COGNITIVE BEHAVIORAL TREATMENTS FOR YOUTH EXPOSED TO TRAUMATIC EVENTS : A META - ANALYSIS EXAMINING VARIABLES MODERATING AND MEDIATING TREATMENT OUTCOMES By Justina Yohannan A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of School Psychology Doctor of Philosophy 2020 ABSTRACT COGNITIVE BEHAVIORAL TREATMENTS FOR YOUTH EXPOSED TO TRAUMATIC EVENTS: A META - ANALYSIS EXAMINING VARIABLES MODERATING AND MEDIATING TREATMENT OUTCOMES By Justina Yohannan Due to the negative impact of exposure to traumatic events (DePrince, Weinzierl, & Combs, 2009), effective treatments are nece ssary to prevent/improve negative outcomes. Cognitive behavioral therapy (CBT) is considered an efficacious treatment for youth exposed to traumatic events (American Psychological Association [APA], 2008). Past meta - analyses showed larger effect sizes for youth who received general CBT and trauma - focused CBT (TF - CBT) when compared to control groups (Gutermann et al., 2016; Kowalik, Weller, Venter, & Drachman, 2011) and other forms of treatment (e.g., play therapy; Silverman et al., 2008; Slade & Warne, 2016 ). Despite the varying meta - analyses available examining trauma treatments, there is a paucity of research examin ing moderating and mediating variables that may impact treatment outcomes. This meta - analytic CBT study address ed those limitations by examining the moderating effects of study design, research study setting, trauma type, and cultural (i.e., demographic) variables (i.e., race, age, gender) on youth posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes. In addition, the mediating effect of treatment components on youth outcomes (e.g., paren tal involvement, treatment delivery , inclusion of other treatment techniques) within CBT studies was also examined . A search using PsycINFO, EBSCO, ERIC, and ProQuest Dissertations and These s identified 94 CBT studies with 97 relevant effect sizes for children and adolescents exposed to traumatic events. Consistent with prior meta - analytic studies, CBT was an effective treatment for youth exposed to traumatic events . P osttraumati c stress symptoms ( d = - .57, p < .001), anxiety symptoms ( d = - .40, p < .001), and depression symptoms ( d = - .40, p < .001) were all found to be positively impacted by CBT . CBT subtreatments did not produce significantly diffe rent results from one another (posttraumatic stress symptoms: p = .073; depression symptoms: p = .296) . All subtreatments , except for Game - Based CBT ( d = - .38, p = .117) , resulted in significant reductions in symptoms . M oderators significantly impacting CBT treatment outcomes for posttraumatic stress symptoms were trauma type (i.e., Q = 24.09, p = .004) and gender (i.e., Q = 10.68, p = .005) while moderators impacting treatment outcomes for depression were study design (i.e., Q = 10.95, p = .00 4) and treatment setting (i.e., Q = 10.98, p = .004). None of the variables examined moderated anxiety symptom outcomes. Further, no mediators were found to significantly impact posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes . The implications of these findings for research and practice are discussed. ACKNOWLEDGEMENTS I would like to thank my advisor and dissertation chair, Dr. John S. Carlson , for the support he has provided me throughout all my research and practice endea vors. With his feedback and guidance, I was able to test my limits. Further, project via the Michigan State University (MSU) College of Education Dissertation Completion Fellowship and MSU Department of C ounseling, Educational Psychology, and Special Education (CEPSE) Research Fellowship made it financially possible for me to complete this project. I would also like to thank my dissertation committee members, Dr. Martin Volker, Dr. Kristin Rispoli, and Dr. Marisa Fisher, for their feedback and suggestions. Their input contributed to making this study possible, and I appreciate their guidance in my tr ansformation as a scholar. Additionally, I want to acknowledge Hope Akaeze and the MSU Center for Statistical Training and Consulting for the instruction provided in meta - analysis . I am so grateful to my family and friends who have held me up during the la st six years. I have learned and laughed with all my classmates, but I am particularly thankful for my cohort for all that they have done to help me survive and thrive in graduate school. I am also grateful for my love d ones who have always had faith in my abilities and kept my spirits high. I would like to thank my extended family, who have always been there for me with love , the occasional snarky comment, and great food . Most importantly, I owe the world to my parents, Ghevarghese Yohannan and Jolly Yohan nan, and my brothers, Jermyn Yohannan and Joshua Yohannan. Your love and encouragement have helped me reach where I am today. Thank you for walking this path alongside me. v TABLE OF CONTENTS LIST OF TABLES .. . i LIST OF FIGURES ... viii CHAPTER 1 INTRODUCTION CHAPTER 2 .. 1 2 LITERATURE REVIEW ..1 2 Defining Trauma and Trauma Prevalence ... 1 2 Outcomes for Traumatized Youth 3 Defining What Makes Treatments Evidence - Based 7 Theories Behind Cognitive Behavioral Ther ap y 9 The Use of Cognitive Behavioral Treatments 2 Trauma - focused cognitive behavioral therapy (TF - CBT) 2 4 Cognitive behavioral intervention for trauma in schools (CBITS) 4 Other treatments that use cognitive behavioral techniques 5 Need for Meta - 6 Research on Cognitive Behavioral Therapy for Children and Adolescents 6 7 Cultural/demographic characteristi 8 9 29 Age 30 3 1 1 Treatment components ... 3 2 Research Questions 3 CHAPTER 3 9 METHODS 9 Literature Search 9 40 Data Coding 4 2 Measures used to determine change 4 3 Moderator and mediator 44 Data Analysis 4 5 CHAPTER 4 8 RESULTS 8 Meta - Analyses 4 8 Question 1: Overall effect and subt reatment effectiveness 4 9 vi Question 2: Moderator analyses 51 Moderator Analyses for Studies with Posttraumatic Stress Symptom .. . 5 1 Moderator Analyses for Studies with Anxiety Symptom Outcomes 5 3 Moderator Analyses for Studies with Depression 4 5 6 Mediator Analyses for Studies with Posttraumatic Stress Symptom Out 5 6 Mediator Analyses for Studies with Anxiety Symptom Outcome s ... . 5 8 Mediator Analyses for Studies with Depression Symptom Outcomes .. 5 9 Publication B ias ... 60 CHAPTER 5 61 DISCUSSION 61 Strengths and Limitations of the Meta - Analysis 70 Conclusion 7 4 APPENDICES ..76 APPENDIX A: ..77 APPENDIX B : Coding Sheet 1 4 0 APPENDIX C 44 vii LIST OF TABLES Table 1 . Variables in Past Meta - Analyses/Systematic Reviews Table 2 . Outcomes for Past Meta - . 80 Table 3 . Cognitive Behavioral Treatments .. ..82 Table 4 . Moderating and Mediating Factors Examine . ...... ...84 Table 5 . Table 6 . Table 7 . ..94 Table 8 . Outcome Measures Used in Meta - . 02 Table 9 . Meta - Analyses Data for Posttraumatic Stress Symptoms in Cognitive Behavioral ....1 07 Table 10 . Meta - Analyses Data for Anxiety Symptoms in Cognitive Behavioral Treatments.... . 1 14 Table 11 . Meta - Analyses Data for Depression Symptoms in Cognitive Behavioral Treatments 1 17 Table 12 . Summa ry of Initial Meta - .. 22 Table 13 . Moderator . . 1 23 Table 14 . Populations that Need Examining in Future Research .. 3 1 viii LIST OF FIGURES Figure 1 . Conceptual Framework on Factors That Impact Traumatized Youth Mental Health Problems .. ...1 3 2 Figure 2 . Cyclic Influence of Emotion, Cognition, and Behavior (Tolin, 2016) ... ..1 33 Figure 3 . Flow Chart of Meta - Analysis Phases ( Moher et al., 2009) ..1 34 Figure 4 . Funnel Plot to Determine Publication Bias for Posttraumatic Stress Symptom Data . .1 35 Figure 5 . Funnel Plot to Determine Publication Bias for Posttraumatic Stress Symptom Data (Without Outlier) .1 36 Figure 6 . Funnel Plot to Determine Publication Bias for Anxiety Symptom ... 37 Figure 7 . Funnel Plot to Determine Publication Bias for Depression Symptom Data .. 38 Figure 8 . Funnel Plot t o Determine Publication Bias for Depression Symptom Data (Without 39 1 CHAPTER 1 INTRODUCTION The prevalence of exposure to at least one traumatic event among children and adolescents is high, with rates among national samples ranging from 41% (Zinzow et al., 2009) to 83% (Ford, Elhai, Connor, & Frueh, 2010). According to the American Psychiatric Assoc iation (2013), a traumatic event is p. 271 ). Similarly, the National Child Traumatic Stress Network (NCTSN, 2003) defines a traumatic event as an acute or chronic event that thre - being. Research has shown that youth who are exposed to traumatic events are at risk for negative outcomes. For example, youth exposed to traumatic events may have a variety of mental health concerns (e.g., anxiety, depression ; De Young, K enardy, & Cobham, 2011 ). They are also at risk for low academic performance and poor executive functioning (DePrince, Weinzierl, & Combs, 2009 ), lower IQ scores, and lower language abilities (Perfect, Turley, Carlson, Yohannan, & Pfenninger Saint Gilles, 2 016). In certain cases, significant negative outcomes may result in trauma and stress - related disorders such as post - traumatic stress disorder (PTSD). According to a review of 43 independent samples of trauma - exposed children and adolescents , 15% of youth met diagnostic criteria for PTSD ( Alisic et al., 2014). Due to the range of negative outcomes, appropriate evidence - based treatments for youth exposed to traumatic events are necessary. There are numerous treatments (e.g., psychological , psychopharmacological) used with youth who have been exposed to traumatic events and struggle with resulting mental health challenges . To effectively target the negative outcomes youth experience after traumatic event exposure, treatment must link with o utcome etiolog y . As seen in Figure 1, there are psychological, biological, and ecological factors that impact poor youth outcomes after exposure 2 to traumatic events ( Masten & Narayan, 2012; McKeever & Huff, 2003; Taylor & Asmundson, 2008) ; trauma treatment must address these factors to improve youth functioning . For example, to address ecological factors, psychoeducation to key stakeholders might be helpful, and to address biological factors, providing relaxation training or medication might be appropriate. Examining recent meta - analyses of treatment studies to determine the current state of the literature can aid mental health professionals in deciding what treatments to use . In Strawn and - analysis, which highlighted the state of psy chopharmacologic research, there was limited evidence for the use of such treatment in youth with PTSD , despite this tackling the biological challenges caused by traumatization . Treatment g uidelines for treat ing youth with PTSD [ American Academy of Child a nd Adolescent Psychiatry (AACAP) , 2010 ] highlight selective serotonin reuptake inhibitors as a second - line treatment (i.e., following a lack of response to psychological treatments ) given the limited efficacy support for symptom improvement in youth with PTSD. The AACAP Official Action report (2010) on treatment guidelines recommends that psychological treatment should be the initial course of action in treating youth PTSD . Through the past couple decades, multiple meta - analyses have been conducted supporting the practice guidelines regarding the use of psychological treatments for trauma in children and adolescents. In prior meta - analyses (see Table 1) examining trauma treatments, psychological treatments were examined and found to be effective , including, but not limited to, cognitive behavioral therapy (CBT), cognitive processing therapy (CPT), narrative therapy, and eye - movement desensitization and reprocessing (EMDR ) . For example , a - analysis of 135 studies showed the positive impact of psychological treatments (e.g., CBT , EMDR hypnotherapy , psychodynamic psychotherapy ) for youth with PTSD ; similarly, Morina, 3 - analysis examined 39 psychological treat ments, including CBT , EMDR , mind - body skills groups, and prolonged exposure treatment. While practice guidelines (AACAP, 2010) promote the use of psychological treatments generally, the American Psychological Association ( APA , 2008 ) specifically names CBT techniques as an effective treatment for youth exposed to traumatic event s. Additionally, while prior meta - analyses f ound that various psychological treatments were efficacious , meaning treatment studies were replicated and were prov en to improve youth outcomes (Chambless & Hollon, 1998) , the evidence base for psychological treatment for youth exposed to traumatic events indicated that CBT and therapies consisting of CBT components produced the strongest outcomes (Gillies, Taylor, Gra Gutermann et al., 2016 ; Kowalik, Weller, Venter, & Drachman , 2011 ; Silverman et al, 2008; Slade & Warne, 2016 ) . As seen in Table 2, t he studies which examined the general efficacy of psychological treatments found that effect sizes for CBT were higher than for control groups and that the treatments affected various psychological constructs. For example, in Gillies and colleagues (2012) study, PTSD symptoms scores ( SMD = - 1.34) and depression scores ( SMD = - 0.80) that exposure to CBT provided statistically significant effect sizes for internalizing symptoms ( g = - .31; p = .001), externalizing sy mptoms ( g = - .19; p = .040), and total problem scores ( g = - .3 3 ; p = .003) , thus favoring the use of CBT over control groups . Some prior meta - analyses also found CBT to be more effective than other trauma treatment s . a - analysis , the authors found that CBT had greater effect sizes than non - CBT treatments for posttraumatic stress symptoms ( d = .50 vs d = .19), depression symptoms ( d = .29 vs d = .08) , and externalizing symptoms ( d = .24 vs d = .02) . Additionally, Slade and Warne (2016) found that 4 trauma - focused cognitive behavioral therapy (TF - CBT), a specific treatment manual using CBT techniques, had greater effect sizes than play therapy in the areas of global symptoms ( d = .21 vs d = .095) and int ernalizing symptoms ( d = .2 3 vs d = .096) . As can be seen, prior research has focused on a variety of differing outcome measures. However, since there is an overlap in the symptomatology of PTSD, anxiety, and depression and there is evidence of high rates of comorbidity between these associated mental health disorders [American Psychiatric Association ( APA ) , 2013 ] , it is crucial for mental health practitioners to understand the impact of treatment on these specific symptoms. Further, prior meta - analyses hav e examined these specific outcomes (i.e., Gillies et al., 2012; Gutermann et al., 2016; Morina et al., 2016; Silverman et al., 2008) , supporting the examination of these outcome measures . While prior meta - analyses examined the general efficacy of CBT, the re is limited research on the moderating and mediating factors that could impact CBT treatment outcomes for youth exposed to traumatic events . Moderator variables refer to variables that describe the circumstances under which something occurs, and they typ ically refer to the population or setting in which therapeutic change occurs (Holmbeck, 1997; Silverman & Hinshaw, 2008). Mediators , on the other hand, refer to the mechanisms of change, meaning the variables which cause therapeutic change to occur (Kazdin & Nock, 2003). By examining moderator and mediator variables, it can be determined for whom and under what conditions treatment is effective and change occurs, thus determining generalizability (Holmbeck, 1997; Silverman & Hinshaw, 2008). One need for th is meta - analysis is to address specifically for whom CBT is effective . Table 1 shows a comprehensive summary of the traumatic events that youth were exposed to in prior meta - analyses conducted . Brown and colleagues (2017) examined the moderating effect of 5 trauma type by study design (i.e., RCT and pre - post) for all psychological treatments, and they found trauma type did not moderate posttraumatic stress symptom outcomes. Additionally, Gutermann and colleagues (2016) found that trauma type did not moderate posttraumatic stress symptom outcomes in their examination of all psychological treatments. Further, Silverman and colleagues (2008) examined the moderating effect of trauma type for all psychol ogical treatments and found that the effect s of posttraumatic stress symptoms and depression symptoms for sexual abuse treatment were significantly higher than for other types of trauma. However, one limitation of prior research is that of the meta - analyse s that examined only CBT, none examined the moderating effect of trauma type on outcomes. It is important to determine if CBT is effective regardless of the type of trauma the youth was exposed to . Research as shown that individuals have poorer outcomes (e .g., posttraumatic stress symptoms, anxiety symptoms, depression symptoms, externalizing symptoms) when exposed to traumatic events that have an early onset, high duration, are interpersonal in nature, and involve multiple types of trauma (Kliethermes, Sch acht, & Drewry, 2014), so it is crucial that mental health practitioners implement effective treatment. The moderating effect of trauma type on posttraumatic stress symptoms, anxiety symptoms, and depression symptoms must be examined to determine if even t hose exposed to the most deleterious types of traumatic events will achieve positive outcomes through CBT. To further determine for whom treatment is effective, an examination of the moderating effect of cultural variables (e.g., racial identity, gender, ag e ) is also important. This can be useful - Jaeger, Kava, Akiba, Lucid, & Dorsey, 2017, p. 231). Unfortunately, research is 6 not always representative of the populations practitioners work with; the majority of research on et it is estimated that soon approximately half of the population will be of Hispanic, African American, Native American, or Asian/Pacific Island descent (Cartledge, Kea, & Simmons - Reed, 2002; Singh, Ellis, Oswald, Wechsler, & Curtis, 1997). Furthermore, r ates of problem behaviors vary based on race and context (Yasui & Dishion , 2007 ). Brown and colleagues (2017) found that age and gender did not moderate posttraumatic stress symptom outcomes in their examination of all psychological treatments for youth ex posed to traumatic events . In their study of all psychological treatments for youth exposed to traumatic events, Gutermann and colleagues (2016) found that age moderated posttraumatic stress symptom outcomes, with older youth having larger effect sizes tha n younger youth. They further found that gender did not moderate posttraumatic stress symptom outcomes. While prior studies examined the moderating effect of the cultural factors of gender, and age on the effectiveness of psychological treatments in genera l , none of the meta - analyses specifically examined the moderating effect of the cultural variables on youth posttraumatic stress symptom , anxiety symptom , and depression symptom outcomes related to CBT . Silverman and colleagues (2008) as well as Garland , Bickman, and Chorpita (2010) recommended that further research on moderators must occur to increase generalizability. CBT research has provided evidence of internal validity for use with traumatized populations, and evidence of external validity (e.g., generalizability) too must be established . While the literature has clearly shown that CBT seems to generally work for youth involved in their studies (e.g., sexual assault; Slade & Warne, 2016), the treatment research is normally conducted in highl y controlled settings (e.g., laboratories, clinical research facilities) with 7 rigorous methodologies (e.g., randomized controlled trials [RCTs]) to ensure treatment efficacy. As can be seen in Table 1, there have been multiple meta - analyses which examined treatment efficacy using RCTs (Cary & McMillen, 2012; Gillies et al., 2012; Kowalik et al., 2011; Morina et al., 2016; Silverman et al., 2008; Slade & Warne, 2016). Such studies can prove efficacy by showing improved outcomes between a treatment and a cont rol group (Chorpita & Regan, 2009; Kratochwill & Shernoff, 2003), but generalization of these studies to everyday life is limited because research conducted in highly controlled environments is not conducive to the individualization that occurs in treatmen t and t his limited individualization not preferred by clinicians (Chorpita, Daleiden, & Weisz, 2005). Additionally, CBT is manualized in nature, so there is an assurance with RCTs that every youth receiving CBT is provided the same aspects of treatment eve ry time it is implemented. Unfortunately, strict adherence to a manual is not necessarily guaranteed in the real world since research has found that therapists were concerned about the influence of manu a ls on individualization of treatment and therapeutic rapport (Chorpita, Daleiden, & Collins, 2014). assurance of its positive impact on youth outcomes (Garland et al., 2010). While some prior meta - analyses have examined less strict study designs (e.g., pre - post, quasi - experimental; Brown et al., 2017; Dorsey et al., 2017; Gutermann et al., 2016; Harvey & Taylor, 2010) in more naturalistic settings (e.g., schools, community clinics; Dorsey et al., 2017; Gillies et al., 2012; H arvey & Taylor, 2010), only Harvey & Taylor (2010) examined the moderating effect of study design. They found that experimental studies had higher effect sizes than quasi - experimental and uncontrolled studies. This study generalized the effect to all psych ological treatments rather than purely examining CBT ; n o studies examined the moderating effect of study setting for CBT . 8 Further, only one study (Harvey & Taylor, 2010) examined treatment setting specifically, and they found that setting did not moderate posttraumatic stress symptom outcomes for psychological treatments. Additionally, Brown and colleagues (2017) found that teachers implementing treatment had lower posttraumatic stress symptom effect sizes than other professionals implementing psychological treatment s . Thus, while there is ample evidence for treatment efficacy in clinic settings , treatment effectiveness in more naturalistic settings is not as certain and understanding these moderating factor s on posttraumatic stress symptoms, anxiety symptoms, and depression symptoms can help practitioners determine which line of treatment to use . An examination of the mediating variables of treatment are also important to understand in treatment research , and to do so, one must examine the components of treatment that cause change . While a variety of treatments are available to mental health professionals, efficacious treatments should be implemented to ensure positive youth outcomes, and these treatments should address the multidimensional challenges associated with traumatic event exposure. CBT is a treatment that addresses the etiological factors (i.e., biological, psychological, ecological) that are associated with negative outcomes. CBT consists of multiple c omponents, including psychoeducation, cognitive processing of thoughts and beliefs, relaxation/coping skills training, and imaginal exposure (Ramirez de Arellano et al., 2014; Silverman et al., 2008); these various components address the complex challenges created by exposure to traumatic events. For example, ecological factors are addressed through psychoeducation with both the youth and the parents, while biological factors are addressed through the instruction of relaxation/coping skills training and the imaginal exposure (i.e., desensitization), and psychological factors are addressed through cognitive processing. 