given the total number of victimization types as well as each type of experience reported. Models were adjusted for demographic characteristics and self-reported honesty in answering survey questions. Then, to identify the influence that polyvictimization had in each model, both the victimization type and a count of victimization types – with the specific type of victimization in question excluded in the count – were estimated together. Transgender girls were not included in this second step because of small sample size.
Weighting was accomplished using iterative proportional fitting to balance the distributions of all variables.36 The use of these weights in statistical analyses ensures that the demographic characteristics of the sample closely approximate those of the national population of youth aged 14-15 years.
Results
Almost 1 in 10 youth (8.8%) were polyvictims, defined as experiencing 18 or more types of victimization. Polyvictim status varied significantly by sexual and gender identity: transgender boys and cisgender sexual minority youth were over-represented, whereas cisgender heterosexual boys were under-represented (Table 1). Polyvictims were also more likely to live in households with self-appraised below-average household income. Youth who reported depressed mood (12.1%) and past-year substance use (13.8%) were also more likely to be polyvictims than youth who did not indicate these problems (2.6% and 4.4%, respectively).
The average number of victimizations experienced varied by sexual and gender identity, ranging from 6.44 (SE=0.22) for cisgender heterosexual boys to 11.87 (SE=0.39) for transgender boys (Table 2). All victimization rates significantly varied by identity across all 8 victimization domains. Transgender boys reported the highest rates of victimization for most types assessed. Detailed information about differences by different forms of victimization within each victimization domains by SGM identity is available in Appendix Table 1.
Table 2.
Self-Reported Victimization by Sexual and Gender Identity (n = 3,756)
| Type of victimization | Cisgender hetero boys (n=621) n (%) |
Cisgender sexual minority boys (n=755) n (%) |
Cisgender hetero girls (n=782) n (%) |
Cisgender sexual minority girls (n=792) n (%) |
Transgender girls a (n=101) n (%) |
Transgender boys a (n=705) n (%) |
P value |
|---|---|---|---|---|---|---|---|
| Cumulative victimization (mean, SE) | 6.44 (0.22) | 8.46 (0.44) b | 7.71 (0.24) b | 9.54 (0.35) b,c,d | 9.48 (1.07) b | 11.87 (0.39) b,c,d,e,f | <0.001 |
| % polyvictims (18+) | 28 (4.7) | 73 (9.4) b | 69 (8.3)b | 108 (13.4)b,d | 12 (14.2)b | 174 (25.5)b,c,d,e | <0.001 |
| Victimization domains | |||||||
| Any sexual harassment | 24 (3.8) | 106 (13.2) b | 146 (18.5) b | 241 (28.9) b, c, d | 29 (36.4) b, c, d | 256 (37.2) b, c, d, e | <0.001 |
| Any sexual victimization | 138 (22.8) | 276 (40.2) b | 327 (40.1) b | 396 (51.5) b, c, d | 46 (41.7) b | 431 (60.3) b, c, d, e | <0.001 |
| Any theft | 253 (40.8) | 266 (38.5) | 310 (40.1) | 325 (44.2) | 40 (42.7) | 318 (45.6) | 0.59 |
| Any physical assault | 110 (18.7) | 120 (16.5) | 62 (7.7) b, c | 77 (11.7) b,d | 17 (23.8) d,e | 127 (18.3) d,e | <0.001 |
| Any peer victimization | 419 (68.2) | 530 (79.6) b | 576 (74.3) b | 606 (79.0) b | 73 (76.5) | 574 (83.1) b, d | <0.001 |
| Any emotional IPV | 201 (33.5) | 290 (41.6) b | 345 (43.0) b | 373 (44.5) b | 37 (38.2) | 377 (53.1) b,c,d,e | <0.001 |
| Any physical IPV | 111 (18.6) | 145 (22.9) | 118 (13.8)b,c | 161 (20.6)d | 21 (22.6) | 197 (27.1) b,d,e | <0.001 |
| Any witnessing violence | 463 (74.8) | 541 (77.7) | 577 (73.3) | 630 (81.0)b,d | 70 (70.1) | 586 (82.7)b,d,f | 0.02 |
Note. Boldface indicates statistical significance (p<0.05).
Data on sexual and gender identity are missing for 82 participants so the table total does not add to that listed in table 1.
M = mean; SE = standard error; SGM = sexual and gender minority
Includes youth who identify their gender as transgender, gender queer, non-binary, pangender, questioning, unsure, and some other identity.
