Children Exposed to Violence

The prevalence of childhood exposure to violence (CEV) in the general population of youth in the United States is measured through a variety of sources [discussed in Data Sources and Definitional Challenges]. The results from these sources show a high prevalence rate of CEV in youth across the country. For more information about specific types of violence, see the Model Programs Guide literature reviews on Bullying, Teen Dating Violence, and Gun Violence and Youth.

Results From the National Survey of Children Exposed to Violence

Data from the three National Survey of Children Exposed to Violence (NatSCEV) waves found that children's exposure to violence is common and that there was very little change from one wave to the next (2008, 2011, 2014) [Finkelhor et al., 2009; Finkelhor et al., 2015a; Finkelhor et al., 2015b].

The most recent NatSCEV included a representative sample of U.S. telephone numbers from Aug. 28, 2013, to April 30, 2014. Information was obtained on 4,000 children 0 to 17 years old, with information about exposure to violence, crime, and abuse provided by youths ages 10 to 17 and by caregivers for children 0 to 9 years old (Finkelhor et al., 2015a). Results include the following lifetime exposure to violence among 14- to 17-year-olds (Finkelhor et al., 2015a):

Experiencing Any Physical Assaults 63.5 percent

• Assault with a weapon: 14.3 percent
• Assault with injury: 22.4 percent
• Assault by an adult: 21.6 percent
• Nonsexual genital assault: 19.8 percent
• Assault by gang or group: 3.1 percent
• Dating violence: 5.2 percent

Experiencing Any Sexual Offense: 21.7 percent

• Sexual assault: 10.2 percent
• Completed rape: 4.5 percent
• Sexual harassment: 20.5 percent

Experiencing Any Maltreatment: 38.1 percent

• Physical abuse: 18.1 percent
• Emotional abuse: 23.9 percent
• Neglect: 18.4 percent

Witnessing Any Violence: 68.1 percent

• Family assault: 32 percent
• Assault in community: 57.9 percent
• Exposure to shooting: 13.4 percent
• Exposure to war: 3.2 percent

The 2014 NatSCEV also identified youths experiencing large numbers of exposures. The results showed that 10 percent of the sample of all youths ages 0–17 had six or more direct exposures of violence, crime, or abuse in just one single year, indicating a highly vulnerable group of children referred to as polyvictims (Finkelhor et al., 2015a). In total, 40.9 percent of the sample had more than 1 direct experience of violence, crime, or abuse; 10.1 percent had 6 or more, and 1.2 percent had 10 or more experiences in just one year.

Results From the Youth Risk Behavior Survey

The results from the 2019 Youth Risk Behavior Survey (YRBS) found that, in the past year, 8.2 percent of high school students had experienced physical dating violence, 8.2 percent had experienced sexual dating violence, 10.8 percent had experienced sexual violence by anyone, 19.5 percent had been bullied on school property, 15.7 percent had been electronically bullied, and 7.4 percent were threatened or injured with a weapon on school property (Basile et al., 2020). Also, in the 12 months before the 2019 survey, 8.7 percent of students reported that they did not go to school because they felt unsafe at school or on their way to or from school. The percentage of students reporting this fear in 2019 nearly doubled from 4.4 percent in 1993.

Results From the National Survey of Children Health

The 2020 National Survey of Children's Health found that 5.3 percent of its sample of children ages 0–17 had seen or heard parents or another adult slapped, hit, kicked, or punched in home, and 4.1 percent had been a victim of or witness to violence in their community. The 2020 survey also found that among 6- to 17-year-olds, 35.1 percent had been bullied, picked on, or excluded by other children in the past 12 months: 25.6 reported being victimized 1–2 times in the past 12 months, 5.2 percent reported victimization 1–2 times per month, 2.5 percent reported victimization 1–2 times per week, and 1.7 percent reported victimization almost every day (Child and Adolescent Health Measurement Initiative, n.d.).

Subgroup Differences

Prevalence of CEV can vary by demographic, geographic, familial, community, or societal factors. Below are findings from some of the studies that have examined prevalence of CEV by various subgroups.

• Race and Ethnicity. Some researchers have found that certain racial and ethnic groups have increased risk of exposure to violence (Sacks and Murphey, 2018; U.S. Department of Health and Human Services, 2021; Yi, Edwards, and Wildeman, 2020; Zimmerman and Messner, 2013; Zimmerman and Posick, 2016). These findings vary across types of violence and are sometimes mixed, likely owing to differences in violence definitions and racial/ethnic categories. For example, the 2016 National Survey of Children's Health found that Black children were more likely than white children to see or hear parents or adults slap, hit, kick, or punch one another in the home. The survey also found that both Black and Hispanic children were more likely than white children to be victims of or witnesses to violence in their neighborhoods (Sacks and Murphey, 2018). Similarly, analysis of data from the Project on Human Development in Chicago Neighborhoods found that the odds of being exposed to violence were 74 percent and 112 percent higher for Hispanic and Black youths, respectively, than for white youths (Zimmerman and Messner, 2013). They defined exposure to violence as the extent to which participants had experienced one or more of the following in the past 12 months: a) seeing someone shoved, kicked, or punched; b) seeing someone attacked with a weapon; c) seeing someone shot at; d) seeing someone shot; e) seeing someone hurt in a serious accident; f) seeing someone chased with the intention of injury; g) seeing someone threatened; h) seeing someone killed; and i) hearing a gunshot.

However, analysis of 2019 YRBS data found that white youths were more likely to be bullied on school property or electronically bullied than Black or Hispanic youths were (Basile et al., 2020). Although not included in many studies because of their small numbers, American Indian and Alaska Native youths may also to be at increased risk of exposure to violence. For example, official data from the U.S. Department of Health and Human Services (2021) showed that, out of 1,000 American Indian/Alaska Native children, 14.8 were substantiated or indicated victims of child maltreatment, compared with rates of 13.8 for Black children, 8.1 for Hispanic children, 7.8 for white children, and 1.7 for Asian children.

• Gender. When examining gender differences in CEV, some studies find that girls are more at risk, while others find that boys are at greater risk. This is likely due to various definitions of violence. Overall, researchers generally find that girls are at higher risk for dating violence and sexual violence victimization (Basile et al., 2020; Finkelhor et al., 2009; Finkelhor et al., 2015a; Tharp et al., 2017), while boys are at higher risk of gun violence and community violence victimization and exposure (Bottiani et al., 2021; Estrada et al., 2021; Finkelhor et al., 2009; Lambert et al., 2005; Zona and Milan, 2011). The child abuse literature also shows differences by gender: girls have a higher child abuse and neglect victimization rate than boys (9.4 per 1,000, compared with 8.4 per 1,000), but boys are more likely than girls to suffer child abuse fatalities (U.S. Department of Health and Human Services, 2021).

• Sexual Orientation. Lesbian, gay, bisexual, transgender, and questioning (LGBTQ) youths are disproportionately affected by multiple forms of violence, including peer violence, physical abuse, sexual abuse, dating violence, and sexual assault (Dank et al., 2014; McCormick, Scheyd, and Terrazas, 2018). A study using data from the 2019 YRBS found that, compared with heterosexual students, LGB students and students unsure of their sexual identity had the highest-prevalence estimates across all five of the examined violence victimization types: 1) physical dating violence, 2) sexual dating violence, 3) sexual violence by anyone, 4) bullied on school property, and 5) electronic bullying victimization (Basile et al., 2020). They also were more likely to not go to school because they felt unsafe there or on their way to or from school. A meta-analysis (a method of combining data and results across studies to determine an overall effect) of school-based studies on the likelihood of childhood abuse found that youths who experience same-sex attractions; self-label as gay, lesbian, or bisexual; or engage in same-sex sexual activity were more likely to be exposed to several types of violence. For example, they were on average 3.8, 1.2, 1.7, and 2.4 times more likely to experience sexual abuse, parental physical abuse, or assault at school, or to miss school through fear, respectively, compared with youths who did not (Friedman et al., 2011).

