Soc Psychiatry Psychiatr Epidemiol DOI 10.1007/s00127-014-0965-2
ORIGINAL PAPER
Community violence exposure and severe posttraumatic stress in suburban American youth: risk and protective factors Sandra Lo¨fving–Gupta • Frank Lindblad Andrew Stickley • Mary Schwab-Stone • Vladislav Ruchkin
•
Received: 18 June 2014 / Accepted: 4 October 2014 Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose The psychological effects of community violence exposure among inner-city youth are severe, yet little is known about its prevalence and moderators among suburban middle-class youth. This study aimed to assess the prevalence of community violence exposure among suburban American youth, to examine associated posttraumatic stress and to evaluate factors related to severe vs. less severe posttraumatic stress, such as co-existing internalizing and externalizing problems, as well as the effects of teacher support, parental warmth and support, perceived neighborhood safety and conventional involvement in this context. Method Data were collected from 780 suburban, predominantly Caucasian middle-class high-school adolescents in the Northeastern US during the Social and Health Assessment (SAHA) study. Results A substantial number of suburban youth were exposed to community violence and 24 % of those victimized by community violence developed severe posttraumatic stress. Depressive symptoms were strongly
S. Lo¨fving–Gupta F. Lindblad V. Ruchkin (&) Child and Adolescent Psychiatry Unit, Department of Neuroscience, Uppsala University, 751 85 Uppsala, Sweden e-mail: [email protected] A. Stickley Stockholm Centre on Health of Societies in Transition (SCOHOST), So¨derto¨rn University, Huddinge, Sweden M. Schwab-Stone V. Ruchkin Child Study Center, Yale University Medical School, New Haven, CT 06520, USA V. Ruchkin Sa¨ter Forensic Psychiatric Clinic, 783 27 Sa¨ter, Sweden
associated with higher levels and perceived teacher support with lower levels of posttraumatic stress. Conclusion Similar to urban youth, youth living in suburban areas in North American settings may be affected by community violence. A substantial proportion of these youth reports severe posttraumatic stress and high levels of comorbid depressive symptoms. Teacher support may have a protective effect against severe posttraumatic stress and thus needs to be further assessed as a potential factor that can be used to mitigate the detrimental effects of violence exposure. Keywords Community violence exposure Internalizing problems Protective factors Adolescents
Introduction The term community violence is used to signify acts by individuals or groups intended to harm others, such as chasing, threatening, beating, robbing, mugging, raping, shooting, stabbing or killing, and experienced outside home [1]. Direct exposure implies being a victim or a target of violence oneself, whereas indirect exposure means presence at the scene of violence and direct witnessing of it [1, 2]. During the 1990s community violence was described as an epidemic of national proportions in the United States [1]. Although it has since decreased [2] it is still regarded as a major public health problem [3–5]. According to a recent national survey [6] 60 % of children and youth in the US had experienced violence during the past year in their homes, schools or communities. However, the prevalence of community violence tends to vary dramatically depending on the geographical location, socio-economical status, age, race and gender, with higher rates of exposure
123
Soc Psychiatry Psychiatr Epidemiol
to community violence in urban, low-income and ethnic minority communities [2]. Little is known about the prevalence and moderators of community violence among youth living in suburban areas [7], as most studies concerning community violence have focused on so-called high-risk populations, such as ethnic minorities living in impoverished areas [4], with a plethora of additional risk factors that may affect child development and the ability to cope with stress [8]. Given this, it remains unclear to what extent the previous research findings on community violence exposure can be applied to youth living in suburban and more affluent areas. Exposure to community violence has been linked to internalizing outcomes such as anxiety and depression [9, 10], and posttraumatic stress [11, 12], but also to externalizing behavior such as substance use, school failure [4, 10, 13], delinquency, and aggressive behavior [14–16]. Research has also demonstrated that outcomes may differ depending on gender [11, 17, 18] age [2] proximity to violence [19] and other factors. A number of studies have attempted to identify factors within the social environment that may decrease the risk of experiencing the negative consequences of community violence. Findings regarding the extent to which family and parenting characteristics moderate the effects of community violence on psychopathology have been inconsistent. Some studies have indicated that a lack of family support may increase the risk of internalizing symptoms while high family support does not seem to buffer the negative effects of exposure to violence [20]. Other studies suggest that a supportive family does help when the exposure is mild, but when it is more extensive or frequent, such support and caring have little or no effect [21]. Longitudinal studies [22, 23] have indicated that high family support may buffer against an increase in internalizing symptoms even where there is an increase in violence exposure and that a high level of family cohesion has a moderating effect on delinquent behavior among adolescents exposed to community violence [24]. There is similar uncertainty about the role of school factors. Several studies have suggested that school connectedness is linked to lower levels of emotional distress and better psychological outcomes and that school support, defined by attachment to school, teacher support and academic motivation, has a protective role among those students who have not been exposed to violence [25, 26]. However, other studies have concluded that, among those exposed to community violence, perceived school support does not moderate symptoms of depression and posttraumatic stress [12, 23]. The findings regarding the role of the neighborhood have also been inconsistent. Youths’ perception of the neighborhood as dangerous and unsafe has been linked to more symptoms of depression, anxiety and conduct
123
disorder [27], while the perception of neighborhood safety mediates the association between exposure to violence and posttraumatic stress symptoms [28]. Among youth living in areas with high levels of violence exposure, a higher degree of perceived neighborhood cohesion and safety is associated with less internalizing symptoms. Yet, a positive perception of the neighborhood did not protect against the adverse effects of community violence exposure [21]. Unstructured socializing and unmonitored free time with no authority figures present has been positively associated with violence exposure, internalizing symptoms [29] and externalizing symptoms [30, 31]. In contrast, access to various resources such as neighborhood youth organizations has been described as protective against the impact of community violence exposure [32]. Moreover, some research has suggested that participation in communitybased activities is negatively associated with risk factors for community violence exposure such as delinquency and substance abuse [33, 34]. Hence in this study, we will assess the prevalence of community violence exposure among adolescents in a suburban middle-class community, as well as evaluate posttraumatic stress associated with such exposure in this group. Further, we will also examine the potential risk (coexisting internalizing and externalizing problems) and protective (family and teacher support, neighborhood perception and conventional involvement) factors associated with the development of severe posttraumatic stress among those victimized by violence. Youth coming from suburban areas would be expected to report lower levels of exposure to community violence than their urban counterparts. However, those, who are exposed would report symptoms of posttraumatic stress in a fashion similar to that described in other studies of community violence exposure, that is, those directly victimized would report higher levels of posttraumatic stress than those who have only witnessed community violence [11], and those who report higher levels of posttraumatic stress would also report higher levels of comorbid anxiety and depressive symptoms. Although we expect that family, school, neighborhood and community involvement may protect against the development of posttraumatic stress symptoms, considering the equivocal findings from previous research, we will refrain from testing any specific hypotheses in this regard and limit our efforts to exploratory analyses.
Methods Participants Data were collected from a large-scale study of risk and protective factors for problem behaviors in youth, the
Soc Psychiatry Psychiatr Epidemiol
Social and Health Assessment (SAHA). The survey methodology has been described previously [13, 15] although earlier reports concerned inner-city youth in New Haven, CT, USA. Although Connecticut has long ranked at or near the top of the US most prosperous states, New Haven remains one of the country’s poorest cities, with 30 % of its population living beneath the poverty line, according to 2011 census data (median household income $38,963). In contrast, the current study focused on youth residing in the same geographic area, but living in four small towns around New Haven with higher income status (mean household income $93,410). Data were collected in the spring of 2003 in all public schools in four suburban municipalities from the greater New Haven area, where a total of 780 sixth and eighth graders completed the SAHA. Of those, 23 students were excluded from the analyses because of inconsistent or incomplete reporting on the scales of interest. In comparison with the study group, no significant differences were obtained on any variable of interest, except that the excluded group perceived their schools (19.65 ± 3.71 vs. 16.54 ± 4.74) and their neighborhoods (22.91 ± 4.13 vs. 21.04 ± 3.74) as being less safe. Participants (N = 757) ranged in age from 13 to 16 years (M = 14.47 ± 1.18). The composition of the final sample was 53.7 % female (N = 405) and the ethnic composition was predominantly Caucasian (84.5 % Caucasian, 6.2 % Asian American, 3.2 % African American, 1.7 % Hispanic American, 0.3 % American Indian and 4.0 % Other), an accurate reflection of the local public school population. Nearly 85 % of youth reported their parents as being married or remarried. The majority of parents had college education (78.5 % of fathers and 81.5 % of mothers) and were full-time or part-time employed (92.1 % of fathers and 82.5 % of mothers). Measures The Social and Health Assessment (SAHA) [15, 35, 36] served as the basis for the survey. The scales used in the present study are described in detail below. Community violence exposure was assessed by a modified version of the instrument developed by Richters and Saltzman [37], and includes two subsets of items, one assessing direct victimization and the other witnessing of community violence, using a 5-point scale response format (from 0 times = 0 to 10 or more times = 4). Victimization by violence was assessed by six items asking whether in the past year in their community respondents had been beaten up or mugged, threatened with serious physical harm by someone, shot, attacked or stabbed with a knife, chased by gangs or individuals, or seriously wounded in an incident of violence, with a total score ranging from 0 to 24.
