Community Violence Exposure and Stress Reactivity in African American and Non-Latino White Adolescents With Overweight/Obesity

Introduction

Within the United States there is a staggering number of children and adolescents exposed to community violence each year (Wilson et al., 2009). Community violence is defined as “an exposure to intentional acts of interpersonal violence intended to cause physical harm against a person or persons in a community” (Cooley-Quille et al., 1995). These intentional acts encompass a variety of behaviors including knife attacks, verbal and physical aggression, and gun violence (David-Ferdon et al., 2016). Data from the National Survey of Children’s Exposure to Violence estimates 60% of youth have experienced at least one direct (victim) or indirect (witnessed) violent assault in the previous year (Finkelhor et al., 2009). Approximately 30% of youth reported moderate or frequent involvement in bullying at school and those who experienced this form of violence also experienced higher rates of victimization in the home and community compared to their peers (Cuevas et al., 2013). Children who were exposed to even one type of violence, both within the past year and over their lifetimes, were at increased risk of experiencing other types of violence (Holt et al., 2007).

Violence exposure is not randomly distributed across communities but is instead socially patterned with education, income, wealth, and neighborhood conditions strongly linked to violence (Egerter et al., 2011). While violence occurs across the socioeconomic spectrum, the risk of exposure to violence is greatest for children and adolescents in the most socioeconomically disadvantaged groups and communities (Stein et al., 2003). As illustrated in Figure 1, the high prevalence of crime, disorder, and violence in these communities is a direct consequence of structural racism (Chen et al., 2016; McMahon et al., 2013; Voisin & Neilands, 2010). Hence, the same social factors that contribute to socioeconomic disadvantage are also strongly linked to community violence.

OPEN IN VIEWER

Figure 1.

Conceptual model linking community violence exposure and biological stress responses (based on Voisin et al., 2010). Empirically validated relationships are represented by solid lines; hypothesized relationship tested in this study is represented by a bolded arrow.

Because of the racial inequities in economic opportunities across the US, many African American adolescents are disproportionately housed in disadvantaged communities making them more vulnerable to community violence exposure compared to their more affluent, non-Latino white counterparts (Braveman et al., 2010; Chauhan & Reppucci, 2009; Reeves et al., 2016). Data from the National Longitudinal Study of Adolescent and Adult Health noted that 60% of African Americans reported exposure to community violence compared to only 36% of non-Latino whites (Chen et al., 2017). Lambert and colleagues confirmed that African American youth who live in low-income neighborhoods are more likely to witness shootings, stabbings, and killings in their neighborhoods compared to their non-Latino white or middle-class peers (Assari et al., 2016; Lambert et al., 2010); this is especially true for African American male youth. Left unaddressed, exposure to violence has serious consequences for adolescent health and well-being and may contribute to the health disparities observed in African American adolescent populations.

Measuring both the direct and indirect exposures to community violence has important implications for child and adolescent health (Lynch, 2003). Direct exposures of violence (i.e., victimization) constitute a threat to one’s mental and physical health, whereas indirect exposures (i.e., witnessing violence) are classified as psychosocial threats to one’s well-being (American Psychological Association, 2017; Lazarus & Folkman, 1984). Both types of threats can engender emotions of fear and anxiety and influence a child’s perception of the resources and support available in their community to help them cope with violence (Assari et al., 2016). Indirect exposures also provide a measure of how much violence is occurring in a community and a context wherein direct exposures of violence take place (Lynch, 2003). Hence, examining the direct and indirect exposures of community violence on children’s health and development is relevant and important.