• Geographic Region. Some research suggests that children living in certain regions of the United States are more likely to experience violence, compared with children in other regions. However, this varies by data collection methods and definitions. The 2016 National Survey of Children's Health found that children from Washington, DC; Hawaii; and Nevada were statistically significantly more likely to be victims or witnesses of violence in their neighborhoods, compared with the national average, and that children from Arizona, Arkansas, Hawaii, Mississippi, and New Mexico were statistically significantly more likely to have seen or heard parents or other adults slap, hit, kick, or punch one another in the home, compared with the national average (Sacks and Murphey, 2018). Also, reported rates of child abuse and neglect vary substantially among states: the lowest rates were reported in Pennsylvania (1.8 per 1,000), North Carolina (2.4 per 1,000), Washington (2.5 per 1,000), and New Jersey (2.6 per 1,000), while the highest rates were recorded in Massachusetts (18.5 per 1,000), West Virginia (18.7 per 1,000), and Kentucky (20.1 per 1,000) [U.S. Department of Health and Human Services, 2021]. Finally, the National Survey of Children's Health found that the percentage of rural students who reported being bullied at school was higher (28 percent) than among students enrolled in schools in other locales (22 percent each for those enrolled in schools in cities and in towns, and 21 percent for those enrolled in schools in suburban areas) [NCES, 2022].

Researchers have found that youths in the juvenile justice system have higher levels than youths in the general population of CEV (Baglivio et al., 2014; Malvaso et al., 2021; Wood et al., 2002). To measure prevalence of CEV in juvenile justice–involved populations, researchers use findings from several tools, including the Childhood Trauma Questionnaire (CTQ), the five-item traumatic experiences subscale of the Massachusetts Youth Screening Instrument, the Exposure to Violence Inventory, the Positive Achievement Change Tool (PACT), and the National Stressful Events Survey PTSD Short Scale (Baglivio et al., 2014; Farina et al., 2018; Hoskins et al., 2019; Malvaso et al., 2021; Moore, 2021; Tsang, 2018). For example, the CTQ includes questions such as "Prior to the age of 18, did you have a traumatic sexual experience (raped, molested, etc.)?" and "Prior to the age of 18, were you the victim of violence (child abuse, mugged or assaulted—other than sexual)?" Some questions included in the PACT are "How often did a parent, stepparent, or adult living in your home push, grab, slap, or throw something at you?" "How often did a parent, stepparent, or adult living in your home hit you so hard that you had marks or were injured?" "Was your mother (or stepmother) sometimes, often, or very often kicked, bitten, hit with a fist, or hit with something hard?" (Baglivio et al., 2014).

A study of ACEs in a sample of more than 64,000 youths involved with the Florida Department of Juvenile Justice (FDJJ), which were measured using the PACT, found that more than 80 percent of the sample had been exposed to family violence (Baglivio et al., 2014). FDJJ data also revealed that girls were more likely than boys to have been exposed to each type of violence (see Figure 1). However, other studies have found that, although justice-involved youth girls reported higher levels of exposure to sexual assault and interpersonal victimization, boys reported higher rates of witnessing violence (Cauffman et al., 1998; Wood et al., 2002).

The Pathways to Desistence Study [discussed in Data Sources and Definitional Challenges] measured violence victimization and witnessing of violence by using a modified version of the Exposure to Violence Inventory It has provided data for several articles examining the consequences and correlates of exposure to violence (Baskin–Sommers and Baskin, 2016; Davis et al., 2017; Moore, 2021; Tsang, 2018). Some publications have also used these data to measure prevalence rates. For example, Davis and colleagues (2017) measured the prevalence of victimization in a sample of 1,100 youths ages 14 to 17 at baseline with early onset problem behaviors, using the six-item victimization subscale of the Exposure to Violence Inventory. Example items were "have you been sexually assaulted?"; "have you been chased when you thought the person chasing you would hurt you in the past 12 months?"; and “have you been attacked with a weapon?" The mean level of victimization for youths in the sample was 3.8, out of a high score of 6.0.

Studies also have examined this issue internationally. A systematic review of 124 studies, published between 2000 and 2019, examined justice-involved young people across 13 countries. The review estimated odds ratios related to specific types of violence, comparing youths who had offended with nonoffending youths: the youth who had offended were 8.3 times as likely to have experienced emotional abuse, 3.4 times as likely to have experienced domestic violence, 2.5 times as likely to have experienced physical abuse, and 1.7 times as likely to have experienced sexual abuse (Malvaso et al., 2021).

Risk Factors for Exposure to Violence

Risk factors are personal traits, characteristics, or conditions at the psychological, biological, family, peer, community, or cultural level that precede and are associated with the likelihood of negative outcomes, such as delinquency, substance misuse, mental and physical health problems, violence, relationship challenges, and educational and employment struggles (Murray and Farrington, 2010; SAMHSA, 2019). There are several identified risk factors for exposure to violence as a child. Much of this research either examines risk factors for specific types of violence like child abuse, bullying, or community violence (e.g., Assink et al., 2019; Mulder et al., 2018; Cook et al., 2010; Zimmerman and Posick, 2016; Zych, Ortega–Ruiz, and Del Rey, 2015) or risk factors for adversities and potentially traumatic experiences more generally, which include exposure to violence (e.g., Copeland et al., 2007; Ports et al., 2021; Walsh et al., 2019). Given the wide scope of experiences that are considered violence, the influence of some of these risk factors varies across studies, types of violence, and risk characteristics. This section briefly summarizes the research literature on risk factors for CEV specifically. For more information on risk factors, see the Model Programs Guide literature review on Risk Factors for Delinquency.

Individual-Level Factors

Researchers have identified several individual-level risk factors related to exposure to violence, including exposure to community violence, dating violence, bullying, and child abuse.

Depression and Other Internalizing Conditions. Internalizing conditions can be risk factors for several types of violence exposure. For example, depressive symptoms are strongly related to teen dating violence victimization in both boys and girls (Johnson et al., 2014; Goncy, Farrell, and Sullivan, 2016). A review of 61 empirical studies that investigated risk factors for violence in romantic relationships of 12- to 24-year-old girls and young women found that overall negative mental health—such as depressive symptoms, suicidal behavior, and low self-esteem—was associated with and increased risk of girls becoming victims of physical or sexual violence (Vézina and Hébert, 2007). Depression can also influence exposure to other types of violence. A study of 582 adolescents (mentioned above) found that anxious and depressive symptoms in sixth grade predicted community violence victimization among girls in later grades. Also, among aggressive boys, depressive symptoms exacerbated their risk for witnessing community violence (Lambert et al., 2005). Finally, a meta-analysis of 152 articles published between 1970 and 2006 that examined predictors of bullying found that having an internalizing problem, such as being withdrawn, depressive, anxious or avoidant, was one of the most influential predictors of bullying victimization (Cook et al., 2010).

Aggression. Several studies have found that aggression in early childhood predicts exposure to community violence during adolescence (Boyd et al., 2003; Lambert et al., 2005). For example, a study of urban middle school students found that aggressive behavior in sixth grade was associated with concurrent witnessing of community violence and later exposure to community violence in grades 7 and 9 (Lambert et al., 2005).