Significantly different from cisgender heterosexual males.
Significantly different from cisgender SGM males.
Significantly different from cisgender heterosexual girls.
Significantly different from cisgender SGM females.
Significantly different from transgender girls.
All SGM youth as well as cisgender heterosexual girls had significantly elevated odds of depressed mood compared with cisgender heterosexual boys (Table 3). As shown in Table 4, the odds of depressed mood increased incrementally with the number of victimization experiences. Higher odds of depressed mood were also noted for youth who experienced each type of victimization than for those who had not had that specific experience. To understand the relative contribution that each type of victimization had in predicting the odds of depressed mood over and above the overall number of victimization types experienced, a count variable – excluding the victimization domain in question – was added to the model. Once this count variable was included, the association between each type of victimization and depressed mood attenuated and/or lost statistical significance in most cases. Of exception, peer victimization continued to be significantly related to depressed mood for all youth, as did witnessing violence for almost all youth.
Table 3.
Unadjusted Odds of Depressed Mood and Any Substance Use by Youth and Victimization Characteristics (n = 3,756)
| Depressed mood | Any substance use | |||
|---|---|---|---|---|
| Construct | OR (95% CI) | P - value | OR (95% CI) | P - value |
| Gender and sexual identity | ||||
| Cisgender heterosexual boys | 1.00 (ref) | --- | 1.00 (ref) | --- |
| Cisgender sexual minority boys | 1.52 (1.08, 2.13) | 0.01 | 1.07 (0.77, 1.49) | 0.67 |
| Cisgender heterosexual girls | 2.09 (1.67, 2.62) | <0.001 | 1.09 (0.87, 1.35) | 0.45 |
| Cisgender sexual minority girls | 3.41 (2.55, 4.56) | <0.001 | 1.41 (1.09, 1.82) | 0.009 |
| Transgender girls a | 2.35 (1.11, 4.98) | 0.03 | 0.88 (0.46, 1.69) | 0.70 |
| Transgender boys a | 5.07 (3.71, 6.93) | <0.001 | 1.53 (1.19, 1.96) | 0.001 |
| Victimization domains | ||||
| Sexual harassment | 3.05 (2.25, 4.12) | <0.001 | 1.75 (1.40, 2.20) | <0.001 |
| Sexual victimization | 2.31 (1.88, 2.84) | <0.001 | 2.69 (2.24, 3.23) | <0.001 |
| Theft | 1.98 (1.62, 2.41) | <0.001 | 1.71 (1.43, 2.05) | <0.001 |
| Physical assault | 1.87 (1.36, 2.56) | <0.001 | 2.98 (2.25, 3.94) | <0.001 |
| Peer victimization | 3.17 (2.59, 3.88) | <0.001 | 2.39 (1.94, 2.93) | <0.001 |
| Emotional IPV | 1.96 (1.61, 2.39) | <0.001 | 2.80 (2.33, 3.36) | <0.001 |
| Physical IPV | 1.82 (1.39, 2.38) | <0.001 | 2.69 (2.12, 3.42) | <0.001 |
| Witnessing | 2.62 (2.12, 3.22) | <0.001 | 2.78 (2.25, 3.45) | <0.001 |
| Count of types of victimization | 1.14 (1.11, 1.16) | <0.001 | 1.12 (1.10, 1.14) | <0.001 |
| Polyvictim | 5.14 (3.12, 8.46) | <0.001 | 3.47 (2.51, 4.81) | <0.001 |
Note. Boldface indicates statistical significance (p<0.05).
Ref = reference category; OR = Odds ratio; IPV = interpersonal violence
Includes youth who identify their gender as transgender, gender queer, non-binary, pangender, questioning, unsure, and some other identity.
Table 4.
