Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Dec 4.
Published in final edited form as: J Trauma Stress. 2019 Dec 4;32(6):890–898. doi: 10.1002/jts.22473

Polytraumatization, Mental Health, and Delinquency Among Adolescent Gang Members

Liesl A Nydegger 1, Katherine Quinn 2, Jennifer L Walsh 2, Maria L Pacella-LaBarbara 3, Julia Dickson-Gomez 4
PMCID: PMC6938535  NIHMSID: NIHMS1060865  PMID: 31800141

Abstract

Adolescent gang members are at high risk for polytraumatization (i.e., experiencing two or more types of trauma), which may contribute to behavioral problems, such as delinquency or drug distribution, and mental health symptoms, such as posttraumatic stress disorder (PTSD) and depression. The present study examined the associations between polytraumatization and these behavioral and mental health outcomes. We hypothesized that increased polytraumatization would lead to increased (b) mental health symptoms, (b) delinquency, and (c) drug distribution. Participants included 441 adolescent gang members (57.8% male; age range: 14–19 years) from a midsized city in the Midwestern United States. A path model was used to test hypotheses. A total of 88.0% of participants experienced polytraumatization, such as physical and sexual assault, involvement in accidents, or witnessing a death or injury. Polytraumatization was uniquely and positively associated with depressive and PTSD symptoms, delinquency, and drug distribution, βs = .25–.50, ps < .001, explaining an additional 5.9%–22.5% of the variance in these outcomes beyond covariates. Untreated traumatic exposure among adolescent gang members may subsequently lead to poor behavioral and mental health outcomes. These results may inform prevention and intervention efforts focused on mental health and social justice among a high-risk adolescent population.

Young adult gang members experience disproportionately high levels of trauma during their lives (Barnes, Boutwell, & Fox, 2012; Loeber, Kalb, & Huizinga, 2001), including childhood trauma (De La Rue & Espelage, 2014; Petering, 2016), sexual assault, dating violence, and serious injury from physical fighting (Gover, Jennings, & Tewksbury, 2009). Gang members are at particularly high risk for polytraumatization (Barnes et al., 2012; Loeber et al., 2001), a term used to capture the multiple types of trauma individuals may experience, including direct and vicarious victimization. Vicarious victimization may include injuries, accidents, illnesses, and incidents that occur to loved ones (Gustafsson, Nilsson, & Svedin, 2009).

Polytraumatized children and adolescents report higher levels of psychological symptoms (e.g., depression, anger, anxiety, emotional symptoms, conduct problems, hyperactivity or inattention, and peer problems) than those who have experienced a single traumatic event (Gustafsson et al., 2009). Specifically, previous studies have shown that polytraumatization contributes to poor mental health outcomes among delinquent youth (Loeber et al., 2001) and gang members (Coid et al., 2013). Compared to non–gang members, gang members have been shown to report higher levels of violence-related traumatic stress (Kerig, Chaplo, Bennett, & Modrowski, 2016), symptoms of posttraumatic stress (Harris et al., 2013; Petering, 2016), and depression (Petering, 2016), and are more likely to engage in delinquent (e.g., police involvement, fights, graffiti) and violent behavior (Harper, Davidson, & Hosek, 2008).

There are also links between traumatization and the perpetration of violence (Hartinger-Saunders et al., 2011; Maschi, 2006; Ross & Arsenault, 2018; Shaffer & Ruback, 2002), drug use (Lin, Cochran, & Mieczkowski, 2011; Loeber et al., 2001), and drug distribution among adolescents. Previous studies have demonstrated that gang members are more likely to engage in delinquent behaviors (De La Rue & Espelage, 2014; Harper et al., 2008) and drug distribution (Bjerregaard, 2010) than non–gang members. Drug distribution has rarely been examined as an outcome variable; in past research, it was either not included (Harper et al., 2008; Hartinger-Saunders et al., 2011; Maschi, 2006; Shaffer & Ruback, 2002) or was included as part of delinquency (Begle et al., 2011). Although drug distribution is a form of delinquency, it is a behavior common among gangs, and, as such, we explored the association between polytraumatization and drug distribution separately from delinquency.