The health consequences of community violence exposure are well documented (Nofziger & Kurtz, 2005). Mental health consequences include increased risk of anxiety, depression, and posttraumatic stress disorder symptoms (Aneshensel & Sucoff, 1996; Cooley-Quille et al., 2001; Kliewer et al., 1998; Terr, 1991). Behavioral health consequences resulting from exposure to violence include increased antisocial behavior, lower academic performance, and higher risk of suicide (Berenson et al., 2001; Schwartz & Proctor, 2000; Wilson et al., 2009). Even in the absence of direct exposure, witnessing violence has been associated with lower outdoor physical activity due to fears related to neighborhood safety (Lenhart et al., 2017). The metabolic health consequences of community violence exposure can include increased risk of obesity during adolescence which may accelerate progression towards cardiometabolic diseases later in life (Ehrlich et al., 2016; Gustafsson et al., 2012). Violence exposure, however, is not deterministic (Gorman-Smith & Tolan, 2003; Jain et al., 2012); the majority of youth who experience direct and indirect forms of violence subsequently develop into healthy and productive adults (Benard, 2004; Werner & Smith, 2001). Previous research has confirmed that community- and family-level protective factors that promote positive youth development include positive social norms, high levels of social cohesion, family support, and rewards for prosocial community involvement (Hardaway et al., 2012; Jain et al., 2012; Jain & Cohen, 2013).

The link between community violence exposure and health, however, is not only influenced by community- and family-level factors but also individual characteristics including one’s appraisal of violence as a stressor which can lead to subsequent activation of the biological stress response (Lazarus & Folkman, 1984). The biological stress response is controlled by the hypothalamic pituitary adrenal (HPA) axis and autonomic nervous system (ANS), both of which play an important role in the maintenance and control of metabolic and immunological processes (Chrousos & Gold, 1992; Goldstein, 2010; Nater & Rohleder, 2009; Schommer et al., 2003). Activation of the HPA axis triggers the release of cortisol, which in healthy individuals has a diurnal rhythm with an early morning cortisol peak and decline in cortisol throughout the day (Chrousos & Gold, 1992). When an individual experiences a physical or psychosocial threat, the short-term release of cortisol provides the necessary energy to cope with and focus on the stressor at hand. After the stressor is resolved, inactivation of the HPA axis occurs with cortisol quickly returning to basal levels. Simultaneously, the ANS which is part of the nervous system is activated and exerts both parasympathetic and sympathetic effects. These effects can include heightened vigilance, heart rate, blood pressure, and increased secretion of salivary alpha-amylase (Goldstein, 2010; Nater & Rohleder, 2009; Schommer et al., 2003). Like cortisol, after the stressor is resolved, salivary alpha-amylase quickly returns to pre-stressor levels. An increase in cortisol or alpha-amylase release is adaptive when individuals are faced with a short-term stressor, however, frequent, and prolonged activation of this response can lead to maladaptation (Chrousos & Gold, 1992).

The biological stress system may become dysregulated when an individual experiences chronic stress beyond his or her control and lacks the resources or support with which to mitigate the effects of such stressors (McEwen, 1998; Miller et al., 2007; Yehuda et al., 1993). In the case of community violence, children and adolescents who report experiences of direct or indirect violence and do not have community or family support to help them cope with these experiences may exhibit maladaptive stress responses. Specifically, cortisol and alpha-amylase stress responses can become increasingly altered to the degree that the short-term release of these stress biomarkers are either atypically high or low. This is consistent with the Attenuation Hypothesis that suggests under conditions of chronic stress the HPA axis and ANS may initially adapt by upregulating cortisol and alpha-amylase secretion resulting in hypersecretion (Susman, 2006; Trickett et al., 2010). Over time however, the HPA axis and ANS can produce a maladaptive response by downregulating cortisol and alpha-amylase secretion resulting in hyposecretion. Both hyper- and hypo-secretion have been linked with increased vulnerability to cardiometabolic diseases and mental health disorders (Chrousos & Gold, 1992).

Given that African American adolescents have greater exposure to community violence and other chronic stressors including racial discrimination and socioeconomic disadvantage compared to non-Latino whites (Caldwell et al., 2002; Lambert et al., 2010; Morsy & Rothstein, 2019; Reeves et al., 2016; Turner & Avison, 2003), it is plausible that the biological stress response to community violence exposure may differ between these two racial groups. Geronimus has proposed a Weathering Hypothesis linking chronic stress and health, which suggests that African Americans develop deleterious health outcomes beginning in young adulthood due to cumulative socioeconomic disadvantage, prejudice, social marginalization, and psychosocial stress (Geronimus, 1992). In support of this hypothesis, Geronimus and colleagues observed increased biological dysregulation (i.e., allostatic load) in African American compared to non-Latino white individuals aged 18–64 (Geronimus et al., 2006). These findings suggest African American adolescents may exhibit greater dysregulation in cortisol and alpha-amylase stress responses (i.e., hyposecretion) to community violence exposure.