Relation between Victimization Domains and Depressed Mood and Any Substance Use for Youth with Different Sexual and Gender Identities
| Depressed mood | Any substance use | |||
|---|---|---|---|---|
| Construct | Individual victimization domain only | Add cumulative victimization count | Individual victimization domain only | Add cumulative victimization count |
| OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| Cisgender sexual minority boys (n=755) | ||||
| Cumulative victimization only | 1.17 (1.10, 1.25) *** | --- | 1.09 (1.04, 1.14) *** | --- |
| Individual victimization domains with and without cumulative victimization | ||||
| Sexual harassment | 6.07 (2.09, 17.64) *** | 2.98 (0.97, 9.19) | 1.50 (0.64, 3.53) | 0.84 (0.30, 2.31) |
| Sexual victimization | 1.37 (0.73, 1.81) | 0.74 (0.36, 1.54) | 1.86 (1.02, 3.41) * | 1.34 (0.68, 2.63) |
| Theft | 1.39 (0.73, 2.65) | 0.82 (0.39, 1.69) | 2.87 (1.53, 5.38) *** | 2.37 (1.20, 4.66) ** |
| Physical assault | 2.25 (0.98, 5.16) * | 1.11 (0.43, 2.87) | 2.85 (1.25, 6.53) ** | 2.12 (0.88, 5.10) |
| Peer victimization | 3.25 (1.72, 6.15) *** | 2.21 (1.15, 4.27) * | 2.20 (1.20, 4.03) ** | 1.65 (0.87, 3.15) |
| Emotional IPV | 2.01 (1.06, 3.82) * | 1.23 (0.59, 2.57) | 2.81 (1.55, 5.99) *** | 2.25 (1.17, 4.34) ** |
| Physical IPV | 3.33 (1.36, 8.16) ** | 1.55 (0.58, 4.15) | 2.44 (1.13, 5.26) * | 1.59 (0.67, 3.76) |
| Witnessing | 2.36 (1.23, 4.52) ** | 1.36 (0.71, 2.63) | 2.72 (1.28, 5.78) ** | 2.04 (0.93, 4.51) |
| Cisgender sexual minority girls (n=792) | ||||
| Cumulative victimization only | 1.11 (1.05, 1.17) *** | --- | 1.11 (1.07, 1.15) *** | --- |
| Individual victimization domains with and without cumulative victimization | ||||
| Sexual harassment | 2.27 (1.21, 4.28) ** | 1.66 (0.82, 3.35) | 1.49 (0.96, 2.31) | 0.97 (0.60, 1.58) |
| Sexual victimization | 1.88 (1.12, 3.15) * | 0.99 (0.58, 1.71) | 2.77 (1.83, 4.19) *** | 1.62 (1.02, 2.59) * |
| Theft | 2.11 (1.24, 3.59) ** | 1.53 (0.90, 2.58) | 1.90 (1.26, 2.87) ** | 1.29 (0.82, 2.04) |
| Physical assault | 4.61 (1.98, 10.72) *** | 2.55 (0.96, 6.81) | 5.37 (2.49, 11.57) *** | 2.88 (1.29, 6.45) ** |
| Peer victimization | 2.35 (1.41, 3.91) *** | 1.76 (1.05, 2.95) * | 2.03 (1.23, 3.36) ** | 1.40 (0.82, 2.38) |
| Emotional IPV | 2.05 (1.21, 3.48) ** | 1.23 (0.72, 2.10) | 1.87 (1.25, 2.78) ** | 1.07 (0.70, 1.64) |
| Physical IPV | 1.92 (0.89, 4.12) | 0.99 (0.43, 2.29) | 2.67 (1.56, 4.56) *** | 1.41 (0.76, 2.62) |
| Witnessing | 3.93 (2.30, 6.72) *** | 3.02 (1.67, 5.45) *** | 3.93 (2.30, 6.72) *** | 3.02 (1.67, 5.45) *** |
| Transgender boys (n=705) a | ||||
| Cumulative victimization only | 1.11 (1.06, 1.17) *** | --- | 1.09 (1.06,1.12) *** | --- |
| Individual victimization domains with and without cumulative victimization | ||||
| Sexual harassment | 4.01 (1.83, 8.78) *** | 2.38 (1.03, 5.51) * | 1.88 (1.22, 2.89) ** | 1.07 (0.65, 1.75) |
| Sexual victimization | 2.34 (1.37, 4.01) ** | 1.36 (0.76, 2.43) | 2.41 (1.63, 3.58) *** | 1.50 (0.98, 2.30) |
| Theft | 1.86 (1.04, 3.30) * | 1.05 (0.56, 1.97) | 1.71 (1.15, 2.54) ** | 0.97 (0.62, 1.52) |
