Final Exam discussion post

Journal of Adolescent Health 65 (2019) 371e377

www.jahonline.org

Original article

Race-Related Traumatic Events Online and Mental Health Among Adolescents of Color

Brendesha M. Tynes, Ph.D. a,*, Henry A. Willis, M.A. b, Ashley M. Stewart, M.A. a, and Matthew W. Hamilton, M.A.T. a a Center for Empowered Learning and Development with Technology, Rossier School of Education, University of Southern California, Los Angeles, California bDepartment of Psychology and Neuroscience, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina

Article history: Received October 30, 2018; Accepted March 14, 2019 Keywords: Collective trauma; PTSD; Depressive symptoms; Internet; Immigrants; Police brutality; Viral videos

A B S T R A C T IMPLICATIONS AND

Purpose: Viral videos of the detainment of undocumented immigrants in cages and police killings of unarmed citizens are two of the most pressing traumatic events facing adolescents of color. However, little is known about whether these online experiences are linked to mental health outcomes. This study examines the association between exposure to such events online and mental health in a sample of African American and Latinx adolescents. Methods: Data were collected from a national sample of 302 African American and Latinx adolescents aged 11e19 years. Participants reported their exposure to traumatic events online (TEO), depressive symptoms, post-traumatic stress disorder (PTSD) symptoms, and other socio- demographics. Multiple regression analyses were conducted. Results: Analyses indicated a significant association between TEO and both PTSD symptoms and depressive symptoms, indicating that more frequent experiences of TEO were associated with higher levels of PTSD symptoms and depressive symptoms. In addition, regression analyses further indicated that girls reported higher PTSD and depressive symptoms than boys. Latinx participants also reported increased depressive symptoms. Conclusions: This study extends recent research that suggests police killings, as well as viewing distressing news directed at members of one’s own racialeethnic group or those who share the same immigration status, are related to poor mental health outcomes. Researchers should also explore what protective factors may buffer youth against the outcomes typically associated with these events.

� 2019 Society for Adolescent Health and Medicine. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Conflicts of interest: The authors have no conflicts of interest to disclose. Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the fund. * Address correspondence to: Brendesha M. Tynes, Ph.D., USC Rossier School

of Education, 3470 Trousdale Pkwy, WPH 600D, Los Angeles, CA 90089. E-mail address: [email protected] (B.M. Tynes).

1054-139X/� 2019 Society for Adolescent Health and Medicine. Published by Elsevier Inc. This is an open access article un creativecommons.org/licenses/by-nc-nd/4.0/). https://doi.org/10.1016/j.jadohealth.2019.03.006

CONTRIBUTION

Forty-five percent of teens report being online “almost constantly,” and video-sharing social me- dia platforms are among the most commonly visited sites for this popu- lation. Education, medical, and mental health pro- fessionals should discuss adolescent exposure to traumatic content and help youth to cope through technological strategies such as ensuring autoplay on videos is disabled for social media platforms. Findings also suggest a need for in- terventions to help youth critique and cope with race-related messages.

Scholarship on the impacts of witnessing traumatic events has grown exponentially since the September 11th attacks, with reviews of this literature citing associations between exposure to

terrorist events, symptoms of post-traumatic stress disorder (PTSD), and anxiety disorders in children [1]. Similarly, studies examining collective traumadthe recollections or psychological reactions to terrible events that happened to a group or society [2]dsuch as the Boston Marathon bombings suggest that repeated media exposure to these events can contribute to acute stress [3]. Although the literature demonstrates the effects that traumatic events and national disasters can have on mental health and well-being, fewer studies have focused specifically on

der the CC BY-NC-ND license (http://

B.M. Tynes et al. / Journal of Adolescent Health 65 (2019) 371e377372

racialeethnic minority populations. Furthermore, to date, there is no study that explores repeated exposure to these traumatic events online (TEO) among minority youth populations. This is particularly important to investigate among adolescent populations, given that recent research suggests Latinx parents of young people this age show increased psychological distress with recent immigration actions and news [4]. In addition, viral videos of the detainment of undocumented immigrants in cages and police killings of unarmed citizens are two of the most pressing traumatic events facing adolescents of color. This study examines the relationship between exposure to such events online and associations withmental health in a sample of African American and Latinx adolescents.