The increased risk of mental health symptoms (Harris et al., 2013; Petering, 2016), violent and nonviolent delinquency (De La Rue & Espelage, 2014; Harper et al., 2008), and drug distribution (Bjerregaard, 2010) among gang members compared to non–gang members may be explained by polytraumatization. The present study examined these relations. Existing research among gang members has focused primarily on recent or specific kinds of trauma, or has not included vicarious traumas. The present study sought to fill these gaps by investigating the associations among polytraumatization, overall delinquency, drug distribution, and poor mental health in a high-risk sample of adolescent gang members in the Midwestern U.S. city of Milwaukee, one of the most racially segregated cities in America and a city that faces significant social and economic disparities (Logan & Stults, 2011). In 2015, Milwaukee’s homicides were concentrated in majority-Black zip codes in high-poverty neighborhoods situated in the central and northwest areas of the city (Gordon, 2016). A study demonstrated that the rate of incarceration for Black men in Wisconsin was the highest in the country and nearly double the national average. More than half of all Black men in their 30s and 40s living in Milwaukee had a history of incarceration (Pawasarat & Quinn, 2013). High schools in Wisconsin suspended Black students at a higher rate than anywhere else in the country, and Milwaukee suspended Black high school students at a rate nearly double the national average (Losen, Hodson, Keith, Morrison, & Belway, 2015). We hypothesized that among our sample, more extensive polytraumatization would be associated with increased (a) symptoms of mental illness, (b) delinquency, and (c) involvement in drug distribution among adolescent gang members. This research may offer keys to interventions aimed at decreasing delinquent involvement and improving early trauma-informed intervention.

Method

Participants and Procedure

The present study took place in the U.S. city of Milwaukee, WI between 2013 and 2015. Participants were active youth gang members representing 71 different street gangs. Eligibility criteria for the present study included being between 14 and 19 years of age, a member of a Milwaukee gang, and able to provide informed consent or assent. Of 461 participants originally enrolled in the study, 441 participants (57.8% boys, 60.3% Black, 26.8% Latino, 12.9% other races, M age = 17.39 years, SD = 1.53; 73.3% from single-parent families) who provided at least some data on key variables were included in the current analyses.

The institutional review board at the Medical College of Wisconsin approved all study procedures. Participants were recruited from Milwaukee using direct street outreach at street festivals, parks, and other areas with known gang activity as well as through referrals from community organizations and schools serving active gang members. We also used participant referrals; individuals who completed the survey received two referral cards to aid in recruitment of other gang members and received a $10 (USD) incentive for each referral. Research assistants conducting outreach, screening, and enrollment had extensive knowledge of community gangs, were known and trusted by many gang leaders, and had strong ties to youth-serving organizations. Interested individuals were screened by phone or in person to verify gang membership and eligibility, confirming gang status by demonstrating knowledge of specific gang names, signs, and leaders. Self-identification as a gang member has been demonstrated to be a strong predictor of embeddedness in gangs (Decker, Pyrooz, Sweeten, & Moule, 2014) and an effective measurement of gang status (Esbensen, Winfree Jr., He, & Taylor, 2001; Thornberry, Krohn, Lizotte, & Smith, 2003). Eligible participants interested in participating completed a written informed consent. A waiver of parental consent was obtained for participants who were under 18 years of age; thus, informed assent (i.e., the same as the informed consent) was collected from participants who were minors. This was to protect the identities of participants whose parents may have been unaware of their gang involvement. Participants completed a cross-sectional computerized survey using audio computer-assisted self-interview software in a community-based setting and received $30 in compensation.

Measures

Control variables.

The following potentially confounding variables were included in all analyses: sex (boy/girl), age, racial/ethnic background (non-Hispanic Black [reference category], Latinx, and other), single-parent versus two-parent households, and neighborhood risk. The seven-item self-report Neighborhood Risk measure (Jones, Forehand, Brody, & Armistead, 2002) was included as a covariate due to the close association between neighborhood risk and gang involvement (Gilman, Hill, & Hawkins, 2014). Example items included “How often are there physical fights without the use of weapons in your neighborhood?” and “How often are there shootings in your neighborhood?” Responses were given using a scale ranging from 1 (almost never) to 4 (almost always). Items were averaged, with higher scores indicating more exposure to neighborhood risk. The reliability of the total scale had adequate reliability in the present sample, Cronbach’s α = .81.

Polytraumatization.

The polytraumatization assessment was adapted from the Exposure to Violence Questionnaire scale (Gladstein & Slater, 1988) and included 17 items. Participants reported whether they had ever experienced various traumatic events or forms of direct or vicarious victimization. Dichotomous items were summed, with higher scores indicating more traumatic experiences. The reliability of the total polytraumatization scale from the adapted Exposure to Violence Questionnaire was high in the present sample, Cronbach’s α = .92.

Depression.

Participants completed the nine-item Patient Health Questionnaire (PHQ-9; Kroenke, Spitzer, & Williams, 2001) to measure depressive symptoms over the last 2 weeks. Example items included “How often have you had little interest or pleasure in doing things?” and “How often have you felt tired or had little energy?” Answers were given using a scale of 0 (not at all) to 3 (nearly every day). Items were summed, with higher scores indicating more symptoms of depression. For descriptive purposes only, a score higher than 10 was considered an indicator of major depression (Kroenke et al., 2001). In the present sample, the internal consistency of the total depression scale from the PHQ was high, Cronbach’s α = .93.

Posttraumatic stress disorder (PTSD).