| Physical assault | 2.60 (1.05, 6.44) * | 1.11 (0.41, 3.01) | 2.74 (1.60, 4.69) *** | 1.45 (0.79, 2.67) |
| Peer victimization | 2.30 (1.30, 4.06) ** | 1.62 (0.86, 3.04) | 2.27 (1.42, 3.63) *** | 1.57 (0.98, 1.51) |
| Emotional IPV | 1.71 (1.00, 2.93) * | 0.90 (0.51, 1.57) | 2.15 (1.46, 3.15) *** | 1.32 (0.86, 2.02) |
| Physical IPV | 3.72 (1.85, 7.45) *** | 1.67 (0.71, 3.91) | 2.53 (1.57, 4.06) *** | 1.26 (0.74, 2.16) |
| Witnessing | 3.72 (2.00, 6.91) *** | 2.52 (1.37, 4.62) ** | 1.77 (1.04, 3.03) * | 1.12 (0.66, 1.87) |
| Cisgender heterosexual boys (n=621) | ||||
| Cumulative victimization only | 1.08 (1.04, 1.13) *** | --- | 1.16 (1.11, 1.21) *** | --- |
| Individual victimization domains with and without cumulative victimization | ||||
| Sexual harassment | 1.07 (0.43, 2.65) | 0.70 (0.26, 1.86) | 1.87 (0.77, 4.54) | 0.97 (0.36, 2.61) |
| Sexual victimization | 1.23 (0.81, 1.85) | 0.85 (0.54, 1.34) | 2.29 (1.53, 3.45) *** | 1.40 (0.91, 2.16) |
| Theft | 1.77 (1.25, 2.49) *** | 1.38 (0.96, 2.00) | 1.85 (1.31, 2.60) *** | 1.11 (0.75, 1.63) |
| Physical assault | 1.49 (0.95, 2.35) | 1.02 (0.62, 1.69) | 3.83 (2.41, 6.11) *** | 2.29 (1.39,3.77) *** |
| Peer victimization | 2.95 (2.04, 4.27) *** | 2.70 (1.84, 3.97) *** | 2.78 (1.90, 4.08) *** | 1.94 (1.31,2.88) *** |
| Emotional IPV | 1.74 (1.21, 2.49) ** | 1.37 (0.94, 2.01) | 2.84 (1.97, 4.08) *** | 1.93 (1.31, 2.85) *** |
| Physical IPV | 1.26 (0.81, 1.97) | 0.80 (0.47, 1.34) | 2.72 (1.74, 4.23) *** | 1.49 (0.91, 2.45) |
| Witnessing | 1.80 (1.22, 2.67) ** | 1.48 (0.99, 2.21) * | 3.30 (2.15, 5.05) *** | 2.52 (1.63, 3.91) *** |
| Cisgender heterosexual girls (n=782) | ||||
| Cumulative victimization only | 1.18 (1.13, 1.23) *** | --- | 1.12 (1.09, 1.16) *** | --- |
| Individual victimization domains with and without cumulative victimization | ||||
| Sexual harassment | 2.29 (1.40, 3.75) *** | 1.09 (0.64, 1.86) | 1.56 (1.07, 2.28) * | 0.78 (0.51,1.19) |
| Sexual victimization | 2.86 (1.99, 4.10) *** | 1.55 (1.05, 2.31) * | 3.62 (2.64, 4.96) *** | 2.42 (1.71,3.42) *** |
| Theft | 2.45 (1.71, 3.50) *** | 1.48 (1.00, 2.19) * | 1.37 (1.01, 1.85) * | 0.78 (0.55, 1.11) |
| Physical assault | 12.56 (2.88, 54.71) *** | 5.01 (1.18, 21.28) * | 2.53 (1.38, 4.63) ** | 1.07 (0.53, 2.13) |
| Peer victimization | 2.99 (2.09, 4.26) *** | 2.19 (1.52, 3.17) *** | 2.44 (1.71, 3.49) *** | 1.81 (1.25, 2.61) ** |
| Emotional IPV | 1.62 (1.16, 2.26) ** | 0.91 (0.63, 1.32) | 3.03(2.23, 4.13) *** | 2.19 (1.58,3.03) *** |
| Physical IPV | 2.88 (1.64, 5.09) *** | 1.06 (0.56, 1.99) | 2.99 (1.90, 4.71) *** | 1.45 (0.86, 2.44) |
| Witnessing | 3.04 (2.14, 4.31) *** | 2.09 (1.45, 3.02) *** | 2.65 (1.86, 3.76) *** | 1.85 (1.28,2.66) *** |
Note. Boldface indicates statistical significance (
p<0.001;
p<0.01;
p<0.05).
Data for transgender girls are not provided owing to a lack of power. Italicized cells are included to illustrate the association between total victimization count with depressed mood and any substance use while adjusting for youth demographic characteristics for different SGM youth.