Theoretical framework

We use García Coll and colleagues’ [5] integrative model for the study of developmental competencies in minority children to inform the present study. The integrative model centers the unique experiences of racial-ethnic minority and immigrant youth and their normative development. It asserts that an emphasis on the race-related experiences of racial-ethnic mi- nority youth in a majority-dominated society is critical for research on their development. In addition, at the foundation of the integrative model is a focus on the social position held by youth of color, in particular, and the social mechanisms and forces that impact social, cognitive, and emotional develop- mental processes. Notably, the authors cite racism, discrimina- tion, and prejudice as directly affecting children’s developmental processes through social interactions in specific inhibiting or promoting environments [5]. Thus, we extend this model to the study of exposure to race-related traumatic events in online environments, and its mental health correlates among African American and Latinx adolescents.

Impact of local and regional traumatic events

Several studies describe how traumatic collective events correlate with PTSD and/or depressive symptoms in the local and regional communities where they occur. In their study on the impact of the September 11th attacks 5 to 8 weeks after the event, Ahern et al. [6] found that Manhattan residents living south of Canal St. had higher rates of PTSD than people living between Canal and 110th street. Building on this finding, Blanchard et al. [7] found that college students from New York and its close suburbs also had higher rates of acute stress dis- order and PTSD than students from upstate New York or students at colleges in two other states. They used the Posttraumatic Stress Disorder Checklist to measure re-experiencing, avoidance, numbing, and hyperarousal, and they concluded that students had PTSD if their scores were �3 [7].

Bor et al. [8] extend research on collective trauma, focusing on associations between police killings of unarmed African Americans in given regions and mental health outcomes. Employing a quasi-experimental design, the authors combined data on police killings with data from a nationally representative survey to examine the association between police killings and themental health of African American adults, specifically. African American survey respondents who lived in states where the police had killed an unarmed black person in the previous 3 months reported increased mental health challenges in the form of more poor mental health days. Bor et al. [8] argue the

association evidenced in this study mirrors that of the associa- tion between diabetes and mental health. They also suggest that police violence should be treated as a public health issue. Boyd [9] further argues that these findings fill an important gap in the literature by showing how police violence contributes to mental health inequity at the population level in addition to the indi- vidual level. She also highlights how their study contributes to naming the “built harm” of racism as a cause of racial health inequities. Given the historical context of pervasive racism in the U.S., it is expected that the effects of seeing African Americans being killed by police are deleterious for African Americans writ large, a phenomenon the authors name a “spillover effect” [8].

Indirect exposure to traumatic events through television and mainstream media

Studies on the relationship between media coverage of traumatic events and mental health suggest that this spillover effect might extend beyond the local regions where traumatic events occur, as people in other regions are indirectly exposed to the events through media coverage. Researchers have defined indirect exposure in general terms that may include hearing about a traumatic event through a relative or friend and discussing these events [10,11]. They may also include experi- ences through watchingmedia such as television, listening to the radio, or reading the newspaper, also called vicarious exposure [7] or media exposure [3]. Although the DSM-V criteria for PTSD exclude exposure to traumatic events through media [12], viewing coverage of 9/11 augmented the local and regional associations of the event, as measured in the studies mentioned previously; these studies reported correlations between post- attack media viewing and increased PTSD symptoms [6,7]. In one study,17.4% of peoplewho viewed TV coverage of individuals jumping out of the twin towers reported depressive symptoms [6]. In a separate study, watching at least 12 hours of 9/11 anniversary TV coverage was associated with a 3.4-fold increase in the risk of probable new-onset PTSD [13]. Although these studies only show correlation, not causation, Hopwood and Schutte conducted a meta-analysis of 18 experimental studies that suggest a causal relationship. With an overall sample size of 1,634 and a large effect size (Hedges’ g of 1.61), their results show that media exposure to large-scale public violence and disasters can cause negative psychological outcomes in the general population [14].

Several studies show that these associations between media viewing and PTSD symptoms hold true among youth of color in particular. One study of 141 fourth- through eighth-grade students in New Orleans evaluated the students over time to see how PTSD symptoms developed after Hurricane Katrina and then after Hurricane Gustav [15]. They found that the amount of time children spent watching Hurricane Gustav TV coverage was significantly correlated with PTSD symptoms, and this associa- tion was stronger for children with pre-existing symptoms. Students in this sample self-reported as 96% African American, 3.5% mixed African American or other, and .5% Hispanic [15]. In addition, a survey of middle school students’ responses to media coverage of the Oklahoma City bombing found that watching this coverage was directly related to post-traumatic stress symptoms of intrusion, avoidance, and arousal among youth (n¼ 2,381, 43% self-reporting as African American, 36% as Caucasian, 11% as Hispanic, 5% as American Indian, and 5% as other) [16]. Television exposure accounted for more variance in symptom measures

B.M. Tynes et al. / Journal of Adolescent Health 65 (2019) 371e377 373

than physical exposure (hearing or feeling the explosion) or emotional exposure (knowing someone who was killed or injured), and there was a significant association between television exposure and symptoms among students with no physical or emotional exposure [16].