Similarly, the witnessing violence index consisted of six items asking the respondents whether or not they had seen someone being victimized by the same types of violence in their community, with the total score also ranging from 0 to 24. These indices have demonstrated good psychometric properties with American inner-city youth [38]. Posttraumatic stress was measured by the Child PostTraumatic Stress Reaction Index (CPTS-RI), a 20-item scale, assessing posttraumatic stress symptoms in schoolaged children and adolescents after exposure to trauma [39, 40]. The frequency of the symptoms in the past month is assessed on a Likert-type five-point rating scale ranging from ‘‘never’’ (0) to ‘‘most of the time’’ (4), with a total score range from 0 to 80. The scale is highly correlated with the DSM-based diagnosis of posttraumatic stress and has well-established clinical cutoffs. A score between 12 and 24 indicates mild posttraumatic stress, a score of 25–39 indicates moderate posttraumatic stress, 40–59 severe posttraumatic stress, and a score of 60 and above indicates very severe posttraumatic stress [40]. The Cronbach a obtained for the scale was adequate (0.86). Depressive symptoms were assessed using an adaptation of the Center for Epidemiological Studies-Depression Scale (CES-D) [41]. Both the CES-D e.g. [42] and its modified versions [43] have demonstrated excellent psychometric properties with adolescents. In the present study, the questionnaire consisted of ten items (e.g. not liking oneself; loss of interest in other people or things). Students reported on symptom presence for the past month on a three-point scale [Not True (0), Somewhat True (1), Certainly True (2)]. The scale had good internal consistency (a = 0.80). Anxiety symptoms were assessed by a 12-item scale [44], reflecting cognitive–affective and behavioral modes of anxiety, including worrisome, preoccupying thoughts or unpleasant feelings (e.g. worrying about being liked; feeling nervous when called on in class; worrying about the future). Students reported on the presence of anxiety symptoms on a three-point scale (Not True = 0, Somewhat True = 1, Certainly True = 2). The scale had a good internal consistency (a = 0.86). Violent and aggressive behavior of the respondents in the past year was assessed by six items from the SAHA, including starting a fist fight or shoving match, hurting someone so badly in a fight that treatment by a doctor was needed, carrying a gun, being in a gang or posse fight, being arrested by the police, and carrying a blade, knife or gun in school. These six items were selected from a larger pool of conduct problems and antisocial behavior items to reflect the types of violence also assessed by the witnessing and victimization measures. The selection of these six items was further supported by factor analysis. Responses were tallied on a five-point scale (ranging from 0 times to 5
123
Soc Psychiatry Psychiatr Epidemiol
or more times), producing a total score range from 0 to 24. The scale had good internal consistency (Cronbach a = 0.81). Perceived teacher support was assessed by eight items (e.g. teachers are willing to help students, teachers notice when I am doing a good job; teachers show concern when a student is absent from school, etc) adapted from Hawkins et al. [45] and developed by the creators of the original survey, Weissberg et al. [35]. The respondents were asked to report on a four-point scale how true for them each of the statements was [ranging from ‘‘definitely not true’’ (1) to ‘‘definitely true’’ (4)]. Cronbach a for the scale was 0.71. Parenting included two scales, assessing parental warmth and parental involvement with each scale item being scored on a four-point scale, ranging from ‘‘Never’’ (1) to ‘‘Often’’ (4). The scales were developed by the SAHA Research Evaluation Team [44]. Parent Involvement consists of six items, assessing youth perceptions of the degree to which their parents and/or primary guardians are involved and interested in their lives (e.g., the parent asks about her/his life, encourages interest in different things, gives good advice, etc). A good level of internal consistency was obtained (a = 0.80). Parental warmth includes five items that ask about students’ perceptions of their parents’ warmth and support for them (e.g. parents or guardians being kind to him/her, hugs or kisses him/her, shows their love for him/her, etc). A good level of internal consistency was obtained (a = 0.82). Conventional involvement was assessed by a six-item scale that asked the respondent about his/her participation in various kinds of group spare time activities. The respondents rated how many times during a regular week they participated in such activities as organized sports, dance, art, youth clubs, after school programs, etc. The frequency was assessed on a five-point scale ranging from ‘‘None’’ (0) to ‘‘6–7 times’’ (4). The scale was developed by the SAHA Research Evaluation Team [44]. Perception of the neighborhood was assessed by the Attachment to Neighborhood Scale [44] that inquires about individual perceptions of the neighborhood in which the respondent lives. The scale consists of seven positively scored items (examples include ‘‘My neighborhood looks nice’’; ‘‘I like spending time in my neighborhood’’; ‘‘People on the streets in my neighborhood are friendly’’). All items are scored on a four-point scale ranging from ‘‘Definitely not true’’ (1) to ‘‘Definitely true’’ (4). Cronbach’s alpha for the present sample was 0.82. Proxy for SES was computed as a sum, based on students’ reports regarding the family status of their parents (two parent families, 1/0), employment status of each parent (1/0), and education level of each parent (college education, 1/0), hence, potentially ranging from 0 to 5. This method has been used for assessing SES previously, e.g. [13].
123
Procedure Parents were informed of the survey at the time of school registration, received a letter about the survey 2 weeks prior to the administration of the questionnaire, and were offered the opportunity to decline participation. The passive informed consent procedure was approved by the university’s IRB and considered as an appropriate ethical procedure by the legislation of the State of Connecticut. Prior to survey administration, students were read a detailed assent form outlining their participation with assurances of confidentiality, and were asked to sign it to indicate assent (parent and child refusals were \1 %). Students completed the survey in a classroom setting during one class period during the regular school day. Trained administrators read all questions aloud while students followed along with their copies of the survey, reading questions to themselves and circling responses in the booklet. A second administrator was available, providing help to individual students as requested. The teacher was also present in the class, which minimized any problems related to classroom management. Make-up administrations were performed at each school within 1 month of the initial administration for those who had been absent. Data analysis The data were analyzed using the Statistical Package for Social Sciences (SPSS-22.0). Chi-square tests and one-way ANOVA tests were employed to assess the differences between the groups. Binary logistic regression analyses were conducted to identify the variables that could discriminate between the groups with severe vs. less severe posttraumatic stress. A proxy for SES, gender, age and ethnic minority status (1/0) was entered in Step 1. Depressive and anxiety symptoms, as well as aggressive and violent behavior were entered in Step 2. Considering previous research about the dose-related effects of trauma [46], the victimization score was added into the equation in Step 3, followed by Step 4, in which all of the environmental factors were entered, including those relating to school (school safety and teacher support), parenting (parental warmth and parental support), and perception of the neighborhood. In an attempt to reduce the number of variables in the equation, only those variables that were significantly different between the groups were used in the model and hence, conventional involvement was excluded from the analysis.
Results The study group (N = 757) was divided according to the reported severity of exposure. Those who did not report
Soc Psychiatry Psychiatr Epidemiol Table 1 Prevalence of different types of witnessing and victimization [N (%)] Witnessing group (261)
Victimization group (182)
v2
Total study group (757)
156 (20.6)
Witnessing (in the past year I have seen…) Someone else being chased by gangs or individuals Someone else get threatened with serious physical harm Someone else getting beaten up or mugged Someone else being attacked or stabbed with a knife A seriously wounded person after an incident of violence Someone else get shot or shot at with a gun Victimization (in the past year I have been…)
70 (26.8)
86 (47.3)
194.9
190 (72.8)
145 (79.7)
501.3
335 (44.3)
72 (27.6)
86 (47.3)
194.5
158 (20.9)
8 (3.1)
25 (13.7)
60.7
33 (4.4)
46 (17.6)
61 (33.5)
124.6
107 (14.1)
12 (4.6)
22 (12.1)
42.8
34 (4.4)
Chased by gangs or individuals Threatened with serious physical harm by someone Beaten up or mugged
52 (28.6)
52 (6.9)
135 (74.2)
135 (17.8)
38 (20.9)
38 (5.0)
Attacked or stabbed with a knife
14 (7.7)
14 (1.8)
Seriously wounded in an incident of violence
17 (9.3)
17 (2.2)
Shot or shot at with a gun
17 (9.1)
17 (2.2)
All Chi-square test comparisons were significant at p \ 0.001
any witnessing or victimization episodes were considered as the non-exposed group (41.5 %). Those, who reported at least one episode of witnessing, but no episodes of victimization were considered as the witnessing group (34.5 %). Finally, those, who reported at least one episode of victimization, were considered the victimization group (24.0 %). The prevalence of different types of exposure is described in detail in Table 1. As regards the prevalence of severe posttraumatic stress between the groups, the victimization group reported the highest prevalence of severe posttraumatic stress 43/182 (23.6 %), when compared with the nonexposed group 5/314 (1.6 %) and witnessing group 8/261 (3.1 %) (v2 = 92.19, p \ 0.001). Since the last two groups did not differ significantly from each other on the prevalence of severe posttraumatic stress (v2 = 1.37, p = 0.242) the following analysis focused only on the victimization group (N = 182). Hence, those youths who reported at least one episode of victimization by community violence during the last year were divided into two groups based on their reported CPTS-RI score, with scores of 40 or higher (indicative of severe or very severe posttraumatic stress), or 39 and below (indicative of moderate or lower degree of posttraumatic stress) being used as cutoff points. Binary logistic regression analyses were conducted with posttraumatic stress (severe vs. less severe) as the dependent variable in four steps. As shown in Table 2 (where only significant predictors are shown; p \ 0.05), Step 1 demonstrated that minority status and gender had a moderate predictive power, explaining 7.7 % of the variance. In
Step 2 depressive symptoms and aggressive and violent behavior predicted posttraumatic stress, with the predictive power of minority status becoming non-significant. In Step 3 victimization further added 2 % to the predictive power of the model. At this step, the predictive power of aggressive and violent behavior became non-significant. Finally, environmental factors increased the predictive power of the model by an additional 5 %. The final model consisted of the following factors: gender, depressive symptoms, victimization, and teacher support. This model fits the data well (v2 = 97.44; df = 6; p \ 0.001), explaining 43 % of the variance (Cox & Snell R2).