Previous researchers have linked maladaptive stress responses to community violence exposure in African American adolescents. Aiyers and colleagues examined the cumulative effect of violence exposure on cortisol responsivity in 266 African American youth (Aiyer et al., 2014). They determined that African Americans, particularly males, who were exposed to violence during adolescence exhibited a more attenuated cortisol pattern in early adulthood. Busso and colleagues demonstrated in a cohort of 169 adolescent teens that exposure to interpersonal violence (maltreatment, community violence) was associated with blunted sympathetic reactivity (heart rate variability) in response to a laboratory stressor (Busso et al., 2017). Using a longitudinal study design, Kliewer examined cumulative risk associated with demographic and psychosocial stress exposure (including community violence) and its relation to ANS reactivity in 205 African American adolescents (Kliewer & Robins, 2017). Increased exposure to psychosocial stressors was associated with a blunted salivary alpha-amylase response to a stress task (social competence interview). Collectively, these data suggest that long-term exposure to community violence may result in hyposecretion of stress hormones in African American adolescents.

To date, racial differences in the relationship between community violence exposure and biological stress responses have been under-explored. Examining these differences may have important implications for understanding how chronic stress may accelerate the progression of disease in African American adolescents via physiological mechanisms. More importantly, understanding the physiological mechanisms that contribute to health disparities in communities of color is important for the design and implementation of targeted intervention approaches that enable African American adolescents to effectively cope with chronic stress. Therefore, the purpose of this study was to examine the relationship between community violence exposure and biological stress responses among African American and non-Latino white adolescents. Our hypotheses were threefold: (1) exposure to violence would be greater among African Americans compared to non-Latino whites; (2) violence exposure would be associated with dysregulation in the HPA axis and ANS in both racial groups; and (3) African Americans would exhibit an attenuated stress response compared to a heightened stress response in non-Latino whites, thereby supporting the Attenuation Hypothesis and Weathering Hypothesis (Geronimus, 1992; Susman, 2006). To overcome the confounding effects of weight status on stress responses, we have chosen to specifically focus our analysis on adolescents with overweight and obesity. Previous research has identified different physiological mechanisms by which healthy weight versus overweight/obese adolescents respond to stress (i.e., heightened stress reactivity observed in overweight/obese youth) (Incollingo Rodriguez et al., 2015; Lucassen & Cizza, 2012). Therefore, this study is centered on violence exposure and stress responses in adolescents with overweight/obesity.

Methods

Stress Condition

The timeline of the stress condition visit is displayed in Figure 2. The stress condition consisted of all participants completing an acute laboratory stressor, the Trier Social Stress Test (TSST) (Allen et al., 2017), a standardized and validated means to induce stress and subsequent glucocorticoid responses in children and adults (Kirschbaum & Hellhammer, 1994). The TSST resembles real-world cognitive and psychosocial stressors experienced by adolescents and is the most widely used stress paradigm in stress research (Kirschbaum & Hellhammer, 1994). The TSST is 20 minutes in duration and prior to the TSST, an instructional and preparation period was included. To control for confounding effects of body position on the physiological stress responses, participants were instructed to maintain an upright posture during the TSST (Nair et al., 2015). Following the stress condition, participants were taken to a separate room and were instructed to rest with the option to eat at their leisure for 1 hour.

OPEN IN VIEWER

Figure 2.

Timeline of stress condition visit. Square denotes salivary cortisol sample; Star denotes pre- and post-TSST questionnaire, respectively.