Self-Control. The Project on Human Development in Chicago Neighborhoods (the PHDCN) [discussed in Data Sources and Definitional Challenges] revealed that youth respondents with lower levels of self-control had higher levels of exposure to community violence, which included being shoved, kicked, or punched; being attacked with a weapon; being shot; being shot at; being chased with intent for injury; and being threatened (Gibson, 2012; Zimmerman and Posick, 2016).

Street Efficacy. Street efficacy refers to the ability to avoid getting into fights in the neighborhood, to do things safely in the neighborhood with their friends, to avoid gangs in the neighborhood, to feel safe in the neighborhood, and to go places within a few blocks of home safely (Zimmerman and Posick, 2016). The PHDCN revealed that youth respondents with lower levels of street efficacy also had higher levels of exposure to violence.

Neighborhood Characteristics

Researchers have examined how community-level risk factors, such as economic disadvantage, lack of services, and immigrant population, may predict childhood exposure to violence (Coulton et al., 2007; Gibson, 2012; Molnar et al., 2003; Zimmerman and Posick, 2016). One study that used data from the PHDCN, found that youths living in neighborhoods with higher levels of disadvantage and a lack of youth services were more likely to be exposed to violence. They used 15 items to measure CEV in the home, school, or community, such as being attacked with a weapon, being shot at, being chased with intent for injury, finding a dead body, and hearing a gunshot (Zimmerman and Posick, 2016). Another study using the PHDCN dataset examined how neighborhood factors may influence parent-to-child physical aggression (Molnar et al., 2003). The researchers found that, while family differences account for much more variation than neighborhood differences, some neighborhood factors were influential: higher levels of concentrated disadvantage and community violence resulted in higher levels of parent-to-child physical aggression; higher levels of immigrant concentration resulted in lower levels of parent-to-child physical aggression (Molnar et al., 2003). Finally, another study using the PHDCN dataset found that although youths with low self-control had an increased likelihood of violent victimization (see above), this finding was attenuated by levels of neighborhood concentrated disadvantage: youths with low self-control were more likely to report being victims of violence in more-affluent neighborhoods, but there was no statistically significant association between low self-control and violent victimization for youths in the most-disadvantaged neighborhoods (Gibson, 2012).

In addition to findings from PHDCN data, several other sources have identified relationships between neighborhood characteristics and exposure to violence. For example, a study of violence exposure among rural youths in Ohio found that higher levels of poverty were significantly related to higher levels of direct exposure to violence in school (Carlson, 2006). Also, a systematic review of 25 studies using various data sources examining the influence of neighborhoods on child maltreatment concluded that living in disadvantaged neighborhoods increased the likelihood of child maltreatment cases (Coulton et al., 2007), although the authors did not believe the relationship between child maltreatment and neighborhood characteristics was well understood based on the data they examined.

Peer and Family Factors

There is a large body of evidence examining family-related risk factors for child maltreatment (Assink et al., 2019; Mulder et al., 2018). Meta-analytic research has concluded that the strongest predictors of child neglect can be found in parental characteristics, such as a history of antisocial/criminal offending, a history of mental/psychiatric problems, experiences of abuse in the parent’s own childhood, and a low educational level (Mulder et al., 2018). Another meta-analysis of 84 studies examining intergenerational continuity in child maltreatment found that the odds of child maltreatment were almost three times as high for children of parents who experienced maltreatment themselves (compared with children whose parents did not have that history) [Assink et al., 2019].

Family factors have also been found to influence bullying victimization. A meta-analysis of 70 studies found that bullying victims were more likely to have experienced maladaptive parenting and to have overprotective parents, compared with youths who were not bullied. Victims also were less likely to 1) have authoritative parents, 2) have good parent–child communication, 3) have parents who were involved and supportive, 4) receive supervision, and 5) have warm and affective parents (Lereya, Samara, and Wolke, 2013).

Finally, analysis of data from the PHDCN found that respondents with higher levels of unstructured socializing were more likely to be exposed to community violence (Zimmerman and Posick, 2016). To measure unstructured socializing, respondents were asked how often they ride around in a car for fun, hang out with friends, go to parties, go out at night, and go out with a date. When analyzing the PHDCN, Gibson (2012) also found an association between violent victimization and unstructured time spent with peers: youths who reported spending more time in unstructured activities with peers were statistically significantly more likely to report being victims of violence.

Prior Violence Exposure

Being exposed to one type of violence predicts continued exposure to that type of violence and exposure to other types (Assink et al., 2019; Brunson and Miller, 2009; Goncy, Farrell, and Sullivan, 2016; Zimmerman and Posick, 2016). For example, a study of middle school students found that witnessing community violence in sixth grade was a significant predictor of witnessing community violence in later grades (Lambert et al., 2005). Analysis of data from the PHDCN found that prior exposure to violence was one of the strongest correlates of subsequent exposure to violence (Zimmerman and Posick, 2016). Similarly, in examining six victimization types (assaults and bullying, sexual victimization, maltreatment by a caregiver, property victimization, witnessing violence, and other violence exposure), the NatSCEV found that each exposure type increased the likelihood of another exposure type (Finkelhor et al., 2015a; Finkelhor et al., 2015b).

This effect also is found in the bullying literature. For example, a meta-analysis of cyberbullying research found that the strongest predictor of cyberbullying victimization was traditional bullying victimization, indicating that youths who are bullied face-to-face are also likely to be bullied online (Kowalski et al., 2014). Further, a meta-analysis of 70 published studies found that bullying victims were more likely to have experienced child abuse (Lereya, Samara, and Wolke, 2013). Finally, a synthesis of 72 studies examining risk factors for child sexual abuse found that the strongest predictor of victimization was prior victimization (Assink et al., 2019).

Reciprocal Nature of Risk Factors and Consequences

The complicated nature of the research on childhood exposure reveals several reciprocal relationships between risk factors and consequences. For example, while some studies find depression to be a risk factor for exposure to violence (Copeland et al., 2007), other studies identify depression as a consequence of exposure to violence (e.g., Schoeler et al., 2018; Ttofi et al., 2011; Zona and Milan, 2011). Some researchers have found that the relationship between gang membership and CEV is reciprocal. For example, an examination of data from 2,450 girls in the Pittsburgh (PA) Girls Study found that there were significant paths from gang involvement at ages 11–12 to peer victimization at age 13, from peer victimization at age 13 to gang involvement at age 14, and, finally, from gang involvement at age 14 to peer victimization at age 15 (Gilman et al., 2017). Other studies discuss associations between CEV and other variables without attempting to identify directionality (Robert and Klein, 2003).

For more detail on risk factors for specific times of violence exposure, go to the Model Programs Guide literature reviews on Bullying, Teen Dating Violence, and Gun Violence and Youth.