9 The use of all these CBT components appears to be a necessary factor for effective treatment , as seen through Table 2 . The most common CBT treatments studied are TF - CBT and cognitive behavioral intervention for trauma in schools (CBITS). Cary and McMillen (2012) examined both TF - CBT and CBITS in their meta - analysis while Slade and Warne (2016) specifically examined TF - CBT , and they found the treatment to be effective for traumatized youth as measured by global, internalizing, externalizing, sexual, and parent report outcomes . However, as can be seen in Table 3 , there are several other treatments available that use the components of CBT. The CB T meta - analyses that limited their examination to only TF - CBT and/or CBITS did not examine the se other potentially effective treatments that use CBT techniques to promote positive outcomes. As Garland and colleagues (2010) noted, research needs to be condu cted on the treatment processes and outcomes to determine treatment impact. Additionally, Kazdin (2008) noted that research needs to examine the processes in treatment that cause, not simply correlate , with outcomes. A meta - analysis could provide seminal i nformation on the similarities and differences between CBT - only treatments and CBT treatments that include other treatment components (e.g., narrative therapy techniques) through a separation of CBT data by treatment manual (e.g., TF - CBT vs CBITS) , which w as missing in the prior meta - analyses since they only looked at one specific treatment manual (e.g., TF - CBT) or aggregated the CBT data .. The mediating effect of other treatment components also needs to be examined to determine which aspects of CBT are ne cessary for positive outcome change in traumatized youth . Some, but not all, CBT manuals include a parental component to treatment to allow for reinforcement and generalizability of skills to settings outside the treatment setting. Silverman and colleagues (2008) examined the mediating effect of parental involvement in all 10 psychological treatments, and they found that while parental involvement did not moderate posttraumatic stress symptoms, parental involvement in treatment had larger effects on depression symptom and anxiety symptoms than no parental involvement. Harvey and Taylor (2010) also found that family involvement in treatment resulted in higher effect sizes than no use of family in psychological treatments as measured by posttraumatic stress symptoms. Similarly, Gutermann and colleagues (2016) found psychological treatments with parental involvement had larger effects as measured by posttraumatic stress symptoms than those without parents. None o f the meta - analyses examined the mediating effect of only CBT on youth outcomes. A thorough analysis of the type of treatment and treatment components provided through a meta - analysis would support the use of all the CBT modules and could help practitioner s determine if parental involvement would be best for their traumatized youth by treatment outcome ( i.e., posttraumatic stress symptoms, anxiety symptoms, depression symptoms) . Treatment delivery (e.g., group, individual) is another mediating factor that needs to be further examined in the literature . Past research has shown that group treatment had lower posttraumatic stress symptom effect sizes than individual treatment for overall psychological treatments (Brown et al., 2017; Gutermann et al., 2016 ). Ho wever, the mediating effect of treatment delivery has not been examined for CBT in relation to posttraumatic stress symptoms, anxiety symptoms, and depression symptoms. Additionally, s ome other variables that were found to not mediate psychological treatme nts that were examined in prior meta - analyses include session number (Brown et al., 2017; Harvey & Taylor, 2010) and session length (Harvey & Taylor). However, as these meta - analyses did not examine CBT specifically, an analysis of these variables as measu red by posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes would add to the current literature. 11 Overall, this meta - analytic review of the CBT literature for youth examine d the moderat ing and mediating factors of treatment (see Table 4 ) by empirically evaluat ing the current evidence base regarding CBT efficacy versus effectiveness , thus filling the current gaps in the CBT treatment literature for youth exposed to traumatic events . I t examine d for whom C BT is an effective treatment and under what conditions CBT is effective. This, in turn, could provide researchers with new avenues to explore in the CBT literature and it could strengthen research methodology and design through the inclusion of groups who were not included in prior research . For mental health practitioners, such as school psychologists, this meta - analytic review would provide the current state of evidence for CBT use with potentially traumatized youth from a variety of backgrounds in various contexts . It should be noted that not all youth exposed to traumatic events present with symptoms severe enough to meet criteria for PTSD or other traumatic stress - related disorders (Carter, 2007), but still experience levels of distress that impact their well - being . For example, in both - - analysis, participants were included in the studies if the youth were exposed to traumatic events, and PTSD diagnoses were not a prerequisite. This meta - analysis also examine d youth who have not necessarily been diagnosed with PTSD but experience negative symptoms due to their exposure to traumatic events. 12 CHAPTER 2 LITERATURE REVIEW In order to address the need for this meta - analytic study, the following sections provide a foundational review of the literature : 1) the definition and prevalence of trauma, 2) the social - emotional - behavioral outcomes for youth exposed to traumatic events , 3) defining what makes a treatment evidence - based , 4) the theoretical basis of cognitive behavioral therapy (CBT), 5) CBT as a child mental health treatment , 6) the need for meta - analysis, and 7 ) the current research on moderators and mediators of CBT with youth who have experienced traumatic events . Defin ing Trauma and Trauma Prevalence The Substance Abuse and Mental Health Services Administration (SAMHSA, 2014) defines a traumatic event experienced by an individual as physically or emotional ly harmful or life threatening and that has last ing spiritual well - There are a variety of traumatic events ranging from non - interpersonal events to in terpersonal events and chronic events to acute events. A non - inte rpersonal event does not involve an individual intentionally harming another individual (e.g., medical procedure, car accident, natural disaster) whereas an interpersonal even t occurs between individuals and can sometimes involve a malicious intent to inflict harm upon an individual (e.g., sexual abuse, physic al abuse, traumatic loss; Lilly, Valdez, & Graham - Bermann, 2011). An acute event is a one - time event whereas a chronic event is one in which e xposure to the trauma occurs re peatedly over time . It is important for practitioners and resea rchers to know and s tudy the impact of trauma and evidence - based treatment for youth because of the high prevalence rates of traumatic 13 exposure. Finkelhor , Ormro d, and Turner (20 07) found in their study of a national sample of youth that 71 % of children in their sample were victimized within the past year. R ecognizing that there are a variety of traumatic events a youth can be exposed to and the rate at whi ch these events occur can help practitioners identify potentially traumatized youth who might need treatment . In their nationally sampled study of traumatic event prevalence, McLaughlin and colleagues (2013) found that almost 62% of youth we re exposed to a traumatic event (i.e., interpersonal violence, accident, network/witnessing an event) at least once in their lives while Ford and colleagues (2010) found that 83% of their nationally representative sample experienced at least one traumatic event. With such high prevalence rates, appropriate treatments are needed t o reduce the likely high prevalence of negative outcomes that might occur due to traumatic exposure. Outcomes for Traumatized Youth Finkelhor, Ormrod, and Turner (2007) found that polyvictimiz ation correlated with negative outcomes and trauma - related symptoms. Additionally, chronic and interpersonal trauma is associated with detrimental outcomes ( Kliethermes, Schacht, & Drewry, 2014). As seen in Figure 1, there are various etiological factors , such as biology, to consider when examining outcomes for youth exposed to traumatic events. The biological model, as described by Ford (2015), explains the symptoms that occur in youth exposed to traumatic events. The amygdala, which is the part of the bra in associated with emotions and is connected to the hormonal stress amygdala (i.e., the alarm center) is hyperactivated, so that any potentially dangerous event can be detected. This means that the brain is constantly in a state of physiological and emotional arousal to prevent harm from occurring to the individual, and this hyper - vigilance and more easily 14 trigged survival instinct can result in some of the behavioral s ymptoms that are seen. On the flip side, hypoarousal can occur, which is when others see the youth engaging in emotional numbing or dissociation. In a non - traumatized brain, the amygdala reacts to challenges and opportunities by storing and retrieving rele vant information and memories to engage in appropriate cognitions and behaviors; however, in the brain of a youth exposed to traumatic events, survival and threat - related memories are at the forefront, so they are more likely to access this information. W hile most youth exposed to traumatic events do not end up meet ing diagnostic criteria for PTSD (Alisic et al., 2014) , the outcomes for youth exposed to traumatic events is poorer than for youth who do not experience a traumatic event (Carter, 2007) . Social - emotional functioning can be negatively impacted by trauma, and consequently, behavior may be affected . However, because trauma can manifest as either internalizing or externalizing behaviors, simply loo king for obvious behavioral symptoms , such as avoidance, is not enough. Instead, one must look for both physical symptoms and affective symptoms (Little & Akin - Little, 2013) . For example, hyper - vigilance, which is when a youth is constantly alert because h er or his brain is looking to identify and process threat, may occur. However, this can look like inattention, leading to misdiagnoses of attention deficit hyperactivity disorder (ADHD) . Posttraumatic stress disorder (PTSD) specifically is generally associ ated with symptoms of re - experiencing (e.g., nightmares and flashbacks), avoidance and numbing (e.g., being unable to recall the traumatic memories, restricted affect), and hyper - arousal (e.g., insomnia, poor concentration ; Carrion , Wong, & Kletter, 2013). Hyperarousal looks like disturbed sleep, high levels of irritability, fussiness and temper tantrums, constant alertness to potential danger, an exaggerated startle response, poor concentration, and hyperactivity (De Young, Kenardy , & Cobham, 2011) , which is also similar to the symptoms of ADHD or other executive functioning disorders. There are also overlapping 15 symptoms between posttraumatic stress - related disorders, depression - related disorders, and anxiety - related disorders. Furthe rmore, there is typically comorbidity between these three disorders (APA, 2013). De Young and colleagues (2011) go into detail about the symptoms associated with trauma in young children, which is relevant because young children are at high risk of exper iencing maltreatment, with 56% of victims being under the age of seven. Young children express themselves differently from adolescents or adults, since they do not have the language skills or cognitive awareness that adolescents or adults can use to verbal ize their experiences and concerns. Young children can re - experience trauma through posttraumatic play, intrusive recollections of the trauma, and distressing ni ghtmares . While intrusive recollections of the trauma can occur at any age, young children expe rience it slightly differently, and may end up expressing this experience through play. Also, while distressing nightmares are not exclusive to young children, it is an important way of recognizing the fear that the child is facing. Avoidance may also occu r in young children, which is witnessed through avoiding exposure to anything that may remind the child of the trauma, social withdrawal, and restricted ex ploratory behavior and play (De Young et al., 2011) . As a coping mechanism, the young child will avoi d any situation that might expose them to reminders of the traumatic event, and restricted behavior and play may be limited to behavior and play surrounding the events of the trauma. A young child is also more likely to display behaviors such as aggression is likely to increase due to trauma. While a child who is being fussy may be viewed should look more into the function of the behavior. Also, as was mentioned, hyper - vigilance and difficulties with concentration may 16 occur because the child is on constant alert for threat ; t tasks they should be completing, such as academic tasks. R esear ch has indicated that childhood adversity and trauma is associated with other mental health disorders, such as mood disorders, anxiety, substance use, and is also associated with disruptive behaviors, and health risk behaviors, such as smoking and suicide attempts in adulthood (De Young et al. , 2011). Not only are the short - term effects of trauma detrimental, but the long - term effects of traumatic events can cause further distress to the adolescent and can even be deadly. The effects of trauma do not sudden ly disappear after childhood, and if not treated appropriately, they will last into adulthood. This can then affect not only levels of distress, but general functioning. As such, examining potentially comorbid mental health symptoms (e.g., anxiety symptoms , depression symptoms) in addition to posttraumatic stress symptoms when determining impact of treatment on outcomes is necessary. Another factor to consider when thinking about outcomes is potential risk factors that can further intensify the negative outcomes of traumatic exposure. For example, being from a low socioeconomic status (SES) background can amplify the effects of trauma, and about 20% of youth in the United States live in poverty and 46% live in stressful housing situations (Federal Interagency Forum on Child and Family Statistics, 2013). Furthermore, poverty rates are higher for racial minority youth than racial majority youth (Macartney, Bishaw, & Fontenot, 2013), and youth from low - income and racial minority families are more likely to experience a higher number of traumatic events (United States Department of Health and Human Services [HHS], 2013). Additionally, Robert s, Gilman, Breslau, Breslau, and Koenen (2011) found that adults of racial majority backgrounds sought out treatment at higher rates than those of racial minority backgrounds, meaning that the individuals who are more likely to be affected by trauma are th e 17 least likely to receive the services they need to reduce their negative outcomes. Because of these poor outcomes , it is important to recognize the symptoms of trauma to provide the youth with the treatment that is needed . Defining What Makes Treatments Evidence - Ba sed To ensure a treatment will work for the individual it is being used with, there must be research supporting it s efficacy and effectiveness. To prove intervention efficacy, studies are conducted in controlled settings, such as labo ratories or clinical research facilities, and they use strong and precise methodology (Kratochwill & Shernoff, 200 3 ). Efficacy - based research, research designed and conducted in a laboratory setting, promotes internal validity of study findings due to rigo rous methodology ( Chorpita , Daleidan , & Weisz, 2005). Such research is typically conducted using randomized controlled trails (RCTs) that show how one treatment works better than another through comparison groups. F or a treatment to be considered efficacio us , replication is key . To be considered efficacious, treatment must show positive effects in at least two studies conducted by independent research teams (Chambless & Hollon, 1998 ) . If treatment efficacy has only been shown through one study or one research team conducted all the treatment research, the treatment is labeled as possibly efficacious . The two levels were updated to fo u article (i.e., well - established, probably efficacious, possibly efficacious, experimental) , and Southam - Gerow and Prinstein (2014) added one additional , lower level (i.e., tested). All these levels refer to the levels of efficacy of treatment based on available research. However, even if a treatment is proven to be efficacious, that does not necessarily mean it efficacy (studied in highly controlled situations) of interventions than on their effectiveness (implemented 18 treatment effectiveness research, it is not evident if a treatment will be effective in real world settings (i.e., external validity) , like in schools or non - laboratory clinics. Effectiveness research also provides information on the feasibility of treatment use (i.e., if it works in practice ) . For clinicians, however, there are a variety of factors that must be consid ered when determining treatment effectiveness since most of the research is on treatment efficacy. In order for a clinician to determine if an evidence - based treatment will actually be effective in practice, they must recognize whether or not the research demographics reflect client demographics, if the treatment is replicable according to the manual/research procedures, if the conditions of implementation (e.g., length, frequency, location of sessions) are raining is similar to the training described in the study (Kratochwill & Shernoff, 2003). Research must reflect that the treatment is both efficacious and effective by examining these conditions. This can help determine if a treatment is evidence - based. A treatment that is evidence - based is one which provides strong evidence of efficacy (studied in highly controlled settings such as clinics with control groups ) and effectiveness (implemented in natural contexts, as when examining treatment outcomes in scho ols ; Ingraham & Oka, 2006 ; Kratochwill & Shernoff, 2003). However, generalization of treatment efficacy and effectiveness is limited to the samples studied. As Chambless and Hollon (1998) noted, researchers must clearly define the populations that were examined in the research and this can help determine if a treatment is efficacious or effective for a specific population. Furthermore, to classify treatment as evidence - based, practitioners and res earchers examining the research base need to be able to answer the question for whom and under what condition is the treatment found to be effective. This is a critical question as it impacts the transportability of the treatment 19 to other settings and th e generalizability of the treatment to populations beyond those that were examined in research studies (Ingraham & Oka, 2006). This then provides the best internal and external validity for treatment outcome studies. As noted earlier, p sychological treatme nts , as opposed to other treatments (i.e., psychotropic medication), should b e the first line of treatment for youth exposed to traumatic events (Motta, 2015). Of the evidence - based treatments available for mental health issues , cognitive behavioral therap y ( CBT ) has been found to be an effective treatment for youth with various mental health issues (e.g., PTSD, anxiety, depression; Kendall & Choudhury, 2003). Theories Behind Cognitive Behavioral Therapy CBT consists of psychoeducation, relaxation and coping techniques (e.g., deep breathing, visualization), cognitive training on how to process thoughts and beliefs, and gradual, imaginal exposure (Silverman et al., 2008) . This approach has been found to be an effective mental health treatm ent for youth with varying mental health concerns. As can be seen in Figure 2 , CBT is a treatment that focuses on the belief psychological problems can mutually influence Tolin, 2016, p. 13) . These elements of psychological problems t ie in with the multidimensional etiology associated with trauma (see Figure 1) . There are theories that support the multidimensional etiology of trauma and support the various elements of CBT. Various theo ries have influenced the development of CBT , including respondent condition ing theor y, operant learning theory, social learning theory, bioinformational theory, and information processing theory (Benjamin et al., 2011 ; Gosch, Flannery - Schroeder, Mauro, & Compton, 2006 ). According to respondent condition ing theory, a neutral conditioned stimulus is followed by a n unconditioned stimulus that causes an unconditioned response (e.g., fear, 20 anxiety) . After repetition of the condition ed stimulus and unconditioned stimulus pairing (i.e., habituation) , the conditioned stimulus alone can elicit the unconditioned response . Treatment then tackles this fear response through habituation and extinction . Extinction of the conditioned response o ccurs through repeated exposure to the conditioned stimulus in the absence of the unconditioned stimulus . Habituation then occurs since the individual is exposed to the feared stimulus (e.g., situational reminders of trauma, memory of the trauma) without h arm ( Watson & Rayn e r, 1920 ). This theory is reflected through the gradual, imaginal exposure component of CBT. However, this theory is not enough to describe the impact of trauma on a child since an acute , one - time event can elicit negative outcomes. Operant learning theory has also influenced the development of CBT. According to ( Skinner, 1969 ) . Thus, reinforcement, wheth er positive or negative, influences behavior. For example, positive reinforcement may occur if a youth receives attention when engaging in certain behaviors, thus shaping the child to engage in these behaviors to receive attention. On the flip side, indivi appropriate , desired behaviors. These types of reinforcement behaviors from stakeholders (e.g., parents, teachers) can then shape the youth to engage in behaviors like hypervigilance to threat ( Gos ch et al., 2006) . It is through reinforcement from the mental health practitioner and parent that desired behaviors are increased, and undesired behaviors are reduced. Social learning theory states that learning occurs through both observation and experi ence ; this implies that fear responses can be learned (Bandura, 1977) . This is important because youth may experience a traumatic event with another individual, such as a parent, sibling, or - efficacy to cope with a traumatic event i s dependent on both on 21 ability to cope with the traumatic event. Teaching and modeling of appropriate coping techniques is thus an important part of CBT as it build s self - efficacy regarding the use of such tec hniques ( Gosch et al., 2006). Bioinformational theory states that memory is stored as an image with affect attached to it; the affective image is not purely a raw observation of the event, but rathe r, it contains interpretive elements like emotions and cognition (Lang, 1977) . Thus, through the careful identification of emotions and cognitions surrounding an event, the emotions and cognitions that interfere with functioning can be interrupted. The CBT component of exposure plays a major part in this interruption while also providing habituation of the event so that it no longer provides that sense of fear. Lastly, information processing theory has influenced the development of CBT. According to infor mation processing theory, biases occurring at the level of perception, encoding, interpretation, and retrieval of information 2006, p. 251). Due to processing being biased by these negative cogni tions , attention and memory are impacted. This, in turn, causes youth to be hypervigilant towards the detection of potential danger, and it makes it more likely that youth will interpret innocuous events as threatening. CBT focuses on recognizing how one is feeling , what one is thinking, and how one behaves due to activation of fear. It then us response while also using other cognitive tools to reframe maladaptive cognitions. It is through this combined theoretical understanding of emotions, cognitions, and behavior, CBT has developed into a treatment that focuses on these three elements to build youth well - being and functioning. 