After the BostonMarathon bombing, another study compared media viewing versus direct exposure to the event, looking at the association of both with acute stress symptoms (avoidance, re-experiencing, arousal/anxiety, and dissociation) [3]. People were classified as directly exposed if theywere at or near the site, if someone close to them was at or near the site, or if they knew someone who was injured or killed. Researchers surveyed samples of people from Boston (n ¼ 846), New York City (n ¼ 941), and the rest of the U.S. (n ¼ 2,888). Symptoms in the Boston and New York samples were comparable, but the national sample showed lower symptoms than the Boston one. Moreover, researchers found that�6 hours a day of bombing-related media coverage was associated with a higher measure of acute stress than direct exposure to the bombing itself [3].

Indirect exposure through social media and other nontraditional media

Although the link between television exposure to trauma and adverse mental health outcomes has been established, much less is known about how social media and other forms of online media come into play. Pew Internet Research’s most recent (2018) survey of adolescent technology usage [17] shows that 95% of youth have access to smartphones, and 45% say they are online “almost constantly.” Furthermore, the survey notes that the most popular Web site for 13- to 17-year-olds is YouTube, with Instagram as a close second [17]. The use of smartphones, video sharing, and social media platforms allow for an unprec- edented level of access to information as it relates to police brutality and other critical issues. This increased level of aware- ness has led to mass movements centering the voices of marginalized populations and the implementation of account- ability measures such as body cameras. However, the perma- nence of online content allows for the voluntary or involuntary (e.g., videos that automatically play when scrolling) repeated exposure to traumatic events. Furthermore, given the amount of time African American and Latinx youth spend online, they may be at a heightened risk for the potential effects of these experi- ences [18], with 54% of Latinx youth going online “almost constantly” compared with 41% of white counterparts [17].

Given the studies on direct and indirect exposure to collective traumas and the recent research showing negative mental health outcomes for those experiencing police killings in a given region, we focus on whether similar outcomes would result when such events are experienced via the Internet. Thus far, journalistic discussions about the psychological effects of viewing videos of police brutality online focus on the ways these videos might affect racialeethnic groups of color [19,20]. However, to date, there has not been a published empirical study examining the association between viewing viral videos with race-related traumatic content and mental health outcomes specifically for racialeethnic minority groups. Although scholars have yet to empirically study the impact of race-related online traumatic events, evidence of the negative effects of online racial discrim- ination [21] suggests this should be the next level of inquiry. This research has shown that African American adolescents and other adolescents of color experience both individual and vicarious

online racial discrimination at high rates. These experiences are associated with adverse mental health outcomes such as depressive symptoms and anxiety over and above offline expe- riences [21].

The present study

The current literature suggests that exposure to trauma both directly and through media outlets has implications for psycho- logical health. There is also some evidence to suggest that exposure to specific race-related trauma such as police brutality uniquely impacts people of color [8]. This study builds on these findings and examines whether exposure to TEO is related to mental health outcomes for African American and Latinx ado- lescents. In addition, adolescence is a formative period in which racialeethnic identity is becoming more salient [22,23]. As such, adolescents are becoming more equipped to perceive their experiences and those of others as racialized, suggesting an urgent need for research exploring offline consequences of the racial messages that pervade their online experiences. Our study examined the following questions:

� RQ1. Is there an association between exposure to race-related TEO and depressive symptoms? We hypothesized that TEO would be positively associated with depressive symptoms.

� RQ2: Is there an association between exposure to race-related TEO and symptoms of PTSD?We hypothesized that TEO would also be related to increased symptoms of PTSD.