IPV = interpersonal violence; SGM – sexual and gender minority.
Includes youth who identify their gender as transgender, gender queer, non-binary, pangender, questioning, unsure, and some other identity.
When considering gender identity, sexual harassment persisted as an explanatory factor for transgender boys. When considering sexual identity, sexual victimization, theft, and physical assault remained influential in explaining the odds of depressed mood for cisgender heterosexual girls.
As the number of victimization experiences increased, so too did the odds of substance use. When each type of victimization was examined singly, each was significantly related to elevated odds of substance use across sexual and gender identities (Table 4). Transgender boys and cisgender sexual minority girls had elevated odds of past-year substance use compared with cisgender heterosexual boys. Further adjusting for the total number of victimization types experienced eliminated statistically significant relationships for transgender boys. In other models, between 2 and 4 victimization types persisted. For cisgender sexual minority boys, experiencing theft and emotional IPV continued to be significantly related to substance use. For cisgender sexual minority girls, this was true for sexual victimization, physical assault, and witnessing violence. Cisgender heterosexual youth had the most independent victimization types related to substance use after adjusting for cumulative victimization.
Discussion
This study adds to a growing body of work that shows that SGM youth experience higher levels of polyvictimization than their non-SGM counterparts. Indeed, most SGM youth in this study experience at least 1 form of assault, sexual harassment, peer victimization, theft, sexual assault, and IPV; and also witnessed violence.4,10 That said, there were more similarities than differences across sexual and gender identity when assessing the relationships between polyvictimization, and depressed mood and substance use.
Consistent with other studies,3 relationships between depressed mood and most individual victimization domains were eclipsed by cumulative victimization. Some exceptions were noted: the relationship between peer victimization, which includes bullying, and depressed mood, persisted within the context of polyvictimization for all youth, regardless of sexual and gender identity. Perhaps this is because research has found that this type of victimization is experienced by many youth both online and in-person,37 making it feel unrelenting and inescapable.
Witnessing violence appears to have independent effects on the relative odds of depressed mood above and beyond the influence of cumulative victimization for most youth across sexual and gender identities. Witnessing violence included a range of exposures, from physical to sexual, and both within the home and among peers. Witnessing violence can result in not only feeling threatened and afraid for oneself at the moment,38 but can also affect more enduring beliefs and attributions (e.g., belief in a just world) that may in turn contribute to depression.39 This may be why this type of victimization experience continues to be influential above and beyond polyvictimization. As such, intervention and prevention initiatives for depressed mood should include screening for and addressing both witnessed and directly experienced violence.
Some important differences by sexual and gender identity also should be noted. Transgender boys emerged as the group with the highest average number of victimization experiences: 1 in 4 were polyvictims. Moreover, sexual harassment persisted as an explanatory factor for depressed mood for transgender boys within the context of polyvictimization. Gender policing, or the enforcement of normative gender expression and roles,40 could partially explain this finding; research suggests that sexual harassment may be used as a form of control for gender non-conforming self-expression.41 Yet, cumulative victimization eclipsed the relative odds of substance use observed for each type of victimization assessed for these youth. Further understanding of the impact of polyvictimization on different mental and behavioral health concerns is warranted.
Girls are consistently found to have higher rates of depression given assumed gender differences in rumination, rejection sensitivity, and hormonal factors.42 However, little known about how gender identity contextualizes these findings. Indeed, transgender boys and cisgender sexual minority girls had higher odds of substance use than cisgender heterosexual boys, and their relative odds of depressed mood were also higher than that of other genders. Such elevated risk requires proactive attention from healthcare providers.
Emotional IPV had a particularly strong relationship with substance use for cisgender heterosexual and sexual minority boys as well as cisgender heterosexual girls. Previous research finds that substance use can be a strategy to cope with post-traumatic stress in female survivors of intimate partner violence;43 this may be true for other youth as well. IPV also often occurs within the context of alcohol or drug use. 44 Substance use can place victims at higher risk for revictimization.45 Clearly, future work will be important to elucidate pathways and causal associations.
Limitations
Rather than limiting victimization types to one reference period, we chose to use the most comprehensive scope of victimization possible in the current survey. As such, some measures referred to past year. We may be undercounting lifetime events in these cases. The measure for depressed mood is broad and should not be considered as a marker for clinical depression. More research is needed that includes a more thorough assessment of mental health indicators. Other factors that could further contextualize violence exposure, such as racism, identity-based discrimination, and marginalization, were not measured. The cross-sectional nature of this study limits inferences about temporal associations. Combining non-binary and binary transgender youth may have obscured important differences. Finally, the smaller number of transgender girls precluded their inclusion in sub-analyses owing to power considerations.