22 The Use of Cognitive Behavioral Treatments The conceptualization of CBT is that the psychological problems associated with traumatic events occur due to both internal and ex ternal factors, such as emotions, physiological reactions, cognitions, behaviors, information processing biases, and behavioral skills deficits (Tolin, 2016). To counteract the negative impacts of traumatic events, treatment must address these internal and external factors, which CBT does. For example, research has shown that - eliciting experiences and situations can . This means that the strong, negative emotions that are associated with inappropriate behaviors and/or harmful cognitions can be limited. Since it is difficult for youth who are emotionally dysregulated to regain control through cognitions (Ford, 2015), e motion regulation through techniques taught in CBT, like deep breathing, play a crucial role in treatment and overall youth outcomes. Originally, CBT was used to target youth with externalizing behaviors (e.g., aggression ; Kendall & Choudhury, 2003). Early CBT models and research targeted impulsivity and self - control ( Kendall & Braswell, 1982 ) and aggression and antisocial behaviors ( Tolan, Guerra, & Kendall, 1995 ) . Once CBT was determined to be efficacious for externalizing disorders and concerns, focus of research shifted to using CBT for internalizing disorders . Depression and anxiety were the two internalizing disorders that were examined early on . CBT was found to be more eff icacious than systemic behavior family therapy and individual nondirective supportive therapy for youth with depression ( Brent, Holder, & Kolko, 1997) , and youth with anxiety maintained positive long - term gains after CBT ( Kendall & Southam - Gerow, 1996). It was only 23 after the impact of CBT was examined on these disorders that the focus turned to trauma and its associated disorders (i.e., posttraumatic stress symptoms) . According to the American Psychological Association ( [ APA ] , 2008 ) , CBT h as been found to be an efficacious psychological treatment fo r youth exposed to traumatic events . Furthermore, the American Academy of Child and Adolescent Psychiatry ( [AACAP], 2010 ) recommended trauma - focused psychological treatments , such as CBT , be the first line of treatment for youth with PTSD , meaning those who have some of the most severe negative reactions to traumatic event exposure . Additionally, as seen in Table 2, the current literature on CBT for youth exposed to traumatic events has shown that CBT, such as trauma - focused cognitive - behavioral therapy ( TF - CBT ) , is more impactful on various measures (e.g., posttraumatic stress, depression, anxiety) than play therapy, standard community - based therapy, supportive therapy, and no treatment (Borntrager , Chorpita, Higa - McMillan, Daleiden, & Starace, 2013 ; Brown et al., 2017; Harvey & Taylor, 2010; Silverman et al., 2008; Slade & Warne, 2016 ). The consensus is that the best evidence - based treatments for youth w ho have experienced trauma tic even ts are those which use cognitive behavioral techniques in a trauma - focused manner, such as through TF - CBT ( Silverman et al., 2008 ). Prior meta - analyses consolidated the CBT data into CBT versus non - CBT, but they did not go into detail about the CBT subtrea tments . For example, t reatments such as cognitive behavioral intervention for trauma in schools ( CBITS ) , a cognitive behavioral treatment developed for use specifically in the schools , were listed as possibly efficacious (Silverman e t al . , 2008), but they have not been compared to other CBT subtreatments . Lastly, prior meta - analyses have not examined the impact of treatments using general cognitive behavioral techniques (e.g., Prolonged Exposure ) . Understanding the treatment components and what works helps since it informs which processes 24 cause treatment effects and which do not, which can then be used to make treatment more streamlined and cost effective (Kazdin & Weisz, 1998). Trauma - focused cog nitive behavioral therapy (TF - CBT) . TF - CBT is a well - established (Dorsey et al., 2017) manualized evidence - based treatment that can be used with youth who have experienced trauma. TF - CBT was developed by Cohen, Mannarino, and Deblinger (2006) , and it - sensitive interventions, cognitive - behavioral principles, as well as aspects of attachment, developmental neurobiology, family, empowerment, Cohen et al., 2006, p. 32). There are approximately 12 stru ctured child, parent, and conjoint sessions implemented weekly for 90 minutes , a trained therapist goes through eight components of treatment through the PRACTICE model. This PRACTICE model includes: psychoeducation and parenting skills (P), relaxation tec hniques (R), affective modulation (A), cognitive coping and processing (C), trauma narrative (T), in vivo exposures to the trauma (I), conjoint child - parent session (C), and enhancing safety and future child development (E; Cary & McMillen, 2012; Cohen et al., 2006). Cognitive behavioral intervention for trauma in schools (CBITS) . CBITS is another manualized evidence - based treatment that can be used with youth who have experienced trauma , which has been determined to be probably efficacious (Dorsey et al. , 2017). T his treatment was developed for impl ementation in schools . As mentioned previously, research has shown that individuals from racial minority backgrounds are less likely to seek out services ( Roberts et al., 2011) , but in schools, disparities in mental health service delivery rates can be decreased. This is d ue to access to youth i n the s chools , so it is an ideal location for practitioners to implement treatment for youth who have experienced traumatic events. 25 C BITS was developed to address violence exposure to recent immigrant students between a partnership between the RAND Corporation, the University of California, Los Angeles, and the Los Angeles Unified School District. The treatment was originally developed for youth from fourth through eighth grades, though it has been adapted for older youth . I t includes ten group sessions of six to eight youth , one to three individual sessions, and two psychoeducation meetings for parents (Jaycox, Kataoka, Stein, Langley, & Wong, 2012 ) . CBITS consists of many elements like those in TF - CBT . These consistent elements include psychoeducatio n , relaxation training, recognition and challenging of cognitive distortions , approaching trauma reminders and triggers instead of avoiding them , safety assessment , developing a trauma narrative, and problems - solving ( Jaycox et al., 2012). However, CBITS was specifically developed for use in the schools with racial minority students while TF - CBT is typically used in clinic al settings. Silverman and colleagues (2008) examined CBT effectiveness - analysis only examined TF - CBT effectiveness for youth exposed to abuse. Other treatments that use cognitive behavioral techniques . According to the NCTSN (201 8 ), Blueprints for Healthy Youth Development (2018), and the California Evidence - Based Clearinghouse for Child Welfare ( [CEBC], 2018) , there are other, lesser known treatments which use cognitive behavioral techniques that also h ave evidence supporting their use amongst youth with trauma. Table 3 provides a summary of the other lesser known treatments with more a more limited research base that use cognitive behavioral techniques. Prior research has not labeled the evidence base l evels (e.g., well - established, probably efficacious, experimental) of studies including these lesser known treatments . 26 Need for Meta - Analyses To determine the current state of literature on treatment, meta - analyses are an excellent methodological tool which consolidate s data of all the past research on a specific treatment ( Lipsey & Wilson, 2000). As noted by Cheung, Ho, Lim, and Mak (2012), a meta - analysis is a method used to combine and compare effect sizes from various related studies ; these effect sizes measure the strength and direction of the effect (e.g., treatment effect ) of studies . Meta - analyses are important for health care practit ioners as it provides a snapshot of the current research, which allows them to stay updated on the current best practices ; it also provides systems - level stakeholders and organizations (e.g., APA) evidence for which they can develop practice guidelines ( Ox man, Cook, & Guyatt, 1994). Meta - analyses can also help label the level of evidence for a treatment (e.g., well - established, probably efficacious; Southam - Gerow & Prinstein, 2014) , and determine treatment generalizability through a thorough examination of the populations studied in the research . T he best practices for conducting a meta - analysis is provided through the PRISMA statement (Moher et al., 2009) , which lists the expectati ons for meta - analyses and steps for conducting meta - an alyses ; by using these guidelines, researchers are likely to find the most accurate effect sizes, thus providing practitioners with the most up to date research. Using these guidelines can also help res earchers answer other questions about treatment beyond effect size, such as the variables that impact treatment outcomes. Research on Cognitive Behavioral Therapy for Children and Adolescents There is still a great deal of research that needs to be conducted regarding evidence - based treatment, especially regarding culturally competent treatment , and to do so, there needs to be a thorough examination of the moderator and mediator variables (Cohe n, Deblinger, & Mannarino, 2018 ; Silverman et al, 2008 ). Understanding the moderator and mediator variables that impact 27 treatment outcome s is important since these variables can play a part in whether a treatment is effective . This, in turn, informs resear chers and practitioners if the treatment should be administered to that population or in that setting . Moderators specify the circumstances or conditions under which a specific effect occurs (Holmbeck, 1997). On the other hand, to understand how treatment works, there needs to be an understanding of the mechanisms of change, which refers to the processes that cause therapeutic change (Kazdin & Nock, 2003). Thus, m ediation refers to the occurrence of therapeutic change while moderation involves for whom and under what conditions this therapeutic change occurs. In psychological treatment research, the variables that lead to and cause change are called mediators. In their evidence update of psychological treatments for depression, Weersing and colleagues ( 2017) found that while CBT was the treatment most frequently used and met criteria as a well - established intervention, results were poorer for particular populations and settings (e.g., child guidance clinics) . A n analysis of the trauma literature regarding CBT outcomes has not yet occurred that examines moderating and mediating factors to have a clearer understanding of for whom CBT works and under what conditions . Trauma type. Knowing the type of traumatic event a youth experien ces is important in understanding the risk for negative outcomes because research has shown that although non - interpersonal trauma is more common, interpersonal trauma is more likely to lead to PTSD symptoms in adults (Kessler, Sonnega, Bromet, & Hughes, 1 995). Additionally, although research has focused on acute single traumatic events, chronic events can potentially have a cumulative effect on outcomes ( Goenjian et al., 2001; Salloum & Overstreet, 2008 ) . Because of this cumulative effect and the possibili ty that chronic and interpersonal traumatic events could still be occurring to the youth, there is a chance that treatment may not be as effective for youth 28 exposed to chronic , interpersonal traumatic events. Research should ideally show that CBT has a positive impact for youth experiencing all type s of trauma exposures . Prior research on overall psycho logical treatments has provided evidence that sexual abuse treatment had higher posttraumatic stress symptom and depression symptom e ffects than other trauma types (Silverman et al., 2008) . Additionally, prior research has specifically examined CBT impact on youth exposed to sexual abuse/assault ( Harvey & Taylor, 2010; Slade & Warne, 2016 ) . However, meta - analysis research has not examined the impact of CBT by the type of trauma to which youth are exposed in regard to posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes . Cultural/d emographic characteristics . To ensure that research is generalizabl e to all populations of youth, it is important that the samples examined in a study reflect the population for which the intervention is meant to be used. Thus, it is i mportant to identify any moderating cultural (i.e., demographic) factors of the groups e xamined in the literature . The factors that are typically examined are age, gender, socioeconomic status, and race (Nock, 2003); however, this meta - analysis examine d some of these factors in addition to other factors that could potentially have a moderating effect on outcome s. identity is defined by several variables such as race, socioeconomic status, age , and gender (Ingraham & Oka, 2006), so research studies must examine individuals who have diverse cultural factors to d etermine if there are certain factors that impede or promote treatment efficacy or effectiveness. As mentioned in the introduction, while the moderating effect of cultural factors have been examined in the past, prior meta - analyses have not examined these factors in CBT for youth exposed to traumatic events. 29 Race . Research has shown that African Ameri can, Native American, and Latino youth are more likely to experience traumatic events than European American or Asian American youth (Woodbridge et al., 2016). The majority of TF - CBT studies include data on the racial background of the sample, but most of the youth in the studies were either White/ European American or Black/ African American (Cohen & Mannarino, 19 96; Cohen , Mannarino, & Knudsen, 2004; Cohen , Mannarino, & Iyengar , 2011; Scheeringa , Weems, Cohen, Amaya - Jackson, & Guthrie, 2011). In comparison, CBITS was developed for immigrant youth ; there have been studies examining CBITS effectiveness amongst Latino, Native American, and African American youth ( Jaycox et al., 2009; Morsette, van del Pol, Schuldberg, Swaney, & Stolle, 2012; Stein et al., 2003) . However, the number of youth involved in these studies were small, limiting generalizability. This limited research on racial minority youth is concerning since race is a predictor of treatment engagement ; r acial minority youth are more likely t o drop out of treatment than European American yout h , and they are less likely to access mental health services ( Atdjian & Vega, 2005 ; Fraynt et al., 2014 ). Because of factors such as attrition and buy - in with youth of color, there is a chance that treatment will not be as effective for non - European Americ an youth . Thus, further research is need ed to understand the potential moderating effect of race on intervention effective ness for youth of all racial backgrounds (e.g., African American, Asian American , Arab American , European American , Hispanic/ Latino, Native American, youth of mixed racial backgrounds ). Gender . Gender can impact the prevalence of experiencing a traumatic event. For example, p rior research has shown that females exper ienced childhood sexual abuse at higher rates than male s, and that females who experienced childhood sexual abuse were more likely to develop PTSD than males who 30 experienced childhood sexual abuse ( Walker, Carey, Mohr, Stein, & Seedat, 2004). Despite the d ifferent types of traumatic event exposures by gender, s ince CBT has been shown to be effective for victims of childhood sexual abuse, it is likely that males and females will both have better outcomes after exposure to treatment. Past research has shown t hat for overall psychological studies, gender did not moderate treatment outcomes ( Brown et al., 2017), but further research needs to examine the impact of gender on CBT posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes. Age . age . This is important to consider when implementing treatment since 2003, p. 98). For example, younger children may not have the cognitions necessary to fully access certain components of treatment, thus making it less effective. According to Cary and McMillen (2012), a great deal of the research on TF - CBT and CBITS effectiveness has been conducted with youth aged eight through 14 or 15 years old. Scheeringa and colleagues (2011) implemented TF - CBT with children between the ages of three and six, and Cohen and Mannarino (1996) worked with preschoolers. However, due to small sample sizes, at the time, generalizability of this research was limited. Past meta - analyses have shown mixed information regarding age as a moderator, with Brown and coll eagues (2017) finding age did not moderate psychological treatment posttraumatic stress symptom outcomes and Gutermann and colleagues (2016) finding that age did moderate psychological treatment posttraumatic stress symptom outcomes. Further research needs to be conducted to determine the moderating effect of age on posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes. 31 Study design. Another moderator variable that needs to be examined is the study design. Most research has examin ed efficacy rather than effectiveness, and while CBT and other evidence - based treatments may be efficacious, it may be difficult to individualize these treatments, thus making them less preferred by therapists (Chorpita, Daleidan, & Weisz, 2005). Research needs to be conducted in less controlled settings to determine if the intervention is effective in practical settings (Nock, 2003). However, conducting such research can be difficult since it involves balancing a strong methodology with research that is ac ceptable, cost effective, and considers the context of implementation (Kratochwill et al., 2012). A TF - CBT study that evidenced effectiveness was the research conducted by Cohen, Mannarino, and Iyengar (2011). While this was a randomized controlled trial, shelter by social workers. The researchers collected data related to treatment integrity to ensure TF - CBT was implemented with fidelity since the TF - CBT literature showed that when implemented according to the man ualized instructions, TF - CBT is beneficial. This study provided a more practical application of TF - CBT, showing effectiveness of the treatment with this population as long as the treatment was followed exactly as prescribed according to the TF - CBT manual. However, as seen in Table 1, only three prior meta - analyses have examined studies beyond RCTs (Brown et al., 2017; Dorsey et al., 2017; Harvey & Taylor 2010), and none of these studies specifically examined the effectiveness of CBT on posttraumatic stress symptoms, anxiety symptoms, and depression symptoms with youth exposed to traumatic events. Setting . Yet a nother moderator variable that needs to be examined is the setting in which treatment occurs. For example, a school - based intervention may increase the likelihood of reaching more students due to accessibility and it could reduce the stigma associated with mental health treatment (Myschailyszyn, 2015). However, a clinical setting may provide more control 32 and may allow for more time for treatment work than a school setting due to the academic commitments a student needs to meet in school. Transportation of treatment across settings requires evaluation (Kendall & Choudhury, 2003). One way of examining this is by looking at TF - CBT literature in compariso n to CBITS literature. Since TF - CBT is generally done in a more clinical setting and CBITS was specifically designed to be implemented in a school setting, a comparison of these two types of treatments may shed light on CBT treatment posttraumatic stress s ymptom, anxiety symptom, and depression symptom outcome differences by location of study implementation (i.e., school vs. clinic setting). Prior research has found setting to not moderate treatment outcomes for overall psychological treatments, but underst anding posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes for CBT would further the literature on treatment transportability. Treatment components. To determine the mechanisms of change, the treatment components for CBT must be examined. It is necessary for practitioners to know exactly which components result in positive outcomes, as it provides crucial information regarding what must be implemented (Kaufman et al., 2005). For example, parental involvement is sometimes a part o f CBT. Knowing if youth who have parents involved (i.e., a mediator) in treatment have better outcomes is important as it will inform practitioners about the potential need for parental involvement. Another important component to consider is the trauma na rrative exposure (i.e., mediator) component of treatment. While all evidence - based trauma treatments include some - reported therapy techniques, only 14 - 22% of their youth sample engag ed in exposure techniques. Additionally, many youth and parents refuse to engage in treatment that involves a thorough discussion of the traumatic event (Connor, Ford, Arnsten, & Greene, 2015). While prior research has examined 33 mediators of CBT for adolesc ents with other mental health disorders (e.g., depression and conduct disorder; Kaufman et al., 2005), no prior meta - analyses has examined the mediating - analysis provided a narr ative description of the literature available on treatment components, it solely examined the trauma narrative component. This is not sufficient; as Weersing and Weisz (2002) noted, self - talk had a mediating role for certain outcome variables in a CBT anxi ety treatment. It is necessary to know if the same is true for CBT trauma treatment. Additionally, while prior meta - analyses have examined mediator variables (e.g., treatment delivery, session number, session length) for general psychological treatments wi th youth exposed to traumatic events (Brown et al., 2017; Gutermann et al., 2016; Harvey & Taylor, 2010; Silverman et al., 2008) , these studies did not examine the mediating effects of these variables for CBT only in relation to posttraumatic stress symptoms, anxiety symptoms, and depression symptoms. R esearch Questions Table 4 lists the moderating and mediating variables investigated to fill in gaps in the current CBT research related to what works for whom under what conditions . This meta - analysis examined who best responds to CBT by examining trauma type, race, gender, and age as moderator s of treatment outcomes. Lastly, this meta - analysis expands the current literature by including all treatments using CBT techniques and comparing their impact on posttraumatic stress symptom, anxiety symptom, and depression symptom out comes . Through a meta - analysis , this study answer ed the following questions: 1. What is the overall impact of CBT and CBT subtreatments on improving mental health outcomes in children and adolescents ? 34 a. Does CBT reduc e symptoms of posttraumatic stress, anxiety, and depression as examined by effect size, when compared to other trauma treatments (e.g., EMDR, play therapy) , waitlist controls , or pre - post design ? Based on prior research showing CBT impact over other treatment types (e. g., Slade & Warne, 2016), it was hypothesized that CBT would be more impactful than other trauma treatments or waitlist control groups. b. Are there any differences in CBT subtreatment impact (e.g., TF - CBT, CBITS) in children and adolescents as examined by t he effect sizes of various psychological construct s (i.e., reductions in posttraumatic stress, anxiety, or depression symptoms ) ? It was hypothesized tha t there would be a significant decrease in symptomatology for youth in all CBT subtreatments after exposure to CBT since it has been listed as effective treatment for youth exposed to trauma (APA, 2008) and prior research with diverse samples ( Gillies et al., 2012; Gutermann et al., 2016; Kowalik et al., 2011; Silverman et al., 2008) has found CBT to be an effective treatment . It was hypothesized that regardless of CBT sub treatment used , insignificant difference s in treatment efficacy would be found. The rationale for this hypothesis is that all the components of CBT that are nec essary to address the biological, ecological, and psychological fac tors impacted by exposure to traumatic events are included in every sub treatment (Gutermann et al., 2016). 2. Are there specific factors that moderate the effectiveness of CBT treatment for ch ildren and adolescents? a. Does trauma exposure type impact CBT treatment outcomes (i.e., reductions in posttraumatic stress, anxiety, or depression symptoms)? 