Participants completed the 17-item Child PTSD Symptoms Scale (Foa, Johnson, Feeny, & Treadwell, 2001), which reflects past 2-week levels of PTSD symptoms, based on criteria from the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), including having upsetting thoughts, feeling upset when thinking about an event, feeling irritable, and being jumpy. Response options were modified for the current study and ranged from 0 (not at all) to 3 (almost always). Items were summed, with higher scores indicating more symptoms of PTSD. For descriptive purposes only, a score higher than 11 was considered an indicator of PTSD (Foa et al., 2001). In the present sample, the reliability of the total PTSD scale from the Child PTSD Symptoms scale was good, Cronbach’s α = .96.

Delinquency.

Past-year delinquency was measured with seven items from the Communities that CARE Youth Survey (Substance Abuse and Mental Health Services Administration [SAMHSA], 2004). Examples of items included “How many times in the past 12 months have you been arrested?” and “How many times in the past 12 months have you attacked someone with the idea of seriously hurting them?” One item was added for the present study: “How many times in the past 12 months have you stolen anything other than a motor vehicle?” Items were dichotomized to indicate any involvement with each delinquent activity (0 = no, 1 = yes). Dichotomous items were summed, with higher scores indicating more delinquency. In the present sample, the reliability from of the total delinquency scale from the CARE Youth Survey was adequate, Cronbach’s α = .82.

Drug distribution.

The measure of drug distribution was adapted from the Drug Distribution Roles measure (Sherman & Latkin, 2002), which reflects whether participants steered customers to or advertised drugs, held drugs or drug money, acted as a lookout, acted as security, assisted in preparing drugs, sold or rented paraphernalia, or sold drugs in the last 6 months. Three items were added to the original scale to reflect whether participants bought drugs for another person, transported large quantities of drugs, or delivered drugs to a customer. Response options were yes and no. Items were summed, with higher scores indicating more involvement with drug distribution. In the present sample, the internal consistency of the total drug distribution scale from the adapted Drug Distribution Roles measure was high, Cronbach’s α = .92.

Data Analysis

Of the 461 participants who enrolled in the larger study, 4.3% (n = 20) were missing data on all key variables in the current study and were thus excluded from analyses. Excluded participants were more likely to be Latino than those included in analyses. For the 441 high-risk participants included in the current investigation, rates of missing data ranged from 0.0% to 23.8% for the key study variables; overall, 14.8% of data on key variables was missing. Reports of delinquency and drug distribution were most likely to be missing. Participants with missing data (40.1% of the sample, n = 177) were more likely than those with complete data to be boys (65.0% vs. 53.0%), Latino (43.5% vs. 15.5%), and from two-parent families (35.4% vs. 21.1%); they were also younger (M age = 17.03 vs. M age = 17.63) and reported lower summed scores on measures of polytraumatization (M = 7.68 vs. M = 11.36), depression (M = 2.34 vs. M = 4.66), PTSD (M = 4.89 vs. M = 11.14), and delinquency (M = 1.77 vs. M = 2.85), and higher scores on the neighborhood risk assessment (M = 2.67 vs. M = 2.50), ps < .001–p = .014. We used multiple imputation, including all study variables, to replace missing values (Shafer, 1997). Multiple imputation avoids biases associated with using only complete cases or with single imputations (Shafer, 1997). We conducted analyses with 100 imputed datasets (Graham, Olchowski, & Gilreath, 2007), and parameter estimates were pooled using the imputation algorithms in Mplus 7 (Muthén & Muthén, 2015).

Our hypotheses were tested using a path model in Mplus, which included paths from polytraumatization to depressive symptoms, PTSD symptoms, delinquency, and drug distribution. This model controlled for neighborhood risk by including a path from neighborhood risk to the four outcomes and for demographic covariates (sex, age, race/ethnicity, and being from a single parent family) by including paths from these covariates to polytraumatization, neighborhood risk, and the four outcomes. We allowed residual covariances between victimization and neighborhood risk as well as between all outcome variables. This model was fit with a full information maximum likelihood estimator robust to nonnormality (MLR estimator; Muthén & Muthén, 2015). Given that study participants were able to refer other gang members to the study, we accounted for potential nonindependence between participants using TYPE=COMPLEX in Mplus, which applied a sandwich estimator (Muthén & Muthén, 2015) to adjust standard errors. We report standardized linear regression or correlation coefficients and standard errors. For binary predictors, we report STDY estimates from Mplus, which can be interpreted as the change in the outcome variable in standard deviation units when the categorical covariate changes from 0 to 1 (Muthén & Muthén, 2015). Notably, because our path model was saturated (i.e., all paths are included), no fit indices are available; instead, the key statistics of interest in our analysis are the linear regression coefficients and significance values for the paths from polytraumatization to depressive symptoms, PTSD symptoms, delinquency, and drug distribution, as well as the variance explained in each outcome.