Research has demonstrated that childhood exposure to violence (CEV) in any stage of childhood can have deleterious effects, even during infancy (Bogat et al., 2006; Carlson et al., 2019; National Scientific Council on the Developing Child, 2012). Although much of this research examines ACEs, which combine varies types of adverse experiences, including exposure to violence (Felitti et al., 1998; Felitti, 2002; Gilbert et al., 2015; Mersky, Topitzes, and Reynolds, 2013; Rasmussen et al., 2020; Sonu, Post, and Feinglass, 2019), some studies isolate violence victimization and exposure specifically. For example, CEV is related to increased likelihood of experiencing the following:

• Structural changes in brain neurocircuitry, such as reduced hippocampal volume, increased amygdala volume, and reduced activation in the dorsal anterior cingulate cortex and frontal pole (Weissman et al., 2020; van Rooij et al., 2020).
• Trauma symptoms and posttraumatic stress disorder (Cecil et al., 2014; Copeland et al., 2007; Fowler et al., 2009; Hong et al., 2021; Ozer and McDonald, 2006), even among infants (Bogat et al., 2006).
• Depression, anxiety, suicidal ideation and behaviors, and other internalizing mental health problems (Cecil et al., 2014; Copeland et al., 2007; Holt et al., 2015; Polanin et al., 2021; Rosario et al., 2008; Zona and Milan, 2011).
• Externalizing problems, such as anger control, behavior problems, and aggression (Cecil et al., 2014; Holmes et al., 2015; Moore, 2021; Veira et al., 2014).
• Physical health problems, such as headaches, chronic pain, allergies, and asthma (Kuhlman, Howell, and Graham–Berman, 2012; McLaughlin et al., 2016).
• Hospitalizations (Orr et al., 2020).
• Risky sexual behavior during adolescence and teen pregnancy (Garwood et al., 2015; Hong et al., 2021; James et al., 2018).
• Substance misuse (Afifi et al., 2020; James et al., 2018).
• Poor educational outcomes and academic achievement (Burdick–Will, 2018; Mitchell, Becker–Blease, and Soicher, 2021; Polanin et al., 2021; Romano et al., 2015).
• Cruelty to animals (Currie, 2006; DeGrue and DeLillo, 2009).
• Less attachment to parents during adolescence (Sousa et al., 2011).
• Delinquency, violence, gun carrying, and gang involvement (see below).

Further, similar to the research on ACEs (Felitti, 2002; Garrido, Weiler, and Taussig, 2018; Johnson, 2018; Mersky, Topitzes, and Reynolds, 2013; Zarse et al., 2019), much of the research on CEV finds that cumulative exposures are associated with cumulative adverse effects across multiple domains (e.g., Copeland et al., 2007; James et al., 2018). In other words, greater exposure to violence in a youth’s history can lead to worsening outcomes. Some studies call this a “dose-response relation” (Graham–Bermann and Seng, 2005). For example, an examination of 2,684 adolescents from the Fragile Families and Child Well-Being Study found that each additional point on the community violence scale was associated with 8 percent higher odds of risky sexual behavior, while each additional point on the family violence scale was associated with 20 percent higher odds of substance use (James et al., 2018). A smaller body of research examines whether high levels of exposure have less effect on negative outcomes, as a result of desensitization (Gaylord–Harden, Cunningham, and Zelenick, 2011).

Also, research has shown that different types of violence exposure may result in different consequences (e.g., Cecil et al., 2014; Estrada et al., 2021). For example, a study of the impact of both childhood maltreatment and community violence concluded that, while childhood maltreatment was associated with increased likelihood of symptoms across a broad range of mental health domains (internalizing, externalizing, and trauma), the impact of community violence was more constrained (affecting only externalizing and trauma symptoms), suggesting that these two types of exposure to violence differentially influence mental health functioning (Cecil et al., 2014). Similarly, another study that categorized more than 28,000 individuals into different profiles based on their rates of exposure to child maltreatment and community violence found several differences in outcomes. For example, they found that individuals with high levels of child maltreatment and low levels of exposure to community violence reported increased rates of depression and other mental health diagnosis while individuals with elevated exposure to community violence and low levels of children maltreatment had higher rates of serious disease (Estrada et al., 2021).

Impact of CEV on Delinquency, Violence, Gun Carrying, and Gang Involvement

Research has examined the relationships between CEV and delinquency, violent behaviors, gang involvement, and juvenile justice system involvement. For example, using the National Longitudinal Study of Adolescent to Adult Health (Add Health) to examine exposure to violence among youths in grades 7–12 on subsequent offending, Farrell and Zimmerman (2018) found that exposure to violence predicted youths’ violent offending 1 year later, 5–7 years later, and 12 years after exposure to violence, and also their current level of violent offending. Similarly, a small study of 71 Chinese American urban adolescents found that exposure to violence predicted perpetration of violence (Ozer and McDonald, 2006). This study assessed exposure to violence using eight items from the Direct Exposure subscale of the Children's Report of Exposure to Community Violence, which indicated the frequency of experiencing violent events, such as being beaten up and seeing others physically assaulted, shot, stabbed, or killed, and measured perpetration of violence using measures from the Add Health study such as fighting and injuring another person in a fight.

Another study of 349 Black youths ages 9–15 living in public housing communities in an eastern metropolis found that youths who were involved in gangs were also significantly more likely to have been victims of both drug-related and non–drug-related violence and to have witnessed violence against others and/or the use of deadly force or death (Li et al., 2002). Finally, a small study of adolescent mothers and their children found that children who had higher levels of violence exposure in their homes, neighborhoods, or schools before age 10 (e.g., “seeing someone else being beaten,” “being told by someone that they were going to hurt you”) were more likely to subsequently exhibit delinquent and violent behaviors (Weaver, Brokowski, and Whitman, 2008).

Researchers also have isolated gun carrying and gun violence as consequences of CEV. Analysis of data from the Mobile Youth Survey [discussed in Data Sources and Definitional Challenges] found that youths who had experienced violent victimization were more likely to initiate gun carrying when they were older (Spano and Bolland, 2013). Youths were considered to have experienced violent victimization if they had been threatened with a knife or gun during the 90 days before the first survey was given, cut badly enough to see a doctor in the past year, or shot at during the past year. Also, the Flint Adolescent Study found that exposure to violence indirectly increased the risk for firearm carriage by decreasing youths’ beliefs that they would graduate from high school and continue education after high school (Lee et al., 2020). This study measured exposure to violence as the frequency of violence victimization and violence observation in the 12 months before the first survey was given. Finally a review of literature related to gun violence and youth found that exposure to violence and availability of firearms were two of the most influential risk factors for engaging in youth gun violence (Development Services Group, Inc., 2016).

Differences by Gender

Some studies have examined how the relationship between CEV and later outcomes varies by demographic factors. In terms of gender, studies have found mixed results. For example, a study of the role of intervening processing between exposure to community violence and internalizing symptoms in a sample of middle school students found that female youths who had been exposed to violence were more likely to internalize their symptoms and were less likely than male youths to cope in positive manners, leading to other individual risk factors (Rosario et al., 2008). Another study examining gender differences in the longitudinal impact of exposure to violence in a sample of urban youth found similar results: girls experiencing violence were more likely than boys experiencing violence to demonstrate dissociative internalizing and externalizing symptoms (Zona and Milan, 2011), which are disorders often characterized by escaping reality in ways that are involuntary and unhealthy and that cause problems with functioning in everyday life. However, some studies have found that gender does not moderate this relationship (Gaylord–Harden, Cunningham, and Zelencik, 2011).

Impact of Different Types of CEV

Some researchers examine whether specific forms of exposure to violence and victimization increase the risk of specific forms of offending and other negative outcomes. A study using data from the Pathways to Desistence project found that violence victimization predicted early behavior problems but that being a witness to violence did not (Moore, 2021). Another study found that both violence victimization and witnessing violence led to internalizing disorders, but only violence victimization was related to conduct problems (Ward et al., 2007). Further, a meta-analysis of studies examining the effect of exposure to violence on mental health outcomes of children and adolescents found that witnessing community violence had a greater effect than hearing about violence on externalizing problems (which are disruptive, harmful, or problem behaviors that are directed to persons and/or thins), but both types of exposure (witnessing and hearing about violence) made equal impacts on internalizing problems, such as depression and anxiety (Fowler et al., 2009).

Childhood exposure to violence (CEV) continues to predict negative outcomes for youth involved in the juvenile justice system. Much of this research is from analyses of Pathways to Desistence data [discussed in Data Sources and Definitional Challenges].