Conclusions
SGM youth experience a disproportionate number of victimizations across multiple domains. Exposure to polyvictimization among youth who identify as a sexual or gender minority may be one of the key factors in elevated mental health concerns and substance misuse among youth experiencing such marginalization. Given recent literature suggesting that violence prevention programs are less effective for SGM youth,46 programs may benefit from increased focus on polyvictimization that is experienced by youth, particularly those who are SGM.
Supplementary Material
Acknowledgements
All phases of this study were supported by an NIH grant, R01 HD083072b. The funder did not participate in the work. The authors have no conflicts of interest to disclose. No financial disclosures have been reported by the authors of this paper.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Turner H, Finkelhor D, Ormrod R. Polyvictimization in a national sample of children and youth. Am J Prev Med. 2010; 38(3): 323–330. 10.1016/j.amepre.2009.11.012. [DOI] [PubMed] [Google Scholar]
- 2.Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The adverse childhood experiences (ACE) study. American Journal of Preventive Medicine. 2019. 10.1016/j.amepre.2019.04.001. [DOI] [PubMed] [Google Scholar]
- 3.Finkelhor D, Ormrod RK, Turner HA. Polyvictimization and trauma in a national longitudinal cohort. Development and Psychopathology. 2007; 19(1): 149–166. https://doi.org/10.10170S0954579407070083. [DOI] [PubMed] [Google Scholar]
- 4.Sterzing PR, Gartner RE, Goldbach JT, McGeough BL, Ratliff GA, Johnson KC. Polyvictimization prevalence rates for sexual and gender minority adolescents: Breaking down the silos of victimization research. Psychol Violence. 2019; 9(4): 419. 10.1037/vio0000123. [DOI] [Google Scholar]
- 5.Coker TR, Austin SB, Schuster MA. The health and health care of lesbian, gay, and bisexual adolescents. Anna Rev Public Health. 2010; 31: 457–477. 10.1146/annurev.publhealth.012809.103636. [DOI] [PubMed] [Google Scholar]
- 6.Johns MM, Lowry R, Andrzejewski J, et al. Transgender identity and experiences of violence victimization, substance use, suicide risk, and sexual risk behaviors among high school students—19 states and large urban school districts, 2017. Morbidity and Mortality Weekly Report. 2019; 68(3): 67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Johns MM, Lowry R, Rasberry CN, et al. Violence victimization, substance use, and suicide risk among sexual minority high school students—United States, 2015–2017. Morbidity and Mortality Weekly Report. 2018; 67(43): 1211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Dank M, Lachman P, Zweig JM, Yahner J. Dating violence experiences of lesbian, gay, bisexual, and transgender youth. J Youth Adolesc. 2014; 43(5): 846–857. 10.1007/s10964-013-9975-8. [DOI] [PubMed] [Google Scholar]
- 9.Puhl RM, Himmelstein MS, Watson RJ. Weight-based victimization among sexual and gender minority adolescents: Findings from a diverse national sample. Pediatr Obes. 2019; 14(7): e12514. 10.1111/ijpo.12514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sterzing PR, Ratliff GA, Gartner RE, McGeough BL, Johnson KC. Social ecological correlates of polyvictimization among a national sample of transgender, genderqueer, and cisgender sexual minority adolescents. Child Abuse Negl. 2017; 67: 1–12. 10.1016/j.chiabu.2017.02.017. [DOI] [PubMed] [Google Scholar]
- 11.Schwab-Reese LM, Currie D, Mishra AA, Peek-Asa C. A comparison of violence victimization and polyvictimization experiences among sexual minority and heterosexual adolescents and young adults. J Interpers Violence. 2018: 0886260518808853. 10.1177/0886260518808853. [DOI] [PubMed] [Google Scholar]