Discussion In this study of public school students from a suburban middle-class area we have found a substantial number of youth reporting at least some degree of exposure to community violence and a large proportion (24 %) of those victimized by community violence also reported severe levels of posttraumatic stress. Depressive symptoms were strongly associated with severe posttraumatic stress. Perceived teacher support was associated with lower levels of posttraumatic stress. In urban settings earlier research has reported that up to one-third of children have been directly victimized and almost all children have witnessed community violence [47]. In our study, the prevalence of exposure was lower than that in urban areas, yet substantial, with one-quarter of youth reporting that they were victims of violence during
123
Soc Psychiatry Psychiatr Epidemiol Table 2 Binary logistic regression analysis predicting severe posttraumatic stress in the victimized group (N = 182)
Step 1
Variable
R2
Gender (male)
0.077
R2D
Ethnicity status (minority) Only those variables that were significant at different steps of analysis are depicted (p \ 0.05). Victimization (dose) refers to a sum of the number and frequency of each experienced episode of exposure
2
123
OR (95 % CI) at entry
Final OR (95 % CI)
-1.48 (0.665)
0.343 (0.165–0.714)
0.227 (0.062–0.836)
0.120 (0.639)
2.45 (1.12–5.34)
1.13 (0.322–3.95)
0.355 (0.065)
1.36 (1.22–1.50)
1.43 (1.25–1.62)
-0.090 (0.134)
1.23 (1.06–1.43)
0.914 (0.703–1.19)
1.21 (1.03–1.43)
1.24 (1.04–1.48)
0.79 (0.690–0.898)
0.79 (0.690–0.898)
Depressive symptoms Aggressive and violent behavior
0.354
3
Victimization (dose)
0.376
0.022
0.214 (0.089)
4
Teacher support
0.427
0.051
-0.239 (0.067)
the past year. However, these figures are also lower than in previous studies from suburban areas. In a study of sixth to eighth graders in a suburban private school, more than twothirds of the children reported victimization during the past year [7] Similarly, 55 % of the middle-class youth aged 12–24 in a study by Gladstein et al. [18] reported experiencing victimization during the past year. These differences might in part, be explained by the use of a broader definition of community violence in the previous studies. Another factor contributing to this difference is that the present study concerned younger adolescents, who are generally less likely to have been exposed to violence [2]. Severe levels of posttraumatic stress symptoms were reported by 6.7 % of the study sample, indicating that potential PTSD levels in this group were higher than the average rate of PTSD observed among adolescents in the US general population, (3.7 % for boys and 6.3 % for girls) [48]. However, the present study focused on the selfreported prevalence of posttraumatic symptoms only, without any clinical assessment of the symptoms themselves or the functional impairment, needed for establishing a clinical diagnosis, which may explain the differences in the prevalence rates. Previous research has suggested that both violent victimization and witnessing violence are associated with posttraumatic stress among urban youth [11]. In the present study, the highest prevalence of severe posttraumatic stress (24 %) was reported by youth directly victimized by violence, which is in line with earlier findings. We found no significant difference between the witnessing group and the non-exposed group regarding the prevalence of severe posttraumatic stress. The amount of violence exposure, both in terms of frequency and the number of different types of exposure, also correlated with an increased number of posttraumatic stress symptoms, which is in line with previous studies about the dose-related effect of violence on the severity of posttraumatic stress [46].
0.277
Final B (SE)
Among the demographic variables neither age nor socioeconomic status was significantly associated with severe posttraumatic stress. Consistent with previous research [49] there was a gender difference in the association with severe posttraumatic stress with a higher prevalence of severe posttraumatic stress being found among girls. Previous studies [50] have suggested that belonging to an ethnic minority can represent a risk factor for severe posttraumatic stress. We also found that ethnic status was initially related to severe posttraumatic stress in the regression analysis. However, when symptoms of depression and aggressive behavior were added into the model, this association with posttraumatic stress became non-significant. This finding can be potentially explained by the results from previous research which indicates that there are generally higher levels of internalizing and externalizing problems in ethnic minority youth [50, 51]. Depressive symptoms were associated with more severe posttraumatic stress symptoms suggesting comorbidity [52, 53] which seems to be especially evident in youth exposed to violence [48]. Some research [54] suggests that individuals with posttraumatic stress tend to perceive comorbid depressive symptoms as being more frequent and/or distressing, and the environment as less safe and less supportive [54], and yet, perceived teacher support was negatively associated with posttraumatic stress, even when depressive symptoms were controlled for, suggesting its potential protective effect against posttraumatic stress. Another important risk factor that was examined in the analysis was aggressive and violent behavior. As suggested by Gorman-Smith and Tolan [14] those youth who commit violent acts themselves are more likely to report higher levels of exposure to violence. They are also at higher risk to become victims of violence and hence are at an increased risk of developing posttraumatic stress. Indeed, in the present study we initially found a significant association between delinquent behavior and posttraumatic
Soc Psychiatry Psychiatr Epidemiol
stress, which became non-significant, however, when the degree of exposure was taken into account. Perceived neighborhood safety was not associated with severe posttraumatic stress. However, the relationship between perception and the actual nature of a neighborhood may be weak [59], which may have obscured an association between neighborhood safety and posttraumatic stress. There was no significant difference between the three study groups regarding participation in conventional activities and neither was conventional involvement associated with posttraumatic stress. Richards et al. [29] reported that the benefits of conventional involvement are particularly pronounced for children in low-income neighborhoods with high crime rates, implying that the effects in our sample from an affluent area may be limited. Family support, as measured by parental involvement and warmth, was not associated with fewer posttraumatic stress symptoms. There may be several reasons for this. The adolescents in this study were at a transitional age when it is developmentally expected to become less family and more peer oriented [57], hence diminishing the role of parental support. Longitudinal research similarly suggests that the resilience obtained by parental support declines with time, as school becomes a more prominent factor in a young person’s life [26]. In addition, parental unawareness of children’s exposure to community violence has been reported previously [58] and may have contributed to the lack of influence from parental warmth and involvement. One main finding was that perceived teacher support had a protective effect against severe posttraumatic stress. To our knowledge, this relationship has not been demonstrated before. Rather, Ozer and Weinstein [12] previously found a lack of correlation between teacher support and posttraumatic stress. A possible explanation for this difference is that Ozer and Weinstein’s study was conducted with younger adolescents, in a school-based, ethnically diverse urban setting. There may also be a difference in the nature of the teacher–student relationship in suburban middleclass schools compared to in inner-city schools, where the student population is often more challenging, as suggested by higher teacher attrition rates in schools in low-socioeconomic and minority areas [55, 56]. It is notable that our findings concern a natural course of events and not the effects of any specific intervention. Targeted intervention studies involving teacher support would probably have yielded an even stronger protective effect against posttraumatic stress. Limitations The cross-sectional design of this study prevents us from drawing any conclusions about causality. Hence, it remains unclear whether prior depressive symptomatology
predisposes victimized youth to have posttraumatic stress reactions or whether depressive symptomatology is simply a response to victimization that is comorbid with posttraumatic stress. Another limitation of this study is the reliance on adolescents’ self-reports as a single source of information. When it comes to the reporting of exposure to violence, however, adolescents themselves tend to be the best informants, as parents may often underestimate the extent of violence exposure [38]. Although, many important variables affecting posttraumatic stress were taken into account, this study could not address the issues of proximity to violence or the identity of the victim, which can also affect the psychological response to witnessing violence [19, 20, 39]. In addition, domestic violence which correlates with community violence [47] was not measured. Clinical implications Similar to urban youth, many young people living in suburban areas in North American settings may be affected by community violence and require clinical attention, since a substantial proportion of exposed youth report severe posttraumatic stress. Additional clinical attention is called for in the presence of depressive symptoms, since there is a strong association between such symptoms and posttraumatic stress among victimized youth. This finding is of particular importance considering that comorbid depression substantially increases the risk for suicide among individuals with posttraumatic stress [53, 60]. Teacher support seems to be an important protective factor, but its role needs to be further evaluated using prospective study designs. Future interventions targeting the negative effects of community violence should consider teacher-based approaches. Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest. Ethical standards The study has been approved by the appropriate ethics committee and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.
References 1. Buka SL, Stichick TL, Birdthistle I, Earls FJ (2001) Youth exposure to violence: prevalence, risks, and consequences. Am J Orthopsychiatry. doi:10.1037/0002-9432.71.3.298 2. Stein BD, Jaycox LH, Kataoka S, Rhodes HJ, Vestal KD (2003) Prevalence of child and adolescent exposure to community violence. Clin Child Fam Psychol Rev. doi:10.1023/B:CCFP. 0000006292.61072.d2 3. Aisenberg E, Herrenkohl T (2008) Community violence in context: risk and resilience in children and families. J Interpers Violence. doi:10.1177/0886260507312287
123
Soc Psychiatry Psychiatr Epidemiol 4. Cooley-Strickland M, Quille TJ, Griffin RS, Stuart EA, Bradshaw CP, Furr-Holden D (2009) Community violence and youth: affect, behavior, substance use, and academics. Clin Child Fam Psychol Rev. doi:10.1007/s10567-009-0051-6 5. Dahlberg LL, Mercy JA (2009) History of violence as a public health problem. Virtual Mentor. doi:10.1001/virtualmentor.2009. 11.2.mhst1-0902 6. Finkelhor D, Turner H, Ormrod R, Hamby SL (2009) Violence, abuse, and crime exposure in a national sample of children and youth. Pediatrics. doi:10.1542/peds.2009-0467 7. Kuther TL, Fisher CB (1998) Victimization by community violence in young adolescents from a suburban city. J Early Adolesc. doi:10.1177/0272431698018001003 8. Lynch M (2003) Consequences of children’s exposure to community violence. Clin Child Fam Psychol Rev. doi:10.1023/B: CCFP.0000006293.77143.e1 9. Cooley-Quille M, Boyd RC, Frantz E, Walsh J (2001) Emotional and behavioral impact of exposure to community violence in inner-city adolescents. J Clin Child Adolesc Psychol 30(2): 199–206 10. Freeman LN, Mokros H, Poznanski EO (1993) Violent events reported by normal urban school-aged children: characteristics and depression correlates. J Am Acad Child Adolesc Psychiatry. doi:10.1097/00004583-199303000-00025 11. Fitzpatrick KM, Boldizar JP (1993) The prevalence and consequences of exposure to violence among African-American youth. J Am Acad Child Adolesc Psychiatry. doi:10.1097/00004583199303000-00026 12. Ozer EJ, Weinstein RS (2004) Urban adolescents’ exposure to community violence: the role of support, school safety, and social constraints in a school-based sample of boys and girls. J Clin Child Adolesc Psychol. doi:10.1207/s15374424jccp3303_4 13. Henrich CC, Schwab-Stone M, Fanti K, Jones SM, Ruchkin V (2004) The association of community violence exposure with middle-school achievement: a prospective study. J Appl Dev Psychol. doi:10.1016/j.appdev.2004.04.004 14. Gorman-Smith D, Tolan P (1998) The role of exposure to community violence and developmental problems among inner-city youth. Psychopathol Dev. doi:10.1017/S0954579498001539 15. Schwab-Stone M, Chen C, Greenberger E, Silver D, Lichtman J, Voyce C (1999) No safe haven. II: The effects of violence exposure on urban youth. J Am Acad Child Adolesc Psychiatry 38(4):359–367 16. Singer MI, Anglin TM, Song LY, Lunghofer L (1995) Adolescents’ exposure to violence and associated symptoms of psychological trauma. JAMA. doi:10.1001/jama.1995.035203000 51036 17. Foster JD, Kuperminc GP, Price AW (2004) Gender differences in posttraumatic stress and related symptoms among inner-city minority youth exposed to community violence. J Youth Adolesc 33:59–69 18. Gladstein J, Rusonis EJ, Heald FP (1992) A comparison of innercity and upper-middle class youths’ exposure to violence. J Adolesc Health. doi:10.1016/1054-139X(92)90159-9 19. Zinzow HM, Ruggiero KJ, Resnick H, Hanson R, Smith D, Saunders B, Kilpatrick D (2009) Prevalence and mental health correlates of witnessed parental and community violence in a national sample of adolescents. J Child Psychol Psychiatry. doi:10.1111/j.1469-7610.2008.02004.x 20. White KS, Bruce SE, Farrell AD, Kliewer W (1998) Impact of exposure to community violence on anxiety: a longitudinal study of family social support as a protective factor for urban children. J Child Fam Stud. doi:10.1023/A:1022943216319 21. Kliewer W, Cunningham JN, Diehl R, Parrish KA, Walker JM, Atiyeh C, Neace B, Duncan L, Taylor K, Mejia R (2004) Violence exposure and adjustment in inner-city youth: child and
123
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37. 38.