Methods

Participants

Data for this study were drawn from a national online survey of adolescents’ critical media literacy skills when faced with race-related messages via the Internet. The sample was collected using the GfK Group’s (now Ipsos Public Affairs) Knowledge- Panel, the largest online panel that relies on probability-based sampling techniques for recruitment. The panel is designed to be representative of the U.S. and is based on addresses from the U.S. Postal Service. Panel members are randomly selected so that survey results can properly represent the U.S. population with a measurable level of accuracy. This representation is achieved not only with respect to a broad set of geodemographic indicators but also for hard-to-reach adults (such as those without Internet access or Spanish-languageedominant Hispanics) who are recruited in proper proportions.

Participants were 302 African American (N¼ 200) and Latinx/ Hispanic (N ¼ 102) youth between the ages of 11 and 19 years residing in the U.S. Of the Latinx participants, 67 were Mexican, Mexican American, or Chicano, 13 Puerto Ricans, 3 Cubans, and 19 from “other” Spanish/Latinx or Hispanic groups. The sample was 54.6% male (N ¼ 165) and 45.4% female (N ¼ 137) with an average age of 14.89 years (standard deviation ¼ 2.64). Partici- pants represented a range of socioeconomic backgrounds with 20.2% (N ¼ 61) reporting a household income of �$19,999, 17.9% (N ¼ 54) $20,000e$39,999, 18.5% (N ¼ 56) $40,000e$59,999, 17.6% (N ¼ 53) $60,000e$84,999, 4.6% (N ¼ 14) $85,000e $99,999, and 21.2% (N¼ 64)>$100,000. One thousand thirty-five households were sampled, and 428 completed the screener interview. Of this number, 302 met sampling criteria and had an African American or Latinx adolescent between the ages of

B.M. Tynes et al. / Journal of Adolescent Health 65 (2019) 371e377374

11e19 years who was available and willing to complete the survey. This is a qualification rate of 70.6%.

This study was approved by the University of Southern California’s Institutional Review Board before data collection, which was completed in July and August of 2018. GfK adhered to all ethical guidelines and followed the research protocol outlined in the institutional review board application. For this study, we were awarded funding that allowed us to include items that would take a maximum of 10 minutes to complete. As a result, all the measures with the exception of the three-item TEO and the critical media literacy assessments (not included in this study) were cut to meet time limits.

Procedure

GfK conducted the study on behalf of the university. The survey consisted of two stages: initial screening for African- American or Hispanic households with teens between the ages of 11 and 19 years and the distribution of the main survey with eligible respondents. Parents were initially screened using a screener interview. If they met the criteria, they were asked to provide consent for their child to complete the study survey. Once assigned to a survey, participants received a notification email letting them know there is a new survey available for them to complete. The survey took a median time of 14 minutes to complete.

Measures

Traumatic events online

The frequency of participants’ experiences of witnessing traumatic race-related events online (e.g., on Internet news sites, Facebook/Twitter, etc.) over the past year was assessed with three self-report items that corresponded to seeing images or videos of others from their ethnic group being beaten (item 1), arrested or detained (item 2), or viral video of a black person being shot by a police officer (item 3, note this item asked spe- cifically about seeing black people being shot as these were the majority of people in the viral videos). A sample item was “During the past year, I have seen images or videos of people frommy ethnic group being arrested or detained (e.g., immigrant children being held in shelters).” The frequency of each traumatic event was rated from 0 ¼ never to 5 ¼ every day. Scores were averaged to create mean (i.e., composite) scores, and higher scores corresponded to more frequent experiences of witnessing traumatic race-related events online. Reliability analyses revealed a high internal consistency, as shown by a Cronbach’s alpha of .85.

Depressive symptoms

Four items from the self-report Children’s Depression In- ventory Second EditioneShort (CDI-S) [24] were used to assess current symptoms of depression. Participants were asked to rate the severity of different symptoms of depression, with each symptom being presented as a series of three phrases (e.g., I am sad once in a while, I am sad many times, I am sad all the time). Scores ranged from 0 to 8, with higher scores corresponding to higher levels of depressive symptoms. Symptoms included being sad, feeling like crying, feeling alone, and feeling like they had friends. The CDI-S has demonstrated internal consistency and

has been shown to be strongly associated with the full version of the CDI [25]. In the present study, the four items used from the CDI-S demonstrated moderate internal consistency, as shown by a Cronbach’s alpha of .56. This modest alpha coefficient is a reflection of the small number of items used from the overall scale. Items used align with studies of online victimization con- ducted by the Crimes Against Children Research Center that assess emotional impact by asking participants if they feel sad, for example [26].