Results

Sample Characteristics

Descriptive statistics for study variables are included in Table 1. The majority of participants were Black (60.0%) and indicated that their primary caregiver was a single parent. Trauma exposure was high: On average, adolescent gang members had been exposed to 10 of the 17 traumatic experiences (see Table 2 for types of traumas) and only 9.9% and 2.1% of the sample had experienced no traumas or one trauma, respectively. Of the remaining sample, 2.3% of participants experienced two traumas, 2.0% experienced three, 1.8% experienced four, 1.8% experienced five, and 80.1% experienced six or more traumas. A total of 16.5% of participants exceeded the PHQ-9 scale cutoff of 10 for major depression, whereas over one-third (39.9%) of participants exceeded the Child PTSD scale cutoff of 11 for PTSD.

Table 1.

Sample Characteristics

Variable Range M SD %
Age (years) 14–19 17.39 1.53
Sex
 Boys 57.8
 Girls 42.2
Race/ethnicity
 Black 60.0
 Latino 27.0
 Other 13.0
Single parent family 73.3
Polytraumatizationa 0–17 10.41 5.35
Depressive symptomsb 0–27 4.22 5.48
Met cutoff for major depressionb 16.5
PTSD symptomsc 0–51 10.01 11.06
Met cutoff for PTSDc 39.9
Delinquencyd 0–7 3.08 2.40
Drug dealinge 0–10 4.56 3.80
Neighborhood riskf 0–4 2.57 0.62
Open in a new tab

Note. N = 441.

a

Adapted Exposure to Violence Questionnaire: Summation of dichotomous responses to various traumatic events or forms of direct or vicarious victimization.

b

Patient Health Questionnaire-9: Items were summed, with higher scores indicating more symptoms of depression; scores higher than 10 was considered an indicator of major depression.

c

Child PTSD Symptoms Scale: Items were summed, with higher scores indicating more symptoms of PTSD; a score higher than 11 was considered an indicator of PTSD.

d

Adapted from the Communities that CARE Youth Survey): Summation of dichotomous responses to involvement in delinquent acts.

e

Adapted from the Drug Distribution Roles: Summation of dichotomous responses to involvement in various roles of selling drugs.

f

Neighborhood Risk Assessment: Average of items regarding different exposures to neighborhood disorder and risk.

Table 2.

Prevalence of Each Type of Trauma Among Adolescent Gang Members

Polytraumatization Item %
1. Someone close died 80.5
2. Saw someone slapped, punched, or hit 77.2
3. Saw someone getting beaten up 76.8
4. Someone close very sick or injured 75.5
5. Saw someone pointing a gun at someone else 72.7
6. Saw someone being told they were going to get hurt 68.1
7. Got slapped, punched, or hit 68.1
8. Saw someone being shot at or shot 67.4
9. Seen a serious accident 66.4
10. Been told they were going to get hurt 63.6
11. Saw someone being attacked or stabbed with knife 56.4
12. Had a serious injury or illness or was rushed to hospital 53.7
13. Been in serious accident 49.5
14. Been beaten up 48.4
15. Had to be separated from parent/guardian for more than a few days 42.6
16. Been attacked by dog or other animal 42.4
17. Thought that self or other would get badly hurt during natural disaster 33.9
Open in a new tab

Note. N = 441.

Path Model

We tested a path model, which included paths from polytraumatization to depressive symptoms, PTSD symptoms, delinquency, and drug distribution, and controlled for demographic covariates and neighborhood risk. All path coefficients for a model that constrained paths from polytraumatization to outcomes to 0 (Step 1) and the final model (Step 2) are reported in Table 3, and Figure 1 depicts key model paths. The path model supported our hypothesis such that polytraumatization was associated with mental health symptoms and behavioral problems. Specifically, the associations between polytraumatization and depressive symptoms, β = .25, SE = .04, p < .001; and PTSD symptoms, β = .30, SE = .05, p < .001, were both positive and significant. Polytraumatization also had a significant, positive association with delinquency, β = .40, SE = .05, p < .001; and drug distribution, β = .50, SE = .05, p < .001.

Table 3.

Associations Between Polytraumatization, Neighborhood Risk, and Demographic Covariates and Depressive Symptoms, Posttraumatic Stress Disorder (PTSD) Symptoms, Delinquency, and Drug Distribution Among Adolescent Gang Members