39.
40.
caregiver emotion regulation skill, caregiver-child relationship quality, and neighborhood cohesion as protective factor. J Clin Child Adolesc Psychol. doi:10.1207/s15374424jccp3303_5 Hammack PL, Richards MH, Luo Z, Edlynn ES, Roy K (2004) Social support factors as moderators of community violence exposure among inner-city African American young adolescents. J Clin Child Adolesc Psychol. doi:10.1207/s15374424jccp3303_3 Ozer EJ (2005) The impact of violence on urban adolescents longitudinal effects of perceived school connection and family support. J Adolesc Res. doi:10.1177/0743558404273072 Barr SC, Hanson R, Begle AM, Kilpatrick DG, Saunders B, Resnick H, Amstadter A (2012) Examining the moderating role of family cohesion on the relationship between witnessed community violence and delinquency in a national sample of adolescents. J Interpers Violence. doi:10.1177/0886260511416477 Resnick BP (1997) Protecting adolescents from harm: findings from the national longitudinal study on adolescent health. JAMA. doi:10.1001/jama.1997.03550100049038 O’Donnell DA, Schwab-Stone ME, Muyeed AZ (2003) Multidimensional resilience in urban children exposed to community violence. Child Dev. doi:10.1111/1467-8624.00471 Aneshensel CS, Sucoff CA (1996) The neighborhood context of adolescent mental health. J Health Soc Behav. doi:10.2307/ 2137258 Overstreet S, Braun S (2000) Exposure to community violence and post-traumatic stress symptoms: mediating factors. Am J Orthopsychiatry 70(2):263–271 Richards MH, Larson R, Miller BV, Luo Z, Sims B, Parrella DP, McCauley C (2004) Risky and protective contexts and exposure to violence in urban African American young adolescents. J Clin Child Adolesc Psychol. doi:10.1207/S15374424JCCP3301_13 Osgood DW, Wilson JK, O’Malley PM, Bachman JG, Johnston LD (1996) Routine activities and individual deviant behavior. Am Sociol Rev. doi:10.1037/h0087828 Pettit GS, Bates JE, Dodge KA, Meece DW (1999) The impact of after-school peer contact on early adolescent externalizing problems is moderated by parental monitoring, perceived neighborhood safety, and prior adjustment. Child Dev. doi:10.1111/ 1467-8624.00055 Gardner M, Brooks-Gunn J (2009) Adolescents’ exposure to community violence: are neighborhood youth organizations protective? J Community Psychol. doi:10.1002/jcop.20310 Gottfredson DC, Gerstenblith SA, Soule´ DA, Womer SC, Lu S (2004) Do after school programs reduce delinquency? Sci Prev. doi:10.1177/0886260508314297 Hoffmann JP, Xu J (2002) School activities, community service, and delinquency. Crime Delinquency. doi:10.1177/00111280 2237130 Weissberg RP, Voyce CK, Kasprow WJ, Arthur MW, Shriver TP (1991) The social and health assessment. Yale Child Study Center, Chicago Schwab-Stone M, Ayers TS, Kasprow W, Voyce C, Barone C, Shriver T, Weissberg RP (1995) No safe haven: a study of violence exposure in an urban community. J Am Acad Child Adolesc Psychiatry. doi:10.1097/00004583-199510000-00020 JE Richters, W Saltzman (1990) Survey of exposure to community violence: self-report version. JE Richters Richters JE, Martinez P (1993) The NIMH community violence project: Vol I. Children as victims of and witnesses to violence. Psychiatry 56:7–21 Pynoos RS, Frederick C, Nader K, Arroyo W, Steinberg A, Eth S, Nunez F, Fairbanks L (1987) Life threat and posttraumatic stress in school-age children. Arch Gen Psychiatry. doi:10.1001/arch psyc.1987.01800240031005 Pynoos RS, Goenjian A, Tashjian M, Karakashian M, Manjikian R, Manoukian G, Steinberg AM, Fairbanks LA (1993) Post-
Soc Psychiatry Psychiatr Epidemiol
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
traumatic stress reactions in children after the 1988 Armenian earthquake. Br J Psychiatry 163:239–247 Radloff LS (1977) The CES-D scale A self-report depression scale for research in the general population. Appl Psychol Meas. doi:10.1177/014662167700100306 Roberts RE, Lewinsohn PM, Seeley JR (1991) Screening for adolescent depression: a comparison of depression scales. J Am Acad Child Adolesc Psychiatry. doi:10.1097/00004583-19910 1000-00009 Andrykowski JSCMA, Mary JWLAH (1998) Psychometrics for two short forms of the Center for Epidemiologic StudiesDepression Scale. Issues Ment Health Nurs. doi:10.1080/0161 28498248917 Ruchkin V, Schwab-Stone M, Vermeiren R (2004) Social and health assessment (SAHA): Psychometric development summary. Yale Univ, N. Hav Hawkins JD, Catalano RF, Miller JY (1992) Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: implications for substance abuse prevention. Psychol Bull 112:64 African-American youth, Fitzpatrick KM (1993) Exposure to violence and presence of depression among low-income. J Consult Clin Psychol. doi:10.1037/0022-006X.61.3.528 Margolin G, Gordis EB (2000) The effects of family and community violence on children. Annu Rev Psychol. doi:10.1146/ annurev.psych.51.1.445 Kilpatrick DG, Ruggiero KJ, Acierno R, Saunders BE, Resnick HS, Best CL (2003) Violence and risk of PTSD, major depression, substance abuse/dependence, and comorbidity: results from the National Survey of Adolescents. J Consult Clin Psychol 71:692–700 Breslau N (1997) Sex differences in posttraumatic stress disorder. Arch Gen Psychiatry. doi:10.1001/archpsyc.1997.018302300 82012 Anderson ER, Mayes LC (2010) Race/ethnicity and internalizing disorders in youth: a review. Clin Psychol Rev . doi:10.1016/j. cpr.2009.12.008
51. Basch CE (2011) Aggression and violence and the achievement gap among urban minority youth. J Sch Health. doi:10.1111/j. 1746-1561.2011.00636.x 52. Garber J, Weersing VR (2010) Comorbidity of anxiety and depression in youth: implications for treatment and prevention. Clin Psychol. doi:10.1111/j.1468-2850.2010.01221.x 53. Oquendo M, Brent DA, Birmaher B, Greenhill L, Kolko D, Stanley B, Zelazny J, Burke AK, Firinciogullari S, Ellis SP, Mann JJ (2005) Posttraumatic stress disorder comorbid with major depression: factors mediating the association with suicidal behavior. Am J Psychiatry 162(3):560–566 54. Haden SC, Scarpa A (2008) Community violence victimization and depressed mood: the moderating effects of coping and social support. J Interpers Violence. doi:10.1177/0886260508314297 55. Borman GD, Dowling NM (2008) Teacher attrition and retention: a meta-analytic and narrative review of the research. Educ Res Rev. doi:10.3102/0034654308321455 56. Shen J (1997) Teacher retention and attrition in public schools: evidence from SASS91. J Educ Res. doi:10.1080/00220679 709597525 57. Fuligni AJ, Eccles JS (1993) Perceived parent-child relationships and early adolescents’ orientation toward peers. Psychol Dev. doi:10.1037/0012-1649.29.4.622 58. Hill HM, Jones LP (1997) Children’s and parents’ perceptions of children’s exposure to violence in urban neighborhoods. J Natl Med Assoc 89:270 59. Cunningham M (1999) African American adolescent males’ perceptions of their community resources and constraints: a longitudinal analysis. J Community Psychol. doi:10.1002/ (SICI)1520-6629(199909)27:5\569:AID-JCOP5[3.0.CO;2-6 60. Panagioti M, Gooding PA, Tarrier N (2012) A meta-analysis of the association between posttraumatic stress disorder and suicidality: the role of comorbid depression. Compr Psychiatry. doi:10.1016/j.comppsych.2012.02.009
123
ORIGINAL PAPER
Community violence exposure and severe posttraumatic stress in suburban American youth: risk and protective factors Sandra Lo¨fving–Gupta • Frank Lindblad Andrew Stickley • Mary Schwab-Stone • Vladislav Ruchkin
•
Received: 18 June 2014 / Accepted: 4 October 2014 Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose The psychological effects of community violence exposure among inner-city youth are severe, yet little is known about its prevalence and moderators among suburban middle-class youth. This study aimed to assess the prevalence of community violence exposure among suburban American youth, to examine associated posttraumatic stress and to evaluate factors related to severe vs. less severe posttraumatic stress, such as co-existing internalizing and externalizing problems, as well as the effects of teacher support, parental warmth and support, perceived neighborhood safety and conventional involvement in this context. Method Data were collected from 780 suburban, predominantly Caucasian middle-class high-school adolescents in the Northeastern US during the Social and Health Assessment (SAHA) study. Results A substantial number of suburban youth were exposed to community violence and 24 % of those victimized by community violence developed severe posttraumatic stress. Depressive symptoms were strongly
S. Lo¨fving–Gupta F. Lindblad V. Ruchkin (&) Child and Adolescent Psychiatry Unit, Department of Neuroscience, Uppsala University, 751 85 Uppsala, Sweden e-mail: [email protected] A. Stickley Stockholm Centre on Health of Societies in Transition (SCOHOST), So¨derto¨rn University, Huddinge, Sweden M. Schwab-Stone V. Ruchkin Child Study Center, Yale University Medical School, New Haven, CT 06520, USA V. Ruchkin Sa¨ter Forensic Psychiatric Clinic, 783 27 Sa¨ter, Sweden
associated with higher levels and perceived teacher support with lower levels of posttraumatic stress. Conclusion Similar to urban youth, youth living in suburban areas in North American settings may be affected by community violence. A substantial proportion of these youth reports severe posttraumatic stress and high levels of comorbid depressive symptoms. Teacher support may have a protective effect against severe posttraumatic stress and thus needs to be further assessed as a potential factor that can be used to mitigate the detrimental effects of violence exposure. Keywords Community violence exposure Internalizing problems Protective factors Adolescents