Results

Racial differences in participant characteristics are displayed in Table 1. Sixty eligible participants completed the study however nine participants were later excluded during data analyses due to insufficient saliva samples (i.e., did not have a baseline value, post-test value, and/or a recovery value to properly calculate AUC). In total, 51 participants remained eligible for analysis, all of whom completed both experimental conditions and all study questionnaires. A post hoc power analysis indicated that a sample size of 51 participants was sufficient to detect a medium effect size (d = 0.39) with a power of 0.78 and a large effect size (d = 0.59) with a power of 0.98. African American adolescents were, on average, significantly younger and at a higher BMI percentile than their white counterparts (age: 16.8 ± 0.2 vs. 17.7 ± 0.3 years, p = 0.02; BMI percentile: 95.5 ± 0.7 vs. 91.7 ± 1.1, p = 0.003). Consistent with hypothesis 1, African Americans reported significantly higher community violence scores as compared to non-Latino white participants (violence: 3.6 ± 0.5 vs. 1.8 ± 0.4, p = 0.003). Perceived stress scores were similar among both racial groups (23.8 ± 1.4 vs. 24.2 ± 1.3, p = 0.85).

OPEN IN VIEWER

Table 1.

Participant demographics.

	African American (n = 28)			White (n = 23)				Sig
	n	Mean SE	%	n	Mean SE	%		(p-value)
Sex	28	—	100	23	—	100		.304
Female	15	—	54	9	—	61		—
Male	13	—	46	14	—	39		—
Age (years)	27	16.8 ± 0.23	—	19	17.7 ± 0.28	—		.019*
Weight (kg)	26	92.5 ± 3.94	—	20	86.6 ± 3.00	—		.265
Height (cm)	25	168.5 ± 2.13	—	20	171.5 ± 1.35	—		.268
BMI (kg/m2)	27	32.4 ± 1.11	—	20	29.4 ± 0.94	—		.061
BMI percentile	27	95.5 ± 0.72	—	18	91.7 ± 1.05	—		.003*
Community violence	19	3.6 ± 0.47	—	24	1.75 ± 0.35	—		.003*

Confirmation of the Stress Response During the TSST

Confirmation of the stress response during the TSST is displayed in Figure 3. There were significant differences across conditions for stress reactivity (the period from baseline to the end of the TSST task) for both salivary cortisol and salivary alpha-amylase (p < .01). There was a significant difference across conditions for stress recovery (end of the TSST task to the end of the 1-hour recovery period) for salivary cortisol (p < .01). There was not however a significant difference across conditions for alpha-amylase stress recovery (p = 0.18).

OPEN IN VIEWER

Figure 3.

Salivary cortisol and alpha-amylase stress reactivity confirmation.

Figure 4 displays the relationship between community violence exposure, cortisol, and alpha-amylase AUC among African American and non-Latino white participants. Table 2 displays the linear regressional analysis for violence exposure, cortisol, and alpha-amylase AUC. Inconsistent with hypothesis 2, violence exposure was not a significant predictor of alpha-amylase AUC (p = 0.08). Race, however, was a significant predictor of alpha-amylase reactivity (p = .0006), with a higher alpha-amylase AUC observed in African American compared to non-Latino white adolescents. Consistent with hypothesis 3, there was a significant difference in the relationship between community violence exposure and alpha-amylase AUC by race (β = −3561 ± 1226, p = .007). At similar increases in violence exposure, African Americans demonstrated a significant decline in alpha-amylase AUC while non-Latino whites demonstrated a significant increase in alpha-amylase AUC. Inconsistent with hypotheses 2 and 3, neither race nor violence exposure were significant predictors of cortisol AUC and there were no significant differences in the relationship between community violence exposure and cortisol AUC by race (all p’s > .05).

OPEN IN VIEWER

Figure 4.

Relationships between community violence exposure, cortisol and alpha-amylase area under the curve (AUC) among African American and non-Latino white participants.

OPEN IN VIEWER

Table 2.

Linear regression analysis for violence exposure, cortisol, and alpha-amylase AUC.