Step 1
Depressive Symptoms
 PTSD Symptoms
 Delinquency
 Drug Distribution
β SE β SE β SE β SE
Boys −.33*** .09 −.35*** .10 .10 .10 .17 .09
Age .09 .05 .04 .05 .08 .05 .09 .05
Latino −.07 .11 −.14 .12 −.20 .13 .04 .14
Other race .09 .13 .07 .14 .01 .14 −.11 .14
Single parent family .16 .11 .14 .11 .29* .12 .20 .13
Neighborhood risk .04 .05 .09 .05 .13* .05 .27*** .05
Polytraumatization -- -- -- -- -- -- --
R2 .05** .02 .06* .02 .06** .02 .11*** .03
Model BIC 13,890.46
Model LL −6,783.87
Change in LL --
Step 2
Depressive Symptoms PTSD Symptoms Delinquency Drug Distribution
β SE β SE β SE β SE
Boys −.38*** .09 −.41*** .09 .02 .10 .07 .09
Age .07 .05 .01 .05 .04 .05 .03 .04
Latino .00 .11 −.05 .11 −.07 .12 .20 .12
Other race .04 .13 .01 .14 −.07 .15 −.22 .14
Single parent family .09 .11 .06 .10 .18 .10 .07 .11
Neighborhood risk .01 .05 .05 .05 .08 .05 .21*** .05
Polytraumatization .25*** .04 .30*** .05 .40*** .05 .50*** .05
R2 .11*** .03 .14*** .03 .20*** .04 .33*** .04
Model BIC 13,740.52
Model LL −6,696.72
Change in LL 87.15***
Open in a new tab

Note. For binary predictors, we report STDY estimates from Mplus, which can be interpreted as the change in the outcome variable in standard deviation units when the categorical covariate changes from 0–1 (Muthén & Muthén, 2015). N = 441. BIC = Bayesian information criterion; LL = log likelihood.

*

p < .05.

**

p < .01.

***

p < .001.

Figure 1.

Figure 1.

Open in a new tab

Path model representing associations between polytraumatization, depressive symptoms, posttraumatic stress disorder (PTSD) symptoms, delinquency, and drug distribution in a sample of adolescent gang members (N = 441). This model controls for demographic covariates sex, age, race/ethnicity, and being from a single parent family; not depicted is neighborhood risk. Standardized coefficients are presented with standard errors. All reported values are averages over 100 multiple imputation datasets. **p < .01. ***p < .001.

In terms of covariates, male gang members had higher polytraumatization scores than did female gang members, β = .20, SE = .09, p = .034. Polytraumatization was also higher for older adolescents, β = .12, SE = .05, p = .010; and for those from single-parent families, β = .27, SE = .12, p = .023. Participants identifying as other races experienced less neighborhood risk than did Black participants, β = −.36, SE = .16, p = .019. Neighborhood risk was correlated with polytraumatization, β = .12, SE = .05, p = .009; and had a positive, significant association with drug distribution, β = .21, SE = .05, p < .001. Male gang members reported fewer depressive symptoms, β = −.38, SE = .09, p < .001; and fewer symptoms of PTSD, β = −.41, SE = .09, p < .001, than female gang members.

A likelihood ratio test showed that the model including polytraumatization was a significant improvement over a model with polytraumatization paths constrained to 0, χ2(4, N = 441) = 174.30, p < .001. The addition of polytraumatization to the model explained 5.9% of the variance in depressive symptoms, 8.2% of the variance in PTSD symptoms, 14.4% of the variance in delinquency, and 22.5% of the variance in drug distribution, over and above the covariates (i.e., sex, age, racial/ethnic background, single-parent vs. two-parent household, and neighborhood risk). This model as a whole explained 11.0% of the variance in depressive symptoms, 13.8% of the variance in PTSD symptoms, 20.2% of the variance in delinquency, and 33.3% of the variance in drug distribution for adolescent gang members. Although gender differences were not a key focus of the current study, a supplementary analysis showed that polytraumatization was a positive, significant predictor of all four outcomes for both male and female adolescent gang members.

Discussion

The present study was a novel examination of the impact of polytraumatization, rather than a discrete traumatic event, on mental health and behavioral outcomes among an understudied population of adolescent gang members. Almost 90% of participants experienced polytraumatization (i.e., two or more traumas), which is significantly higher than what was reported in a national U.S. survey of children aged 2 to 17 years, 66.0% of whom reported polytraumatization (Turner, Finkelhor, & Ormrod, 2010). The analyses confirmed our hypotheses: polytraumatization was associated with both depression and PTSD, even after controlling for demographic variables and neighborhood risk. These results are consistent with previous literature demonstrating associations between traumatization and psychiatric disorders, depression, and PTSD among a national sample of adolescents (Ford, Elhai, Connor, & Frueh, 2010).

Interestingly, only 16.5% of participants exceeded the cutoff for major depression, which is similar to U.S. national averages for adolescents; in 2017, 14.5% of 14–15-year-olds and 17.7% of 16–17-year-olds reported having at least one major depressive episode in the past year (SAMHSA, 2018). In research among gang members in Great Britain, depression rates were lower among gang members than non–gang members (Coid et al., 2013). It is possible that the social support from the gang alleviates depressive symptoms (Harper et al., 2008); however future research should explore depression and potential protective factors among this high-risk population. On the other hand, 39.9% of participants in the present study exceeded the cutoff for displaying PTSD symptoms, which is much higher than the U.S. national average of approximately 5.0% among 13–18 year (National Center for PTSD, 2016). The high rates of PTSD symptoms among adolescent gang members were expected given the high levels of polytraumatization reported.