Increased Levels of Delinquency

Examination of 1,354 serious delinquent adolescents from the Pathways to Desistance study found that violence victimization or witnessing violence was uniquely associated with increased self-reported delinquent behavior (Tsang, 2018). Another study using the same dataset found that youths who had more chronic and direct exposure to community violence were more likely to remain involved in violent criminal behavior, independent of other known risk factors (Baskin and Sommers, 2013). Finally, DeLisi, Piquero, and Cardwell (2016) used this same dataset and found that greater exposure to violence was one of the only statistically significant predictors of being charged with some type of homicide as an adolescent.

Further, studies have examined the effect of CEV on recidivism in juvenile justice populations in several states. Most of this evidence comes from studies of ACEs overall (e.g., Weber and Lynch, 2021; Wolff et al., 2020), but there are some studies looking specifically at violence. For example, a small study of 122 girls in secure custody in Virginia found that girls who had witnessed neighborhood violence were more likely than girls who had not witnessed neighborhood violence to be rearrested (Chauhan, Reppucci, and Turkheimer, 2009). Closer examination showed differences by race: witnessing neighborhood violence was associated with recidivism for Black girls, while parental physical abuse was associated with recidivism for white girls.

As with the general youth population, studies have documented the influence of CEV on gun carrying among juvenile justice system–involved youth (Beardslee et al., 2018; Reid et al., 2017b). For example, analyses of more than 1,000 male youths who committed serious offenses at ages 14 to 19 who participated in the Pathways to Desistence study found that exposure to violence was statistically significantly related to gun carrying (Reid et al., 2017b). However, another study of this same sample found that these male youths were more likely to carry a gun following exposure to gun violence, but not after exposure to nongun violence (Beardslee et al., 2018). Yet, exposure to nongun violence does have an effect in some studies. An analysis of a smaller sample of these same male youths with a history of shooting found that both witnessing nongun violence and being victimized by nongun violence statistically significantly increased the likelihood of their engaging in gun violence (Rowan et al., 2019).

Substance Use and Psychiatric Outcomes

Analysis of data from the Pathways to Desistance study found that adolescents who committed serious offenses who were exposed to community violence during adolescence and emerging adulthood were statistically significantly more likely to have internalizing symptoms, compared with adolescent who committed serious offenses who were not exposed (Donnelly and Holzer, 2018). Internalizing problems included measures of obsession–compulsion, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism. Another study of first-time justice-involved youths found that those who experienced more types of childhood abuse (e.g., emotional, physical, or sexual abuse) were more likely to use alcohol and experience negative consequences for using alcohol or marijuana (e.g., “The quality of my schoolwork has suffered because of my marijuana use”), and to exhibit internalizing (e.g., atypicality, anxiety, depression), externalizing (e.g., aggression, hyperactivity), and trauma symptoms (Folk et al., 2021).

Several studies have examined the relationship between CEV and psychopathy among juveniles who offend. For example, research using data from the Pathways to Desistance study found that psychopathic traits mediated the relationship between exposure to violence and violent offending (Baskin–Sommers and Baskin, 2016). In other words, youths with psychopathic traits who are exposed to violence were more likely to commit violent offenses than youths who are exposed to violence without psychopathic traits.

Future Orientation

Future orientation is a multicomponent process through which individuals orient themselves to anticipated future events and objectives, which results in individuals considering future consequences of their actions (Nurmi, 1991; Steinberg et al., 2009). Several studies have found that youths with higher levels of future orientation are less likely to experience undesirable outcomes such as externalizing behaviors, violence, delinquency, substance use, and engaging in sex under the influence of substances (Cedeno et al., 2010; Robbins and Bryan, 2004; So et al., 2018; Stoddard et al., 2011). However, exposure to violence has been found to negatively affect future orientation. Analysis of data from 1,354 youths who committed serious offenses in the Pathways to Desistence study found that higher exposure to violence was associated with lower levels of future orientation at age 15 and with suppressed development of future orientation from age 15 to 25 (Monahan et al., 2015).

Human Trafficking

A study using the Florida Department of Juvenile Justice data examined the difference between youths with a history of human trafficking abuse reports and those without those reports (Reid et al., 2017a). This study examined six ACEs (emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, and family violence). Researchers found that each of the ACEs was more prevalent among youths who had trafficking reports. Sexual abuse was the strongest predictor of human trafficking for both boys and girls: the odds of human trafficking was 2.5 times as great for girls who experienced sexual abuse (compared with girls who did not experience sexual abuse) and 8.21 times as great for boys (compared with boys who did not experience sexual abuse). The researchers concluded that maltreated youths were more susceptible to exploitation in human trafficking (Reid et al., 2017a).

Protective factors (sometimes referred to as “buffers”) are characteristics or conditions of the child, family, and/or broader social environment that reduce the likelihood of a) exposures to violence and adversity or b) negative child and youth development behaviors and outcomes (including delinquency and adult offending) if the child is exposed to violence and adversity (Vanderbilt–Adriance and Shaw, 2008; Development Services Group, Inc., 2015). Resilience is defined as a characteristic of those who master normative developmental tasks despite their experiences of significant adversity (Jaffee et al., 2007). This section concentrates on protective factors that lessen the negative effect of adversity on child outcomes.

A meta-analysis of 118 studies that assessed exposure to violence, positive adaptation, and protective factors involving 101,592 participants found that self-regulation, family support, school support, and peer support were the strongest predictors of resilience for children exposed to violence (Yule, Houston, and Grych, 2019). This protection was consistent across the three types of violence included in the study: 1) maltreatment, 2) intimate-partner violence, and 3) community violence. The authors concluded that efforts to target the identified protective factors are likely to have beneficial effects for children regardless of which type of violence they experienced. Below are specific findings from the 2019 meta-analysis with regard to self-regulation, family support, school support, and peer support.

• Self-Regulation reflects children’s capacity to manage their emotions, impulses, and behavior. Eighteen studies in the meta-analysis examined self-regulation. For example, one of the included studies looked at children ages 9–13 who had been exposed to intimate-partner violence; the study authors found that children with a higher capacity for emotion regulation were more likely to report a better quality of life (Grip et al., 2014). Another study using data from the Environmental Risk Longitudinal Study examined more than 1,000 twin pairs, finding that maltreated children with well-adjusted temperaments (i.e., sociable and self-controlled) were more likely to be resilient than maltreated children with less–well-adjusted temperaments (i.e., frustrated and hostile) [Jaffee et al., 2007]. However, when they had been exposed to multiple stressors, temperament was not a statistically significant predictor. The authors defined resiliency using a measure of antisocial behavior scores at ages 5 and 7 (after maltreatment).
• Family Support is characterized by measures of parental warmth and acceptance, family cohesion and structure, and perceived support from family members. Sixty-nine studies in the meta-analysis examined family support. A study of low-income African American sixth graders found that witnessing community violence was positively associated with anxiety among youths who reported low levels of maternal closeness, but not among youths who reported high levels of maternal closeness (Hammack et al., 2004). Also, a study of 1,599 urban middle school students found that parent support may offer boys who witness violence some protection against committing subsequent acts of violence, but this finding was not true for girls (Brookmeyer, Henrich, and Schwab–Stone, 2005; Brookmeyer, Fanti, and Henrich, 2006). A more-recent study using the Resilience Protective Factors Checklist found that, among individuals who have experienced more adverse childhood experiences, those with higher levels of family protective factors had positive social relationship outcomes, while those with lower levels of family protective factors had negative social relationship outcomes (Powell, Rahm–Knigge, and Conner, 2021). Social relationships were measured using 3 items from the 24-item World Health Organization Quality of Life Brief (e.g., “How satisfied are you with youth personal relationships?”).
• School Support included measures of the extent to which students feel supported and valued by teachers and staff, and a sense of security at school. Nineteen studies in the meta-analysis examined school support, including a study of 457 youths in Pennsylvania, which found that commitment to school lowered rates of lifetime violence, delinquency, and status offenses among adolescents who were physically abused as children (Herrenkohl et al., 2005). Also, a study of 391 low-income youths ages 13–17 found that high levels of participation in extracurricular activities acted as a protective factor, weakening the association between exposure to violence and externalizing problems (Hardaway, McLoyd, and Wood, 2012).
• Peer Support and Friendship Attachment included measures assessing emotional support, social support, relationship satisfaction, and level of attachment with friends, classmates, and peers. Twenty-four studies in the meta-analysis examined peer support and friendship attachment. For example, findings from a study of urban high school students found that secure attachment to peers may serve as a protective factor against depression and anxiety within the context of exposure to violence during adolescence (Heinze et al., 2018). The study found that exposure to violence during adolescence predicted later anxiety and depression. However, adolescents who were strongly attached to friends reported faster decreases in mental health symptoms after exposure to violence, compared with students who were insecurely attached to peers.