	β ± SE	p-value
Model 1: Cortisol awakening response AUC
Race	−4.9 ± 6.0	.42
Violence	−0.03 ± 1.8	.99
Race*Violence	−0.92 ± 2.01	.65
Age	−1.9 ± 1.6	.24
Gender	0.4 ± 4.0	.92
BMI percentile	0.3 ± 0.4	.45
Perceived stress	0.04 ± 0.26	.88
Model 2: Alpha-amylase awakening response AUC
Race	10740 ± 3665	.006*
Violence	2008 ± 1110	.08
Race*Violence	−3561 ± 1226	.007*
Age	−628 ± 967	.52
Gender	3353 ± 2430	.18
BMI percentile	103 ± 271	.71
Perceived stress	229 ± 157	.16

Note. AUC=area under the curve. BMI = body mass index. Asterisk denotes significance at p < .05.

Discussion

Community violence encompasses a variety of behaviors including gun violence, knife attacks, as well as verbal and physical aggression (David-Ferdon et al., 2016). These stressors can cause a mounting biological response in adolescents, which over time can result in maladaptive stress responses. Consistent with our hypotheses, African American adolescents reported greater exposure to community violence compared to their non-Latino white peers and two maladaptive stress responses were observed in the present study. In non-Latino white adolescents, an amplified ANS response with heightened alpha-amylase reactivity was observed with increased exposure to community violence. In African American adolescents, a blunted ANS response with attenuated alpha-amylase reactivity was observed with increased exposure to violence. Both forms of altered autonomic activity (heightened vs. blunted reactivity) have been linked to cardiometabolic risk factors (Chrousos & Gold, 1992). This finding provides partial support for the Attenuation Hypothesis, which posits when chronically activated, biological stress systems can initially produce a heightened stress response. Over time however, a heightened stress response can ultimately lead to further dysregulation evidenced by an attenuated stress response (Susman, 2006). Contrary to our hypothesis, community violence was not associated with cortisol reactivity in either racial group. It is important to note that with a sample size of 51 participants this exploratory analysis was underpowered to detect all but substantial effect sizes. Hence, these findings should be confirmed using a larger sample size. If these findings are indeed replicated, longitudinal studies examining the cumulative effects of violence exposure on stress reactivity may provide a better understanding of the mechanisms by which violence exposure differentially impacts biological stress responses by race.

The dysregulated alpha-amylase responses observed in African American and non-Latino white participants in response to the TSST is likely the result of both parasympathetic withdrawal and increases in sympathetic activity. Previous research has confirmed that parasympathetic nerves play a significant role in salivary alpha amylase release. For example, several salivary glands, like the sublingual and minor glands, are almost exclusively under parasympathetic nervous system control (Bosch et al., 2011). Experimental studies have also demonstrated that the sympathetic effects on salivary alpha amylase release are moderated by parasympathetic nervous system activity (Proctor & Carpenter, 2007). Hence, the heightened alpha-amylase reactivity noted in non-Latino whites who reported exposure to community violence may be displaying dysregulation in both components of the autonomic nervous system. The blunted alpha-amylase response observed in the African American participants may reflect an even greater dysregulation of these systems. While specific markers of parasympathetic and sympathetic activity (e.g., vagal tone, heart rate variability, and norepinephrine) were not assessed in the present study, additional research should continue to assess whether one or both components of the autonomic nervous system are driving the racial differences in salivary alpha amylase reactivity in response to community violence exposure.

The blunted autonomic response observed in African American adolescents was hypothesized to be due in part to the increased stress exposure previously observed in this racial group. Data from the Economic Policy Institute suggest a disproportionate number of African American adolescents live in disadvantaged communities and experience greater socioeconomic stressors (low family income, inadequate household wealth, low parental educational levels, and low parental occupational status), compared to their non-Latino white counterparts (Morsy & Rothstein, 2019). Increased exposure to racial discrimination among African American youth has also been observed in this racial group (Turner & Avison, 2003). In a community-sample of 198 African American youth residing in southeast Michigan, 83% reported exposure to some form of racial discrimination in the past 6 months (Nelson et al., 2018). This racial group is also more likely to be exposed to violent police practices, racial profiling, and unwarranted attention by police (Boyd et al., 2016). Compared to their non-Latino white peers, African American adolescents are also more likely to experience discriminatory practices at school (suspended more frequently, for longer periods of time, and receive greater punishment for similar infractions) (Nance, 2015).