Our results also demonstrated significant associations between polytraumatization and delinquency and drug distribution. Previous research among gang members has only identified the association between trauma and perpetration of violence, rather than including other delinquent acts (Ross & Arsenault, 2018). Although directionality was not specified in our study, it has been suggested that the relation between trauma and delinquency among adolescent gang members may be reciprocal, as one study demonstrated that trauma exposure increased the risk for delinquency, and gang membership increased the risk for additional traumatization (Begle et al., 2011). Additional longitudinal, prospective research is needed to determine these associations among adolescent gang members.

There were some limitations to the present study. The results presented herein reflect the experiences of adolescent gang members in a single urban city, and findings may not be generalizable to gangs in other cities. Additionally, given that this study used a cross-sectional survey, we cannot determine causality, and future longitudinal studies that can establish temporal precedence are warranted. Responses to mental health symptomology were skewed toward zero. It is possible that gang members may express mental health symptomology in different ways. Researchers may need to develop scales specifically for gang members to reflect their experiences. It is also possible such symptoms were underreported, which may reflect the continuing stigma around mental health, especially for male participants. Additionally, some participants may have had other mental health diagnoses that were not assessed in the present study. Another limitation was the significant amount of missing data, potentially due to the sensitivity of the questions, the secrecy and stigma around gang activities, and potential distrust of the study or study team members.

Further, although we used established measures employed in previous research, the time frames of these measures varied; for example, mental health symptoms were assessed based on the prior 2 weeks, and behavioral outcomes were recalled over 6- and 12-month periods. This may have obscured associations between constructs. Future research examining these issues should work to align the timeframes to better understand the relations between trauma, delinquency, and mental health. In addition, the polytraumatization scale did not account for the time in the participants’ lives during which these events occurred. Traumatic events that occurred during early childhood versus teenage years may have had more or more severe long-term consequences (Turner et al., 2010); thus, future research should consider inquiring about the timeframes of traumatic events.

The present results can be used to inform prevention and intervention efforts. First, gang members are often viewed as delinquents rather than as individuals who need treatment due to exposure to polytraumatization. They should be targeted for early intervention and treatment instead of involvement in the criminal justice system. Social workers or others who encounter adolescent gang members may consider assessing for exposure to multiple traumas (Gustafsson et al., 2009). Trauma-informed care emphasizing strengths, resources, and provision of appropriate services that address the needs of trauma survivors is essential (De La Rue & Espelage, 2014). Additionally, work is needed to address the drivers of marginalization that contribute to experiences of trauma. As noted earlier, Milwaukee is one of the most racially and economically segregated metropolitan areas in the country. Black men and women face numerous health and psychosocial disparities, including disparities in education, employment, and criminal justice involvement. Systems and policy-level interventions aimed at improving environments in which young people live are needed to reduce experiences and consequences of trauma. Finally, early prevention from a holistic perspective may help prevent polytraumatized children from joining gangs. For example, Koffman and colleagues (1999) reported that a program focusing on reducing depression, resisting delinquency, boundary setting, increasing self-esteem and empowerment, promoting leadership skills and social responsibility, and teaching parenting skills was successful in decreasing depression symptoms and delinquent behavior among children at risk for joining gangs.

The present study extends the prior research among adolescent gang members to highlight their high levels of polytraumatization and to demonstrate the associations among polytraumatization and mental health symptoms, delinquency, and drug distribution. Untreated trauma may lead to subsequent delinquency and mental health symptoms. The present research also reinforces the need to examine polytraumatization rather than discrete traumatic events to understand the full burden of trauma faced by adolescent gang members. Similarly, interventions ought to address cumulative life traumatic experiences and consider how such trauma exposure contributes to and is enhanced by gang membership. In sum, gang members represent a population at high risk of experiencing the adverse mental and behavioral effects of cumulative trauma exposure; the development and modification of early interventions are essential to potentially reduce exposure to further trauma, mental health symptoms, and delinquency among adolescent gang members.

Acknowledgments

This research was supported by the National Institute on Drug Abuse (R01-DA027299); the National Institute of Mental Health (T32–0MH19985, K01-MH099956, P30-MH52776); the Population Research Center at The University of Texas at Austin, funded by Eunice Kennedy Shriver National Institute of Child Health and Human Development (P2CHD042849); and the National Institute of Arthritis and Musculoskeletal Skin Diseases (K01-AR073300–09A1).