35 It was hypothesized that there would be a significant decrease in symptomatology for youth expos ed to all trauma types after exposure to CBT since it has been listed as effective treatment (APA, 2008) and prior research with diverse samples (Gillies et al., 2012; Gutermann et al., 2016; Kowalik et al., 2011 ) . Prior research has indicated that trauma type may be a potential moderating factor (Silverman et al., 2008) . Thus, it was hypothesized that trauma type would moderate posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes, and sexual abuse/assault would have higher effect sizes than other trauma types. b. Are there cultural variables (e.g., racial identity, gender, age) that impact CBT tr eatment outcomes (i.e., reductions in posttraumatic stress, anxiety, or depression symptoms)? It was hypothesized that there would be a significant decrease in symptomatology for youth of varying racial identities , gender s , and age s after exposure to CBT since it has been listed as effective treatment for youth exposed to trauma (APA, 2008) and prior research with diverse samples (Gillies et al., 2012; Gutermann et al., 2016; Kowalik et al., 2011; Silverman et al., 2008) has found CBT to be an effective t reatment. Since different cultural populations experience different types of traumatic events, there may be cultural factors that impact treatment outcomes. However, past CBT research (e.g., CBITS; Jaycox et al., 2009; Morsette et al., 2012; Stein et al., 2003) has found treatment to be effective for youth of varying racial identities . Thus, it was hypothesized that treatment w ould be effective regardless of youth racial background and that there would be significant reductions in outcomes after implementat ion of treatment . Young children might be less likely to experience the benefits of treatment due to limited cognitive abilities (Kendall & Choudhury, 2003) , and prior research has shown older youth had larger 36 effect sizes than younger youth (Gutermann et al., 2016) , so it was hypothesized that older youth would have more positive outcomes from treatment than young children. c. Does study design (e.g., RCT versus pre - post and quasi - experimental) impact the outcome (i.e., reductions in posttraumatic stress, anxiety, or depression symptoms) for CBT treatments? It was hypothesized that there would be a significant decrease in symptomatology regardless of study design after exposure to CBT since it has been listed as effective treatment for youth exposed to tra uma (APA, 2008) and prior research with diverse samples (Gillies et al., 2012; Gutermann et al., 2016; Kowalik et al., 2011; Silverman et al., 2008) has found CBT to be an effective treatment. Furthermore, past research (Harvey & Taylor , 2010 ) has shown that pre - post and quasi - experimental studies had lower effects than experimental studies, so it was hypothesized that RCTs would have larger effect sizes than non - RCTs. d. Does CBT treatment setting (e.g., laboratory/clinic versus school) impact treatment outcomes (i.e., reductions in posttraumatic stress, anxiety, or depression symptoms)? It was hypothesized that there would be a significant decrease in symptomatology for youth in all treatment settings after exposure to CBT sinc e it has been listed as effective treatment for youth exposed to trauma (APA, 2008) and prior research with diverse samples (Gillies et al., 2012; Gutermann et al., 2016; Kowalik et al., 2011; Silverman et al., 2008) has found CBT to be an effective treatm ent. Regarding treatment setting, while school settings have benefits ( Myschailyszyn, 2015) , the clinical setting might provide more time and resources than the school setting . Further, prior research has shown that teachers had lower effect sizes than 37 oth er mental health professionals (Brown et al., 2017), so it was hypothesized that school settings would have lower effect sizes than other treatment settings (e.g., clinics). 3. Are there specific factors that mediate the effectiveness of CBT treatment for children and adolescents? a. Does parental involvement impact the effectiveness of CBT treatment on mental health outcomes (i.e., reductions in posttraumatic stress, anxiety, or depression symptoms)? It was hypothesized that there would be a significant dec rease in symptomatology for youth with or without parental involvement in treatment after exposure to CBT since it has been listed as effective treatment for youth exposed to trauma (APA, 2008) and prior research with diverse samples (Gillies et al., 2012; Gutermann et al., 2016; Kowalik et al., 2011; Silverman et al., 2008) has found CBT to be an effective treatment. It was hypothesized that specific components of treatment would impact the effectiveness of CBT treatment. Furthermore, it was hypothesized t hat f actors such as parental involvement would be correlated with positive outcomes since prior research has shown that parental involvement had larger effects on depression symptom and anxiety symptom outcomes (Silverman et al., 2008), and treatment with parents had larger effects than treatment without parents (Gutermann et al., 2016) b. Are there other mediating factors (e.g., length and frequency of treatment) that impact tre atment outcomes (i.e., reductions in posttraumatic stress, anxiety, or depression symptoms)? It was hypothesized that there would be a significant decrease in symptomatology for youth despite the various mediating factors after exposure to CBT since it has been listed as effective treatment for youth exposed to trauma (APA, 2008) and prior research with diverse samples 38 (Gillies et al., 2012; Gutermann et al., 2016; Kowalik et al., 2011; Silverman et al., 2008) has found CBT to be an effective treatment. S ince prior rese arch has found individual treatment to have higher effect sizes than group treatment (Gutermann et al., 2016; Harvey & Taylor, 2010), it was hypothesized that treatment delivery would mediate outcomes. Furthermore, while length and frequency of treatment m ay allow for more intensive work with youth, prior research has shown these factors do not mediate treatment outcomes (Harvey & Taylor, 2010). Thus, it was hypothesized that while these factors would be correlated with positive outcomes , they would not med iate treatment o utcomes . 39 CHAPTER 3 METHOD S The current study used a meta - analysis to investigate the impact of cognitive behavioral therapy (CBT) on youth outcomes , and the moderating and mediating variables of treatment . While empirical research studi es involve working with participants for data collection purposes, a meta - analysis involves collecting past studies as secondary data and synthesizing this data ( Lips ey & Wilson, 20 00) . The Preferred Reporting Item s for Systematic Review s and Meta - Analyses ( PRISMA ) guidelines ( Moher et al., 2009) , a set of itemized guidelines aimed to assist researchers in conducting systematic reviews and meta - analyses, were used in the writ ing of this dissertation study . Literature Search In order to condu ct a comprehensive literatu re se arch and find an appropriate number of studies for meta - analytic purposes, a thorough search of relevant databases for studies that met inclusion criteria was completed (Lipsey & Wilson, 2000). For the purpose of this meta - a nalysis, the following databases were searched: Psy cINFO , EBSCO, ERIC, and ProQuest Dissertations and Theses . ProQuest Dissertation and Theses was included for the purpose of find ing any unpublished studies that may be relevant for the purposes of this study. From these databases, the following key words were used in the literature search: 1. child * OR adolesc * OR youth OR teen * OR pediatric OR young 2. PTSD OR posttrauma * OR post - trauma* OR post trauma* OR trauma* 3. cognitive behavior * OR CBT OR treatment OR intervention OR therapy OR psychotherapy 40 4. OR neglect OR maltreat* OR mistreat* OR refugee OR war OR hurricane OR tsunami OR tornado OR earthquake OR flood OR victim Each asterisk above indicates that all terms with that root were utilized as a keyword. Furthermore, the reference sections of articles found through the keyword search were examined to identify other potential studies. Prior meta - analytic stu were also examined. Lastly, direct requests via email were sent to 14 promin ent researchers in the field of CBT among youth exposed to traumatic events to find studies that have not yet been , or will not be , published. A researcher was considered prominent if they designed the specific subtreatment or if they had at least two studies of CBT for youth exposed to traumatic events published. As seen in Figure 3, the initial search produced 22,334 articles , of which 3,062 were duplicate s . Eighteen relevant article s were found outside the initial search through examining the references of prior meta - analyses ( k = 3) and included research studies ( k = 15) . No articles were included due to the communications with researchers. Inclusion criteria. To be included in this study, articles needed to : (a) contain original data (i.e., be a treatment outcome study, not a description of treatment and not a literature review of treatment ) ; (b) involve individuals 18 years or younger in the study sample ; (c) involve participants who have been exposed to traumatic events ; and (d) involve treatment that uses cognitive behavioral techniques (i.e., psychoeducation, relaxati on and coping techniques, cognitive training, and gradual, imaginal exposure) . All components of CBT had to be included to be included in the study , as seen per Figure 3 . If a study only e xamine d certain pieces of 41 cognitive behavioral techniques, such as engagement in only cognitive therap y approaches , this was excluded from the meta - analysis. Studies examining both efficacy (i.e., RCTs) and effectiveness (i.e., naturalistic, pre - post design) were included in the me ta - analysis . Only articles written in English were included in this meta - analysis. Articles included were not limited to a specific time period . The PsycINFO search resulted in studies from 1907 onwards, the EBSCO search resulted in studies from 1918 onwar ds, the ERIC search resulted in studies from 1969 onwards, and ProQuest Dissertations and Theses search resulted in studies from 1953 onwards. Ninety - four articles met all criteria for inclusion in the meta - analysis . Three of the 94 studies involved more t han one relevant comparison group in relation to the purpose of this meta - analysis, resulting in three of the studies contributing a second effect size estimate. Each effect size estimate was independent of the others. Thus, the equivalent of the total sample of 97 effect sizes (i.e., distinct data points from 94 studies ) w ere available for the meta - analysis. In Jaycox an behavioral treatment (i.e., TF - CBT and CBITS) were used, and thus this study was treated as two te studies because one treatment group included parental involvement and the other group did not have studies because one treatment group included parental invol vement and the other did not have parent involvement. Of the 97 data points from 94 studies , 95 had posttraumatic stress symptom outcome data , 39 had anxiety symptom outcome data, and 65 had depression symptom outcome data . While a number of the included studies had data regarding other outcome variables (e.g., externalizing symptoms, emotion regulation), only posttraumatic stress symptom , anxiety symptom , and depression symptom data were investigated within this meta - analytic study . 42 Data Coding All data coding procedures were tracked using the PRISMA flowchart (see Figure 3) and Microsoft Excel. The data coding involve d two steps. In the first step, an initial search and screening was completed by the primary author to exclude studies that d id not meet i nclusion criteria , as seen in Figure 3 . To do so, the article title and abstract were read . A random selection of 10% of the articles found were examined by a second coder to determine if they met inclusion criteria for inter - rater reliability purposes without correction for chance agreement. There was 99% agreement between raters. Areas associated with disagreement included incorrectly labeling a study as a treatment study when it was a literature review , incorrectly identifying a treatment as CBT, and disagreement about sample age. Reasons for exclusion from the meta - analysis , such as noting that the study did not include youth or study did not involve trauma, were noted . Studies that appeared to meet inclusion criteria were coded independently by the primary author and a second coder for accuracy and reliability purposes. The articles to be included in the study were coded using a coding sheet in Microsoft Excel . Data code d for studies meeting inclusion criteria include d variables such as type of analysis, demographic data, information on the youth outcomes, type of treatment, and type of study (i.e., clinical versus naturalistic). The data coding manual is attached in Appe ndix B . A random selection of 10% of the articles that were to be included in the meta - analysis were coded by the second coder for inter - rater reliability purposes . There was 96% agreement between raters . In instances where disagreement occurred, analysis of the pattern of differences and discussion resulted in resolution. Areas associated with the disagreement included mislabeling of the comparison group, disagreement with the type of study design, and spe cifics regarding trauma type . 43 Measures used to determine change . Y outh mental health functioning and behavior changes for trauma treatments were measured diversely across studies. A variety of mental health constructs including posttraumatic stress sympto ms, anxiety symptoms, and depression symptoms , and a diverse group of measures used across studies were used to answer research question 1 . Table 8 shows the specific outcome measures used and the specific study that used the outcome measure to determine the impact of CBT on mental health functioning. Self - report measures were typically used in the treatment studies; if a self - report measure was not used, a parent report was used. To highlight the diversity of outcome measures used across stu dies, th e following were those used to measure posttraumatic stress symptoms: Anxiety Disorders Interview Schedule for DSM - IV (Silverman & Albano, 1996) , Child and Adolescent Trauma Screen (Sachser et al., 2017) , Child Post - Traumatic Stress Disorder Reacti on Index ( Pynoos et al., 1987 ) , Child PTSD Symptom Scale (Foa et al., 2001) , Child Report of Post - Traumatic Stress Symptoms (Greenwald & Rubin, 1999) , Child Revised Impact of Events Scale (Smith et al., 2003) , Child Stress Scale (Lipp & Lucarelli, 1998) , C Revised (Wolfe et al., 1991) , - Traumatic Stress Reaction Index (Nader & Fairbanks, 1994) , Clinician - Administered PTSD Scale Child and Adolescent Version (Nader et al., 1994) , Diagnostic Interview Schedule for Children (Shaffer et al., 2000) , DSM - III - R PTSD Symptoms (APA, 1987) , DSM - IV Interview (APA, 2003) , Impact of Events Scale Revised (Weiss, 2004) , Post - Traumatic Stress Symptoms in Children (Ahmad et al., 2000) , Preschool Age Psychiatric Asse ssment (Egger et al., 2006) , Schedule for Affective Disorders and Schizophrenia for School - Age Children (Kaufman et al., 1997) , Trauma Symptom Checklist for Children (Briere, 1996) , Trauma Symptom Checklist for Young Children (Briere, 2005) , UCLA PTSD Reac tion Index for 44 DSM - IV (Steinberg et al., 2004) , Young Child PTSD Checklist (Scheeringa et al., 2010) , and Youth Self - Report (Achenbach, 1991) . To highlight the diversity of outcome measures used across studies, the following were those used to measure an xiety symptoms : Multidimensional Anxiety Scale for Children (March et al., 1997) (Reynolds & Richmond, 1985) , Screen for Child Anxiety Related Disorders (Birmaher et al., 1999) (Sp ence, 1998) , State - Trait Anxiety Inventory for Children (Biaggio & Spielberger, 1983) , Trauma Symptom Checklist for Children (Briere, 1996) , and Trauma Symptom Checklist for Young Children (Briere, 2005) . To highlight the diversity of outcome measures used across studies, the following were those used to measure depression symptoms : Beck Depression Inventory (Beck et al., 1961) , The Behavior Assessment System for Children, 2 nd Edition (Reynolds & Kamphaus, 2004), , Center for Epidemiological Studies Depression Scale (Radloff, 1977) (Kovacs, 1992) , Depression Self - Rating Scale (Birleson, 1981) , Montgomery - Asberg Depression Rating Scale (Svanborg & Asberg, 1994) , Moods and Feel ings Questionnaire (Angold et al., 1995) , Preschool Age Psychiatric Assessment (Egger et al., 2006) , Reynolds Adolescent Depression Scale (Reynolds, 1987) , Trauma Symptom Checklist for Children (Briere, 1996) , and Trauma Symptom Checklist for Young Childre n (Briere, 2005) . Moderator and mediator variables. Treatment moderators (e.g., trauma type, race , study design ) and mediators (e.g., stakeholder involvement , inclusion of other treatment techniques such as narrative therapy techniques) were coded per the approach described in Appendix B . This coding approach was used to guide the data analysis of potential moderating and mediating effects of treatment. While the terminology and conceptualization of moderators 45 and mediators were used throughout thi s paper since it is how it is described in the literature, the analysis for these two factors (i.e., as covariates) were the same , meaning that the distinction between moderators and mediators is purely conceptual, not statistical. Data Analysis The data - analysis package, Comprehensive Meta - Analysis , Version 3 (CMA; Borenstein et al., 2005) was used to conduct all analyses. For R CTs, to calculate the standardized mean difference, d , post - test differences between control and experimental groups were examined using the following formula: Where is the control group mean and is the experimental group mean, and SD within is the within groups standard deviation, which were calculated as follows: Where is the sample size of group 1 and is the sample size of group 2, and where SD 1 is the standard deviation of group 1 and SD 2 is the standard deviation of group 2. For studies which do not report means and standard deviations and for studies that used designs without a control group , the standardized mean difference were ca lculated using the t - value or p - value . These were used to answer research question 1 . To find the effect sizes of non - RCT studies , the standardized mean gain score was calculated by dividing the unstandardized mean score (i.e., post - test mean minus pre - test mean) with the pooled SD of pre - test and post - test scores (Littell, Corcoran, & Pillai, 2008) . The random - effects model was used to examine true effect size of each study , and it was calculated as follows: 46 Where is the heterogeneity of variance, is the total number of effect sizes, and is the inverse of the effect size. The inverse of the effect size is calculated by squaring the standard error to find the variance, and then dividing the variance by 1. The standard error is calculated as follows: Where cores, is the sample size of each study, and is the between - studies variance (Borenstein, Rothstein, Hedges, & Higgins, 2009). Each was recorded, and overall effects were calculated through the programming tool Comprehensive Meta - Analysis (Borenstein et al., 2005). A random - effects mixed - effects model (i.e., meta - regression) was used to examine the moderating and mediating effect on variance . This is calculated through t he following: i = 0 + 1 x i1 + . . . + p x ip + u i W here u i N (0, ) , and x ij is the value of the j th moderator variable for the i th study (Viechtbauer, 201 0). Since non - significant study results are less likely to be published than significant study results (i.e., file drawer problem), a forest plot analysis through a fail - safe N was calculated to mitigate this effect (Lipsey & Wilson , 2000). This was calculated throu gh the following: - 1 47 Where k is the number of studies found in the meta - analysis, ESk is the weighted effect size, ESc is the criterion effect size level, and is the number of effect sizes with a value of zero . Like the other analyses, the forest plot analysis was conducted through CMA ( Borenstein et al., 2005). 48 CHAPTER 4 RESULTS Specific treatment details of the 94 studies are in Table 7. In the 94 studies included in this meta - analysis, there were a diverse array of samples included (see Table 6). Forty - four studies occurred in the United States while 49 studies occurred internationally. One study did not report location. The international studies occurred in a range of countries including Australia , Brazil, Canada, China , DR Congo, Germany, Greece, Ira n, Israel , Japan , Jordan, Nepal, the Netherlands , New Zealand, Norway, Palestine, Rwanda , Scotland, Sri Lanka, Sweden, Tanzania, Thailand, Turke y , the United Kingdom , and Zambia . As the construct of race varies across countries, racial identity was only examined in United States - based samples. Three United States - based studies did not report predominant racial identity data. Of the 41 studies in the United States that reported predominant racial identity data , the predominant race in the sample was Biracial in one study , Black/African American in 14 studies, Hispanic/Latinx in six studies, Native American in two studies, and White/European American in 18 studies. Overall, there were Asian/Asian Am erican participants in seven studies, Biracial participants in 19 studies, Black/African American participants in 38 studies, Hispanic/Latinx participants in 33 studies, Native American participants in 10 studies, and White/European American participants in 34 studies. Twelve studies examined female - only samples, four studies examined male - only samples, and 77 studies examined mixed gender samples. One study did not report gender data. Youth age ranged fro m three years to 18 years old in the 94 studies . Meta - Analyses Standardized mean differences were found for the 97 data points from the 94 studies through the Comprehensive Meta - Analysis statistical software . Table 9 shows effect sizes and 49 variance data for individual studies that had posttraumatic stress symptom outcome measures, Table 10 shows effect sizes and variance data for individual studies that had anxiety symptom outcome measures, and Table 11 shows effect sizes and variance data for individual studies that had depression symptom outcome measures . Analyses were separated by outcome measure: posttraumatic stress symptoms , anxiety symptoms , and depression symptoms . Question 1: O verall effect and subtreatment effectiveness. Because higher scores on outcome measures indicate higher symptomatology and lower scores on outcome measures indicate a higher level of functioning, negative effect sizes indicate a better outcome for youth exposed to cogn itive behavioral therapie s when compared to a control group or pre - treatment outcomes. As seen in Table 12 , the overall standardized mean difference indicated a medium effect for posttraumatic stress symptoms ( k = 95; d = - 0.57; 95% CI = - 0.66, - 0.48), and a small effect for anxiety symptoms ( k = 39; d = - 0. 40 ; 95% CI = - 0.5 1 , - 0.2 9 ) and depression symptoms ( k = 65; d = - 0.40; 95% CI = - 0.47, - 0.33 ) . A dditionally, the p - values for the effect sizes (i.e., the test of the null) were all less than 0.001, meaning that there was a significant difference for treatment out comes (i.e., post traumatic stress, anxiety, depression symptoms) between pre - test and post - test and between treatment groups and comparison groups (e.g., waitlist, treatment as usual) . Heterogeneity was considered substantial in the effect sizes that examined posttraumatic stress symptoms ( I 2 = 85.04 ; Q = 628.35; p < 0.001; k = 95 ), anxiety symptoms ( I 2 = 69.50 ; Q = 124.60; p < 0.001; k = 39), and depressive symptom s ( I 2 = 51.36 ; Q = 131.5 9 ; p < 0.001 ; k = 65 ) . This means that 85% of the observed variance in posttraumatic stress symptoms reflects real differences in study effects, 70% of the observed variance in anxiety symptoms reflects real differences in study effects, and 5 1 % of the observed variance in depressive symptoms reflects real differences in study effects . Additionally, b ecaus e th ere was statistical significance for th e 50 Q - statistics ( p < .001) , dispersion is likely not due to random error, and there were likely real differences in the study effects. Thus, moderator analyses were used to determine variables that may be impacting outcomes. Of the 94 studies that produced 97 effect sizes, 78 examined 23 specific subtreatments, and 16 examined general cognitive behavioral treatments. Data included in the narrative summaries are in Tables 6 and 7, and Appendix C provide s a detailed narrative summary of the CBT studies includ ed in this meta - analysis. As seen in Table 13 , the subtreatments examined for posttraumatic stress symptom s were CBITS ( k = 5 ; ; d = - .53, p = .002 ) , CPC - CBT ( k = 3 ; d = - 1.23, p < .00 1 ) , ERASE Stress ( k = 4 ; d = - .51, p = .009 ) , GB - CBT ( k = 3 ; d = - .38, p = .117 ) , PE - A ( k = 5 ; d = - .46, p = .023 ) , TRT ( k = 12 ; d = - .32, p = .009 ) , and TF - CBT ( k = 28 ; d = - .66, p < .001 ) ( k = 35 ; d = - .56, p < .001 ) due to small sample sizes. The standardized mean effects ranged from a small to large effect. Based on the results, the mean effects of the subtreatments were not significantly different from one another ( Q = 12.97, p = .073). The subtreatments examined for depression symptom outcomes were CBITS ( k = 5 ; d = - .41, p < .001 ) , TRT ( k = 10 ; d = - .25, p = .004 ) , and TF - CBT ( k = 17 ; d = - .44, p < .001 ) . The remaining subtreatments were consolidated into the ( k = 33 ; d = - .42, p < .001 ) due to small sample sizes. The standardized mean effects indicated a small effect size for all subtreatments. Based on the results, the mean effects of CBITS, TRT, TF - CBT, and other treatments on reducing posttraumatic stress and depressive symptoms were not significantly different from one another ( Q = 3.70, p = .296). Subtreatment analyses were not conducted on anxiety symptom outcomes due to the limited number of studies available per subtreatment. 