Several other protective factors were identified in the meta-analysis as potential predictors of resilience, including individual factors such as positive self-perceptions (Tlapek et al., 2017), coping (Daigneault et al., 2007), future orientation (So et al., 2018; Stoddard, Zimmerman, and Bauermeister, 2011), and cognitive ability (Klika et al., 2012); family factors such as parental effectiveness (David et al., 2015); and community factors such as religious involvement (Pearce et al., 2003) and community cohesion (Browning et al., 2014).

Several studies have found that exposure to an increasing number of protective factors results in greater protection and resilience (Herrenkohl et al., 2005; Sattler and Font, 2018; Turner et al., 2007). A study investigating protective factors among abused children found the overall number of protective factors was significantly associated with each outcome; that is, the more protective factors (0–4) to which children were exposed during adolescence, the lower involvement they had in each behavior examined, including delinquency, status offenses, and violence (Herrenkohl et al., 2005).

The protective factors identified for children exposed to violence are similar to those for children in the general population (Development Services Group, Inc., 2015). One study explained that this finding debunks the idea that children who overcome extreme adversity have qualities that make them different from other children (who do not have to overcome extreme adversity), and that instead children exposed to abuse, violence, and other risk factors respond similarly to protective factors as children without these experiences do. The differentiating factor (between children who overcome adversity and those who do not have to) is that children exposed to violence may require higher levels of exposure to protective factors to buffer the effects of their adversity histories (Herrenkohl et al., 2005).

For more information on protective factors, see the Model Programs Guide literature review on Protective Factors Against Delinquency.

There is a growing body of literature on how childhood exposure to violence (CEV) and other adverse childhood experiences (ACEs) relate to trauma (Bogat et al., 2006; Copeland et al., 2007; Fowler et al., 2009; Hong et al., 2021; Ozer and McDonald, 2006) and on how youth-serving organizations can best meet the needs of children who suffer from trauma-related disorders (Bunting et al., 2019; Ellis and Dietz, 2017; Forkey et al., 2021; Marsac et al., 2016; Maynard et al., 2019; McCormick, Scheyd, and Terrazas, 2018; Sprick et al., 2021). Though much of this research includes experiences in addition to violence (e.g., a serious accident, fire, experiencing a natural disaster, and diagnosis of physical illness [Copeland et al., 2007]), CEV is a key exposure.

While there is a lack of consensus on the definition of trauma and disagreement about its meaning (Bartlett and Sacks, 2019; Krupnik, 2019; SAMHSA, 2014), the construct of trauma is often linked to posttraumatic stress disorder (PTDS) and Criterion A of trauma-related disorders, as defined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (American Psychiatric Association, 2013). It is sometimes called “toxic stress” or a response to toxic stress (Cohen, Groves, and Kracke, 2009; Schaeffer, 2017; Shern, Blanch, and Steverman, 2014). Some researchers also suggest that the construct of trauma should be grounded in the general theory of stress, where trauma is considered a particular kind of stress response alongside adversity and normative stress (Krupnik, 2019). PTSD is distinct from most psychiatric disorders because it requires an initiating stressor (American Psychiatric Association, 2013; Copeland et al., 2007).

Trauma-Informed Care in Juvenile Justice

Several models of trauma-informed care have been proposed for youth in the juvenile justice system (Baglivio et al., 2014; Branson et al., 2017; Griffin, Germain, and Wilkerson, 2012). Trauma-informed approaches are especially important in secure residential programs where youths are particularly likely to have histories of complex trauma resulting from abuse and family violence, polyvictimization, and negative experiences that compromise core attachments with caregivers (Ford et al., 2012). One study examining children living in residential programs describes these children as “trauma organized” as their “attempts to cope with unmanageable conditions had led to maladaptive coping skills and the development of problematic habits” (Bloom, 2014:109).

Trauma-informed models of juvenile justice differ from both the traditional punishment model and the mental health model. The punishment model views youths as rational actors who can learn to stop inappropriate behaviors through graduated systems of punishments and rewards. The mental health model concentrates on the treatment of targeted symptoms and behaviors, which are identified through comprehensive assessments and diagnoses (Griffin, Germain, and Wilkerson, 2012). Though a trauma-informed model is more like a mental health model than a punitive model, there are three primary differences. First, a trauma-informed model understands that the youth is reacting to external events rather than battling an inherent mental illness. Second, a trauma-informed model takes more of a strength-based approach and does not concentrate solely on removing mental health symptoms or negative behaviors. It focuses on self-regulation skills and safety rather than diagnosis and medication management. It also concentrates more on brain development. Third, the line staff has a larger role to play in a trauma-informed model, taking advantage of treating a child within the context of a 24-hour-a-day milieu. Within a mental health model, the juvenile justice staffers are used more for referrals and behavior compliance while clinicians provide the treatment. In a trauma model, the juvenile justice line staff members are an essential part of the treatment (Griffin, Germain, and Wilkerson, 2012).

A systematic review of literature that defines trauma-informed care and recommends specific policies and practices in the juvenile justice system found there is relative consensus around the core domains of a trauma-informed juvenile justice system related to clinical services, agency context, and overall systems (Branson et al., 2017). Ten publications met the review requirements, which in total provided 71 different recommendations or policies. The core domains are 1) screening and assessment, 2) services and interventions, 3) cultural competence, 4) youth and family engagement, 5) staff training and support, 6) promoting a safe agency environment, 7) agency-level policies, procedures, and leadership, 8) cross-system collaboration, 9) system-level policies and procedures, and 10) quality assurance and evaluation. Although there is relative consensus in the published literature on the inclusion of these 10 domains in the juvenile justice system, there appears to be much less agreement on specific policies and practices (Branson et al., 2017). For example, while there is agreement that staff should receive training on trauma, there is less consensus around the specific topics that training should cover.

One example of a framework for trauma-informed care in residential facilities is the Sanctuary Model. There are seven values emphasized in this model, which are called Sanctuary commitments. These commitments are 1) nonviolence, 2) emotional intelligence, 3) social learning, 4) open communication, 5) democracy, 6) social responsibility, and 7) growth and change (Bloom, 2014). A study of the Sanctuary Model in a secure, residential program for girls found that implementation of the model resulted in increases in safety for both residents and staff, as measured by use of physical restraints, use of isolation/confinement, assaults and fights among youth, youth assaults on staff, injuries to staff, staff grievances filed, youth grievances filed, and youth reporting that they feared for their safety in the facility (Elwyn, Esaki, and Smith, 2015).