References

  1. Barnes JC, Boutwell BB, & Fox KA (2012). The effect of gang membership on victimization: A behavioral genetic explanation. Youth Violence and Juvenile Justice, 10, 227–244. 10.1177/1541204011429948 [DOI] [Google Scholar]
  2. Begle AM, Hanson RF, Danielson CK, McCart MR, Ruggiero KJ, Amstadter AB, … Kilpatrick DG (2011). Longitudinal pathways of victimization, substance use, and delinquency: Findings from the National Survey of Adolescents. Addictive Behaviors, 36, 682–689. 10.1016/j.addbeh.2010.12.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bjerregaard B (2010). Gang membership and drug involvement: Untangling the complex relationship. Crime & Delinquency, 56, 3–34. 10.1177/0011128707307217 [DOI] [Google Scholar]
  4. Coid JW, Ullrich S, Keers R, Bebbington P, DeStavola BL, Kallis C, … Donnelly P (2013). Gang membership, violence, and psychiatric morbidity. American Journal of Psychiatry, 170, 985–993. 10.1176/appi.ajp.2013.12091188 [DOI] [PubMed] [Google Scholar]
  5. De La Rue L, & Espelage DL (2014). Family and abuse characteristics of gang-involved, pressured-to-join, and non–gang-involved girls. Psychology of Violence, 4, 253–265. 10.1037/a0035492 [DOI] [Google Scholar]
  6. Decker SH, Pyrooz DC, Sweeten G, & Moule RK (2014). Validating self-nomination in gang research: Assessing differences in gang embeddedness across non-, current, and former gang members. Journal of Quantitative Criminology, 30, 577–598. 10.1007/s10940-014-9215-8 [DOI] [Google Scholar]
  7. Esbensen F-A, Winfree LT Jr., He N, & Taylor TJ (2001). Youth gang and definitional issues: When is a gang a gang, and why does it matter? Crime & Delinquency, 47, 105–130. [Google Scholar]
  8. Foa E, Johnson KM, Feeny NC, & Treadwell KRH (2001). The Child PTSD Symptom Scale: A preliminary examination of its psychometric properties. Journal of Clinical Child Psychology, 30, 376–384. 10.1097/00004703-200112000-00032 [DOI] [PubMed] [Google Scholar]
  9. Ford JD, Elhai JD, Connor DF, & Frueh BC (2010). Poly-victimization and risk of posttraumatic, depressive, and substance use disorders and involvement in delinquency in a national sample of adolescents. Journal of Adolescent Health, 46, 545–552. 10.1016/j.jadohealth.2009.11.212 [DOI] [PubMed] [Google Scholar]
  10. Gilman AB, Hill KG, & Hawkins JD (2014). Long-term consequences of adolescent gang membership for adult functioning. American Journal of Public Health, 104, 938–945. 10.2105/AJPH.2013.301821 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gladstein J, & Slater EJ (1988). Inner city teenagers’ exposure to violence: A prevalence study. Maryland Medical Journal, 37, 951–954. [PubMed] [Google Scholar]
  12. Gordon S (2016, May 10). How Milwaukee’s economic, social disparities correspond with gun violence: Measures of poverty, education, housing align with city’s segregation. National Public Radio. Retrieved from http://www.wpr.org/how-milwaukees-economic-social-disparities-correspond-gun-violence [Google Scholar]
  13. Gover AR, Jennings WG, & Tewksbury R (2009). Adolescent male and female gang members’ experiences with violent victimization, dating violence, and sexual assault. American Journal of Criminal Justice, 34, 103–115. 10.1007/s12103-008-9053-z [DOI] [Google Scholar]
  14. Graham JW, Olchowski AE, & Gilreath TD (2007). How many imputations are really needed? Some practical clarifications of multiple imputation theory. Prevention Science, 8, 206–213. 10.1007/s11121-007-0070-9 [DOI] [PubMed] [Google Scholar]
  15. Gustafsson PE, Nilsson D, & Svedin CG (2009). Polytraumatization and psychological symptoms in children and adolescents. European Child & Adolescent Psychiatry, 18, 274–283. 10.1007/s00787-008-0728-2 [DOI] [PubMed] [Google Scholar]
  16. Harper GW, Davidson J, & Hosek SG (2008). Influence of gang membership on negative affect, substance use, and antisocial behavior among homeless African American male youth. American Journal of Men’s Health, 2, 229–243. 10.1177/1557988307312555 [DOI] [PubMed] [Google Scholar]
  17. Harris TB, Elkins S, Butler A, Robles B, Kwok S, Simpson S, … Sargent AJ (2013). Youth gang members: Psychiatric disorders and substance use. Laws, 2, 392–400. 10.3390/laws2040392 [DOI] [Google Scholar]
  18. Hartinger-Saunders RM, Rittner B, Wieczorek W, Nochajski T, Rine CM, & Welte J (2011). Victimization, psychological distress and subsequent offending among youth. Children and Youth Services Review, 33, 2375–2385. 10.1016/j.childyouth.2011.08.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Jones DJ, Forehand R, Brody G, & Armistead L (2002). Psychosocial adjustment of African American children in single-mother families: A test of three risk models. Journal of Marriage and the Family, 64, 105–115. 10.1111/j.1741-3737.2002.00105.x [DOI] [Google Scholar]