Introduction The term community violence is used to signify acts by individuals or groups intended to harm others, such as chasing, threatening, beating, robbing, mugging, raping, shooting, stabbing or killing, and experienced outside home [1]. Direct exposure implies being a victim or a target of violence oneself, whereas indirect exposure means presence at the scene of violence and direct witnessing of it [1, 2]. During the 1990s community violence was described as an epidemic of national proportions in the United States [1]. Although it has since decreased [2] it is still regarded as a major public health problem [3–5]. According to a recent national survey [6] 60 % of children and youth in the US had experienced violence during the past year in their homes, schools or communities. However, the prevalence of community violence tends to vary dramatically depending on the geographical location, socio-economical status, age, race and gender, with higher rates of exposure
123
Soc Psychiatry Psychiatr Epidemiol
to community violence in urban, low-income and ethnic minority communities [2]. Little is known about the prevalence and moderators of community violence among youth living in suburban areas [7], as most studies concerning community violence have focused on so-called high-risk populations, such as ethnic minorities living in impoverished areas [4], with a plethora of additional risk factors that may affect child development and the ability to cope with stress [8]. Given this, it remains unclear to what extent the previous research findings on community violence exposure can be applied to youth living in suburban and more affluent areas. Exposure to community violence has been linked to internalizing outcomes such as anxiety and depression [9, 10], and posttraumatic stress [11, 12], but also to externalizing behavior such as substance use, school failure [4, 10, 13], delinquency, and aggressive behavior [14–16]. Research has also demonstrated that outcomes may differ depending on gender [11, 17, 18] age [2] proximity to violence [19] and other factors. A number of studies have attempted to identify factors within the social environment that may decrease the risk of experiencing the negative consequences of community violence. Findings regarding the extent to which family and parenting characteristics moderate the effects of community violence on psychopathology have been inconsistent. Some studies have indicated that a lack of family support may increase the risk of internalizing symptoms while high family support does not seem to buffer the negative effects of exposure to violence [20]. Other studies suggest that a supportive family does help when the exposure is mild, but when it is more extensive or frequent, such support and caring have little or no effect [21]. Longitudinal studies [22, 23] have indicated that high family support may buffer against an increase in internalizing symptoms even where there is an increase in violence exposure and that a high level of family cohesion has a moderating effect on delinquent behavior among adolescents exposed to community violence [24]. There is similar uncertainty about the role of school factors. Several studies have suggested that school connectedness is linked to lower levels of emotional distress and better psychological outcomes and that school support, defined by attachment to school, teacher support and academic motivation, has a protective role among those students who have not been exposed to violence [25, 26]. However, other studies have concluded that, among those exposed to community violence, perceived school support does not moderate symptoms of depression and posttraumatic stress [12, 23]. The findings regarding the role of the neighborhood have also been inconsistent. Youths’ perception of the neighborhood as dangerous and unsafe has been linked to more symptoms of depression, anxiety and conduct
123
disorder [27], while the perception of neighborhood safety mediates the association between exposure to violence and posttraumatic stress symptoms [28]. Among youth living in areas with high levels of violence exposure, a higher degree of perceived neighborhood cohesion and safety is associated with less internalizing symptoms. Yet, a positive perception of the neighborhood did not protect against the adverse effects of community violence exposure [21]. Unstructured socializing and unmonitored free time with no authority figures present has been positively associated with violence exposure, internalizing symptoms [29] and externalizing symptoms [30, 31]. In contrast, access to various resources such as neighborhood youth organizations has been described as protective against the impact of community violence exposure [32]. Moreover, some research has suggested that participation in communitybased activities is negatively associated with risk factors for community violence exposure such as delinquency and substance abuse [33, 34]. Hence in this study, we will assess the prevalence of community violence exposure among adolescents in a suburban middle-class community, as well as evaluate posttraumatic stress associated with such exposure in this group. Further, we will also examine the potential risk (coexisting internalizing and externalizing problems) and protective (family and teacher support, neighborhood perception and conventional involvement) factors associated with the development of severe posttraumatic stress among those victimized by violence. Youth coming from suburban areas would be expected to report lower levels of exposure to community violence than their urban counterparts. However, those, who are exposed would report symptoms of posttraumatic stress in a fashion similar to that described in other studies of community violence exposure, that is, those directly victimized would report higher levels of posttraumatic stress than those who have only witnessed community violence [11], and those who report higher levels of posttraumatic stress would also report higher levels of comorbid anxiety and depressive symptoms. Although we expect that family, school, neighborhood and community involvement may protect against the development of posttraumatic stress symptoms, considering the equivocal findings from previous research, we will refrain from testing any specific hypotheses in this regard and limit our efforts to exploratory analyses.
Methods Participants Data were collected from a large-scale study of risk and protective factors for problem behaviors in youth, the
Soc Psychiatry Psychiatr Epidemiol
Social and Health Assessment (SAHA). The survey methodology has been described previously [13, 15] although earlier reports concerned inner-city youth in New Haven, CT, USA. Although Connecticut has long ranked at or near the top of the US most prosperous states, New Haven remains one of the country’s poorest cities, with 30 % of its population living beneath the poverty line, according to 2011 census data (median household income $38,963). In contrast, the current study focused on youth residing in the same geographic area, but living in four small towns around New Haven with higher income status (mean household income $93,410). Data were collected in the spring of 2003 in all public schools in four suburban municipalities from the greater New Haven area, where a total of 780 sixth and eighth graders completed the SAHA. Of those, 23 students were excluded from the analyses because of inconsistent or incomplete reporting on the scales of interest. In comparison with the study group, no significant differences were obtained on any variable of interest, except that the excluded group perceived their schools (19.65 ± 3.71 vs. 16.54 ± 4.74) and their neighborhoods (22.91 ± 4.13 vs. 21.04 ± 3.74) as being less safe. Participants (N = 757) ranged in age from 13 to 16 years (M = 14.47 ± 1.18). The composition of the final sample was 53.7 % female (N = 405) and the ethnic composition was predominantly Caucasian (84.5 % Caucasian, 6.2 % Asian American, 3.2 % African American, 1.7 % Hispanic American, 0.3 % American Indian and 4.0 % Other), an accurate reflection of the local public school population. Nearly 85 % of youth reported their parents as being married or remarried. The majority of parents had college education (78.5 % of fathers and 81.5 % of mothers) and were full-time or part-time employed (92.1 % of fathers and 82.5 % of mothers). Measures The Social and Health Assessment (SAHA) [15, 35, 36] served as the basis for the survey. The scales used in the present study are described in detail below. Community violence exposure was assessed by a modified version of the instrument developed by Richters and Saltzman [37], and includes two subsets of items, one assessing direct victimization and the other witnessing of community violence, using a 5-point scale response format (from 0 times = 0 to 10 or more times = 4). Victimization by violence was assessed by six items asking whether in the past year in their community respondents had been beaten up or mugged, threatened with serious physical harm by someone, shot, attacked or stabbed with a knife, chased by gangs or individuals, or seriously wounded in an incident of violence, with a total score ranging from 0 to 24.