51 Question 2: Moderator analyses. Table 13 shows the results of moderator analyses . It should be noted that age was conceptualized as continuous. All other moderator variables were conceptual ized as categorical. To answer the research questions and determine moderators that may be impacting outcomes, there needed to be at least three effect sizes per category to be included in analyses. For posttraumatic stress symptom outcomes, there were 95 effect sizes in the trauma exposure analysis, 37 effect sizes in the race analysis, 94 effect sizes in the gender analysis, 95 effect sizes in the study design analysis, 94 effect sizes in the treatment setting analysis, 85 effect sizes in the parental inv olvement analysis, 95 effect sizes in the treatment technique analysis, 89 effect sizes in the treatment delivery analysis, and 95 effect sizes in the session frequency analysis. For anxiety symptom outcomes, there were 39 effect sizes in the trauma exposu re analysis, 19 effect sizes in the race analysis, 38 effect sizes in the gender analysis, 39 effect sizes in the study design analysis, 39 effect sizes in the treatment setting analysis, 35 effect sizes in the parental involvement analysis, 39 effect size s in the treatment technique analysis, and 36 effect sizes in the treatment delivery analysis. For depression symptom outcomes, there were 65 effect sizes in the trauma exposure analysis, 31 effect sizes in the race analysis, 62 effect sizes in the gender analysis, 65 effect sizes in the study design analysis, 65 effect sizes in the treatment setting analysis, 60 effect sizes in the parental involvement analysis, 65 effect sizes in the treatment technique analysis, 62 effect sizes in the treatment delivery analysis, and 65 effect sizes in the session frequency analysis. Moderator Analyses for Studies with Posttraumatic Stress Symptom Outcomes. The trauma exposures examined were natural disasters ( k = 10; d = - .57 ; p < .001 ) , physical abuse ( k = 4; d = - 1.42 , p < .001 ) , sexual abuse/assault ( k = 28; d = - .59 , p < .001 ) , single incident trauma ( k = 3; d = - 1.02 , p = .001 ) , terrorism ( k = 3; d = - .22 , p = .318 ) , traumatic grief ( k = 4; d = - .84 , p 52 < .001 ) , various traumas ( k = 13; d = - .62 , p < .001 ) , violence ( k = 3; d = - .58 , p = .009 ) , and war - related violence ( k = 20; d = - .38 , p < .001 ) . The remaining trauma exposures were consolidated ( k = 7; d = - .42 , p = .008 ) due to small sample sizes. The standardized mean effects ran ged from small to large , and they varied significantly ( k = 95; Q = 24.09 , p = . 004), with terrorism being associated with the smallest effect si ze and physical abuse being associated with the largest effect size . This indicate d that trauma type was a significant moderator variable in reducing posttraumatic stress symptoms . To examine racial identity, the predominant race in the sample was used, and only Black/African American ( k = 13; d = - .47 , p = .001 ) , Hispanic/Latinx ( k = 6; d = - .58 , p = .004 ) , and White/European American ( k = 18; d = - . 63 , p < .001 ) predominant samples were used. The standardized mean effects were in the small to medium range. The mean effect sizes between the racial identities were not significantly different from one another ( k = 37 ; Q = .79, p = .673) , so results indicated that ra ce was not a significant moderator variable in reducing posttraumatic stress symptoms . Female - only samples ( k = 12; d = - .64 , p < .001 ) , male - only samples ( k = 4; d = - 1.36 , p < .001 ) , and mixed gender samples ( k = 78; d = - . 53 , p < .001 ) were examined in the moderator analyses. The effect sizes were in the medium to large range. They were significantly different from one another ( k = 94; Q = 10.6 8 , p = .005), with female - only and mixed gender samples having similar effect sizes and male - only sa mples having a larger effect size. This indicates gender was a significant moderator variable in reducing posttraumatic stress symptoms. Lastly, age was examined as a moderator through meta - regression analysis , and the results indicated th at age was not a significant moderator variable in reducing posttraumatic stress symptoms ( k = 94; R 2 = - .006, p = .755). The study designs examined in t he moderator analyses were pre - post ( k = 48; d = - .62 , p < .001 ) , quasi - e xperimental ( k = 8; d = - .64 , p < .001 ) , and RCT ( k = 39; d = - .48 , p < .001 ) designs. The 53 effect sizes were in the small to medium range, and they were not significantly different from one another ( k = 95; Q = 2.21, p = .331). The settings examined in the moderator analyses were clinic ( k = 3; d = - .66 , p = .061 ) , community ( k = 32; d = - .63 , p < .001 ) , hospital outpatient ( k = 3; d = - .44 , p = .070 ) , school ( k = 31; d = - . 43 , p < .001 ) , and university ( k = 3; d = - 1.18 , p < .001 ) samples . The remaining settings were ( k = 22; d = - .63 , p < .001 ) due to small sample sizes. The effect sizes ranged from a small to large effect, and they were not significantly different from one another ( k = 9 4 ; Q = 10.11, p = .072) , so results indicated that treatment setting was not a significant moderator variable in reducing posttraumatic stress symptoms . Moderator Analyses for Studies with Anxiety Symptom Outcomes. The trauma exposures examined were sexual abuse/ assault ( k = 18; d = - .41 , p < .001 ) , single incident trauma ( k = 3; d = - .76 , p = .005 ) , various traumas ( k = 3; d = - .57 , p = .011 ) , and war - related violence ( k = 6; d = - .20 , p = .138 ) category ( k = 9; d = - .41 , p < .001 ) due to small sample sizes. The standardized mean effects ranged from small to medium . Based on the results, the mean effects between the different trauma exposures were not significantly different from one another ( k = 39; Q = 4. 49, p = .344) so results in dicated th at trauma type was not a significant moderator variable in reducing anxiety symptoms . To examine racial identity, the predominant race in the sample was used, and only Black/African American ( k = 3; d = - .35 , p = .024 ) , Hispanic/Latinx ( k = 4; d = - .31 , p = .035 ) , and White/European American ( k = 12; d = - .42 , p < .001 ) predominant samples were used. The standardized mean effects were in the small range and the y were not significantly different from one another ( k = 19 ; Q = .49, p = .783) , so results indicated th at race was not a significant moderator variable in reducing anxiety symptoms . Additionally, female - only samples ( k = 4; d = 54 - .55 , p = .001 ) and mixed gender samples ( k = 34; d = - .38 , p < .001 ) were the only gender categories examined in the moderator analysis . The effect sizes were in the small to medium range, and they were not significantly different from one another ( k = 38; Q = .87, p = .350) so results from these limited samples indicated th at gender was not a significant moderator variable in reducing anxiety symptoms . Lastly, age was examined as a moderator through meta - regression analyses, and the results indicated that age was not a significant moder ator variable in reducing anxiety symptoms ( k = 38; R 2 = - .006, p = .755). The study designs examined in the moderator analyses were pre - post ( k = 19; d = - .50 , p < .001 ) , quasi - e xperimental ( k = 4; d = - .47 , p = .005 ) , and RCT ( k = 16; d = - .23 , p = .016 ) designs. The effect sizes were in the small to medium range, and they were not significantly different from one another ( k = 39 ; Q = 4.73, p = .094) , so results indicated that study design was not a significant moderator variable in reducing anxiety symptoms . Due to the low sample sizes for various treatment settings (e.g., hospital, residential treatment facility), only community ( k = 13; d = - .47 , p < .001 ) and school - based ( k = 11; d = - .24 , p = .017 ) samples were used in the moderator anal ysis . The remaining treatment ( k = 15; d = - .48 , p < .001 ) due to small sample sizes. The mean effect sizes were in the small range, and they were not significantly different from one another ( k = 39 ; Q = 3.92, p = .141) , so results indicated th at treatment setting was not a significant moderator variable in reducing anxiety symptoms . Moderator Analyses for Studies with Depression Symptom Outcomes. The trauma exposures examined were natural disasters ( k = 7; d = - .46 , p < .001 ) , physical abuse ( k = 3; d = - .85 , p < .001 ) , sexual abuse/assault ( k = 21; d = - .37 , p < .001 ) , single incident trauma ( k = 3; d = - .66 , p = .005 ) , terrorism ( k = 3; d = - .26 , p = .070 ) , traumatic grief ( k = 4; d = - .47 , p < .001 ) , various traumas ( k = 8; d = - .57 , p < .001 ) , v iolence ( k = 3; d = - .37 , p = .014 ) , and w ar - r elated 55 v iolence ( k = 7; d = - .21 , p = 017 ) . The remaining trauma exposures were consolidated into the ( k = 6; d = - .40 , p < .001 ) due to small sample sizes. The standardized mean effects ranged from small to large. The mean effects between the different trauma exposures were not significantly different from one another ( k = 65; Q = 14.19, p = .116) , so results indicated th at trauma type was not a significant moderator variable in reducing depression symptoms . To examine racial identity , only those represented by a reasonable number of studies were included in the analysis, thus only Black/African American ( k = 9; d = - .38 , p < .001 ) , Hispanic/Latinx ( k = 6; d = - .27 , p = .017 ) , and White/European American ( k = 16; d = - .47 , p < .001 ) predominant samples were used. The standardized mean effects were generally in the small range. The mean effect sizes between the racial identities were not significantly different from one another ( k = 31; Q = 2.67, p = .264) , so results indicated th at race , as reflected in the three ethnic groups available, was not a significant moderator variable in reducing depression symptoms . Additionally, female - only samples ( k = 7; d = - .33 , p = .001 ) and mixed gender samples ( k = 55; d = - .40 , p < .001 ) were the only ones examined in the moderator analyses. The effect sizes were generally in the small range, and they were not significantly different from one another ( k = 62; Q = .63, p = .428) , so results from these limited samples indicated th at gender was not a significant moderator variable in reducing depression symptoms. Finally , age was examined as a moderator through meta - regression analysis , and the results indicated that age was not a significant moderator variable ( k = 64; R 2 = .02, p = .133) in reducing depression symptoms . The study designs examined in the moderator analyses were pre - post ( k = 31; d = - .50 , p < .001 ) , quasi - experimental ( k = 6; d = - .35 , p < .001 ) , and RCT ( k = 28; d = - .26 , p < .001 ) designs. The effect sizes were in the small to medium range, and they were significantly different from one another ( k = 65; Q = 10.95, p = .004), with RCTs having the smallest effect 56 size and pre - post designs having the largest effect size. This indicate d that study design was a significant moderator variable in reducing depression symptoms . Due to the low sample sizes for various treatment settings (e.g., hospital, residential treatment facility), only community ( k = 22; d = - .56 , p < .001 ) and school - ba sed ( k = 21; d = - .31 , p < .001 ) samples were used in moderator analyses. The remaining ( k = 22; d = - .35 , p < .001 ) due to small sample sizes. The effect sizes ranged from a small to medium effect, and they were significantly different from one another ( k = 65; Q = 10.98, p = .004), with school and other settings having similar effects, while community settings had a relatively larger effect. This indicate d that treatment setting was a significant moderator variable in reducing depression symptoms. Question 3 : Mediator analyses. Table 13 shows the results of mediator analyse s . It should be noted that session length and number of sessions were conceptualized as continuous variables . All other variables were conceptualized as categorical. To answer the research questions and determine mediators that may be impacting outcomes, there needed to be at least three effect sizes per category to be included in analyses. For posttraumatic stress sy mptom outcomes, there were 94 effect sizes in the age analysis, 64 effect sizes in the session length analysis, and 83 effect sizes in the session number analysis. For anxiety symptom outcomes, there were 27 effect sizes in the age analysis and 33 effect s izes in the session number analysis. For depression symptom outcomes, there were 64 effect sizes in the age analysis, 40 effect sizes in the session length analysis, and 56 effect sizes in the session number analysis. Mediator Analyses for Studies wit h Posttraumatic Stress Symptom Outcomes. Parental involvement examined parents being included in the treatment ( k = 50; d = - .60 , p < .001 ) and parents not being included in the treatment ( k = 35; d = - .52 , p < .001 ) . The effect sizes 57 for parental involvement were in the medium range, and the effect sizes were not significantly different from one another ( k = 95; Q = - .73, p = .689) , so results indicated th at parental involvement was not a significant mediator variabl e in reducing posttraumatic stress symptoms . These analyses included treatment that used only CBT ( k = 76; d = - .58 , p < .001 ) and treatments that used other techniques in addition to CBT ( k = 19; d = - .55 , p < .001 ) . The effect sizes were in the medium range, and they were not significantly different from one another ( k = 95; Q = .05, p = .831) , so results indicated th at inclusion of other treatment techniques was not a significant mediator variable in reducing posttraumatic stress symptoms . Treatment delivery was predominantly individual treatment ( k = 48; d = - .60 , p < .001 ) or predominantly done in group settings ( k = 41; d = - . 57 , p < .001 ) . The effect sizes were in the small to medium range, and they were not significantly different from one another ( k = 95; Q = 2.82, p = .244) , so results indicated that treatment delivery was not a significant mediator variable in reducing posttraumatic stress symptoms . Session length was examined as a mediator through meta - regression analyses, and the results indicated that session length was not a significant mediator variable ( k = 64; R 2 = - .0007, p = .770) , so results indicated th at session length was not a significant mediator variable in reducing posttraum atic stress symptoms . Treatment frequency categories included in the mediator analyses were biweekly ( k = 5; d = - .17 , p = .443 ) , weekly ( k = 52; d = - .60 , p < .001 ) , and three times a week ( k = 4; d = - .47 , p = .041 ) . Studies involving other ( k = 34; d = - .58 , p < .001 ) due to small sample sizes. The effect sizes were in the small to medium range, and they were not significantly different from one another ( k = 95 ; Q = 3.59, p = .309) , so results indicated th at treatment frequency was not a significant mediator variable in reducing posttraumatic stress symptoms . Number of treatment sessions was examined as a mediator 58 through meta - regression analyses, and the results indicated that number of treatment sessions was not a significant mediator variable ( k = 83; R 2 = - .01, p = .366) in reducing posttraumatic stress symptoms . Mediator Analyses for Studies with Anxiety Sym ptom Outcomes. Parental involvement examined parents being included in the treatment ( k = 21; d = - .39 , p < .001 ) and parents not being included in the treatment ( k = 14; d = - .37 , p < .001 ) . The effect sizes for parental involvement were generally in the small range, and the effect sizes were not significantly different from one another ( k = 39; Q = 1.05, p = .593) , so results indicated that parental involvement was not a significant mediator variable in reducing anxiety symptoms . T hese analyses included treatments that either used only CBT ( k = 32; d = - .40 , p < .001 ) or treatments that used other techniques in addition to CBT ( k = 7; d = - .41 , p = .002 ) . The effect sizes were generally small, and they were not significantly differe nt from one another ( k = 39 ; Q = .01, p = .917) . Thus, results indicated th at inclusion of other treatment techniques was not a significant mediator variable in reducing anxiety symptoms . Treatment delivery was either predominantly individual treatment ( k = 21; d = - .44 , p < .001 ) or predominantly done in group settings ( k = 15; d = - .44) , p < .001 . The effect sizes were generally small, and they were not significantly different from one another ( k = 36 ; Q = 3.77, p = .152) , so results indicated th at treatment delivery was not a significant mediator variable in reducing anxiety symptoms . Session length was examined as a mediator through meta - regression analyses, and session length was not a significant mediator variable ( k = 27; R 2 = - .002, p = .601 ). Therefore, results indicated that session length was not a significant mediator variable in reducing anxiety symptoms . Analyses of the frequency of treatment were not conducted due to the limited number of studies available per category (e.g., weekly, biweekly). However, number of treatment sessions was examined as a 59 mediator through meta - regression analyses, and the results indicate d that the number of treatment sessions was not a significant mediator variable ( k = 33; R 2 = - .0009, p = .962) for reduction of anxiety symptoms . Mediator Analyses for Studies with Depression Symptom Outcomes. Parental involvement examined parents being i ncluded in the treatment ( k = 32; d = - .48 , p < .001 ) and parents not being included in the treatment ( k = 28; d = - .34 , p < .001 ) . The effect sizes for parental involvement were generally in the small range, and they were not significantly different from one another ( k = 65; Q = 3.49, p = .175) . Thus, results indicated th at parental involvement was not a significant mediator variable in reducing depression symptoms . These analyses included treatments used only CBT ( k = 52; d = - .38 , p < .00 1 ) and treatments that used other techniques in addition to CBT ( k = 13; d = - .48 , p < .001 ) . The effect sizes were generally small, and they were not significantly different from one another ( k = 65; Q = 1.33, p = .249) , so results indicated th at inclusion of other treatment techniques was not a significant mediator variable in reducing depression symptoms . Treatment delivery w as predominantly individual treatment ( k = 30; d = - .40 , p < .001 ) or predominantly done in group settings ( k = 32; d = - .38 , p < .001 ) . The effect sizes were small, and they were not significantly different from one another ( k = 62 ; Q = 1.75, p = .417) . Thus, results indicated th at treatment delivery was not a significant mediator variable in reducing depression symptoms . Session length was examined as a mediator through meta - regression analysis , and the results indicated there was no evidence that session length was a significant mediator variable ( k = 40; R 2 = - .001, p = .524). Categories for treatment frequency involved in the mediator analyses were biweekly ( k = 3; d = - .17 , p = .335 ) and weekly ( k = 40; d = - .39 , p < .001 ) . The category ( k = 22; d = - .44 , p < .001 ) due to small sample sizes. The effect sizes were small, and 60 they were not significantly different from one another ( k = 65; Q = 2.21, p = .331) , so res ults indicated th at treatment frequency was not a significant mediator variable in reducing depression symptoms . Number of treatment sessions was examined as a mediator through meta - regression analysis , and the results indicated that there was no evidence that number of treatment sessions was a significant mediator variable ( k = 56 ; R 2 = .004, p = .652). Publication Bias For the main outcome measure analyses (i.e., posttraumatic stress symptom analyses , anxiety analyses, depression analyses, ), vis ual inspection of the symmetry of the funnel plots, seen in Figure 4, Figure 6 , and Figure 7 , suggested no publication bias. Further, as seen in Figure 5 and Figure 8, removal of the outlier s in the initi al funnel plot s resulted in a typical level of symmetry, also suggesting no publication bias. Another examination of potential publication bias was conducted through the analysis of fail - safe N . As seen in Table 12, there would need to be 7 , 078 studies with a treatment effect of zero to lead to a n onsignificant overall result for studies with posttraumatic stress symptom outcomes, there would need to be 1 , 451 studies with a treatment effect of zero to lead to a nonsignificant overall result for studies with anxiety symptom outcomes, and there would need to be 4 , 411 studies with a treatment effect of zero to lead to a nonsignificant overall result for studies with depression symptom outcomes. 61 CHAPTER 5 DISCUSSION This study examined the impact of CBT on treatment outcomes (i.e., reductions in posttraumatic stress symptoms, anxiety symptoms, or depression symptoms ) for youth exposed to traumatic events through a meta - analytic design. A total of 94 studies representing 97 treatment effects were analyzed. The effects of moder ating and mediating variables on CBT out come were also examined through this meta - analysis and provide a unique contribution to the prior CBT meta - analytic literature . Study r esults indicated that CBT for youth exposed to traumatic events significant ly r educ es posttraumatic stress symptoms ( d = - .57, p < .001) , anxiety symptoms ( d = - .40, p < .001), and depression symptoms ( d = - .40, p < .001) across a diverse array of measures when compared to other trauma treatments (e.g., cue - centered therapy, EMDR ) or waitlist control groups. Such improvements in symptoms are consistent with prior meta - analyses of CBT outcome studies specifically examining posttraumatic stress symptoms , anxiety symptoms, and depression symptoms ( Gillies et al., 2012; Silverman et al., 2008) . P rior research ( Gillies et al., 2012) found higher effect sizes for posttraumatic stress symptoms ( SMD = - 1.34 ) and depression symptoms ( SMD = - .80 ) . However, due to the small effect sizes in th at study (i.e., k = 3 ) , t he magnitude of those results may likely be inflated. Contrarily, Silverman and colleagues (2008) found slightly lower effect sizes for posttraumatic stress symptoms ( d = .50) and considerably lower effective sizes for anxiety symptoms ( d = .15), and depression symptoms ( d = .29) than this meta - analysis. Although this meta - analytic study found slightly more positive results than (2008) study , the findi ngs are fairly similar . In line with previous meta - analyses ( e.g., Morina et al., 2016) , which showed higher effects for posttraumatic stress 62 symptoms (i.e., a decrease in symptomatology) than depression symptom s , this meta - analysis provides evidence that mean effects were higher for posttraumatic stress symptoms than other outcomes (i.e., anxiety symptoms , depression symptoms ). Consistent with treatment guidelines (e.g., APA, AACAP), CBT with youth exposed to traumatic events is clearly effective in reducing mental health symptoms associated with traumatic event exposure. Find i n g s from t his meta - analysis clearly provide further evidence for the impact of CBT on children experiencing co morbid mental health issues (i.e., posttraumatic stress, anxiety, depression symptoms ) . In sum, youth experiencing a range of traumatic effects can be treated effectively with CBT . Five prior meta - analyses have examined the impact of CBT on children and a dolescents who have experienced trauma . However, those meta - analyses had not examine d the possibility of subtreatment s differentially impacting posttraumatic stress symptom , anxiety symptom, or depression symptom outcomes ( Dorsey et al., 2017; Kowalik et al., 2011; Harvey & Taylor, 2010 ; Silverman et al., 2008 ; Slade & Warne, 2016 ). The results of this meta - analysis suggest CBT subtreatments all positively impacted posttraumatic stress symptoms with one exception . Game - based cognitive behavioral therapy ( GB - CBT ) did not evidence statistically significant decreases in posttraumatic stress symptom s ( p = .117) . This could be due to Misurell, Springer, and Tryon ( 2011) study , in which there were decreases in posttraumatic stres s symptom outcomes, but these decreases were not significant. Given the multitude of available effective CBT treatments, support for using GB - CBT is quite limited. However, this meta - analysis provided ample evidence on TF - CBT , TRT, and CBITS as first - line treatments for youth exposed to traumatic events given the pos itive impact on primary (i.e., posttraumatic stress symptom) and secondary (i.e., depression symptom) outcomes. 