While a comprehensive assessment of chronic stress exposure was not conducted in the present study, African American adolescents displayed a heightened alpha-amylase response compared to their non-Latino white counterparts even in the absence of community violence exposure. It is plausible that in African Americans, the compounding effects of community violence exposure with other stressors may accelerate the progression from a heightened stress response towards an attenuated stress response and subsequent disease and illness including diabetes, hypertension, and cardiovascular disease later in life in this racial group. This is consistent with the Weathering Hypothesis which posits that African Americans experience early health deterioration because of the cumulative impact of repeated experiences with social or economic adversity and political marginalization (Geronimus, 1992). Geronimus proposes that the stress inherent in living in a race-conscious society that stigmatizes and disadvantages African Americans may cause disproportionate “wear and tear on the body” which accumulates as an individual is exposed to repeated or chronic stress (Geronimus, 1992). This deterioration in health produces racial inequities in health with age through middle adulthood. Future research should continue to examine whether greater violence exposure is contributing to “weathering” in African American adolescents.

The absence of an association between community violence and cortisol reactivity in both racial groups in the present study was inconsistent with our hypotheses and previous research. In a cohort of 124 adolescents, Peckins et al. demonstrated that exposure to violence occurring over the past 12 months was predictive of cortisol reactivity in adolescent males (Peckins et al., 2012). In a cohort of 222 African American adolescents, Peckins et al. also observed that greater levels of violence exposure during early childhood (ages 3–9) was associated with a blunted cortisol response to a laboratory stressor (Socially Evaluated Cold-Pressor task) at age 15. This relationship remained significant even after controlling for social deprivation and other factors known to influence cortisol reactivity (Peckins et al., 2012). The inconsistency between our findings and those of Peckins et al. is likely due to the small sample size included in the present analysis as we were underpowered to detect all but substantial effect sizes. Differences in study design (longitudinal vs. cross-sectional), the length of time between the stressor and data collection (lifetime vs. 6 months), age of participants (children vs. adolescents) and participant weight status (healthy weight vs. overweight/obese) may also have contributed to our disparate findings. In addition, the timing of community violence exposure was unknown in the present study, which may have impeded our ability to detect an association between exposure and cortisol reactivity. Nevertheless, it is important to note that community violence is indeed a toxic stressor for all racial groups and previous research has consistently demonstrated that violence exposure is associated with cortisol dysregulation (Aiyer et al., 2014; Wexler et al., 2020). Indeed, our laboratory recently observed a significant association between community violence exposure and a blunted cortisol awakening response in both African American and non-Latino white adolescents with overweight/obesity (Wexler et al., 2020). Hence, researchers should continue to examine the deleterious effects of community violence exposure on adolescent health.

The present study had many strengths including the use of a laboratory stressor with high internal validity. The use of a comprehensive assessment of acute physiological stress responses (cortisol and alpha amylase) during both a stress and control condition served as another strength. Finally, the inclusion of a racially diverse sample is a notable strength of this study. However, several limitations should be mentioned. First, we ran a linear regression, and our effect of interest was in the interaction. While we did expect a large effect size based on racial differences previously reported in other metabolic parameters (Goran, 2008; Wexler et al., 2020), we were underpowered to test its interaction with only 51 people; sample size of 107 participants would have provided a power of 0.80 to reject an incorrect null hypothesis. Hence our findings should be framed as exploratory, and the generalizability of our study findings cannot be inferred. Second, while we employed a social-evaluative stressor that resembles real-world scenarios, it is recognized that an in-lab stressor is not necessarily reflective of stressors that adolescents experience in their natural environment. Nevertheless, participants did exhibit a biological stress response as illustrated by the cortisol and alpha-amylase reactivity during the TSST. Third, the cross-sectional study design of our study limits inferences on the directionality of the associations between violence exposure and stress reactivity. Fourth, data related to the timing and context of violence exposure was unavailable. Finally, the absence of contextual factors including socioeconomic status and social support measures are notable limitations of this data source. High socioeconomic status can reduce the risk of violence exposure and social support can attenuate the negative impact of exposure to violence on adolescent health whereas low socioeconomic status and social support can exacerbate it (Aiyer et al., 2014). Despite these limitations, we successfully induced a biological stress response during the TSST and observed significant associations between community violence and alpha-amylase reactivity.