PTSD symptoms

PTSD symptoms were assessed using three out of four items from the self-report New York PTSD Risk Score [27]. The items used in the present study corresponded to re-experiencing, hy- perarousal, and numbing, which have been shown to be specific to PTSD and distinct from general psychological distress [28]. The omitted item asked about avoidant behaviors related to the traumatic event such as avoiding places and activities where the event occurred. This would have suggested avoiding the Internet and would have required additional explanation and time. Par- ticipants were asked to respond “yes” or “no” if they had expe- rienced any of the symptoms after being involved in or witnessing traumatic life events. Total scores ranged from 0 to 3, with higher scores corresponding to higher levels of PTSD symptoms. The New York PTSD Risk Score has demonstrated internal consistency and has been shown to be effective in pre- dicting PTSD [27]. In the present study, the three items used from the overall scale demonstrated moderate internal consistency, as shown by a Cronbach’s alpha of .69. This alpha coefficient approached the recommended level of .7, and items represent common symptoms of PTSD for this population [29].

Covariates

The study incorporated covariates that may influence depressive and PTSD symptoms, such as age, sex, ethnicity, and household income.

Results

Preliminary analysis

Descriptive statistics (i.e., correlations, means, and standard deviations) for the key study variables are summarized in Table 1. Bivariate correlations indicate a positive association between TEO and depressive symptoms as well as symptoms of PTSD (p < .05). Each individual item of TEO was also positively correlated with PTSD symptoms. In addition, a positive correla- tion between depressive symptoms and symptoms of PTSD is evidenced.

Analytic plan

Multiple regression was used to examine the role of TEO as a risk factor in the context of depressive and PTSD symptoms. Age, sex, household income, and ethnicity were included as cova- riates. TEO was entered as a main effect predictor of the outcome variables in two separate multiple regression models: depressive and PTSD symptoms. All continuous independent variables and covariates were mean-centered in both models. To assess if in- dividual items of the TEO were differentially associated with

Table 1 Bivariate correlations among study variables

Correlations 1 2 3 4 5 6 7 8 9 10

1. Age d

2. Gender .00 d

3. Ethnicity .04 �.06 d

4. Household income .143* �.07 .07 d

5. TEOdimages/videos of beatings .165** .09 �.208** .07 d

6. TEOdimages/videos of arrests/detainment .217** .132* �.129* .02 .705** d

7. TEOdviral video of black person shot .269** .06 �.167** �.02 .619** .666** d

8. TEOdcomposite score .248** .11 �.191** .03 .884** .902** .857** d

9. Depressive symptoms .08 .151** .127* �.05 .122* .08 .120* .124* d

10. PTSD symptoms .137* .156** .01 �.10 .258** .191** .254** .267** .431** d

Mean 14.89 d d d 1.90 2.43 2.07 2.13 1.00 .41 SD 2.64 d d d 1.58 1.59 1.44 1.35 1.08 .77

PTSD ¼ post-traumatic stress disorder; SD ¼ standard deviation; TEO ¼ traumatic events online. *Correlation is significant at the .05 level (two tailed). **Correlation is significant at the .01 level (two tailed).

B.M. Tynes et al. / Journal of Adolescent Health 65 (2019) 371e377 375

depressive and PTSD symptoms, we conducted exploratory multiple regression. The findings mirrored the correlational an- alyses (Table 1); therefore, we only report the effect of the mean, composite score of TEO on depressive and PTSD symptoms.

The effect of TEO on depressive symptoms

The omnibus test for the analytic model (Table 2) was sig- nificant, (F(5, 293) ¼ 4.05, p < .001; adjusted R2 ¼ .05). In exploring the effect of the covariates, ethnicity was significantly associated with depressive symptoms (b ¼ .37, p < .01). Specif- ically, Hispanic participants had significantly higher depressive scores than African American participants (M ¼ 1.2 and .91, respectively). In addition, sex was significantly associated with depressive symptoms (b ¼ .31, p < .05). Of note, female partici- pants had significantly higher depressive scores than male par- ticipants (M ¼ 1.19 and .85, respectively). With respect to main effects, the composite TEO score was significantly associated with depressive symptoms (b¼ .10, p<.05), indicating that more overall frequent experiences of TEO were associated with higher levels of depressive symptoms.