63 This study is the first appearing in the literature to examine the impact of CBT on a diverse range of moderators ( e.g., gender, age, setting ,) on a broa d set of treatment outcomes (i.e., posttraumatic stress, anxiety, depression symptom s) . One moderator examined was trauma type. Trauma type led to differential treatment outcomes for posttraumatic stress symptoms but not anxiety symptoms or depre ssion symptoms . Specifically, of the effect sizes with larger samples, youth who had been sexual ly abuse d demonstrated higher treatment effects ( d = - .59) compared to other trauma types (e.g., war - related violence ) . This meta - analysis partially aligns with prior research - analytic study, the overall treatment effect for posttraumatic stress symptoms ( d = .43) was lower than for sexual abuse interventions ( d = .46) . However, this meta - analysis did not separate CBT from non - CBT studies, so these results contribute to the literature by providing more concrete information about treatment effects for CBT by trauma type. While sexual abuse/assault showed a medium effect in this study , implying CBT ha s a positive impact for youth exposed to sexual abuse/assault, physical abuse had the highest effect, followed by single incident trauma and traumatic grief (see Table 13) . Further, there were significant decreases in posttraumatic stress, anxiety, and dep ression symptom outcomes for youth exposed to most traumatic events , as seen in prior literature (Brown et al., 2017; Gutermann et al., 2016 ) , but trauma type did not moderate anxiety and depression symptom outcomes as anticipated . However, t his could be e xplained by the small sample size and future research may be needed to determine if this result is due to genuine variance or sample size. Further, youth exposed to terrorism did not experience significant decreases in posttraumatic stress symptoms ( p = .3 18) and depression symptoms ( p = .070), and youth exposed to war - related violence did not experience significant decreases in anxiety symptoms ( p = .138). This can potentially be explained by the small sample sizes for youth 64 exposed to terrorism. However, it could also mean that CBT may not be the ideal treatment for youth exposed to this type of trauma due to the chronic and potentially current nature of the trauma. Further, traumatic events like war - related violence and terrorism also likely imp act (e.g., parents, siblings, peers) who may not be receiving treatment . This experience of collective trauma could potentially impact treatment outcomes , and future research will be needed to determine the impact of CBT for yout h who have experienced these events . However , it should be noted that for youth exposed to war - related violence, t here were significant decreases in posttraumatic stress symptoms and depression symptoms, so practitioners should be aware that while CBT may not appropriately address anxiety symptoms , CBT can still be an impactful treatment for youth exposed to war - related violence . Past research has not thoroughly examined anxiety symptom treatment outcomes for youth exposed to traumatic events, so this meta - analysis adds valuable data to the literature for researchers and practitioners. Study findings indicated that both male and female samples respond positively to CBT. Yet, this meta - analysis identified that for posttraumatic stress symptom outcomes, m ale - only samples had significantly higher effect size outcomes compared to female - only samples or mixed samples. This conflicts with prior meta - analytic research (Gutermann et al., 2016) that suggested gender does not moderate posttraumatic stress symptom outcomes. However, Lindebø Knutsen and colleagues (2020) found that female samples did not respond to TF - CBT as well as male samples, which may support the more positive posttraumatic stress symptom outcomes for males exposed to CBT in this meta - analytic study. Further, similar to the trauma type data, the gender data that was significantly different had smaller sample sizes, with the four studies examining male - only samples while 12 studies examined female - only samples and 78 studied mixed 65 samples. It is likely that the sample sizes could be the source of the significant difference in posttraumatic stress symptom effect sizes, so this result needs to be interpreted with caution. Additionally, male - only samples were not examined in anxiety sy mptom and depression symptom analyses, so further research is needed in relation to anxiety symptoms and depression symptoms to determine if gender may be a moderating variable for male - only samples. This study also found that one moderator impacting trea tment outcomes for depression symptoms was study design . Past research did not examine the impact of study design on depression symptom outcomes , so this meta - analysis contributes to the literature by providing evidence of the moderating effect of study design on depression symptom outcomes. However, it was hypothesized that RCTs would have a larger effect than non - RCTs since Harvey and Taylor (2010) found that pre - post and quasi - experimental studies had significantly lower effects on posttraumatic stress symptoms than experimental studies. The findings of this meta - analysis conflict with those findings since pre - post and quasi - experimental studies had highe r effects than RCTs. One explanation for this outcome is that non - RCTs may provide the opportunity for practitioners to be more individualized and less strict with the implementation of treatment ( Chorpita et al., 2005), which may explain the current findi ngs. Further, this meta - analysis conflicts with prior research since this meta - analysis found study design only moderated depression symptom outcomes and does not moderate posttraumatic stress symptom outcomes. However, Harvey and Taylor (2010) examined al l psychological treatments while this meta - analysis only examined CBT, which may explain results . These findings can help researchers and practitioners better understand in what settings what treatments work for youth exposed to traumatic events. 66 Further there was a significant decrease in almost all outcomes (i.e., posttraumatic stress symptoms, anxiety symptoms, depression symptoms) after exposure to CBT regardless of setting. This meta - analysis further adds t o the literature by providing evidence that treatment setting moderates depression symptom outcomes, with community settings have a higher effect than school settings. Brown and colleagues (2017) found that teachers had lower posttraumatic stress symptom effect sizes than other mental health professionals, which led to the hypothesis that schools would have smaller effects than other settings. The current meta - analysis supports the idea tha t perhaps school settings may be less impactful in regard to depression symptoms, but it also conflicts with prior research since posttraumatic stress symptoms were not moderated by treatment setting. This conflict can potentially be explained by the speci fic treatments examined - analysis (i.e., all psychological treatments ) while this meta - analysis only examined CBT . Despite the moderating effect of study design on depression symptoms , though, the schools are still a set ting in which treatment could be done effectively since results showed small, and not negligible, effects, and school - based studies had lower results that were significant for only depression outcomes . Clinic samples ( p = .061) and hospital outpatient samp les ( p = .070) did not experience significant decreases in posttraumatic stress symptom outcomes, but this may be a result of the small sample sizes for both types of settings. Regardless, this meta - analysis provides valuable information about the settings in which treatment may be more impactful for youth exposed to traumatic events and the symptomology decreases that can be expected. Another way this meta - analysis contributes to the literature is by providing evidence that racial identity did not moderat e posttraumatic stress symptoms, anxiety symptoms, or depression symptoms , as mentioned previously . However, all racial identity groups experienced a reduction 67 in symptom severity after exposure to CBT. This implies that regardless of racial identity, CBT has a positive impact on treatment outcomes. This aligns with the hypothesis that CBT will positively impact treatment outcomes regardless of racial identity status. Since previous meta - analyses did not examine the moderating effect of racial identity on t reatment outcomes, this meta - analysis provides new information to contribute to the literature. This meta - analysis also provides evidence that age does not moderate posttraumatic stress symptoms, anxiety symptoms, or depression symptoms , but there were dec reases in symptom severity , implying that youth of all ages benefit from CBT. This conflicts with the prior literature (Gutermann et al., 2016) that evidenced older youth having significantly larger effects than younger youth on posttraumatic stress symptoms. Prior research was conducted on all psychological treatments, not just CBT , and this may explain the differing results. Overall, though, th is meta - analysis adds to the literature by indicating youth exposed to traumatic events who are of all ages can gain from CBT regardless of primary mental health concern (i.e., posttraumatic stress, anxiety, depression). Furthermore, this meta - analysis in dicated trauma exposure, gender, study design, and treatment setting did not moderate anxiety symptoms. However, although the moderators were not significant, there was a reduction in anxiety symptoms . This implies that there is a positive impact on anxiet y symptoms regardless of trauma exposure, gender, study design, and treatment setting . Prior research has not quantitatively examined variables that moderate anxiety symptom outcomes for youth exposed to traumatic events, so this meta - analysis contributes new data to the literature. Additionally, the inclusion of quantitative data for depression symptoms is new data that adds to the literature by evidencing how trauma exposure, racial identity, and gender do not moderate outcomes while study design and trea tment setting do moderate depression symptom outcomes. Overall, this dissertation provides important information about some 68 moderating factors that impact posttraumatic stress symptoms, anxiety symptoms, and depression symptoms . R esearchers and practitione rs can use this data to determine what works for whom based on primary mental health concerns, allowing for more impactful, evidence - based treatment. Another purpose of this meta - analysis was to determine which variables examined (see Table 4) would medi ate treatment outcomes. None of the mediators examined in this study (i.e., parental involvement, inclusion of other treatment techniques, treatmen t delivery, session frequency, session length , session number ) were found to significantly impact posttraumatic stress symptom, anxiety symptom, or depression symptom outcomes. However, there were decreases in posttraumatic stress symptoms, anxiety symptoms, and depression symptoms. This implies that a diverse array of CBT treatments with varying compo nents (e.g., varying session lengths, session frequencies) all result in positive outcomes and reductions in posttraumatic stress symptoms, anxiety symptoms, and depression symptoms. This conflicts with prior research that found posttraumatic stress sympto ms were mediated by parental involvement ( Gutermann et al., 2016; Silverman et al., 2008) and treatment delivery (Gutermann et al., 2016; Harvey & Taylor, 2010). The past meta - analyses examined all psychological treatments rather than simply CBT, as was do ne in this meta - analysis, and this could explain the current findings. This meta - analysis adds to the literature by providing information mediating factors for posttraumatic stress symptoms, anxiety symptoms, and depression symptoms for CBT specifically, w hich has not been done in prior research. Outside parental involvement and treatment delivery, prior research has mainly examined only session number (Brown et al., 2017; Harvey & Taylor, 2010) and session length (Harvey & Taylor , 2010) as mediating facto rs. The other variables examined in this meta - analysis (i.e., 69 inclusion of other treatment techniques, session frequency) have not been properly examined as mediating factors in the literature. Thus, this meta - analysis adds to the current literature by pro viding evidence that these variables do not mediate posttraumatic stress symptom , anxiety symptom , or depression symptom outcomes for CB T , and regardless of most of these mediating factors, treatment was found to positively impact outcomes. Session frequen cy was not examined as a mediating variable in relation to anxiety symptom outcomes due to the limited data, so further research is needed to rule this out as a mediating factor for CBT. Session frequency has not been frequently examined as a mediating va riable for treatment in youth exposed to traumatic events. Past research has found that biweekly sessions rather than weekly sessions for adult depression have resulted in better outcomes ( Cuijpers, et al., 2013). However, this meta - analysis found that the re was not a significant decrease in posttraumatic stress symptoms ( p = .443) and depression symptoms ( p = .335) for youth exposed to treatment biweekly. Interestingly, weekly and triweekly sessions significantly decreased posttraumatic stress symptoms and weekly sessions significantly decreased depression stress symptoms. This could be in part due to the small sample sizes, but further study is needed to better understand the media ting effect of session frequency on CBT for youth exposed to traumatic events. This contradicts the idea that longer treatment may impact treatment outcomes; instead practitioners may want to focus on the quality of care as the quantity does not determine outcomes . Additionally, although anxiety also showed high heterogeneity, none of the moderators and mediators examined were statistically significant. This means that there may have been a moderator or mediator that was not examined in the study (see Table 14) that may be impacting anxiety outcomes. These results should be interpreted cautiously and generalization about 70 moderator impact may be limited s ince the moderators that were significant varied depending on the outcome measure. Overall, the findings of this study contribute greatly to the literatu re. This study examined the impact of CBT on posttraumatic stress symptom, anxiety symptom, and depres sion symptom outcomes, allowing researchers and practitioners to better understand which treatments work for youth exposed to traumatic events. It also provided more data on CBT subtreatments and how various subtreatments impact treatment outcomes. Further , t his meta - analysis found specific moderators (i.e., trauma type, gender, study design, treatment setting) were statistically significant for only some treatment outcomes (i.e., posttraumatic stress symptoms, depression symptoms) while others (e.g., predo minant race in sample, different age groups) were not found to be statistically significant for any treatment outcomes (i.e., posttraumatic stress symptoms, anxiety symptoms, and depression symptoms). However, though there were statistically significant mo derators, treatment appeared to positively impact youth exposed to traumatic events equally. For example, there were no statistical differences in ra cial identity , suggesting that CBT works equally well for all youth regardless of ty . Similarly, while there were no statistically significant mediators in this meta - analysis, youth receiving CBT experienced positive outcomes . This implies that regardless of certain factors (e.g., parental involvement, session length), CBT works equally well for all youth. With the findings of this meta - analysis, re searchers and practitioners can better understand for whom and under what circumstances CBT positively impacts outcomes for youth exposed to traumatic events. Strengths and Limitations of the Meta - Analysis A strength of this meta - analysis is it supplies evidence on the impact of CBT for youth exposed to traumatic events. However, many studies in the meta - analysis simply stated that they 71 used cognitive behavioral treatment or they referred to previously studied CBT manuals without clarifying exact ly which treatment techniques were used in the ir study . Because of this, it had to be assumed that all components of CBT were used in all 94 studies. This assumption limited the meta - analytic interpretation as it had to be assumed that the 94 studies were truly examining CBT. In future studies, researchers should specify exactly which treatment techniques were used so that researchers and practitioners can better understand the compone nts of treatment that were used in the study and label the treatment accurately . A nother strength of this meta - analysis is that it provides researchers and practitioners with more information on the gaps in research that need to be filled for CBT . For exa mple, sample sizes for anxiety symptom outcomes were smaller than for the other outcome measures, and further research may be needed to address heterogeneity. Additionally, while this meta - analysis was able to determine some moderating variables for posttraumatic stress symptom and depression symptom outcomes, further research is needed since it is likely that there were some other moderators that were not reported in individual studies that could be impacting treatment outcomes (see Table 14) . Yet another strength of this meta - analysis was the examination of moderator variables, generalizability of the treatment research across the overall population in the countries in which the treatments were studied. However, there continue to be gaps in the research (see Table 14 ) . Specifically, there are racial identity (e.g., Asian/Asian American youth, Native American youth), trauma type (i.e., racial trauma , generati onal trauma , neglect ), socioeconomic status, sexual orientation, gender identity, and physical and mental health comorbidity variables that need to be examined. Further, youth involved in the juvenile justice system and in residential 72 placement facilities need to be studied . Future research should focus on examining these population s. Future research should also report demographic variables that were not commonly found in the 94 studies , and thus could not be examined in this meta - analysis (e.g., sexual ori entation, gender identity, consistent socioeconomic status data , specific mental health comorbidity data ). These are important variables to include about the treatment sampl e to better understand for whom treatment works. A limitation of this study was tha t certain moderator variables were examined in a manner that limits interpretation. R ace was examined by examining the predominant race in the study due to convenience for meta - analytic purposes , and individual studies need to be conducted to determine the moderating impact of race on treatment outcomes. Furthermore , the methods used to examine age as a moderator analysis were limited; mean age was used to conduct the analysis, and this likely limi ted the outcomes. Future research should examine age through developmental age units to determine if CBT is developmentally appropriate for youth of varying age groups. While there was data showing inclusion of other treatment techniques (e.g., DBT, bod y - oriented exercise, play therapy), the only analyses that could be done in this meta - analysis was on whether or not other treatment techniques were used in general. More research is needed on specific techniques that are used in addition to CBT that make the treatment more effective. More is also needed on if there is a specific treatment technique outside of CBT that is more effective than the others. This can help both researchers and practitioners better understand the mechanisms of change that impact t reatment outcomes for youth exposed to traumatic events. Furthermore, since practitioner theoretical orientation (e.g., psychodynamic, cognitive behavioral, family systems, eclectic) may vary, it is important that researchers study and know 73 what extra com ponents are included in CBT in naturalistic , real - world settings and how well these techniques work in addition to CBT. Furthermore, a limitation of this meta - analysis is regarding measurement of outcomes. As seen in Table 8, there were 22 measures used to examine posttraumatic stress symptoms, seven measures used to examine anxiety symptoms, and 11 measures used to examine depression symptoms . The lack of consistency in measures lessens the methodological rigor of treatment studies and raises concerns a bout comparing treatment studies. Consistency in measure use should be considered in future research to ensure methodological rigor and enhance understanding of the true impact of treatment on posttraumatic stress symptoms , anxiety symptoms , and depression symptoms. Another factor to consider about this meta - analysis is that the youth involved in treatment were typically those who completed treatment . While minimal data was reported on the youth who did not complete treatment, thus an analysis of these yo uth could not be completed, this is important data to include. The literature has shown that youth who prematurely terminate treatment tend to lose out on man y of the positive impacts of treatment and it can limit treatment effectiveness ( Brand & Jungmann, 2014). Additionally, k nowing about attrition rates and the potential moderators impacting attrition will help practitioners counteract these effects and retain more youth in treatment. For example, prior research has shown that the number of trau matic events experienced may impact attrition from TF - CBT ( Wamser - Nanney & Steinzor, 2017). Thus, further research is needed on the factors that impact attrition for youth exposed to traumatic events who are treated with CBT. Lastly, one limitation of t his study and a focus for future research is the varying samples sizes associated with certain categories. For example, while there were a number of studies that 74 examined youth exposed to sexual abuse/assault, there were fewer studies that examined youth e xposed to physical abuse, limiting the conclusions that can be made about the analyses in this meta - analysis. Future research should focus on expanding on the various populations that are not as commonly studied (e.g., Native American youth, youth exposed to physical abuse, LGBTQ youth) in the CBT literature. Conclusion The purpose of this meta - analysis was to provide a synthesis of the current literature on CBT and the moderating and mediating variables that impact treatment outcomes to better understand what CBT treatments work for whom and under what circumstances . Overall, this meta - analysis found CBT to be an impactful treatment for youth exposed to traumatic events of varying populations. The meta - analysis also found that involvement in almost all CB T subtreatments resulted in positive posttraumatic stress symptom, anxiety symptom, and depression symptom outcomes. Further, the study found that trauma type and gender moderated posttraumatic stress symptoms while study design and treatment setting moder ated depression symptoms. No moderators examined significantly impacted anxiety symptoms. One of the strengths of this study was the thorough examination of the moderators and mediators that impact posttraumatic stress symptoms, anxiety symptoms, and dep ression symptoms , thus supporting generalizability of CBT across various populations and expanding on what works for youth exposed to traumatic events. It also allows for practitioners to choose what specific subtreatment may be best suited for the populat ions with whom they are working. Furthermore, this meta - analysis allows for practitioners to aim treatment at the primary presentation problem (i.e., posttraumatic stress symptoms , anxiety symptoms, depression symptoms ) for youth exposed to traumatic events. While future research is needed to address the 75 gaps in the current literature, this meta - analytic review should provide both researchers and practitioners with salient data regarding CBT for youth exposed to traumat ic events. 