- 12.Marshal MP, Friedman MS, Stall R, et al. Sexual orientation and adolescent substance use: a meta-analysis and methodological review. Addiction. 2008; 103(4): 546–556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Friedman MS, Marshal MP, Guadamuz TE, et al. A meta-analysis of disparities in childhood sexual abuse, parental physical abuse, and peer victimization among sexual minority and sexual nonminority individuals. Am J Public Health. 2011; 101(8): 1481–1494. 10.2105/AJPH.2009.190009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Reisner SL, Greytak EA, Parsons JT, Ybarra ML. Gender minority social stress in adolescence: disparities in adolescent bullying and substance use by gender identity. J Sex Res. 2015; 52(3): 243–256. 10.1080/00224499.2014.886321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Coulter RW, Bersamin M, Russell ST, Mair C. The effects of gender-and sexuality-based harassment on lesbian, gay, bisexual, and transgender substance use disparities. J Adolesc Health. 2018; 62(6): 688–700. 10.1016/j.jadohealth.2017.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Veale JF, Watson RJ, Peter T, Saewyc EM. Mental health disparities among Canadian transgender youth. J Adolesc Health. 2017; 60(1): 44–49. 10.1016/j.jadohealth.2016.09.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.McGeough BL, Sterzing PR. A systematic review of family victimization experiences among sexual minority youth. J Prim Prev. 2018; 39(5): 491–528. 10.1007/s10935-018-0523-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schneeberger AR, Dietl MF, Muenzenmaier KH, Huber CG, Lang UE. Stressful childhood experiences and health outcomes in sexual minority populations: A systematic review. Soc Psychiatiy Psychiatr Epidemiol. 2014; 49(9): 1427–1445. 10.1007/s00127-014-0854-8. [DOI] [PubMed] [Google Scholar]
- 19.Ford JD, Elhai JD, Connor DF, Frueh BC. Polyvictimization and risk of posttraumatic, depressive, and substance use disorders and involvement in delinquency in a national sample of adolescents. J Adolesc Health. 2010; 46(6): 545–552. 10.1016/j.jadohealth.2009.11.212. [DOI] [PubMed] [Google Scholar]
- 20.Davis JP, Dworkin ER, Helton J, et al. Extending polyvictimization theory: Differential effects of adolescents’ experiences of victimization on substance use disorder diagnoses upon treatment entry. Child Abuse Negl. 2019; 89: 165–177. 10.1016/j.chiabu.2019.01.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ybarra ML, Petras H. Groups of sexual violence perpetration in a national sample of youth 13–25 years of age. Prev Sci. 2020: 1–11. 10.1007/s11121-020-01172-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ybarra ML, Thompson RE. Predicting the emergence of sexual violence in adolescence. Prev Sci. 2018; 19(4): 403–415. 10.1007/s11121-017-0810-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.American Association for Public Opinion Research. Response Rate Calculator V4.1. AAPOR. https://www-archive.aapor.org/Standards-Ethics/Standard-Definitions-(1).aspx. Published 2020. Accessed January 17, 2023. [Google Scholar]
- 24.Ybarra ML, Goodman KL, Saewyc E, Scheer JR, Stroem IF. Youth characteristics associated with sexual violence perpetration among transgender boys and girls, cisgender boys and girls, and nonbinary youth. JAMA Network Open. 2022; 5(6): e2215863–e2215863. 10.1001/jamanetworkopen.2022.15863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Koss MP, Gidycz CA. Sexual experiences survey: Reliability and validity. J Consult Clin Psychol. 1985; 53(3): 422. 10.1037//0022-006x.53.3.422. [DOI] [PubMed] [Google Scholar]
- 26.Lipson J, ed Hostile Hallways: Bullying, Teasing, and Sexual Harassment in School. Washington, DC: AAUW Educational Foundation; 2001. [Google Scholar]
- 27.Ybarra ML, Espelage DL, Langhinrichsen-Rohling J, Korchmaros JD. Lifetime prevalence rates and overlap of physical, psychological, and sexual dating abuse perpetration and victimization in a national sample of youth. Arch Sex Behav. 2016; 45(5): 1083–1099. 10.1007/s10508-016-0748-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hamby SL, Finkelhor D, Ormrod RK, Turner HA. The Juvenile Victimization Questionnaire (JVQ): Administration and scoring manual. Durham, NH: Crimes Against Children Research Center;2004. [Google Scholar]