  20. Kerig PK, Chaplo SD, Bennett DC, & Modrowski CA (2016). “Harm as harm”: Gang membership, perpetration trauma, and posttraumatic stress symptoms among youth in the juvenile justice system. Criminal Justice and Behavior, 43, 635–652. 10.1177/0093854815607307 [DOI] [Google Scholar]
  21. Koffman S, Ray A, Berg S, Covington L, Albarran NM, & Vasquez M (2009). Impact of a comprehensive whole child intervention and prevention program among youths at risk of gang involvement and other forms of delinquency. Children and Schools, 31, 239–245. 10.1093/cs/31.4.239 [DOI] [Google Scholar]
  22. Kroenke K, Spitzer RL, & Williams JBW (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16, 606–613. 10.1046/j.1525-1497.2001.016009606.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lin W-H, Cochran JK, & Mieczkowski T (2011). Direct and vicarious violent victimization and juvenile delinquency: An application of general strain theory. Sociological Inquiry, 81(2), 195–222. 10.1111/j.1475-682X.2011.00368.x [DOI] [PubMed] [Google Scholar]
  24. Loeber R, Kalb L, & Huizinga D (2001). Juvenile delinquency and serious injury victimization (NCJ 188676). Washington, DC: U.S: Department of Justice, Office of Juvenile Justice and Delinquency Prevention. [Google Scholar]
  25. Logan JR, & Stults BJ (2011). The persistence of segregation in the metropolis: New findings from the 2010 Census. Retrieved from http://www.s4.brown.edu/us2010
  26. Losen DJ, Hodson C, Keith MA, Morrison K, & Belway S (2015). Are we closing the school discipline gap? Retrieved from the The Center for Civil Rights Remedies website at www.civilrightsproject.ucla.edu/resources/projects/center-for-civil-rights-remedies/school-to-prison-folder/federal-reports/are-we-closing-the-school-discipline-gap/AreWeClosingTheSchoolDisciplineGap_FINAL221.pdf
  27. Maschi T (2006). Unraveling the link between trauma and male delinquency: The cumulative versus differential risk perspectives. Social Work, 51, 59–70. [DOI] [PubMed] [Google Scholar]
  28. Muthén LK, & Muthén BO (2015). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthen & Muthen. [Google Scholar]
  29. National Center for PTSD. (2016). PTSD in children and adolescents. Retrieved from https://www.ptsd.va.gov/professional/treat/specific/ptsd_child_teens.asp#one
  30. Pawasarat J, & Quinn LM (2013). Wisconsin’s mass incarceration of African American males: Workforce challenges for 2013. Milwaukee, WI: Employment and Training Institute, University of Wisconsin-Milwaukee; Retrieved from https://www4.uwm.edu/eti/2013/BlackImprisonment.pdf [Google Scholar]
  31. Petering R (2016). Sexual risk, substance use, mental health, and trauma experiences of gang-involved homeless youth. Journal of Adolescence, 48, 73–81. 10.1016/j.adolescence.2016.01.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ross L, & Arsenault S (2018). Problem analysis in community violence assessment: Revealing early childhood trauma as a driver of youth and gang violence. International Journal of Offender Therapy and Comparative Criminology, 62, 2726–2741. 10.1177/0306624X17734798 [DOI] [PubMed] [Google Scholar]
  33. Shafer JL (1997). Analysis of incomplete multivariate data. New York, NY: Chapman and Hall. [Google Scholar]
  34. Shaffer JN, & Ruback RB (2002). Violent victimization as a risk factor for violent offending among juveniles (NCJ 195737). Washington, DC: U.S: Department of Justice, Office of Juvenile Justice and Delinquency Prevention; Retrieved from https://www.ncjrs.gov/pdffiles1/ojjdp/195737.pdf [Google Scholar]
  35. Sherman SG, & Latkin CA (2002). Drug users’ involvement in the drug economy: Implications for harm reduction and HIV prevention programs. Journal of Urban Health, 79, 266–277. 10.1093/jurban/79.2.266 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Substance Abuse and Mental Health Services Administration. (2004). Communities That Care Youth Survey (CTC020). Retrieved from https://www.communitiesthatcare.net/
  37. Substance Abuse and Mental Health Services Administration. (2018). Results from the 2017 National Survey on Drug Use and Health: Detailed tables. Rockville, MD: U.S: Department of Health & Human Services; Retrieved from https://www.samhsa.gov/data/sites/default/files/cbhsq-reports/NSDUHDetailedTabs2017/NSDUHDetailedTabs2017.htm#tab9-6B [Google Scholar]
  38. Thornberry TP, Krohn MD, Lizotte AJ, & Smith CA (2003). Gangs and delinquency in developmental perspective. New York, NY: Cambridge university Press. [Google Scholar]
  39. Turner HA, Finkelhor D, & Ormrod R (2010). Poly-Victimization in a national sample of children and youth. American Journal of Preventive Medicine, 38, 323–330. 10.1016/j.amepre.2009.11.012 [DOI] [PubMed] [Google Scholar]

RESOURCES