76 APPENDICES 77 APPENDIX A TABLES AND FIGURES Table 1 . Variables in Past Meta - Analyses /Systematic Reviews Demographics Study Design Authors and Date Trauma Type Treatment (s) Age Gender Race SES Settin gs Studies Methodologies Brown et al., 2017 ( N = 37) MPA, ND, T/W CBT, EMDR, KidNET Mean Range: 8.2 - 16.6 years Reported NR NR NR Pre - Post, RCT Cary & McMillen, 2012 ( N = 10) IPV, SA , Various, Violence CBT, CBITS, CCT, OTT, RAP, TF - CBT 3 - 18 years NR NR NR NR RCT Dorsey et al., 2017 ( N = 37) ND , SA , T/W , Various CBITS, CBT, CCPT, CCT, EMDR , ITCT, KidNET, MED - RELAX, Mind - body skills group, PE , RRFT, SSET, TARGET, TF - CBT 4 - 1 9 years Reported 35% of studies had diverse samples NR Community Clinic, School, University or Hospital Clinic Naturalistic, Open Trial, Quasi - experimental, RCT 78 Table 1 . Gillies et al., 2012 * ( N = 14) CSV , DV, ND, PT, SA , Various CBT, E BT , EMDR , FBT, IPT, PDP , S C 6 - 18 years Reported 43% of studies had diverse samples NR Medical Center, Mental Health Clinic, Refugee Camp, School , Youth Correctional Facility RCT Guterman n et al., 2016 * ( N = 135) Accidents, Loss, ND, PA, SA, sickness, T/W CBT, EMDR, Hypnotherapy, Other, PDP, Psychoed , RMI, SM, ST Mean: 12.55 years Reported NR NR NR Pre - Post, RCT Harvey & Taylor, 2010 ( N = 39) SA CBT, CCT, EMDR, IRT, RAP, SC, SIT, TF - CBT NR NR NR NR Community, Inpatient Non - random assignment, Quasi - experimental, RCT Kowalik et al., 2011 * ( N = 8) SA CBT 5 - 17 years 25% reported gender NR NR Clinic RCT Morina et al., 2016 ( N = 41 ) Abuse/ Neglect, MVA, ND, SA, T/W, Various, Violence CCT, CBI, CBT , EMDR, EPSI, KidNET, Mind - body skills group, PDP, PE, RRFT, SC, SSET, TARGET 3 - 18 years NR NR NR NR RCT 79 Table 1 . Silverman et al., 2008 * ( N = 21) CV, DV, MVA, ND, PA, SA , Various CBITS, CBT, CCT, CPP, CPT, EMDR, FBT, PD, RAP, RPT, SGT, TF - CBT 2 - 18 years Reported 100% of studies had diverse samples NR NR RCT Slade & Warne, 2016 * ( N = 10) PA , SA TF - CBT, Play Therapy 4 - 1 2 years Reported 100% of studies had diverse samples NR NR RCT *Note. Studies that aggregated CBT data Note . CBI = Classroom - Based Intervention; CBITS = Cognitive Behavioral Intervention for Trauma in Schools ; CBT = Cognitive Behavioral Therapy; CCPT = Child - Centered Play Therapy; CCT = Cue - Centered Treatment; CPP = Child - Parent Psychotherapy; CPT = Cognitive Pro cessing Therapy; CSV = Civil or Social Violence; CV = Community Violence; DV = Domestic Violence; EBT = Exposure - Based Therapy; EMDR = Eye - Movement Desensitization and Reprocessing; EPSI = Eclectical Psychosocial Intervention, FBT = Family - Based Therapy; I PT = Interpersonal Therapy; IPV = Interpersonal Violence; IRT = Imagery Rehearsal Therapy ; ITCT = Integrative Treatment of Complex Trauma; KidNET = Narrative Exposure Therapy for the Treatment of Traumatized Children and Adolescents; MED - RELAX = Meditation and Relaxation Protocol for Tsunami Survivors Developed in Sri Lanka; MPA = Major Public Accident; MVA = Motor Vehicle Accident; ND = Natu ral Disaster; NR = Not Reported; OTT = Overshadowing the Threat of Terrorism; PA = Physical Abuse; PD = Psychological Debriefing; PDP = Psychodynamic Psyc h otherapy; PE = Prolonged Exposure Therapy ; PT = Physical Trauma ; RAP = Recovering from Abuse Program; RCT = Randomized Controlled Trial; RPT = Resilient Peer Treatment; RMI = Relaxation or Meditation Interventions; RRFT = Risk Reduction through Family Therapy; SA = Sexual Abuse /Assault ; SC = Supportive Counseling; SES = socioeconomic status; SGT = Support Group Therapy; SIT = Stress Inoculation Training; SM = Stress Management; SSET = Support for Students Exposed to Trauma; ST = Supportive Therapy; TARGET = Trauma Affect Regulation: Guide for Education and Therapy ; TF - CBT = Trauma - Focused Cognitive Behavioral Therapy; T/W = Terrorism/War 80 Table 2 . Outcomes for Past Meta - Analyses with Aggregated CBT Data Study Author/Date Independent Variable Dependent Variables/Outcomes Gillies et al., 2012 General CBT ( N = 5 ) PTSD Symptoms: Significant decrease in scores (SMD = - 1.34; 95% CI = - 1.79 , - 0 .89) Depression: Significant decrease in scores (SMD: - 0 .80; 95% CI = - 1.47 to - 0 .13) Anxiety: Not significantly different Loss to Follow - up: Not significantly different Gutermann et al., 2016 General CBT ; TF - CBT ( N = 84 ) Pooled analysis of CBT: large ES ( g = 0 .99; 95% CI = 0 .89 , 1.08) Primarily C ognitive CBT: large ES ( g = 1.27; 95% CI = 0 .65 , 1.89) Primarily E xposure CBT: large ES ( g = 1.29; 95% CI = 0 .99 , 1.58) Primarily Coping/Skills CBT: medium ES ( g = 0 .79; 95% CI = 0 .50 , 1.09) Mixed CBT: large ES ( g = 0 .97; 95% CI = 0 .85 , 1.10) TF - CBT: large ES ( g = 1.15; 95% CI = 0 .92 , 1.38) General CBT vs TAU/active CG : medium ES ( g = 0 .52; 95% CI = 0 .4 , 0 .63) RCTs for CBT: medium E S ( g = 0 .79; 95% CI = 0 .6 , 0 .96) Kowalik et al., 2011 General CBT ( N = 21 ) CBCL TP: S tatistically significant effect size favoring CBT ( g = - .3 3 ; p = .003) CBCL INT: Statistically significant effect size favoring CBT ( g = - .31; p = .001) CBCL EXT: Statistically significant effect size favoring CBT ( g = - .19; p = .04) CBCL TCOMP: CBT did not have statistically significant effect size ( g = - .054; p = .62) Silverman et a l., 2008 General CBT ( N = 11 ) Based on Chambless & - CBT met well - established criteria and CBITS met probably efficacious criteria PTSS: CBT ( d = .50) had greater ES than non - CBT ( d = .19) Depression: CBT ( d = .29) h ad a greater ES than non - CBT ( d = .08) Externalizing : CBT ( d = .24) had a greater ES than non - CBT ( d = .02) Slade & Warne, 2016 TF - CBT ( N = 6 ) Global: TF - CBT ( d = .21) had a greater ES than PT ( d = .095) Internalizing: TF - CBT ( d = .2 3 ) had a greater ES than PT ( d = .096) Sexual Outcomes: TF - CBT ( d = .16) had a greater ES than PT ( d = .042) Parent Report: TF - CBT ( d = .3 6 ) had a greater ES than PT ( d = - .15) 81 Table 2 . Note . CBCL = Child Behavioral Checklist; CBITS = Cognitive Behavioral Intervention for Trauma in Schools ; CBT = Cognitive Behavioral Therapy ; CI = confidence interval; CG = control group ; d d ; EXT = Externalizing; g g ; INT = Internalizing; PT = Play Therapy; PTSD = Posttraumat ic Stress Disorder ; PTSS = posttraumatic stress symptoms; RCT = randomized controlled trial; SMD = standardized mean difference ; TAU = treatment as usual ; TCOMP = Total Competence; TF - CBT: Trauma - Focused Cognitive Behavioral Therapy ; TP = Total Problems 82 Table 3 . Cognitive Behavioral Treatments Name of Program Developers Year Reference Program Components Program Duration Alternatives for Families: Cognitive Behavioral Therapy (AF - CBT) Kolko & Fishman Hicks 1996 The National Child Traumatic Stress Network psychoeducation, skill - building (e.g., emotion regulation, restructuring thoughts, managing behavior, imaginal exposure), family applications (e.g., healthy communication, enhancing safety, solving family problems) 20 sessions, 60 - 90 minutes Child First Lowell 2011 Blueprints for Healthy Youth Development cognitive behavioral training, family therapy, home visitation, parent training, social emotional learning 6 to 12 months Cognitive Behavioral Intervention for Trauma in Schools (CBITS) 1 Escudero, Jaycox, Kataoka, Stein, & Wong 2003 The National Child Traumatic Stress Network psychoeducation, relaxation training, safety assessment, cognitive restructuring, social problem - solving, trauma narrative 10 sessions, 30 - 45 minut es Combined Parent - Child Cognitive - Behavioral Therapy (CPC - CBT) Deblinger & Runyon 2013 The National Child Traumatic Stress Network psychoeducation, coping skill building, family safety, abuse clarification (i.e., trauma narrative) 16 - 20 sessions, 90 - 120 minutes Grief and Trauma Intervention (GTI) for Children Salloum 2015 The California Evidence - Based Clearinghouse for Child Welfare cognitive behavior therapy techniques, narrative therapy techniques, trauma narrative 12 sessions, 60 minutes Preschool PTSD Treatment (PPT) Scheeringa 2015 The California Evidence - Based Clearinghouse for Child Welfare psychoeducation, feelings identification, coping skill building, trauma narrative 12 sessions, 60 minutes 83 Table 3 . Prolonged Exposure Therapy for Adolescents (PE - A) Foa, Chrestman, & Gilboa - Schechtman 2009 The California Evidence - Based Clearinghouse for Child Welfare psychoeducation, breathing retraining exercises, in vivo exposure to trauma reminders, imaginal exposure to trauma memory 8 - 15 sessions, 60 - 90 minutes Stanford Cue - Centered Treatment (CCT) Carrion 2015 The California Evidence - Based Clearinghouse for Child Welfare psychoeducation, coping skills training, strength - building, cue exposure, trauma narrative, cognitive restructuring, parent psychoeducation 15 - 19 sessions, 45 minutes Trauma - Focused Cognitive Behavioral Therapy (TF - CBT) 2 Cohen, Mannarino, & Deblinger 2004 The National Child Traumatic Stress Network psychoeducation and parenting skil ls, relaxation techniques, affective modulation, cognitive coping and processing, trauma narrative, in vivo exposures to the trauma, conjoint child - parent session, and enhancing safety and future child development 12 - 25 sessions, 60 - 90 minutes Trauma - Focused Coping (aka Multimodality Trauma Treatment) Amaya - Jackson & March 1999 The National Child Traumatic Stress Network psychoeducation, emotion regulation, narrative exposure, cognitive processing 14 sessions, 40 - 90 minutes 1 Modified versions of CBITS are Bounce Back and Support for Students Exposed to Trauma (SSET) 2 Modified versions of TF - CBT are Community Outreach Program - Esperanza (COPE) and Culturally Modified Trauma - Focused Treatment 84 Table 4 . Moderating and Mediating Factors E xamined Moderators Examined Mediators Examined Trauma type Parental involvement Race Inclusion of other treatment techniques Gender Treatment delivery Age Session frequency Study design Session length Treatment setting Session number Note . Mediators are conceptual but will be examined using moderator analyses within the context of the meta - analysis 85 Table 5 . Cognitive Behavioral Subtreatments and Classifications Subtreatment Name Number of Studies Study Author and Date Well - Established Treatment General Cognitive Behavioral Treatments 17 Berliner & Saunders, 1996; Brown et al 2006; Deblinger et al 1990; Deblinger et al., 1996; Deblinger et al., 2001; de Roos et al., 2011; Gormez et al., 2017; Graham et al., 201 7; Habigzang et al., 2013; Habigzang et al., 2016; Ito et al., 2016; Jaberghaderi et al., 2004; King et al., 2000; Saltzman et al., 2001; Sezibera et al., 2009; Smith et al., 2007; Wolmer et al., 2003; Wolmer et al., 2013 Teaching Recovery Techniques (TRT) 1 12 Barron & Abdallah, 2017; Barron et al., 2016; Barron et al., 2017; Chen et al., 2014; Ehntholt et al., 2005; Eloranta et al., 2017; Giannopoulou et al., 2006; Kangaslampi et al., 2016; Ooi et al., 2016; Pityaratstian et al., 2015; Qouta et al., Sarkadi et al., 2018 Trauma - Focused Cognitive Behavioral Therapy (TF - CBT) 28 Allen & Hoskowitz, 2017; Bambrah et al., 2018; Bartlett et al., 2018; Cohen et al., 2004; Cohen et al., 2005; Cohen et al., 2016; Costantino et al., 2014; Damra et al., 2014; Deblinger et al., 2006; Deblinger et al., 2017; Dorsey et al., 2014; Feather & Ro nan 2009; Hartman et al., 2011; H é bert &; Daignault, 2015; Jaycox et al., 2010; Jensen et al., 2014; Kameoka et al., 2015; McMullen et al., 2013; Murray et al., Scheeringa et al., 2011; Schottelkorb et al., 2012; Stewart et al., 2017; Thornback & Muller, 2015 Probably Efficacious Treatme nt Cognitive Behavioral Intervention for Trauma in Schools (CBITS) 5 Goodkind e t al., 2010; Jaycox et al., 2010; Kataoka et al., 2003; Morsette et al., 2012; Stein et al., 2003 86 Table 5 Enhancing Resiliency Amongst Students Experiencing Stress (ERASE - Stress) 4 Berger & Gelkopf, 2009; Berger et al., 2012; Gelkopf & Berger, 2009; Shaheen et al., 2016 Narrative Exposure Therapy for Children (KIDNET) 2 Catani et al., 2009; Ruf et al., 2010 Prolonged Exposure Therapy for Adolescents (PE - A) 5 Brownlow et al., 2016; Capaldi et al., 2016; F Gilboa - Schechtman et al., 2010; Kaczkurkin et al., 2016 Possibly Efficacious Treatment Classroom - Based Intervention (CBI) 1 Jordans et al., 2010 Motivation - Adaptive Skills - Trauma Resolution Eye Movement Desensitization and Reprocessing (MASTR - EMDR) 1 Farkas, 2009 Overshadowing the Threat of Terrorism (OTT) 1 Berger et al., 2007 Risk Reduction through Family Therapy (RRFT) 1 Danielson et al., 2012 Sexual Abuse - Specific Cognitive Behavioral Therapy (SAS - CBT) 1 Cohen & Mannarino, 1998 Stress Inoculation Training (SIT) 1 Wolmer et al., 2011 Experimental Treatment Cognitive Behavioral Therapy for Childhood Traumatic Grief (CBT - CTG) 1 Cohen et al., 2004a Combined Parent Child Cognitive Behavioral Therapy for Families at Risk of Child Physical Abuse (CPC - CBT) 3 Kjellgren et al., 2013; Runyon et al., 2009; Runyon et al., 2010 Game - Based Cognitive Behavioral Therapy (GB - CBT) 3 Misurell et al., 2011; Misurell et al., 2014; Springer et al., 2012 Mein Weg 1 Pfeiffer & Goldbeck, 2017 Multi - Modality Trauma Treatment (MMTT) 1 March et al., 1998 PARTNERS with Teens 1 Grefe, 2011 Project LAST 1 Salloum & Overstreet, 2008 Project Sexual Abuse Family Education 1 Hubel et al., 2014 87 Table 5 Real Life Heroes 1 Kagan et al., 2008 SAY Group 1 Sinclair et al., 1995 STEPS 1 Bicanic et al., 2014 1 Modified version of TRT named 88 Table 6 . Demographics of Cognitive Behavioral Treatm ent Studies Study Trauma Type Age Range Location Gender Predominant Race in Sample White / European American Black / African American Hispanic/ Latinx Native American Asian / Asian American Bi - R acial Other Allen & Hoskowitz, 2017 Sexual Abuse 3 - 12 United States Mixed White / European American Yes Yes Yes No No No Yes Bambrah et al., 2018 Various 7 - 12 Canada Mixed --- --- --- --- --- --- --- --- Barron & Abdallah, 2017 Traumatic Grief 10 - 18 Palestine Mixed --- --- --- --- --- --- --- --- Barron et al., 2016 War - Related Violence 11 - 15 Palestine Mixed --- --- --- --- --- --- --- --- Barron et al., 2017 Domestic Trauma 14 - 18 Scotland Mixed --- --- --- --- --- --- --- --- Bartlett et al., 2018 Various 3 - 18 United States Mixed White / European American Yes Yes Yes Yes Yes No No Berger & Gelkopf, 2009 Natural Disaster 9 - 15 Sri Lanka Mixed --- --- --- --- --- --- --- --- Berger et al., 2007 War - Related Violence 7 - 12 Israel Mixed --- --- --- --- --- --- --- --- Berger et al., 2012 War - Related Violence 11 - 13 Israel Mixed --- --- --- --- --- --- --- --- Berliner & Saunders, 1996 Sexual Abuse 4 - 13 United States Mixed White / European American Yes Yes Yes No No No Yes Bicanic et al., 2014 Rape 13 - 18 The Netherlands Female --- --- --- --- --- --- --- --- Brown et al., 2006 Terrorism 8 - 13 United States Mixed Black / African American No Yes Yes No No Yes No Brownlow et al., 2016 Sexual Abuse/ Assault 13 - 18 United States Female Black / African American Yes Yes Yes No No Yes Yes Capaldi et al., 2016 Sexual Abuse/ Assault 13 - 18 United States Female Black / African American Yes Yes Yes No No Yes Yes Catani et al., 2009 Natural Disaster 8 - 14 Sri Lanka Mixed --- --- --- --- --- --- --- --- Chen et al., 2014 Natural Disaster NR China Mixed --- --- --- --- --- --- --- --- 89 Cohen & Mannarino, 1998 Sexual Abuse 7 - 15 United States Mixed White / European American Yes Yes Yes No No Yes No Cohen et al., 2004 Sexual Abuse 8 - 14 United States Mixed White / European American Yes Yes Yes No No Yes Yes Cohen et al., 2004a Traumatic Grief 6 - 17 United States Mixed White / European American Yes Yes No No No No No Cohen et al., 2005 Sexual Abuse 8 - 15 United States Mixed White / European American Yes Yes Yes No No Yes No Cohen et al., 2016 Various 12 - 17 United States Mixed White / European American Yes Yes Yes Yes Yes No Yes Costantino et al., 2014 Terrorism 9 - 11 United States Mixed Hispanic/ Latinx No No Yes No No No No Damra et al., 2014 Physical Abuse 10 - 12 Jordan Male --- --- --- --- --- --- --- --- Danielson et al., 2012 Sexual Assault 13 - 17 United States Mixed Black / African American Yes Yes Yes Yes No Yes No de Roos et al., 2011 Firework Disaster 4 - 18 The Netherlands Mixed --- --- --- --- --- --- --- --- Deblinger et al., 1990 Sexual Abuse 3 - 16 United States Female NR NR NR NR NR NR NR NR Deblinger et al., 1996 Sexual Abuse 7 - 13 United States Mixed White / European American Yes Yes Yes No No No Yes Deblinger et al., 2001 Sexual Abuse 4 - 11 United States Mixed White / European American Yes Yes Yes No No No Yes Deblinger et al., 2006 Sexual Abuse 8 - 14 United States Mixed White / European American Yes Yes Yes No No Yes Yes Deblinger et al., 2017 Sexual Abuse 7 - 17 United States Mixed White / European American Yes Yes Yes No No Yes Yes Dorsey et al., 2014 Various 6 - 15 United States Mixed Biracial Yes Yes No Yes Yes Yes No Ehntholt et al., 2005 War - Related Violence 11 - 15 United Kingdom Mixed --- --- --- --- --- --- --- --- Eloranta et al., 2017 War - Related Violence 10 - 13 Palestine Mixed --- --- --- --- --- --- --- --- Farkas, 2009 Various 13 - 17 Canada Mixed --- --- --- --- --- --- --- --- 90 Feather & Ronan, 2009 Maltreatment 9 - 13 New Zealand Mixed --- --- --- --- --- --- --- --- Foa et al., 2013 Sexual Abuse 13 - 16 United States Female Black / African American Yes Yes Yes No No Yes Yes Gelkopf & Berger, 2009 Terrorism 12 - 14 Israel Male --- --- --- --- --- --- --- --- Giannopoulou et al., 2006 Natural Disaster 8 - 12 Greece Mixed --- --- --- --- --- --- --- --- Gilboa - Schechtman et al., 2010 Various 12 - 18 NR Mixed --- --- --- --- --- --- --- --- Goodkind et al., 2010 Violence 12 - 15 United States Mixed Native American No No No Yes No Yes No Gormez et al., 2017 War - Related Violence 10 - 15 Turkey Mixed --- --- --- --- --- --- --- --- Graham et al., 2017 Natural Disaster 8 - 17 United States Mixed White / European American Yes Yes Yes Yes Yes No No Grefe, 2011 Various 13 - 17 United States Female Black/ African American Yes Yes Yes No No Yes No Habigzang et al., 2013 Sexual Abuse 9 - 16 Brazil Female --- --- --- --- --- --- --- --- Habigzang et al., 2016 Sexual Violence 7 - 16 Brazil Female --- --- --- --- --- --- --- --- Hartman, 2011 Sexual Abuse 8 - 14 United States Mixed Hispanic/ Latinx No Yes Yes No No Yes No H é bert & Daignault, 2015 Sexual Abuse 3 - 6 Canada Mixed --- --- --- --- --- --- --- --- Hubel et al., 2014 Sexual Abuse 6 - 12 United States Mixed White / European American Yes Yes Yes Yes No Yes No Ito et al., 2016 Natural Disaster NR Japan Mixed --- --- --- --- --- --- --- --- Jaberghaderi et al., 2004 Sexual Abuse 12 - 13 Iran Female --- --- --- --- --- --- --- --- Jaycox et al., 2010 Natural Disaster 9 - 14 United States Mixed White / European American Yes Yes Yes No No No Yes Jensen et al., 2014 Various 10 - 18 Norway Mixed --- --- --- --- --- --- --- --- Jordans et al., 2010 War - Related Violence 11 - 14 Nepal Mixed --- --- --- --- --- --- --- --- 91 Kaczkurkin et al., 2016 Sexual Abuse 13 - 18 United States Female Black / African American NR Yes NR NR NR NR NR Kagan et al., 2008 Various 8 - 15 United States Mixed White / European American Yes Yes Yes No No Yes No Kameoka et al., 2015 Various 3 - 17 Japan Mixed --- --- --- --- --- --- --- --- Kangaslampi et al., 2016 War - Related Violence 10 - 13 Palestine Mixed --- --- --- --- --- --- --- --- Kataoka et al., 2003 Community Violence 8 - 14 United States Mixed Hispanic/ Latinx No No Yes No No No No King et al., 2000 Sexual Abuse 5 - 17 Australia Mixed --- --- --- --- --- --- --- --- Kjellgren et al., 2013 Physical Abuse 6 - 14 Sweden Mixed --- --- --- --- --- --- --- --- March et al., 1998 Single Incident Stressor 10 - 15 United States Mixed White / European American Yes Yes No Yes Yes No No McMullen et al., 2013 War - Related Violence 13 - 17 DR Congo Male --- --- --- --- --- --- --- --- Misurell et al., 2011 Sexual Abuse 5 - 10 United States Mixed Black / African American NR Yes Yes NR NR NR NR Misurell et al., 2014 Sexual Abuse 4 - 17 United States Mixed Black / African American Yes Yes Yes No No Yes No Morsette et al., 2012 Violence 10 - 15 United States Mixed Native American Yes No Yes Yes No No Yes Murray et al., 2013 Witnessing Violence 5 - 18 Zambia Mixed --- --- --- --- --- --- --- --- Nixon et al., 2012 Single Incident Trauma 7 - 17 Australia Mixed --- --- --- --- --- --- --- --- O'Callaghan et al., 2013 Sexual Exploitation 12 - 17 DR Congo Female --- --- --- --- --- --- --- --- O'Callaghan et al., 2015 War - Related Violence 8 - 17 DR Congo Mixed --- --- --- --- --- --- --- --- O'Donnell et al., 2014 Traumatic Grief 7 - 13 Tanzania Mixed --- --- --- --- --- --- --- --- Ooi et al., 2016 War - Related Violence 10 - 17 Australia Mixed --- --- --- --- --- --- --- --- Pfeiffer & Goldbeck, 2017 War - Related Violence 14 - 18 Germany Male --- --- --- --- --- --- --- --- Pityaratstian et al., 2015 Natural Disaster 10 - 15 Thailand Mixed --- --- --- --- --- --- --- --- 92 Qouta et al., 2012 War - Related Violence 10 - 13 Palestine Mixed --- --- --- --- --- --- --- --- Ruf et al., 2010 War - Related Violence 7 - 16 Germany Mixed --- --- --- --- --- --- --- --- Ruiz, 2016 Sexual Abuse 8 - 16 United States Mixed Hispanic/ Latinx Yes Yes Yes No No No Yes Runyon et al., 2009 Physical Abuse 4 - 14 United States Mixed Black / African American Yes Yes Yes No No No No Runyon et al., 2010 Physical Abuse 7 - 13 United States Mixed NR NR Yes NR NR NR NR NR Salloum & Overstreet, 2008 Traumatic Grief 7 - 12 United States Mixed Black / African American Yes Yes No No No Yes No Saltzman et al., 2001 Community Violence 11 - 14 United States Mixed Hispanic/ Latinx Yes Yes Yes No No No No Sarkadi et al., 2018 War - Related Violence 13 - 18 Sweden Mixed --- --- --- --- --- --- --- --- Scheeringa et al., 2011 Various 3 - 6 United States Mixed Black / African American Yes Yes No No No No Yes Schottelkorb et al., 2012 War - Related Violence 6 - 13 United States Mixed Black / African American Yes Yes No No Yes No Yes Sezibera et al., 2009 War - Related Violence 15 - 18 Rwanda Mixed --- --- --- --- --- --- --- --- Shaheen & Oppenheim, 2016 War - Related Violence 10 - 14 Palestine Mixed --- --- --- --- --- --- --- --- Sinclair et al., 1995 Sexual Abuse 12 - 18 United States Female White / European American Yes Yes Yes Yes Yes No No Smith et al., 2007 Single Incident Trauma 8 - 18 United Kingdom --- --- --- --- --- --- --- --- --- Springer et al., 2012 Sexual Abuse NR United States Mixed Black / African American Yes Yes Yes No No Yes Yes Stein et al., 2003 Violence NR United States Mixed NR NR NR NR NR NR NR NR Stewart et al., 2017 Various 7 - 16 United States Mixed Hispanic/ Latinx Yes Yes Yes No No No No Thornback & Muller, 2015 Various 7 to 12 Canada Mixed --- --- --- --- --- --- --- --- Wolmer et al., 2003 Natural Disaster 6 - 11 Turkey Mixed --- --- --- --- --- --- --- --- 93 Wolmer et al., 2011 War - Related Violence 9 - 11 Israel Mixed --- --- --- --- --- --- --- --- Wolmer et al., 2013 War - Related Violence 9 - 11 Israel Mixed --- --- --- --- --- --- --- --- Note. Race data was not included for studies outside North America as constructs related to race vary by country . Note. NR = Not Reporte d 94 Table 7 . Treatment Descriptors by Study Study Treatment Type Study Design Setting Inclusion of Other Treatment Techniques Parental Involvement Treatment Delivery Treatment Frequency Session Length Number of Sessions Jordans et al., 2010 CBI RCT School Yes No Group 3 x a Week 60 min 15 Goodkind et al., 2010 CBITS Pre - Post School No Yes Group Weekly NR 10 Jaycox et al., 2010 CBITS Pre - Post School No Yes Group NR NR 1 0 Kataoka et al., 2003 CBITS RCT School No Yes Group Weekly NR 8 Morsette et al., 2012 CBITS Pre - Post School Yes Yes Group Weekly NR 10 Stein et al., 2003 CBITS RCT School No No Group NR NR 10 Cohen et al., 2004a CBT - CTG Pre - Post Outpatient Clinic No Yes Individual Weekly 60 min 16 Kjellgren et al., 2013 CPC - CBT Pre - Post Community No Yes Individual Weekly 120 min 16 Runyon et al., 2009 CPC - CBT Pre - Post University No Yes Individual Weekly 120 min 16 Runyon et al., 2010 CPC - CBT Pre - Post University No Yes Group Weekly 120 min 16 Berger & Gelkopf , 2009 ERASE - Stress Quasi - Experimental