It is also important to note that the lack of a healthy weight control group prevented any conclusive assumptions regarding the deleterious effect of community violence on cortisol and alpha-amylase stress responses in healthy weight adolescents. Previous research has determined that biological stress responses differ by weight status (Champaneri et al., 2013; Walker et al., 2000). Obesity is associated with alterations in cortisol metabolism [i.e., upregulation of the enzyme 11β-Hydroxysteroid Dehydrogenase Type 1 (11-B-HSD1)] resulting in a hyperresponsiveness to stress (Rask et al., 2002). Increased secretion of the hormone leptin and free fatty acids from abdominal fat tissue has also been associated with elevated sympathetic activity (Smith & Minson, 2012). Over time, chronically elevated sympathetic nervous system activity can impair beta-adrenergic signaling thereby contributing to further HPA axis dysregulation (Smith & Minson, 2012). Obesity also represents a distinct psychological stress. Adolescents who are overweight/obese experience all the same stressors as their healthy weight peers while also being exposed to weight-based stigma causing heightened stress (Tomiyama, 2014). Hence, exposure to community violence may have acted to exacerbate biological stress dysregulation in our sample of adolescents. Additional research is needed to understand the unique effects of community violence exposure on biological stress responses in healthy weight adolescents.

Exposure to community violence is a significant problem in the US with negative public health consequences. Results from the current study provide preliminary evidence that the cumulative effects of community violence may contribute to maladaptive stress responses, which may differ by race (heightened vs. blunted alpha-amylase reactivity in non-Latino white and African American adolescents, respectively). Both forms of altered autonomic responses have been linked to cardiometabolic risk factors (Chrousos & Gold, 1992). While prevention work does not typically have access to biological measures, these assessments provide important insight regarding the unique influences of specific toxic stressors including community violence on child health and development. As such, we were interested in examining potential racial differences in the relationship between community violence exposure and biological stress responses to better understand how this toxic stressor “gets under the skin” to accelerate disease progression in adolescents, African Americans in particular. While an assessment of biological mechanisms cannot reduce exposure, it can inform targeted intervention efforts at the individual and community level that are aimed at helping adolescents cope with chronic stress.

The Division of Violence Prevention at the Center for Disease Control and Prevention has identified several strategies at multiple levels that aid in the promotion of positive youth development (Dahlberg & Krug, 2002). Individual strategies include increased education and life skills training. Promoting healthy relationships and building conflict-resolution and problem-solving skills through approaches like mentoring, peer support programs, and family counseling services is another successful prevention strategy. Policymakers and interventionists should also more widely implement and disseminate programs that limit adolescent violence exposure. One program to consider is the Massachusetts Safe and Successful Youth Initiative (Campie et al., 2017), which was launched in 2010 in 11 Massachusetts cities with the highest per capita rates of violent crime. This initiative was designed to enhance individual skills, relational experiences, and employment opportunities to reduce violence and promote healthy development and outcomes among young males who are at the greatest risk for violent offenses and victimization. Cities that implemented the program reported significant reductions in crime compared with 32 other cities in the state that did not have the program available to them among the age group targeted by the program (Campie et al., 2017). Another successful program is the Cure Violence Model, which utilizes community stakeholders to enhance community engagement, prevent retaliatory shootings, and mediate ongoing conflicts (Butts et al., 2015). The Cure Violence model has been implemented in cities including Chicago, Maryland, and Phoenix and has demonstrated a significant decrease in gun violence at implementation sites and in surrounding areas (Butts et al., 2015). Finally, societal approaches that address structural racism and social disadvantage should focus on developing policies that target community economic development, poverty reduction, and job training to reduce and eventually eliminate the social conditions that produce violence (Voisin, 2007). Policies, interventions, and community-based programs centered on positive youth development will be an important step to minimizing the deleterious effects of community violence on adolescent health.