The effect of TEO on PTSD symptoms

The omnibus test for the analytic model (Table 3) was sig- nificant, (F(5, 286) ¼ 6.94, p < .001; adjusted R2 ¼ .09). In exploring the effect of the covariates, sex was significantly associated with PTSD symptoms (b ¼ .19, p < .05). Specifically, female participants had significantly higher PTSD scores than male participants (M ¼ .54 and .304, respectively). Household income was marginally significant (b ¼ �.02, p ¼ .051), such that

Table 2 Summary of multiple regression analysis predicting depressive symptoms from traumatic events online and control variables (N ¼ 302)

Variable B SE b

Intercept .07 .27 d

Sex .31* .12 .14 Ethnicity .37** .13 .16 Age .02 .02 .06 Household income �.01 .01 �.06 Traumatic events onlinedcomposite score .10* .05 .13

B ¼ unstandardized regression coefficient; b ¼ standardized coefficient; SE ¼ standard error of the coefficient. *p < .05; **p < .01.

higher household income was associated with lower levels of PTSD symptoms. With respect to main effects, TEO was signifi- cantly associated with PTSD symptoms (b ¼ .14, p <.001), indi- cating that more frequent experiences of TEO were associated with higher levels of PTSD symptoms.

Discussion

This study is the first nationally representative survey to assess whether viewing race-related TEO such as viral videos of police shootings and immigrant detainment is associated with depressive and PTSD symptoms among adolescents of color. Results show that viewing these events is related to increased levels of both outcomes. In addition, consistent with the litera- ture [30,31], girls had higher depressive and PTSD symptoms than boys. Latinx participants also experienced more depressive symptoms in the face of these events. Findings are consistent with Garcia Coll et al.’s framework [5] and suggest that online environments may become inhibiting, or they may undermine the development of emotional competencies when viral videos of discriminatory and violent practices are viewed.

This study’s findings are also consistent with extant literature, which shows that media exposure to traumatic events is asso- ciated with negative psychological outcomes [16]. It extends previous literature that found those in close proximity to col- lective trauma have the most negative psychological symptoms [6]. For example, New Yorkers simply viewing Sept 11th televi- sion coverage 1 year later had new-onset PTSD in a sample with no probable symptoms at baseline [13]. These findings also hold for adolescents of color, specifically, as literature has shown they

Table 3 Summary of multiple regression analysis predicting PTSD symptoms from trau- matic events online and control variables (N ¼ 302)

Variable B SE b

Intercept �.004 .19 d

Sex .19* .09 .12 Ethnicity .10 .09 .06 Age .03 .02 .09 Household income �.02 .01 �.11 Traumatic events onlinedcomposite score .14*** .03 .24

B ¼ unstandardized regression coefficient; b ¼ standardized coefficient; SE ¼ standard error of the coefficient. *p < .05; **p < .01; ***p<.001.

B.M. Tynes et al. / Journal of Adolescent Health 65 (2019) 371e377376

are more likely to experience PTSD symptoms with increased television viewing of natural disasters [15].

This research further builds on recent work on police shoot- ings that shows that each killing was associated with a .14 day increase in poor mental health days for adults in a given geographical location, a total of 55 million at the population level [8]. This study asked about viewing viral videos of police killings in the past year among adolescents. It may be the case that viewing videos of marginalized groups who experience race or ethnicity-related trauma online may increase rumination about these events. Media exposure has been noted to keep an acute stressor “active and alive in one’s mind” [3]. Youth encountering messages and images that disparage them based on race are also potentially self-identifying with the individuals directly affected by such trauma and perceiving a potential threat based on membership in a particular ethnic racial group [32]. It may also make adolescents keenly aware that individuals in the video could easily have been them, and this may decrease participants’ sense of control overwhat happens to them. Alang et al. [33] note that research has specifically linked police brutality with poor health through physical injury and oppressive structures that cause systematic disempowerment. They argue that each episode has an emotional and physiological effect on individuals and communities. Most importantly, they argue it is a constant reminder of the devaluing of black lives. Similarly, as our model suggests, viral videos in this study were visual reminders of Af- rican American and Latinx individuals’ low social position.

The fact that Latinx participants had increased levels of depressive symptoms associated with traumatic online events aligns with studies that consistently show Latinx adolescents have higher rates of depression than other groups [34,35]. In addition, research has shown that although African Americans experience more PTSD in the face of perceived discrimination, Latinx experience more depressive symptoms [36]. This finding is also consistent with recent research that shows immigrant- related policies and news coverage are associated with psycho- logical distress for Latinx parents [4]. It may be that recent policies that place large populations of Latinxs in precarious positions, the persistent news coverage that demeans this group along with an increase in face-to-face experiences where they are dehumanized by peers with threats to “build a wall” [37], may create a climate for heightened depressive symptoms.