Similarly, the witnessing violence index consisted of six items asking the respondents whether or not they had seen someone being victimized by the same types of violence in their community, with the total score also ranging from 0 to 24. These indices have demonstrated good psychometric properties with American inner-city youth [38]. Posttraumatic stress was measured by the Child PostTraumatic Stress Reaction Index (CPTS-RI), a 20-item scale, assessing posttraumatic stress symptoms in schoolaged children and adolescents after exposure to trauma [39, 40]. The frequency of the symptoms in the past month is assessed on a Likert-type five-point rating scale ranging from ‘‘never’’ (0) to ‘‘most of the time’’ (4), with a total score range from 0 to 80. The scale is highly correlated with the DSM-based diagnosis of posttraumatic stress and has well-established clinical cutoffs. A score between 12 and 24 indicates mild posttraumatic stress, a score of 25–39 indicates moderate posttraumatic stress, 40–59 severe posttraumatic stress, and a score of 60 and above indicates very severe posttraumatic stress [40]. The Cronbach a obtained for the scale was adequate (0.86). Depressive symptoms were assessed using an adaptation of the Center for Epidemiological Studies-Depression Scale (CES-D) [41]. Both the CES-D e.g. [42] and its modified versions [43] have demonstrated excellent psychometric properties with adolescents. In the present study, the questionnaire consisted of ten items (e.g. not liking oneself; loss of interest in other people or things). Students reported on symptom presence for the past month on a three-point scale [Not True (0), Somewhat True (1), Certainly True (2)]. The scale had good internal consistency (a = 0.80). Anxiety symptoms were assessed by a 12-item scale [44], reflecting cognitive–affective and behavioral modes of anxiety, including worrisome, preoccupying thoughts or unpleasant feelings (e.g. worrying about being liked; feeling nervous when called on in class; worrying about the future). Students reported on the presence of anxiety symptoms on a three-point scale (Not True = 0, Somewhat True = 1, Certainly True = 2). The scale had a good internal consistency (a = 0.86). Violent and aggressive behavior of the respondents in the past year was assessed by six items from the SAHA, including starting a fist fight or shoving match, hurting someone so badly in a fight that treatment by a doctor was needed, carrying a gun, being in a gang or posse fight, being arrested by the police, and carrying a blade, knife or gun in school. These six items were selected from a larger pool of conduct problems and antisocial behavior items to reflect the types of violence also assessed by the witnessing and victimization measures. The selection of these six items was further supported by factor analysis. Responses were tallied on a five-point scale (ranging from 0 times to 5
123
Soc Psychiatry Psychiatr Epidemiol
or more times), producing a total score range from 0 to 24. The scale had good internal consistency (Cronbach a = 0.81). Perceived teacher support was assessed by eight items (e.g. teachers are willing to help students, teachers notice when I am doing a good job; teachers show concern when a student is absent from school, etc) adapted from Hawkins et al. [45] and developed by the creators of the original survey, Weissberg et al. [35]. The respondents were asked to report on a four-point scale how true for them each of the statements was [ranging from ‘‘definitely not true’’ (1) to ‘‘definitely true’’ (4)]. Cronbach a for the scale was 0.71. Parenting included two scales, assessing parental warmth and parental involvement with each scale item being scored on a four-point scale, ranging from ‘‘Never’’ (1) to ‘‘Often’’ (4). The scales were developed by the SAHA Research Evaluation Team [44]. Parent Involvement consists of six items, assessing youth perceptions of the degree to which their parents and/or primary guardians are involved and interested in their lives (e.g., the parent asks about her/his life, encourages interest in different things, gives good advice, etc). A good level of internal consistency was obtained (a = 0.80). Parental warmth includes five items that ask about students’ perceptions of their parents’ warmth and support for them (e.g. parents or guardians being kind to him/her, hugs or kisses him/her, shows their love for him/her, etc). A good level of internal consistency was obtained (a = 0.82). Conventional involvement was assessed by a six-item scale that asked the respondent about his/her participation in various kinds of group spare time activities. The respondents rated how many times during a regular week they participated in such activities as organized sports, dance, art, youth clubs, after school programs, etc. The frequency was assessed on a five-point scale ranging from ‘‘None’’ (0) to ‘‘6–7 times’’ (4). The scale was developed by the SAHA Research Evaluation Team [44]. Perception of the neighborhood was assessed by the Attachment to Neighborhood Scale [44] that inquires about individual perceptions of the neighborhood in which the respondent lives. The scale consists of seven positively scored items (examples include ‘‘My neighborhood looks nice’’; ‘‘I like spending time in my neighborhood’’; ‘‘People on the streets in my neighborhood are friendly’’). All items are scored on a four-point scale ranging from ‘‘Definitely not true’’ (1) to ‘‘Definitely true’’ (4). Cronbach’s alpha for the present sample was 0.82. Proxy for SES was computed as a sum, based on students’ reports regarding the family status of their parents (two parent families, 1/0), employment status of each parent (1/0), and education level of each parent (college education, 1/0), hence, potentially ranging from 0 to 5. This method has been used for assessing SES previously, e.g. [13].
123
Procedure Parents were informed of the survey at the time of school registration, received a letter about the survey 2 weeks prior to the administration of the questionnaire, and were offered the opportunity to decline participation. The passive informed consent procedure was approved by the university’s IRB and considered as an appropriate ethical procedure by the legislation of the State of Connecticut. Prior to survey administration, students were read a detailed assent form outlining their participation with assurances of confidentiality, and were asked to sign it to indicate assent (parent and child refusals were \1 %). Students completed the survey in a classroom setting during one class period during the regular school day. Trained administrators read all questions aloud while students followed along with their copies of the survey, reading questions to themselves and circling responses in the booklet. A second administrator was available, providing help to individual students as requested. The teacher was also present in the class, which minimized any problems related to classroom management. Make-up administrations were performed at each school within 1 month of the initial administration for those who had been absent. Data analysis The data were analyzed using the Statistical Package for Social Sciences (SPSS-22.0). Chi-square tests and one-way ANOVA tests were employed to assess the differences between the groups. Binary logistic regression analyses were conducted to identify the variables that could discriminate between the groups with severe vs. less severe posttraumatic stress. A proxy for SES, gender, age and ethnic minority status (1/0) was entered in Step 1. Depressive and anxiety symptoms, as well as aggressive and violent behavior were entered in Step 2. Considering previous research about the dose-related effects of trauma [46], the victimization score was added into the equation in Step 3, followed by Step 4, in which all of the environmental factors were entered, including those relating to school (school safety and teacher support), parenting (parental warmth and parental support), and perception of the neighborhood. In an attempt to reduce the number of variables in the equation, only those variables that were significantly different between the groups were used in the model and hence, conventional involvement was excluded from the analysis.
Results The study group (N = 757) was divided according to the reported severity of exposure. Those who did not report
Soc Psychiatry Psychiatr Epidemiol Table 1 Prevalence of different types of witnessing and victimization [N (%)] Witnessing group (261)
Victimization group (182)
v2
Total study group (757)
156 (20.6)
Witnessing (in the past year I have seen…) Someone else being chased by gangs or individuals Someone else get threatened with serious physical harm Someone else getting beaten up or mugged Someone else being attacked or stabbed with a knife A seriously wounded person after an incident of violence Someone else get shot or shot at with a gun Victimization (in the past year I have been…)
70 (26.8)
86 (47.3)
194.9
190 (72.8)
145 (79.7)
501.3
335 (44.3)
72 (27.6)
86 (47.3)
194.5
158 (20.9)
8 (3.1)
25 (13.7)
60.7
33 (4.4)
46 (17.6)
61 (33.5)
124.6
107 (14.1)
12 (4.6)
22 (12.1)
42.8
34 (4.4)
Chased by gangs or individuals Threatened with serious physical harm by someone Beaten up or mugged
52 (28.6)
52 (6.9)
135 (74.2)
135 (17.8)
38 (20.9)
38 (5.0)
Attacked or stabbed with a knife
14 (7.7)
14 (1.8)
Seriously wounded in an incident of violence
17 (9.3)
17 (2.2)
Shot or shot at with a gun
17 (9.1)
17 (2.2)
All Chi-square test comparisons were significant at p \ 0.001
any witnessing or victimization episodes were considered as the non-exposed group (41.5 %). Those, who reported at least one episode of witnessing, but no episodes of victimization were considered as the witnessing group (34.5 %). Finally, those, who reported at least one episode of victimization, were considered the victimization group (24.0 %). The prevalence of different types of exposure is described in detail in Table 1. As regards the prevalence of severe posttraumatic stress between the groups, the victimization group reported the highest prevalence of severe posttraumatic stress 43/182 (23.6 %), when compared with the nonexposed group 5/314 (1.6 %) and witnessing group 8/261 (3.1 %) (v2 = 92.19, p \ 0.001). Since the last two groups did not differ significantly from each other on the prevalence of severe posttraumatic stress (v2 = 1.37, p = 0.242) the following analysis focused only on the victimization group (N = 182). Hence, those youths who reported at least one episode of victimization by community violence during the last year were divided into two groups based on their reported CPTS-RI score, with scores of 40 or higher (indicative of severe or very severe posttraumatic stress), or 39 and below (indicative of moderate or lower degree of posttraumatic stress) being used as cutoff points. Binary logistic regression analyses were conducted with posttraumatic stress (severe vs. less severe) as the dependent variable in four steps. As shown in Table 2 (where only significant predictors are shown; p \ 0.05), Step 1 demonstrated that minority status and gender had a moderate predictive power, explaining 7.7 % of the variance. In
Step 2 depressive symptoms and aggressive and violent behavior predicted posttraumatic stress, with the predictive power of minority status becoming non-significant. In Step 3 victimization further added 2 % to the predictive power of the model. At this step, the predictive power of aggressive and violent behavior became non-significant. Finally, environmental factors increased the predictive power of the model by an additional 5 %. The final model consisted of the following factors: gender, depressive symptoms, victimization, and teacher support. This model fits the data well (v2 = 97.44; df = 6; p \ 0.001), explaining 43 % of the variance (Cox & Snell R2).