Another example is the Resilience, Opportunity, Safety, Education, Strength (ROSES) program. The program is a community-based, trauma-informed, gender-responsive advocacy intervention for girls ages 11 to 17 who are at risk for or already involved in the juvenile justice system. The goal of ROSES is to strengthen girls’ contexts (e.g., school, family, peers) in a way that promotes their positive development and decreases their engagement with risk-enhancing contexts. Contexts can be informal (e.g., peer, family) or formal (e.g., school, juvenile justice system, child welfare). The program uses an advocacy framework that concentrates on girls’ strengths (rather than their deficits) and aims to increase girls’ access to community resources and make the community (which includes the systems in which girls are involved) more responsive to their needs. ROSES involves a university–community partnership, in which the program is implemented by advocates—specifically advanced undergraduate students from a local university. The advocates work alongside girls, in the girls’ home communities. In addition, the program incorporates trauma-informed principles, such as promotion of safety (by attending to victimization risk and safety planning); trustworthiness (by holding advocates accountable to high standards set by the program); choice (by supporting multiple solutions to meeting girls’ goals and needs); collaboration (by involving girls in the directing and planning of the intervention); and empowerment (by following a rights-based model that asserts each girl has a right to positive youth development) [Javdani and Allen, 2016; Javdani, Singh, and Sichel, 2017]. An evaluation study with 257 girls found that those who participated in ROSES demonstrated statistically significantly less violence perpetration (e.g., engaging in a physical fight), indirect and direct violence exposure, status offending and minor delinquency, and substance misuse, compared with girls in the control group who did not participate. ROSES participants also reported missing fewer days of school and demonstrated significantly fewer symptoms of internalizing (e.g., depression, anxiety) and externalizing (e.g., anger) mental health challenges (Javdani, 2020).

Many studies have examined the effectiveness of interventions serving children who have been exposed to violence and other adverse childhood experiences (ACEs). Many of these interventions try to address the impact of trauma after exposure. Several have demonstrated positive outcomes. Below are examples of evidence-based practices and programs featured on the Model Programs Guide. Most of this section focuses on interventions serving youths who have already been exposed to violence. Some prevention programs also are described.

Therapeutic Practices

Meta-analyses have identified practices that can improve outcomes for children exposed to violence and other adverse experiences. For example, Eye Movement Desensitization and Reprocessing (EMDR) aims to help children recover from posttraumatic stress disorder (PTSD) through short imaginal exposure to the memory and subsequent stimuli to trigger eye movement. The goal of EMDR is to alleviate distress associated with traumatic memories by accessing and processing these memories in brief doses while focusing on an external stimulus (usually therapist-directed eye moments). EMDR is a brief intervention and takes place over the course of eight phases: 1) history taking, 2) preparation, 3) assessment, 4) desensitization (where the child is asked to focus on the traumatic memory and its associated dysfunctional cognition, emotion, and the physical sensations while the therapist tries to stimulate eye movement), 5) installation (where the therapist seeks to connect the child's traumatic memory with their earlier formulated positive, functional cognition), 6) body scan, 7) positive closure, and 8) reevaluation (de Roos et al., 2017; Rodenburg et al., 2009).

Two meta-analyses examined the effect of EMDR on measures of trauma and internalizing behaviors; overall there were mixed results. One meta-analysis by Rodenberg and colleagues (2009) found that the treatment had a positive, statistically significant effect on reducing trauma, while a meta-analysis by the Washington State Institute for Public Policy (2016) found no statistically significant effect on trauma measures. The meta-analysis by the Washington State Institute for Public Policy also found that EMDR had no statistically significant effect on measures of depressive disorders and anxiety.

Another example is therapeutic approaches for sexually abused children and adolescents which aim to reduce the developmental consequences that result from this distinct form of maltreatment. These approaches target children and adolescents ages 18 and under who have experienced sexual abuse. Although males can also experience child sexual abuse and suffer the same consequences, generally females are more often the victims of this specific type of maltreatment. These approaches include a variety of interventions that are designed to treat the negative impacts of child sexual abuse, such as cognitive behavioral therapy, child-centered therapy, EMDR (described above), imagery rehearsal therapy, supportive counseling, and stress inoculation training. Three meta-analyses found that sexually abused children and adolescents who received these therapeutic approaches showed statistically significant reductions in PTSD symptoms, internalizing behaviors, and externalizing behaviors, compared with children and adolescents who did not receive the therapeutic approaches (Harvey and Taylor, 2010; MacDonald et al., 2012; Trask, Walsh, and DeLillo, 2011).

Programs Aimed at Children Exposed to Violence

There are several programs that researchers have found to produce positive outcomes for children exposed to violence. These programs are implemented with families, in homes, and as part of larger service-delivery systems.

Family-Based Interventions. Much research points to the importance of family as a protective and resilience factor in buffering the effect of CEV on negative outcomes (see Protective Factors and Resilience section). Given that influence, there are several interventions that aim to improve outcomes for children and youths by engaging the family. This involvement can take many forms, including individual therapy with the parent, group therapy, role playing, and strategies to improve the parent–child relationship. Most of the programs aim to teach parents new skills such as cognitive strategies to respond calmly and thoughtfully to their child’s challenging behaviors, communication skills, maintaining developmentally appropriate expectations for their child, intrapersonal and interpersonal skills to enhance self-control, and nonviolent disciplinary strategies (see examples below).

Trauma-Focused Cognitive Behavioral Therapy (TF–CBT) is designed to help 3- to 18-year-olds and their parents overcome the negative effects of traumatic life events such as child sexual or physical abuse. The program allows the individual experiencing trauma to talk directly about their traumatic experience in a supportive environment. The goals of TF–CBT are to provide the skills to manage distressing thoughts related to their trauma and to process the events productively (Cary and McMillen, 2012; Cohen et al., 2004). Children who participated in this program had a statistically significant lower number of PTSD and depressive symptoms and problematic behaviors, compared with children in the control group (Cohen et al., 2004; Cohen and Mannarino, 1996; Deblinger, Lippman, and Steer, 1996). Parents in the intervention also had a statistically significant lower number of depressive symptoms and improved measures of parenting skills, compared with parents in the control group (Cohen et al., 2004; Deblinger, Lippman, and Steer, 1996).

Parent–Child Interaction Therapy (PCIT) teaches parents new interaction and discipline skills to reduce children's problem behaviors and to stop parental physical child abuse by improving parent–child relationships and parental responses to difficult child behavior. PCIT targets changing parenting practices and parent–child interactions to help prevent the recurrence of physical abuse in abusive families. Two studies found that parents who participate in PCIT showed statistically significant improvements in both parent and child behavior and a reduction in re-reports of child physical abuse (Chaffin et al., 2004; Nixon et al., 2003).

Alternatives for Families: Cognitive Behavioral Therapy is a family therapy program for children exhibiting behavioral or emotional dysfunction because of exposure to a hostile or physically aggressive family life. The intervention includes several lessons and strategies, including feeling identification, abuse discussions, new ways of thinking, emotional and behavior management, how to get along with others, and problem-solving techniques. Kolko (1996) found statistically significant reductions in family conflict, parent–child violence, physical punishment, and child abuse risk and improvements in discipline, cohesion, and child acceptance. There were also statistically significant reductions in child internalizing and externalizing symptoms. However, there were no statistically significant reductions in re-abuse rates (Kolko, 1996).