The finding that girls had higher levels of PTSD and depressive symptoms than boys also parallels extant literature. For example, reviews of literature on mass shootings show that women and girls are more likely to experience psychological adversity postincident. Researchers argue this may be because women are more likely to use ruminative coping styles, and this, in turn, increases the severity of symptoms [10]. These events also in- crease fears andmake people feel unsafe. This maymakewomen and girls particularly vulnerable. Interestingly, none of the ana- lyses that explored gender by race-ethnicity interactions were significant.

Althoughmeta-analyses suggest that the association between media exposure to traumatic events and negative psychological symptoms may be causal [14], cross-sectional data in this study cannot be used to make such claims. Also, we did not assess viral video viewing immediately following the incident for all partic- ipants, but asked if they had viewed these videos in the past year. This is a potential limitation considering Neria and Sullivan argue that theremay be initial but not extended rates of PTSD following indirect exposure to mass violence via mass media [11]. Future

research should determine the longitudinal implications of race- related TEO. Qualitative and mixed-method studies should also be conducted to examine the meaning-making process with respect to identifying with people in the images/videos and feelings of safety or loss of control. Researchers should also explore what protective factors may buffer youth against poor mental health outcomes related to these events. In addition, mental health professionals should ask questions about adoles- cents’ viewing practices to determine clients’ exposure and meaning-making around race-related trauma via the Internet.

Funding Sources

This work was supported by the James H. Zumberge Faculty Research and Innovation Fund, Office of the Provost at the University of Southern California, United States.

References

[1] Fremont WP. Childhood reactions to terrorism-induced trauma: A review of the past 10 years. J Am Acad Child Adolesc Psychiatry 2004;43:381e92.

[2] Hirschberger G. Collective trauma and the social construction of meaning. Front Psychol 2018;9:1441.

[3] Holman EA, Garfin DR, Silver RC. Media’s role in broadcasting acute stress following the Boston Marathon bombings. Proc Natl Acad Sci U S A 2014; 111:93. Available at: http://www.pnas.org/content/111/1/93.abstract. Accessed September 26, 2018.

[4] Roche KM, Vaquera E, White RMB, Rivera MI. Impacts of immigration ac- tions and news and the psychological distress of U.S. Latino parents raising adolescents. J Adolesc Health 2018;62:525e31.

[5] García Coll G, Lamberty G, Jenkins R, et al. An integrative model for the study of developmental competencies in minority children. Child Dev 1996;67:1891.

[6] Ahern J, Galea S, Resnick H, et al. Television images and psychological symptoms after the September 11 terrorist attacks. Psychiatry 2002;65: 289.

[7] Blanchard EB, Kuhn E, Rowell DL, et al. Studies of the vicarious traumati- zation of college students by the September 11th attacks: Effects of prox- imity, exposure and connectedness. Behav Res Ther 2004;42:191e205.

[8] Bor J, Venkataramani AS, Williams DR, Tsai AC. Police killings and their spillover effects on the mental health of black Americans: A population- based, quasi-experimental study. Lancet 2018;392:302e10.

[9] Boyd RW. Police violence and the built harm of structural racism. Lancet 2018;392:258e9.

[10] Lowe SR, Galea S. The mental health consequences of mass shootings. Trauma Violence Abuse 2017;18:62e82.

[11] Neria Y, Sullivan GM. Understanding the mental health effects of indirect exposure to mass trauma through the media. JAMA 2011;306:1374e5.

[12] American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Publishing; 2013.

[13] Bernstein K, Ahern J, Tracy M, et al. Television watching and the risk of incident probable posttraumatic stress disorder: A prospective evaluation. J Nerv Ment Dis 2007;195:41e7.

[14] Hopwood TL, Schutte NS. Psychological outcomes in reaction to media exposure to disasters and large-scale violence: A meta-analysis. Psychol Violence 2017;7:316e27.

[15] Weems CF, Scott BG, Banks DM, Graham RA. Is TV traumatic for all youths? The role of preexisting posttraumatic-stress symptoms in the link between disaster coverage and stress. Psychol Sci 2012;23:1293e7.

[16] Pfefferbaum B, Nixon SJ, Tivis RD, et al. Television exposure in children after a terrorist incident. Psychiatry 2001;64:202e11.