Discussion In this study of public school students from a suburban middle-class area we have found a substantial number of youth reporting at least some degree of exposure to community violence and a large proportion (24 %) of those victimized by community violence also reported severe levels of posttraumatic stress. Depressive symptoms were strongly associated with severe posttraumatic stress. Perceived teacher support was associated with lower levels of posttraumatic stress. In urban settings earlier research has reported that up to one-third of children have been directly victimized and almost all children have witnessed community violence [47]. In our study, the prevalence of exposure was lower than that in urban areas, yet substantial, with one-quarter of youth reporting that they were victims of violence during
123
Soc Psychiatry Psychiatr Epidemiol Table 2 Binary logistic regression analysis predicting severe posttraumatic stress in the victimized group (N = 182)
Step 1
Variable
R2
Gender (male)
0.077
R2D
Ethnicity status (minority) Only those variables that were significant at different steps of analysis are depicted (p \ 0.05). Victimization (dose) refers to a sum of the number and frequency of each experienced episode of exposure
2
123
OR (95 % CI) at entry
Final OR (95 % CI)
-1.48 (0.665)
0.343 (0.165–0.714)
0.227 (0.062–0.836)
0.120 (0.639)
2.45 (1.12–5.34)
1.13 (0.322–3.95)
0.355 (0.065)
1.36 (1.22–1.50)
1.43 (1.25–1.62)
-0.090 (0.134)
1.23 (1.06–1.43)
0.914 (0.703–1.19)
1.21 (1.03–1.43)
1.24 (1.04–1.48)
0.79 (0.690–0.898)
0.79 (0.690–0.898)
Depressive symptoms Aggressive and violent behavior
0.354
3
Victimization (dose)
0.376
0.022
0.214 (0.089)
4
Teacher support
0.427
0.051
-0.239 (0.067)
the past year. However, these figures are also lower than in previous studies from suburban areas. In a study of sixth to eighth graders in a suburban private school, more than twothirds of the children reported victimization during the past year [7] Similarly, 55 % of the middle-class youth aged 12–24 in a study by Gladstein et al. [18] reported experiencing victimization during the past year. These differences might in part, be explained by the use of a broader definition of community violence in the previous studies. Another factor contributing to this difference is that the present study concerned younger adolescents, who are generally less likely to have been exposed to violence [2]. Severe levels of posttraumatic stress symptoms were reported by 6.7 % of the study sample, indicating that potential PTSD levels in this group were higher than the average rate of PTSD observed among adolescents in the US general population, (3.7 % for boys and 6.3 % for girls) [48]. However, the present study focused on the selfreported prevalence of posttraumatic symptoms only, without any clinical assessment of the symptoms themselves or the functional impairment, needed for establishing a clinical diagnosis, which may explain the differences in the prevalence rates. Previous research has suggested that both violent victimization and witnessing violence are associated with posttraumatic stress among urban youth [11]. In the present study, the highest prevalence of severe posttraumatic stress (24 %) was reported by youth directly victimized by violence, which is in line with earlier findings. We found no significant difference between the witnessing group and the non-exposed group regarding the prevalence of severe posttraumatic stress. The amount of violence exposure, both in terms of frequency and the number of different types of exposure, also correlated with an increased number of posttraumatic stress symptoms, which is in line with previous studies about the dose-related effect of violence on the severity of posttraumatic stress [46].
0.277
Final B (SE)
Among the demographic variables neither age nor socioeconomic status was significantly associated with severe posttraumatic stress. Consistent with previous research [49] there was a gender difference in the association with severe posttraumatic stress with a higher prevalence of severe posttraumatic stress being found among girls. Previous studies [50] have suggested that belonging to an ethnic minority can represent a risk factor for severe posttraumatic stress. We also found that ethnic status was initially related to severe posttraumatic stress in the regression analysis. However, when symptoms of depression and aggressive behavior were added into the model, this association with posttraumatic stress became non-significant. This finding can be potentially explained by the results from previous research which indicates that there are generally higher levels of internalizing and externalizing problems in ethnic minority youth [50, 51]. Depressive symptoms were associated with more severe posttraumatic stress symptoms suggesting comorbidity [52, 53] which seems to be especially evident in youth exposed to violence [48]. Some research [54] suggests that individuals with posttraumatic stress tend to perceive comorbid depressive symptoms as being more frequent and/or distressing, and the environment as less safe and less supportive [54], and yet, perceived teacher support was negatively associated with posttraumatic stress, even when depressive symptoms were controlled for, suggesting its potential protective effect against posttraumatic stress. Another important risk factor that was examined in the analysis was aggressive and violent behavior. As suggested by Gorman-Smith and Tolan [14] those youth who commit violent acts themselves are more likely to report higher levels of exposure to violence. They are also at higher risk to become victims of violence and hence are at an increased risk of developing posttraumatic stress. Indeed, in the present study we initially found a significant association between delinquent behavior and posttraumatic
Soc Psychiatry Psychiatr Epidemiol
stress, which became non-significant, however, when the degree of exposure was taken into account. Perceived neighborhood safety was not associated with severe posttraumatic stress. However, the relationship between perception and the actual nature of a neighborhood may be weak [59], which may have obscured an association between neighborhood safety and posttraumatic stress. There was no significant difference between the three study groups regarding participation in conventional activities and neither was conventional involvement associated with posttraumatic stress. Richards et al. [29] reported that the benefits of conventional involvement are particularly pronounced for children in low-income neighborhoods with high crime rates, implying that the effects in our sample from an affluent area may be limited. Family support, as measured by parental involvement and warmth, was not associated with fewer posttraumatic stress symptoms. There may be several reasons for this. The adolescents in this study were at a transitional age when it is developmentally expected to become less family and more peer oriented [57], hence diminishing the role of parental support. Longitudinal research similarly suggests that the resilience obtained by parental support declines with time, as school becomes a more prominent factor in a young person’s life [26]. In addition, parental unawareness of children’s exposure to community violence has been reported previously [58] and may have contributed to the lack of influence from parental warmth and involvement. One main finding was that perceived teacher support had a protective effect against severe posttraumatic stress. To our knowledge, this relationship has not been demonstrated before. Rather, Ozer and Weinstein [12] previously found a lack of correlation between teacher support and posttraumatic stress. A possible explanation for this difference is that Ozer and Weinstein’s study was conducted with younger adolescents, in a school-based, ethnically diverse urban setting. There may also be a difference in the nature of the teacher–student relationship in suburban middleclass schools compared to in inner-city schools, where the student population is often more challenging, as suggested by higher teacher attrition rates in schools in low-socioeconomic and minority areas [55, 56]. It is notable that our findings concern a natural course of events and not the effects of any specific intervention. Targeted intervention studies involving teacher support would probably have yielded an even stronger protective effect against posttraumatic stress. Limitations The cross-sectional design of this study prevents us from drawing any conclusions about causality. Hence, it remains unclear whether prior depressive symptomatology
predisposes victimized youth to have posttraumatic stress reactions or whether depressive symptomatology is simply a response to victimization that is comorbid with posttraumatic stress. Another limitation of this study is the reliance on adolescents’ self-reports as a single source of information. When it comes to the reporting of exposure to violence, however, adolescents themselves tend to be the best informants, as parents may often underestimate the extent of violence exposure [38]. Although, many important variables affecting posttraumatic stress were taken into account, this study could not address the issues of proximity to violence or the identity of the victim, which can also affect the psychological response to witnessing violence [19, 20, 39]. In addition, domestic violence which correlates with community violence [47] was not measured. Clinical implications Similar to urban youth, many young people living in suburban areas in North American settings may be affected by community violence and require clinical attention, since a substantial proportion of exposed youth report severe posttraumatic stress. Additional clinical attention is called for in the presence of depressive symptoms, since there is a strong association between such symptoms and posttraumatic stress among victimized youth. This finding is of particular importance considering that comorbid depression substantially increases the risk for suicide among individuals with posttraumatic stress [53, 60]. Teacher support seems to be an important protective factor, but its role needs to be further evaluated using prospective study designs. Future interventions targeting the negative effects of community violence should consider teacher-based approaches. Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest. Ethical standards The study has been approved by the appropriate ethics committee and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.
References 1. Buka SL, Stichick TL, Birdthistle I, Earls FJ (2001) Youth exposure to violence: prevalence, risks, and consequences. Am J Orthopsychiatry. doi:10.1037/0002-9432.71.3.298 2. Stein BD, Jaycox LH, Kataoka S, Rhodes HJ, Vestal KD (2003) Prevalence of child and adolescent exposure to community violence. Clin Child Fam Psychol Rev. doi:10.1023/B:CCFP. 0000006292.61072.d2 3. Aisenberg E, Herrenkohl T (2008) Community violence in context: risk and resilience in children and families. J Interpers Violence. doi:10.1177/0886260507312287
123
Soc Psychiatry Psychiatr Epidemiol 4. Cooley-Strickland M, Quille TJ, Griffin RS, Stuart EA, Bradshaw CP, Furr-Holden D (2009) Community violence and youth: affect, behavior, substance use, and academics. Clin Child Fam Psychol Rev. doi:10.1007/s10567-009-0051-6 5. Dahlberg LL, Mercy JA (2009) History of violence as a public health problem. Virtual Mentor. doi:10.1001/virtualmentor.2009. 11.2.mhst1-0902 6. Finkelhor D, Turner H, Ormrod R, Hamby SL (2009) Violence, abuse, and crime exposure in a national sample of children and youth. Pediatrics. doi:10.1542/peds.2009-0467 7. Kuther TL, Fisher CB (1998) Victimization by community violence in young adolescents from a suburban city. J Early Adolesc. doi:10.1177/0272431698018001003 8. Lynch M (2003) Consequences of children’s exposure to community violence. Clin Child Fam Psychol Rev. doi:10.1023/B: CCFP.0000006293.77143.e1 9. Cooley-Quille M, Boyd RC, Frantz E, Walsh J (2001) Emotional and behavioral impact of exposure to community violence in inner-city adolescents. J Clin Child Adolesc Psychol 30(2): 199–206 10. Freeman LN, Mokros H, Poznanski EO (1993) Violent events reported by normal urban school-aged children: characteristics and depression correlates. J Am Acad Child Adolesc Psychiatry. doi:10.1097/00004583-199303000-00025 11. Fitzpatrick KM, Boldizar JP (1993) The prevalence and consequences of exposure to violence among African-American youth. J Am Acad Child Adolesc Psychiatry. doi:10.1097/00004583199303000-00026 12. Ozer EJ, Weinstein RS (2004) Urban adolescents’ exposure to community violence: the role of support, school safety, and social constraints in a school-based sample of boys and girls. J Clin Child Adolesc Psychol. doi:10.1207/s15374424jccp3303_4 13. Henrich CC, Schwab-Stone M, Fanti K, Jones SM, Ruchkin V (2004) The association of community violence exposure with middle-school achievement: a prospective study. J Appl Dev Psychol. doi:10.1016/j.appdev.2004.04.004 14. Gorman-Smith D, Tolan P (1998) The role of exposure to community violence and developmental problems among inner-city youth. Psychopathol Dev. doi:10.1017/S0954579498001539 15. Schwab-Stone M, Chen C, Greenberger E, Silver D, Lichtman J, Voyce C (1999) No safe haven. II: The effects of violence exposure on urban youth. J Am Acad Child Adolesc Psychiatry 38(4):359–367 16. Singer MI, Anglin TM, Song LY, Lunghofer L (1995) Adolescents’ exposure to violence and associated symptoms of psychological trauma. JAMA. doi:10.1001/jama.1995.035203000 51036 17. Foster JD, Kuperminc GP, Price AW (2004) Gender differences in posttraumatic stress and related symptoms among inner-city minority youth exposed to community violence. J Youth Adolesc 33:59–69 18. Gladstein J, Rusonis EJ, Heald FP (1992) A comparison of innercity and upper-middle class youths’ exposure to violence. J Adolesc Health. doi:10.1016/1054-139X(92)90159-9 19. Zinzow HM, Ruggiero KJ, Resnick H, Hanson R, Smith D, Saunders B, Kilpatrick D (2009) Prevalence and mental health correlates of witnessed parental and community violence in a national sample of adolescents. J Child Psychol Psychiatry. doi:10.1111/j.1469-7610.2008.02004.x 20. White KS, Bruce SE, Farrell AD, Kliewer W (1998) Impact of exposure to community violence on anxiety: a longitudinal study of family social support as a protective factor for urban children. J Child Fam Stud. doi:10.1023/A:1022943216319 21. Kliewer W, Cunningham JN, Diehl R, Parrish KA, Walker JM, Atiyeh C, Neace B, Duncan L, Taylor K, Mejia R (2004) Violence exposure and adjustment in inner-city youth: child and
123
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37. 38.