Expanded Early Pathways for Young Traumatized Children aims to improve outcomes for young children with behavioral and emotional problems who have experienced trauma and live in poverty by providing weekly, in-home, caregiver–child therapy. This is an expansion of the Early Pathways program that includes trauma-informed and culturally adapted strategies to meet the needs of families. The goal of the program is to treat and prevent disruptive behaviors in young, traumatized children. Love and Fox (2019) found that children who participated in this intervention experienced statistically significant reductions in several challenging behaviors, such as anxious/withdrawal symptoms (e.g., seeming worried) and fearful symptoms (e.g., sleep disturbances and being easily startled). There were also statistically significant improvements in the quality of caregiver–child relationships.

Systems-Based Interventions (Schools, Hospitals, Courts). As mentioned above (in the Protective Factors and Resilience section), research has identified school support as a protective factor for maltreated youth, buffering the influence of abuse on externalizing problems, violence, delinquency, and status offending (Hardaway, McLoyd, and Wood, 2012; Herrenkohl et al., 2005). Given the large amount of time children and youths generally spend in school, implementing programs there can be effective. Other systems, such as hospitals and courts, have also been identified as promising settings to implement programs for youths exposed to violence (Carey et al., 2010; Cooper et al., 2006; Corbin et al., 2013).

For example, Cognitive Behavioral Intervention for Trauma in Schools (CBITS) was designed for use in schools for children ages 10–15 who have had substantial exposure to violence or other traumatic events and who have symptoms of PTSD. CBITS incorporates cognitive–behavioral therapy skills in a group format and has three main goals: 1) to reduce symptoms related to trauma, 2) to build resilience, and 3) to increase peer and parent support. Two studies demonstrated statistically significant reductions in depressive and PTSD symptoms among the youths who participated in CBITS (Kataoka et al., 2003; Stein et al., 2003). One study also found a statistically significant reduction in parent-reported psychosocial dysfunction in their children; however, there were no statistically significant impacts on classroom behavior problems (Stein et al., 2003).

The Children's Advocacy Center Model is a multidisciplinary, victim-focused approach designed to improve forensic interviewing and the continuity of care for youths who are victims of sexual abuse and assault. The model includes the provision of services (which may vary from site to site) that coordinate with local child-friendly facilities to enable professionals from victim advocacy and child protective services, law enforcement and prosecution, and the medical and mental health fields to work together to investigate, prosecute, and treat child abuse. Participation in this program demonstrated a statistically significant increase in the receipt of physical health examinations and counseling referrals among victims of child maltreatment (Edinburgh, Saewyc, and Levitt, 2008).

Primary Prevention Programs. Several primary prevention programs, which are programs that attempt to prevent exposure to violence before it happens, have been shown to prevent violence victimization among children and youths. Many of these are designed for broad populations and are school based.

The Olweus Bullying Prevention Program is a schoolwide, multicomponent intervention designed to reduce and prevent aggression and bullying among students. An evaluation of this program in three urban public middle schools in the southeastern United States found that this intervention resulted in decreases in levels of teacher-rated verbal and relational victimization of children, though it did not result in decreases in teacher-rated physical victimization or student-rated victimization (Farrell et al., 2018).

Safe Dates is a prevention program for middle and high school students, which is designed to stop or prevent dating violence perpetration and victimization. Its goals are to change adolescent norms regarding dating violence and gender roles, to improve conflict resolution skills for dating relationships, to promote victims’ and perpetrators’ beliefs in the need for help and awareness of community resources for dating violence, to encourage help-seeking by victims and perpetrators, and to develop peer help-giving skills. Participation in the program was associated with statistically significant reductions in psychological, physical, and sexual abuse perpetration, and physical abuse victimization, compared with the control group at the 4-year follow-up; however, there were no significant differences between groups on sexual abuse victimization (Foshee et al., 2005).

IMpower Program for American Indian Girls is a school-based sexual assault prevention program designed to increase resistance skills (i.e., mental, verbal, and physical skills) and empowerment self-defense knowledge (i.e., the ability to recognize risk and use resistance strategies to prevent sexual assault). The primary goal of the program was to reduce sexual assault victimization for American Indian girls. Secondary goals included reducing sexual harassment and physical dating violence. A study of 255 American Indian girls found that those who participated in the program showed statistically significantly reduced rates of sexual assault victimization and sexual harassment victimization, though there was no statistically significant difference in dating violence victimization (Edwards et al., 2021).

Estimates vary by source and definition of violence, but most researchers agree that current levels of childhood exposure to violence (CEV) in the United States are too high (Finkelhor et al., 2015a; Sacks and Murphey, 2018; Yule, Houston, and Grych, 2019). CEV is itself a negative experience. CEV is also related to increased risk for experiencing a multitude of additional negative consequences, including mental and physical health problems, risky sexual behaviors, substance use, poor educational outcomes, delinquency, violence, and juvenile justice system involvement (Afifi et al., 2020; Carlson et al., 2019; Farrell and Zimmerman, 2018; Felitti, 2002; Fowler et al., 2009; Mitchell, Becker–Blease, and Soicher, 2021; Polanin et al., 2021). Youths involved in the juvenile justice system have higher levels of CEV than the general youth population (Baglivio et al., 2014; Malvaso et al., 2021); within the juvenile justice population, a history of CEV is associated with negative outcomes such as increased delinquent behavior, gun carrying, recidivism, and decreased future orientation (Beardslee et al., 2018; Monahan et al., 2015; Reid et al., 2017b; Tsang, 2018). The negative outcomes related to CEV are heightened with multiple types of exposure and higher levels of exposure (Copeland et al., 2007; James et al., 2018). 

Risk factors for CEV vary among children and youth. For example, child and youths who are LGBTQ, suffer from depression, live in disadvantaged communities, or who have parents with histories of antisocial behavior, offending, mental health problems, low educational levels, or were victims of child abuse themselves have higher likelihood of being exposed to several types of violence (Assink et al., 2019; Basile et al., 2020; Cook et al., 2010; Dank et al., 2014; McCormick, Scheyd, and Terrazas, 2018; Mulder et al., 2018; Zimmerman and Posick, 2016). Most significantly, youths who have already been exposed to violence are more likely to have additional exposures (Assink et al., 2019; Goncy, Farrell, and Sullivan, 2016; Zimmerman and Posick, 2016).

Researchers have identified protective factors that may buffer the effects of CEV. The strongest of these factors are self-regulation, family support, school support, and peer support. Other protective factors include individual factors such as positive self-perceptions, coping, and cognitive ability; family factors such as parental effectiveness; and community factors such as religious involvement and community cohesion (Yule, Houston, and Grych, 2019).

Several programs and practices have been evaluated that result in positive outcomes for children and youths who have experienced violence. Many of these try to address the impact of trauma after exposure. For example, therapeutic approaches such as cognitive behavioral therapy, child-centered therapy, eye movement desensitization and reprocessing, imagery rehearsal therapy, supportive counseling, and stress inoculation training have been effective for reducing PTSD symptoms, internalizing behaviors, and externalizing behaviors for sexually abused children and adolescents (Harvey and Taylor, 2010; MacDonald et al., 2012; Trask, Walsh, and DeLillo, 2011). Many programs are family based. For example, participation in Parent–Child Interaction Therapy, which teaches parents skills to reduce children’s problem behaviors and parental physical child abuse, has led to statistically significant improvements in both parent and child behavior and a reduction in re-reports of child physical abuse (Chaffin et al., 2004; Nixon et al., 2003).

Suggested Reference: Development Services Group, Inc. December 2022. “Children Exposed to Violence.” Literature review. Washington, DC: Office of Juvenile Justice and Delinquency Prevention. https://ojjdp.ojp.gov/model-programs-guide/literature-reviews/Children-Exposed-to-Violence

Prepared by Development Services Group, Inc., under Contract Number: 47QRAA20D002V.