[17] Pew Research Center. Teens, social media & technology 2018. Available at: http://www.pewinternet.org/2018/05/31/teens-social-media-technology- 2018/. May 31, 2018. Accessed October 26, 2018.

[18] Lenhart A, Duggan M, Perrin A, et al. Teens, social media & technology overview 2015. Pew Research Center; April 9, 2015. Available at: http:// www.pewinternet.org/2015/04/09/teens-social-media-technology-2015/. Accessed October 26, 2018.

[19] Adetiba L, Almendrala A. Watching videos of police brutality can trauma- tize you, especially if you’re black. Huffington Post; September 20, 2016. Available at: https://www.huffingtonpost.com. Accessed September 2, 2018.

B.M. Tynes et al. / Journal of Adolescent Health 65 (2019) 371e377 377

[20] Downs K. When black death goes viral, it can trigger PTSD-like trauma. PBS NewsHour; July 22, 2016. Available at: https://www.pbs.org. Accessed September 2, 2018.

[21] Tynes BM, Giang MT, Williams DR, Thompson GN. Online racial discrimi- nation and psychological adjustment among adolescents. J Adolesc Health 2008;43:565e9.

[22] Phinney JS. Stages of ethnic identity development in minority group ado- lescents. J Early Adolesc 1989;9:34e49.

[23] Fuligni AJ, Hughes DL, Way N. Ethnicity and immigration. In: Lerner R, Steinberg L, eds. Handbook of Adolescent Psychology. 3rd ed. Hoboken, NJ: John Wiley & Sons; 2009.

[24] Kovacs M. Children’s depression inventory 2. Available at: https://www. pearsonclinical.com/psychology/products/100000636/childrens-depression- inventory-2-cdi-2.html#tab-details. 2010. Accessed October 26, 2018.

[25] Ahlen J, Ghaderi A. Evaluation of the children’s depression inventory-short version. Psychol Assess 2016;29:1157e66.

[26] Mitchell KJ, Jones LM, Turner HA, Wolak J. Technology-involved harass- ment victimization; Placement in a broader victimization context, Sum- mary report, NIJ grant no. 2012-IJ-CX-0024. Durham, NH: Crimes Against Children Research Center; 2015.

[27] Boscarino JA, Kirchner HL, Hoffman SN, et al. The New York PTSD risk score for assessment of psychological trauma: Male and female versions. Psy- chiatry Res 2012;200:827e34.

[28] Foa EB, Riggs DS, Gershunny BS. Arousal, numbing, and intrusion: Symptom structure of PTSD following assault. Am J Psychiatry 1995;152: 116e20.

[29] Kline RB. Principles and practice of structural equation modeling. 2nd ed. New York, NY: Guildford; 2005.

[30] Thapar A, Collishaw S, Pine DS, Thapar AK. Depression in adolescence. Lancet 2012;379:1056e67.

[31] Hyde JS, Mezulis AH, Abramson LY. The ABCs of depression: Integrating affective, biological, and cognitive models to explain the emergence of the gender difference in depression. Psychol Rev 2008;115:291e313.

[32] Schmader T, Block K, Lickel B. Social identity threat in response to stereotypic film portrayals: Effects on self-conscious emotion and implicit ingroup attitudes. J Soc Issues 2015;71:54e72.

[33] Alang S, McAlpine D, McCreedy E, Hardeman R. Police brutality and black health: Setting the agenda for public health scholars. Am J Public Health 2017;107:662e5.

[34] Ramirez AG, Gallion KJ, Aguilar R, Dembeck ES. Mental health and Latino kids: A research review. Salud America; September 12, 2017. Available at: https://salud-america.org/wp-content/uploads/2017/09/FINAL-mental- health-research-review-9-12-17.pdf. Accessed October 26, 2018.

[35] Céspedes YM, Huey SJ. Depression in Latino adolescents. Cultur Divers Ethnic Minor Psychol 2008;14:168e72.

[36] Chou T, Asnaani A, Hofmann S. Perception of racial discrimination and psychopathology across three U.S. ethnic minority groups. Cultur Divers Ethnic Minor Psychol 2012;18:74e81.

[37] Southern Poverty Law Center. SPLC survey: Presidential campaign leading to widespread fear, bullying in schools. Available at: https://www. splcenter.org/news/2016/04/13/splc-survey-presidential-campaign-leading- widespread-fear-bullying-schools. April 13, 2016. Accessed October 26, 2018.