39.
40.
caregiver emotion regulation skill, caregiver-child relationship quality, and neighborhood cohesion as protective factor. J Clin Child Adolesc Psychol. doi:10.1207/s15374424jccp3303_5 Hammack PL, Richards MH, Luo Z, Edlynn ES, Roy K (2004) Social support factors as moderators of community violence exposure among inner-city African American young adolescents. J Clin Child Adolesc Psychol. doi:10.1207/s15374424jccp3303_3 Ozer EJ (2005) The impact of violence on urban adolescents longitudinal effects of perceived school connection and family support. J Adolesc Res. doi:10.1177/0743558404273072 Barr SC, Hanson R, Begle AM, Kilpatrick DG, Saunders B, Resnick H, Amstadter A (2012) Examining the moderating role of family cohesion on the relationship between witnessed community violence and delinquency in a national sample of adolescents. J Interpers Violence. doi:10.1177/0886260511416477 Resnick BP (1997) Protecting adolescents from harm: findings from the national longitudinal study on adolescent health. JAMA. doi:10.1001/jama.1997.03550100049038 O’Donnell DA, Schwab-Stone ME, Muyeed AZ (2003) Multidimensional resilience in urban children exposed to community violence. Child Dev. doi:10.1111/1467-8624.00471 Aneshensel CS, Sucoff CA (1996) The neighborhood context of adolescent mental health. J Health Soc Behav. doi:10.2307/ 2137258 Overstreet S, Braun S (2000) Exposure to community violence and post-traumatic stress symptoms: mediating factors. Am J Orthopsychiatry 70(2):263–271 Richards MH, Larson R, Miller BV, Luo Z, Sims B, Parrella DP, McCauley C (2004) Risky and protective contexts and exposure to violence in urban African American young adolescents. J Clin Child Adolesc Psychol. doi:10.1207/S15374424JCCP3301_13 Osgood DW, Wilson JK, O’Malley PM, Bachman JG, Johnston LD (1996) Routine activities and individual deviant behavior. Am Sociol Rev. doi:10.1037/h0087828 Pettit GS, Bates JE, Dodge KA, Meece DW (1999) The impact of after-school peer contact on early adolescent externalizing problems is moderated by parental monitoring, perceived neighborhood safety, and prior adjustment. Child Dev. doi:10.1111/ 1467-8624.00055 Gardner M, Brooks-Gunn J (2009) Adolescents’ exposure to community violence: are neighborhood youth organizations protective? J Community Psychol. doi:10.1002/jcop.20310 Gottfredson DC, Gerstenblith SA, Soule´ DA, Womer SC, Lu S (2004) Do after school programs reduce delinquency? Sci Prev. doi:10.1177/0886260508314297 Hoffmann JP, Xu J (2002) School activities, community service, and delinquency. Crime Delinquency. doi:10.1177/00111280 2237130 Weissberg RP, Voyce CK, Kasprow WJ, Arthur MW, Shriver TP (1991) The social and health assessment. Yale Child Study Center, Chicago Schwab-Stone M, Ayers TS, Kasprow W, Voyce C, Barone C, Shriver T, Weissberg RP (1995) No safe haven: a study of violence exposure in an urban community. J Am Acad Child Adolesc Psychiatry. doi:10.1097/00004583-199510000-00020 JE Richters, W Saltzman (1990) Survey of exposure to community violence: self-report version. JE Richters Richters JE, Martinez P (1993) The NIMH community violence project: Vol I. Children as victims of and witnesses to violence. Psychiatry 56:7–21 Pynoos RS, Frederick C, Nader K, Arroyo W, Steinberg A, Eth S, Nunez F, Fairbanks L (1987) Life threat and posttraumatic stress in school-age children. Arch Gen Psychiatry. doi:10.1001/arch psyc.1987.01800240031005 Pynoos RS, Goenjian A, Tashjian M, Karakashian M, Manjikian R, Manoukian G, Steinberg AM, Fairbanks LA (1993) Post-
Soc Psychiatry Psychiatr Epidemiol
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
traumatic stress reactions in children after the 1988 Armenian earthquake. Br J Psychiatry 163:239–247 Radloff LS (1977) The CES-D scale A self-report depression scale for research in the general population. Appl Psychol Meas. doi:10.1177/014662167700100306 Roberts RE, Lewinsohn PM, Seeley JR (1991) Screening for adolescent depression: a comparison of depression scales. J Am Acad Child Adolesc Psychiatry. doi:10.1097/00004583-19910 1000-00009 Andrykowski JSCMA, Mary JWLAH (1998) Psychometrics for two short forms of the Center for Epidemiologic StudiesDepression Scale. Issues Ment Health Nurs. doi:10.1080/0161 28498248917 Ruchkin V, Schwab-Stone M, Vermeiren R (2004) Social and health assessment (SAHA): Psychometric development summary. Yale Univ, N. Hav Hawkins JD, Catalano RF, Miller JY (1992) Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: implications for substance abuse prevention. Psychol Bull 112:64 African-American youth, Fitzpatrick KM (1993) Exposure to violence and presence of depression among low-income. J Consult Clin Psychol. doi:10.1037/0022-006X.61.3.528 Margolin G, Gordis EB (2000) The effects of family and community violence on children. Annu Rev Psychol. doi:10.1146/ annurev.psych.51.1.445 Kilpatrick DG, Ruggiero KJ, Acierno R, Saunders BE, Resnick HS, Best CL (2003) Violence and risk of PTSD, major depression, substance abuse/dependence, and comorbidity: results from the National Survey of Adolescents. J Consult Clin Psychol 71:692–700 Breslau N (1997) Sex differences in posttraumatic stress disorder. Arch Gen Psychiatry. doi:10.1001/archpsyc.1997.018302300 82012 Anderson ER, Mayes LC (2010) Race/ethnicity and internalizing disorders in youth: a review. Clin Psychol Rev . doi:10.1016/j. cpr.2009.12.008
51. Basch CE (2011) Aggression and violence and the achievement gap among urban minority youth. J Sch Health. doi:10.1111/j. 1746-1561.2011.00636.x 52. Garber J, Weersing VR (2010) Comorbidity of anxiety and depression in youth: implications for treatment and prevention. Clin Psychol. doi:10.1111/j.1468-2850.2010.01221.x 53. Oquendo M, Brent DA, Birmaher B, Greenhill L, Kolko D, Stanley B, Zelazny J, Burke AK, Firinciogullari S, Ellis SP, Mann JJ (2005) Posttraumatic stress disorder comorbid with major depression: factors mediating the association with suicidal behavior. Am J Psychiatry 162(3):560–566 54. Haden SC, Scarpa A (2008) Community violence victimization and depressed mood: the moderating effects of coping and social support. J Interpers Violence. doi:10.1177/0886260508314297 55. Borman GD, Dowling NM (2008) Teacher attrition and retention: a meta-analytic and narrative review of the research. Educ Res Rev. doi:10.3102/0034654308321455 56. Shen J (1997) Teacher retention and attrition in public schools: evidence from SASS91. J Educ Res. doi:10.1080/00220679 709597525 57. Fuligni AJ, Eccles JS (1993) Perceived parent-child relationships and early adolescents’ orientation toward peers. Psychol Dev. doi:10.1037/0012-1649.29.4.622 58. Hill HM, Jones LP (1997) Children’s and parents’ perceptions of children’s exposure to violence in urban neighborhoods. J Natl Med Assoc 89:270 59. Cunningham M (1999) African American adolescent males’ perceptions of their community resources and constraints: a longitudinal analysis. J Community Psychol. doi:10.1002/ (SICI)1520-6629(199909)27:5\569:AID-JCOP5[3.0.CO;2-6 60. Panagioti M, Gooding PA, Tarrier N (2012) A meta-analysis of the association between posttraumatic stress disorder and suicidality: the role of comorbid depression. Compr Psychiatry. doi:10.1016/j.comppsych.2012.02.009
123