- 29.Ybarra ML, boyd d, Korchmaros JD, Oppenheim JK. Defining and measuring cyberbullying within the larger context of bullying victimization. J Adolesc Health. 2012; 51(1): 53–58. 10.1016/j.jadohealth.2011.12.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Foshee VA Gender differences in adolescent dating abuse prevalence, types and injuries. Health Educ Res. 1996; 11(3): 275–286. [Google Scholar]
- 31.Foshee VA, Bauman KE, Arriaga XB, Helms RW, Koch GG, Linder GF. An evaluation of Safe Dates, an adolescent dating violence prevention program. Am J Public Health. 1998; 88(1): 45–50. 10.2105/ajph.88.1.45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hamby S, Finkelhor D, Ormrod R, Turner H. The Comprehensive Juvenile Victimization Questionnaire. Durham, NH: Crimes against Children Research Center, University of New Hampshire;2004. [Google Scholar]
- 33.Finkelhor D, Ormrod RK, Turner HA. Lifetime assessment of polyvictimization in a national sample of children and youth. Child Abuse Negl. 2009; 33(7): 403–411. 10.1016/j.chiabu.2008.09.012. [DOI] [PubMed] [Google Scholar]
- 34.Centers for Disease Control and Prevention. Youth Risk Behavioral Surveillance System (YRBSS). U.S. Department of Health and Human Services. https://www.cdc.gov/healthyyouth/data/yrbs/index.htm. Published 2019. Accessed August 8, 2020. [Google Scholar]
- 35.Ybarra ML, Espelage DL, Mitchell KJ. The co-occurrence of Internet harassment and unwanted sexual solicitation victimization and perpetration: Associations with psychosocial indicators. J Adolesc Health. 2007; 41(6): S31–S41. 10.1016/j.jadohealth.2007.09.010. [DOI] [PubMed] [Google Scholar]
- 36.Deming WE, Stephan FF. On a least squares adjustment of a sampled frequency table when the expected marginal totals are known. The Annals of Mathematical Statistics. 1940; 11(4): 427–444. [Google Scholar]
- 37.Mitchell KJ, Jones FM, Turner HA, Shattuck A, Wolak J. The role of technology in peer harassment: Does it amplify harm for youth? Psychol Violence. 2016; 6(2): 193. 10.1037/a0039317. [DOI] [Google Scholar]
- 38.Mitchell KJ, Jones FM, Turner HA, Beseler CL, Hamby S, Wade R Jr. Understanding the impact of seeing gun violence and hearing gunshots in public places: Findings from the Youth Firearm Risk and Safety Study. J Interpers Violence. 2021; 36(17-18): 8835–8851. 10.1177/0886260519853393. [DOI] [PubMed] [Google Scholar]
- 39.Collins KS, Koeske GF, Russell EB, Michalopoulos LM. Children’s attributions of community violence exposure and trauma symptomatology. Journal of Child & Adolescent Trauma. 2013; 6(3): 201–216. 10.1080/19361521.2013.811458. [DOI] [Google Scholar]
- 40.Henderson ER, Sang JM, Louth-Marquez W, et al. “Words Aren’t Supposed to Hurt, But They Do”: Sexual and Gender Minority Youth’s Bullying Experiences. J Interpers Violence. 2022; 37(11-12): NP8747–NP8766. 10.1177/0886260520978199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Kaltiala R, Ellonen N. Transgender identity and experiences of sexual harassment in adolescence. Child Abase Review. 2022: e2748. 10.1002/car.2748. [DOI] [Google Scholar]
- 42.Hatzenbuehler ML, McLaughlin KA. Sex, Sexual Orientation, and Depression. The University Oxford Press Oxford; 2017. [Google Scholar]
- 43.Gezinski LB, Gonzalez-Pons KM, Rogers MM. Substance use as a coping mechanism for survivors of intimate partner violence: implications for safety and service accessibility. Violence Against Women. 2019; 27(2): 108–123. 10.1177/10778012198824. [DOI] [PubMed] [Google Scholar]
- 44.Temple JR, Freeman DH Jr. Dating violence and substance use among ethnically diverse adolescents. J Interpers Violence. 2011; 26(4): 701–718. 10.1177/0886260510365858. [DOI] [PubMed] [Google Scholar]
- 45.Ullman SE, Najdowski CJ, Filipas HH. Child sexual abuse, post-traumatic stress disorder, and substance use: Predictors of revictimization in adult sexual assault survivors. J Child Sex Abus. 2009; 18(4): 367–385. 10.1080/10538710903035263. [DOI] [PubMed] [Google Scholar]
- 46.Coker AL, Bush HM, Clear ER, Brancato CJ, McCauley HL. Bystander program effectiveness to reduce violence and violence acceptance within sexual minority male and female high school students using a cluster RCT. Prev Sci. 2020: 1–11. 10.1007/s11121-019-01073-7. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.