School Yes No Group Weekly 90 min 12 Berger et al., 2012 ERASE - Stress Quasi - Experimental School Yes No Group Weekly 90 min 16 95 Gelkopf & Berger, 2009 ERASE - Stress Quasi - Experimental School No No Group Weekly 90 min 12 Shaheen & Oppenheim, 2016 ERASE - Stress Pre - Post School No Yes NR NR 90 min 13 Misurell et al., 2011 GB - CBT Pre - Post Hospital Yes No Group NR 90 min 12 Misurell et al., 2014 GB - CBT Pre - Post Hospital Yes Yes Individual NR 90 min 11.5 Springer et al., 2012 GB - CBT Pre - Post Community Yes No Group Weekly 90 min 12 Berliner & Saunders, 1996 General CBT RCT Sexual Assault Clinic No No Group Weekly NR NR Brown et al., 2006 General CBT Pre - Post School No No Group Weekly NR 10 de Roos et al., 2011 General CBT RCT Community No Yes Individual Weekly 60 min 4 Deblinger et al., 1990 General CBT Pre - Post Community No Yes Individual NR NR 12 Deblinger et al., 1996 General CBT RCT Community No Yes Individual NR 80 min NR Deblinger et al., 2001 General CBT RCT Community No Yes Individual Weekly 120 min 8.52 Gormez et al., 2017 General CBT Pre - Post School No NR Group Weekly 70 min 8 Graham et al., 2017 General CBT Pre - Post School No NR Individual Weekly 55 min NR Habigzang et al., 2013 General CBT Pre - Post NR No No Group Weekly 90 min 16 96 Habigzang et al., 2016 General CBT Quasi - Experimental NR No No Group Weekly 90 min 16 Ito et al., 2016 General CBT Pre - Post School No No Group Once 90 min 1 Jaberghader i et al., 2004 General CBT RCT Clinic No Yes Individual Weekly 45 min 12 King et al., 2000 General CBT RCT Clinic No Yes Individual Weekly 50 min 20 Saltzman et al., 2001 General CBT Pre - Post School No No Group Weekly 50 min 20 Sezibera et al., 2009 General CBT Pre - Post NR No No Individual Weekly 120 min 10 Smith et al., 2007 General CBT RCT NR No Yes Individual Weekly NR 10 Wolmer et al., 2003 General CBT Pre - Post School Yes Yes Group NR 120 min 8 Wolmer et al., 2013 General CBT Pre - Post School Yes NR NR NR NR 14 Pfeiffer & Goldbeck, 2017 Mein Weg Pre - Post Child Welfare Agencies Yes No Group Weekly 90 min 6 Farkas, 2009 MASTR - EMDR RCT Community Yes NR Individual Weekly 90 min 12 March et al., 1998 MMTT Pre - Post School No No Group Weekly NR 18 Catani et al., 2009 KIDNET RCT Refugee Camp No No Individual 3 x a Week 60 min 6 Ruf et al., 2010 KIDNET RCT Outpatient Clinic No No Individual Weekly 90 min 8 Berger et al., 2007 OTT Quasi - Experimental School Yes Yes Group Weekly 90 min 8 97 Grefe, 2011 PARTNERS with Teens Pre - Post University Yes Yes Individual NR NR 18 Salloum & Overstreet, 2008 Project LAST Pre - Post Community Yes Yes Group Weekly 60 min 10 Hubel et al., 2014 Project SAFE Pre - Post Child Advocacy Center No Yes Group Weekly 90 min 12 Brownlow et al., 2016 PE - A RCT Community No No Individual Weekly 60 min 14 Capaldi et al., 2016 PE - A RCT Community No No Individual NR 60 min 12 Foa et al., 2013 PE - A RCT Crisis Center No NR Individual Weekly 60 min 14 Gilboa - Schechtman et al., 2010 PE - A RCT Community No No Individual Weekly 60 min 13.42 Kaczkurkin et al., 2016 PE - A RCT Crisis Center No NR Individual Weekly 60 min 12 Kagan et al., 2008 Real Life Heroes Pre - Post Residential Treatment ; Outpatient Mental Health Clinic No Yes Individual NR NR NR Danielson et al., 2012 RRFT RCT Community Yes Yes Individual Weekly 60 min 23 Sinclair et al., 1995 SAY Group Pre - Post Group Home No No Group Weekly NR 20 Cohen & Mannarino, 1998 SAS - CBT RCT Outpatient Clinic No Yes Individual Weekly 45 min 12 98 Bicanic et al., 2014 STEPS Pre - Post Community No Yes Group Weekly 120 min 9 Wolmer et al., 2011 SIT RCT School No NR NR Weekly 45 min 14 Barron & Abdallah, 2017 TRT Quasi - Experimental School No No Group NR 120 min 7 Barron et al., 2016 TRT RCT School No No Group NR NR 5 Barron et al., 2017 TRT RCT Secure Facility No No Group Biweekly 40 min 14 Chen et al., 2014 TRT RCT School Yes No Group Weekly 60 min 6 Ehntholt et al., 2005 TRT Quasi - Experimental School No No Group Weekly 60 min 6 Eloranta et al., 2017 TRT Quasi - Experimental School No No Group NR NR NR Giannopoul ou et al., 2006 TRT Pre - Post Community No Yes Group Weekly 120 min 6 Kangaslamp i et al., 2016 TRT RCT School No Yes Group Biweekly 120 min 8 Ooi et al., 2016 TRT RCT School No No Group NR 60 min 8 Pityaratstian et al., 2015 TRT RCT NR No No Group 3 Days 120 min 3 Qouta et al., 2012 TRT RCT School No No Group Biweekly 120 min 16 Sarkadi et al., 2018 TRT Pre - Post Community Yes Yes Group NR 90 min 5 99 Allen & Hoskowitz, 2017 TF - CBT Pre - Post Community Yes No NR NR NR 15.7 Bambrah et al., 2018 TF - CBT Pre - Post Community No Yes Individual Weekly 45 min NR Bartlett et al., 2018 TF - CBT Pre - Post Community No Yes Individual NR NR 21 Cohen et al., 2004 TF - CBT RCT Community No Yes Individual Weekly 90 min 12 Cohen et al., 2005 TF - CBT RCT Community No Yes Individual Weekly 90 min 12 Cohen et al., 2016 TF - CBT Pre - Post Residential Treatment No NR Individual NR NR NR Costantino et al., 2014 TF - CBT RCT School No Yes Individual Weekly 90 min 18 Damra et al., 2014 TF - CBT RCT Community No Yes NR Biweekly 60 min 10 Deblinger et al., 2006 TF - CBT RCT NR No Yes NR NR NR NR Deblinger et al., 2017 TF - CBT Pre - Post Community No Yes Individual NR NR NR Dorsey et al., 2014 TF - CBT Pre - Post Community Yes Yes Individual NR NR NR Feather & Ronan, 2009 TF - CBT Pre - Post Community No Yes Individual NR NR 16 Hartman, 2011 TF - CBT Pre - Post Hospital No No Individual Weekly NR NR H é bert &; Daignault , 2015 TF - CBT Pre - Post Clinical No Yes Individual NR NR 12.6 100 Jaycox et al., 2010 TF - CBT Pre - Post Community No Yes Individual NR NR 12 Jensen et al., 2014 TF - CBT RCT Community No Yes Individual Weekly NR 13.5 Kameoka et al., 2015 TF - CBT Pre - Post Community No Yes Individual NR 60 min 14.31 McMullen et al., 2013 TF - CBT RCT School No NR Group NR NR 15 Murray et al., 2013 TF - CBT Pre - Post Community No Yes Individual Weekly 60 min 11 Nixon et al., 2012 TF - CBT RCT University Hospital No Yes Individual Weekly 90 min 6.59 n et al., 2013 TF - CBT RCT Vocational Training Setting No Yes Group 3 x a Week 120 min 15 n et al., 2015 TF - CBT RCT Community No Yes Individual 3 x a Week 90 min 9 et al., 2014 TF - CBT Pre - Post Community No Yes Group Weekly NR 12 Ruiz, 2016 TF - CBT Pre - Post NR No Yes Individual NR NR 12 Scheeringa et al., 2011 TF - CBT RCT NR No Yes Individual NR NR 12 Schottelkor b et al., 2012 TF - CBT RCT School No Yes Individual Biweekly 30 min 17 Stewart et al., 2017 TF - CBT Pre - Post Telehealth No NR Individual Weekly 45 min 14.13 Thornback & Muller, 2015 TF - CBT Pre - Post Community No Yes Individual Weekly NR NR 101 Note . CBI = Classroom - Based Intervention; CBITS = Cognitive Behavioral Intervention for Trauma in Schools; CBT - CTG = Cognitive Behavioral Therapy for Childhood Traumatic Grief; CPC - CBT = Combined Parent Child Cognitive Behavioral Therapy for Families at Risk for Child Physical Abuse; ERASE - Stress = Enhancing Resiliency Amongst Students Experiencing Stress; KIDNET = Narrative Exposure Therapy for Children; GB - CBT = Game - Based Cognitive Behavioral Therapy; MASTR - EMDR = Motivation - Adaptive Skills - Trauma Resolution E ye Movement Desensitization and Reprocessing; MMTT = Multi - Modality Trauma Treatment; NR = Not Reported; OTT = Overshadowing the Threat of Terrorism; PE - A = Prolonged Exposure Therapy for Adolescents; Project LAST = Project Loss and Survival Team; Project SAFE = Project Sexual Abuse Family Education; RCT = Randomized Controlled Trials ; RRFT = Risk Reduction Through Family Therapy ; SAS - CBT = Sexual Abuse - Specific Cognitive Behavioral Therapy; SIT = Stress Inoculation Training; TRT = Teaching Recover y Techniques; TF - CBT = Trauma - Focused Cognitive Behavioral Therapy 102 Table 8 . Outcome Measures Used in Meta - Analysis Outcome Measure Developers k Study Posttraumatic Stress Symptoms Anxiety Disorders Interview Schedule for DSM - IV (ADIS) PTS Scale Silverman & Albano, 1996 1 King et al., 2000 Child and Adolescent Trauma Screen (CATS) Sachser et al., 2017 1 Pfeiffer & Goldbeck, 2017 Child Post - Traumatic Stress Disorder Reaction Index (CPTSD - RI) Pynoos et al., 1987 2 Wolmer et al., 2003; Wolmer et al., 2013 Child PTSD Symptom Scale (CPSS) Foa et al., 2001 14 Brown et al., 2006; Brownlow et al., 2016; Capaldi et al., 2016; Foa et al., 2013; Gilboa - Schechtman et al., 2010 ; Goodkind et al., 2010; Grefe, 2011; Jensen et al., 2014; Jordans et al., 2010; Kaczkurkin et al., 2016; Kataoka et al., 2003; Morsette et al., 2012; Nixon et al., 2012; Smith et al., 2007; Stein et al., 2003 Child Report of Post - Traumatic Symptoms (CROPS) Greenwald & Rubin, 1999 1 Jaberghaderi et al., 2004 Child Revised Impact of Events Scale (CRIES) Smith et al., 2003 10 Barron & Abdallah, 2017; Barron et al., 2016; Barron et al., 2017; Chen et al., 2014; Eloranta et al., 2017; Giannopoulou et al., 2006 ; Kangaslampi et al., 2016; Ooi et al., 2016; Qouta et al., 2012; Sarkadi et al., 2018 Child Stress Scale (CSS) Lipp & Lucarelli, 1998 1 Habigzang et al., 2013 Revised (CITES - R) Wolfe et al., 1991 1 Hubel et al., 2014 - Traumatic Stress Reaction Index (CPTS - RI) Nader & Fairbanks, 1994 3 Feather & Ronan, 2009; Gormez et al., 2017; H é bert & Daignault, 2015 103 Clinician - Administered PTSD Scale - Child a n d Adolescent Version (CAPS - C) Nader et al., 1994 1 March et al., 1998 Diagnostic Interview Schedule for Children (DISC) PTS Scale Shaffer et al., 2000 1 Farkas, 2009 DSM - III - R PTSD Symptoms APA, 1987 1 Deblinger et al., 1990 DSM - IV Interview APA, 2003 1 Habigzang et al., 2016 Impact of Events Scale Revised (IES - R) Weiss, 2004 2 Ehntholt et al., 2005; Ito et al., 2016 Post - Traumatic Stress Symptoms in Children (PTSS C) Ahmad et al., 2000 1 Damra et al., 2014 Preschool Age Psychiatric Assessment (PAPA) PTSD Scale Egger et al., 2006 1 Scheeringa et al., 2011 Schedule for Affective Disorders and Schizophrenia for School - Aged Children (K - SADS) Kaufman et al., 1997 7 Cohen et al., 200 4; Deblinger et al., 1996; Deblinger et al., 2001; Deblinger et al., 2006; Deblinger et al., 2017; Runyon et al., 2009; Runyon et al., 2010 Trauma Symptom Checklist for Children (TSCC) PTS Scale Briere, 1996 11 Bicanic et al., 2014; Cohen et al., 2005; Graham et al., 2017; Hartman, 2011; Kagan et al., 2008; Kjellgren et al., 2013; Misurell et al., 2011; Misurell et al., 2014; Ruiz, 2016; Springer et al., 2012; Thornback & Muller, 2015 Trauma Symptom Checklist for Young Children (TSCYC) PTS Scale Briere, 2005 2 Allen & Hoskowitz, 2017; Bambrah et al., 2018 UCLA PTSD Reaction Index for DSM - IV Steinberg et al., 2004 27 Berger & Gelkopf, 2009; Berger et al., 2007; Berger et al., 2012; Catani et al., 2009; Cohen et al., 2016; Cohen et al., 2004a; Costantino et al., 2014; Danielson et al., 2012; de Roos et al., 2011; Dorsey et al., 2014; Gelkopf & Berger, 2009; Jaycox et al., 2010; Kameoka et al., 2015; McMullen et al., 2013; Murray et al., 2013; 2015; Ruf et al., 2010; Salloum & Overstreet, 104 2008; Saltzman et al., 2001; Schottelkorb et a l., 2012; Sezibera et al., 2009; Shaheen & Oppenheim, 2016; Stewart et al., 2017; Wolmer et al., 2011 Young Child PTSD Checklist (YCPC) Scheeringa, 2010 1 Bartlett et al., 2018 Youth Self - Report (YSR) PTSD Scale Achenbach, 1991 1 Sinclair et al., 1995 Anxiety Multidimensional Anxiety Scale for Children (MASC) March et al., 1997 5 Brown et al., 2006 ; Costantino et al., 2014; de Roos et al., 2011; Goodkind et al., 2010; March et al., 1998 (RCMAS) Reynolds & Richmond, 1985 6 Berliner & Saunders, 1996 ; Ehntholt et al., 2005; Hubel et al., 2014; King et al., 2000; Nixon et al., 2012; Smith et al., 2007 Screen for Child Anxiety Related Disorders (SCARED) Birmaher et al., 1999 7 Berger et al., 2007; Berger et al., 2012 ; Cohen et al., 2004a; Jensen et al., 2014; Jordans et al., 2010; Shaheen & Oppenheim, 2016; Stewart et al., 2017 Spence, 1998 1 Gormez et al., 2017 State - Trait Anxiety Inventory for Children (STAIC) Biaggio & Spielberger, 1983 6 Cohen & Mannarino, 1998; Cohen et al., 2004 ; Deblinger et al., 1990; Deblinger et al., 1996 ; Deblinger et al., 2006; Habigzang et al., 2013; Habigzang et al., 2016 Trauma Symptom Checklist for Children (TSCC) - Anxiety Scale Briere, 1996 9 Bicanic et al., 2014 ; Cohen et al., 2005; Farkas, 2009; Graham et al., 2017; Hartman, 2011; Kje llgren et al., 2013; Misurell et al., 2011; Ruiz, 2016; Springer et al., 2012 Trauma Symptom Checklist for Young Children (TSCYC) Anxiety Scale Briere, 2005 1 Allen & Hoskowitz, 2017 Depression Beck Depression Inventory (BDI) Beck et al., 1961 3 Berger & Gelkopf, 2009; Gelkopf & Berger, 2009; Gilboa - Schechtman et al., 2010 105 The Behavior Assessment System for Children, 2 nd Edition (BASC - 2) Reynolds & Kamphaus, 2004 1 Grefe, 2011 Center for Epidemiological Studies Depression Scale (CES - D) Radloff, 1977 2 Chen et al., 2014; Ito et al., 2016 Kovacs, 1992 26 Berliner & Saunders, 1996; Brown et al., 2006; Cohen & Mannarino, 1998; Cohen et al., 2004; Cohen et al., 2005; Costantino et al., 2014; Damra et al., 2014; Danielson et al., 2012; Deblinger et al., 1990; Deblinger et al., 1996; Deblinger et al., 2006; Dorsey et al., 2014; Foa et al., 2013; Goodkind et al., 2010; Habigzang et al., 2013; Habigzang et al., 2016; Hartman, 2011; Hubel et al., 2014; Jaycox et al., 2010; Kataoka et al., 2003; King et al., 2000; Kjellgren et al., 2013; March et al., 1998; Morsette et al., 2012; Nixon et al., 2012; R unyon et al., 2009; Stein et al., 2003 Depression Self - Rating Scale (DSRS) Birleson, 1981 9 Barron et al., 2016; de Roos et al., 2011; Ehntholt et al., 2005; Eloranta et al., 2017; Giannopoulou et al., 2006; Jordans et al., 2010; Ooi et al., 2016; Qo uta et al., 2012; Smith et al., 2007 Montgomery - Asberg Depression Rating Scale (MADRS - 5) Svanborg & Asberg, 1994 1 Sarkadi et al., 2018 Moods and Feeli ngs Questionnaire (MFQ) Angold et al., 1995 8 Barron & Abdallah, 2017; Barron et al., 2017; Cohen et al., 2016; Cohen et al., 2004a; Jensen et Overstreet, 2008; Stewart et al., 2017 Preschool Age Psychiatric Assessment (PAPA) Depression Scale Egger et al., 2006 1 Scheeringa et al., 2011 Reynolds Adolescent Depression Scale (RADS) Reynolds, 1987 2 Saltzman et al., 2001; Sinclair et al., 1995 106 Table 8. Trauma Symptom Checklist for Children (TSCC) - Depression Scale Briere, 1996 6 Bicanic et al., 2014; Farkas, 2009; Graham et al., 2017; Misurell et al., 2011; Ruiz, 2016; Springer et al., 2012 Trauma Symptom Checklist for Youn g Children (TSCYC) Depression Scale Briere, 2005 1 Allen & Hoskowitz, 2017 107 Table 9 . Meta - Analyses Data for Posttraumatic Stress Symptoms in Cognitive Behavioral Treatments 95% Confidence Interval Study Treatment Type St Diff in Means Standard Error Variance Lower Limit Upper Limit Jordans et al., 2010 CBI - 0.1803 0.1112 0.0124 - 0.3982 0.0376 Goodkind et al., 2010 CBITS - 0.4379 0.2183 0.0476 - 0.8657 - 0.0101 Jaycox et al., 2010 CBITS - 0.6424 0.1455 0.0212 - 0.9275 - 0.3573 Kataoka et al., 2003 CBITS - 0.2937 0.1689 0.0285 - 0.6248 0.0374 Morsette et al., 2012 CBITS - 0.5753 0.1646 0.0271 - 0.8980 - 0.2527 Stein et al., 2003 CBITS - 0.7139 0.1912 0.0366 - 1.0887 - 0.3391 Cohen et al., 2004a CBT - CTG - 1.7163 0.3353 0.1124 - 2.3734 - 1.0592 Kjellgren et al., 2013 CPC - CBT - 1.3517 0.2949 0.0870 - 1.9297 - 0.7737 Runyon et al., 2009 CPC - CBT - 0.6816 0.2866 0.0822 - 1.2434 - 0.1199 Runyon et al., 2010 CPC - CBT - 1.6224 0.2610 0.0681 - 2.1339 - 1.1108 Berger & Gelkopf, 2009 ERASE - Stress - 1.2752 0.1703 0.0290 - 1.6090 - 0.9415 Berger et al., 2012 ERASE - Stress - 0.4858 0.1772 0.0314 - 0.8330 - 0.1385 108 Gelkopf & Berger, 2009 ERASE - Stress - 0.6930 0.1997 0.0399 - 1.0845 - 0.3016 Shaheen & Oppenheim, 2016 ERASE - Stress 0.3889 0.1661 0.0276 0.0634 0.7144 Misurell et al., 2011 GB - CBT - 0.2377 0.2459 0.0605 - 0.7197 0.2443 Misurell et al., 2014 GB - CBT - 0.3231 0.1905 0.0363 - 0.6964 0.0502 Springer et al., 2012 GB - CBT - 0.5769 0.2415 0.0583 - 1.0502 - 0.1036 Brown et al., 2006 General CBT - 0.1024 0.1294 0.0168 - 0.3561 0.1513 de Roos et al., 2011 General CBT 0.0855 0.2775 0.0770 - 0.4583 0.6294 Deblinger et al., 1990 General CBT - 2.2597 0.4324 0.1870 - 3.1073 - 1.4122 Deblinger et al., 1996 General CBT - 0.8760 0.3156 0.0996 - 1.4946 - 0.2573 Deblinger et al, 1996 #2 General CBT - 0.9113 0.3101 0.0961 - 1.5191 - 0.3036 Deblinger et al., 2001 General CBT 0.0656 0.3019 0.0911 - 0.5262 0.6573 Gormez et al., 2017 General CBT - 0.4966 0.1935 0.0374 - 0.8759 - 0.1173 Graham et al., 2017 General CBT - 0.3194 0.0969 0.0094 - 0.5093 - 0.1295 Habigzang et al., 2013 General CBT - 0.4857 0.1510 0.0 228 - 0.781 8 - 0.189 7 109 Habigzang et al., 2016 General CBT - 0.3339 0.1012 0.0102 - 0.5323 - 0.1354 Ito et al., 2016 General CBT - 0.6036 0.2318 0.0537 - 1.0580 - 0.1493 Jaberghaderi et al., 2004 General CBT 0.5191 0.5434 0.2953 - 0.5461 1.5842 King et al., 2000 General CBT - 1.2637 0.5031 0.2531 - 2.2498 - 0.2776 King et al., 2000 #2 General CBT - 1.1122 0.4936 0.2437 - 2.0797 - 0.1447 Saltzman et al., 2001 General CBT - 0.6923 0.2184 0.0477 - 1.1203 - 0.2643 Sezibera et al., 2009 General CBT - 0.8143 0.2460 0.0605 - 1.2965 - 0.3321 Smith et al., 2007 General CBT - 2.4767 0.5426 0.2945 - 3.5403 - 1.4132 Wolmer et al., 2003 General CBT - 0.2350 0.0713 0.0051 - 0.3748 - 0.0952 Wolmer et al., 2013 General CBT - 0.0939 0.0285 0.0008 - 0.1498 - 0.0380 Pfeiffer & Goldbeck, 2017 Mein Weg - 0.6822 0.2062 0.0425 - 1.0863 - 0.2781 Farkas, 2009 MASTR - EMDR - 0.4154 0.3200 0.1024 - 1.0426 0.2118 March et al., 1998 MMTT - 0.9738 0.2945 0.0867 - 1.5509 - 0.3966 Catani et al., 2009 KIDNET - 0.0141 0.3594 0.1292 - 0.7185 0.6903 110 Ruf et al., 2010 KIDNET - 0.6395 0.4104 0.1684 - 1.4438 0.1649 Berger et al., 2007 OTT - 1.0592 0.1792 0.0321 - 1.4105 - 0.7079 Grefe, 2011 PARTNERS with Teens - 1.204 8 0.587 5 0.345 2 - 2.356 3 - 0.0 533 Salloum & Overstreet, 2008 Project LAST - 1.1642 0.1931 0.0373 - 1.5427 - 0.7858 Hubel et al., 2014 Project SAFE - 0.4315 0.1307 0.0171 - 0.6876 - 0.1753 Brownlow et al., 2016 PE - A - 0.7565 0.2651 0.0703 - 1.2761 - 0.2369 Capaldi et al., 2016 PE - A - 0.8424 0.2672 0.0714 - 1.3661 - 0.3186 Foa et al., 2013 PE - A 0.6124 0.2620 0.0687 0.0988 1.1260 Gilboa - Schechtman et al., 2010 PE - A - 0.4879 0.3292 0.1084 - 1.1332 0.1574 Kaczkurkin et al., 2016 PE - A - 0.8261 0.2668 0.0712 - 1.3490 - 0.3032 Kagan et al., 2008 Real Life Heroes - 0.4167 0.1737 0.0302 - 0.7572 - 0.0761 Danielson et al., 2012 RRFT - 0.3801 0.3684 0.1357 - 1.1022 0.3420 Sinclair et al., 1995 SAY Group - 0.4043 0.1605 0.0258 - 0.7188 - 0.0897 111 Bicanic et al., 2014 STEPS - 0.5546 0.1678 0.0281 - 0.8834 - 0.2258 Wolmer et al., 2011 SIT - 0.2570 0.0536 0.0029 - 0.3621 - 0.1519 Barron & Abdallah, 2017 TRT - 0.4255 0.1609 0.0259 - 0.7409 - 0.1102 Barron et al., 2016 TRT - 0.6608 0.1747 0.0305 - 1.0033 - 0.3184 Barron et al., 2017 TRT 0.3809 0.4971 0.2471 - 0.5934 1.3553 Chen et al., 2014 TRT - 0.4654 0.3869 0.1497 - 1.2238 0.2929 Ehntholt et al., 2005 TRT - 0.8753 0.4151 0.1723 - 1.6889 - 0.0617 Eloranta et al., 2017 TRT - 0.1624 0.0997 0.0099 - 0.3579 0.0330 Giannopoulou et al., 2006 TRT - 1.0691 0.3237 0.1048 - 1.7034 - 0.4347 Kangaslampi et al., 2016 TRT - 0.1315 0.0912 0.0083 - 0.3102 0.0473 Ooi et al., 2016 TRT 0.0216 0.2219 0.0493 - 0.4134 0.4566 Pityaratstian et al., 2015 TRT - 0.1493 0.3338 0.1114 - 0.8035 0.5049 Qouta et al., 2012 TRT - 0.0908 0.1192 0.0142 - 0.3243 0.1428 Sarkadi et al., 2018 TRT - 0.3671 0.1523 0.0232 - 0.6657 - 0.0686 112 Allen & Hoskowitz, 2017 TF - CBT - 0.5222 0.0661 0.0044 - 0.6518 - 0.3926 Bambrah et al., 2018 TF - CBT - 0.4766 0.1450 0.0210 - 0.7607 - 0.1925 Bartlett et al., 2018 TF - CBT 0.3318 0.1178 0.0139 0.1009 0.5627 Cohen et al., 2004 TF - CBT - 0.4888 0.1513 0.0229 - 0.7854 - 0.1923 Cohen et al., 2005 TF - CBT - 0.2259 0.3004 0.0902 - 0.8146 0.3629 Cohen et al., 2016 TF - CBT - 0.9122 0.2104 0.0443 - 1.3245 - 0.4999 Costantino et al., 2014 TF - CBT 0.1378 0.2241 0.0502 - 0.3014 0.5770 Damra et al., 2014 TF - CBT - 8.3090 1.4629 2.1400 - 11.1762 - 5.4418 Deblinger et al., 2006 TF - CBT - 0.4935 0.1678 0.0281 - 0.8223 - 0.1647 Deblinger et al., 2017 TF - CBT - 1.0958 0.1010 0.0102 - 1.2937 - 0.8979 Dorsey et al., 2014 TF - CBT - 0.7133 0.2155 0.0465 - 1.1358 - 0.2909 Feather & Ronan, 2009 TF - CBT - 1.1752 0.4597 0.2113 - 2.0762 - 0.2742 Hartman, 2011 TF - CBT - 0.8365 0.3105 0.0964 - 1.4451 - 0.2279 Hébert &; Daignault , 2015 TF - CBT - 0.9265 0.2899 0.0841 - 1.4948 - 0.3582 113 Jaycox et al., 2010 TF - CBT - 0.8205 0.3090 0.0955 - 1.4261 - 0.2149 Jensen et al., 2014 TF - CBT - 0.5018 0.1862 0.0347 - 0.8668 - 0.1369 Kameoka et al., 2015 TF - CBT - 1.3979 0.2377 0.0565 - 1.8637 - 0.9321 McMullen et al., 2013 TF - CBT - 2.7506 0.4027 0.1621 - 3.5398 - 1.9614 Murray et al., 2013 TF - CBT - 0.5494 0.1409 0.0198 - 0.8255 - 0.2733 Nixon et al., 2012 TF - CBT 0.0023 0.4369 0.1909 - 0.8541 0.8587 et al., 2013 TF - CBT - 1.9918 0.3399 0.1155 - 2.6580 - 1.3256 et al., 2015 TF - CBT 0.0468 0.2831 0.0802 - 0.5080 0.6017 al., 2014 TF - CBT - 0.4315 0.1307 0.0171 - 0.6876 - 0.1753 Ruiz, 2016 TF - CBT - 0.3857 0.1619 0.0262 - 0.7030 - 0.0685 Scheeringa et al., 2011 TF - CBT - 1.1076 0.4050 0.1640 - 1.9013 - 0.3139 Schottelkorb et al., 2012 TF - CBT 0.2167 0.3945 0.1557 - 0.5566 0.9900 Stewart et al., 2017 TF - CBT - 2.2825 0.4902 0.2403 - 3.2433 - 1.3216 Thornback & Muller, 2015 TF - CBT - 0.2793 0.1362 0.0186 - 0.5463 - 0.0123 114 Table 10 . Meta - Analyses Data for Anxiety Symptoms in Cognitive Behavioral Treatments 95% Confidence Interval Study Treatment Type St Diff in Means Standard Error Variance Lower Limit Upper Limit Jordans et al., 2010 CBI 0.0 93 6 0.1110 0.0 123 - 0.1240 0.3111 Goodkind et al., 2010 CBITS - 0.4483 0.2187 0.0 478 - 0.8770 - 0.0 19 6 Cohen et al., 2004a CBT - CTG - 0.9551 0.257 3 0.0 66 3 - 1.459 4 - 0.450 9 Kjellgren et al., 2013 CPC - CBT - 0.929 6 0.2551 0.0 65 1 - 1.4296 - 0.4295 Berger et al., 2012 ERASE - Stress - 0.0 874 0.175 1 0.0 306 - 0.430 6 0.255 7 Shaheen & Oppenheim, 2016 ERASE - Stress 0.6958 0.178 5 0.0 318 0.346 1 1.0456 Misurell et al., 2011 GB - CBT - 0.7082 0.2712 0.0 73 6 - 1.2398 - 0.176 6 Springer et al., 2012 GB - CBT - 0.4997 0.2314 0.053 6 - 0.9533 - 0.0 461 Berliner & Saunders, 1996 General CBT 0.0 14 4 0.250 5 0.0 627 - 0.476 6 0.5053 Brown et al., 2006 General CBT - 0.0 977 0.1294 0.0167 - 0.351 4 0.155 9 de Roos et al., 2011 General CBT 0.0 42 4 0.277 4 0.0 769 - 0.501 3 0.5860 115 Table 10. Deblinger et al., 1990 General CBT - 0.5391 0.2455 0.0 60 3 - 1.0203 - 0.0 579 Deblinger et al., 1996 General CBT - 0.4787 0.305 8 0.0 935 - 1.078 1 0.120 7 Deblinger et al, 1996 #2 General CBT - 0.3653 0.2976 0.0 88 6 - 0.9486 0.21 80 Gormez et al., 2017 General CBT - 0.659 4 0.1950 0.0380 - 1.041 7 - 0.277 1 Graham et al., 2017 General CBT - 0.3194 0.0 96 9 0.00 9 4 - 0.509 3 - 0.1295 Habigzang et al., 2013 General CBT - 0.792 9 0.163 8 0.0 268 - 1.113 9 - 0.471 9 Habigzang et al., 2016 General CBT - 0.333 9 0.1012 0.0 102 - 0.5323 - 0.1354 King et al., 2000 General CBT - 0.3231 0.4624 0.213 9 - 1.2294 0.5833 King et al., 2000 #2 General CBT - 0.466 7 0.465 7 0.2168 - 1.379 4 0.44 60 Smith et al., 2007 General CBT - 1.095 8 0.4378 0.191 7 - 1.953 9 - 0.237 7 Farkas, 2009 MASTR - EMDR - 0.67 80 0.325 6 0.10 60 - 1.316 1 - 0.0 39 9 March et al., 1998 MMTT - 0.973 8 0.294 5 0.0 867 - 1.5509 - 0.3966 Berger et al., 2007 OTT - 0.957 4 0.1772 0.0 314 - 1.304 7 - 0.6100 Hubel et al., 2014 Project SAFE - 0.4430 0.134 2 0.0 180 - 0.7060 - 0.180 1 Cohen & Mannarino, 1998 SAS - CBT - 0.0830 0.2933 0.0 860 - 0.6579 0.4918 116 Bicanic et al., 2014 STEPS - 0.554 6 0.167 8 0.0 281 - 0.883 4 - 0.225 8 Ehntholt et al., 2005 TRT - 0.633 7 0.406 6 0.165 3 - 1.4305 0.163 2 Allen & Hoskowitz, 2017 TF - CBT - 0.3758 0.0 64 2 0.00 41 - 0.501 6 - 0.250 1 Cohen et al., 2004 TF - CBT - 0.3678 0.149 1 0.0 222 - 0.660 1 - 0.0 756 Cohen et al., 2005 TF - CBT - 0.258 5 0.300 7 0.0 90 4 - 0.8477 0.3308 Costantino et al., 2014 TF - CBT 0.2434 0.2246 0.0 50 5 - 0.1968 0.683 7 Deblinger et al., 2006 TF - CBT - 0.369 4 0.1649 0.027 2 - 0.6926 - 0.0 46 2 Feather & Ronan, 2009 TF - CBT - 1.630 6 0.5396 0.291 2 - 2.688 2 - 0.57 30 Hartman, 2011 TF - CBT - 0.610 2 0.302 1 0.0 912 - 1.2022 - 0.0 181 Jensen et al., 2014 TF - CBT - 0.299 3 0.1886 0.0 35 6 - 0.66 90 0.0 70 4 Nixon et al., 2012 TF - CBT - 0.0 825 0.4371 0.191 1 - 0.939 2 0.7742 Ruiz, 2016 TF - CBT - 0.3264 0.160 3 0.025 7 - 0.6405 - 0.0 12 3 Stewart et al., 2017 TF - CBT - 0.893 1 0.3280 0.107 6 - 1.53 60 - 0.250 2 117 Table 11 . Meta - Analyses Data for Depression Symptoms in Cognitive Behavioral Treatments 95% Confidence Interval Study Treatment Type St Diff in Means Standard Error Variance Lower Limit Upper Limit Jordans et al., 2010 CBI - 0.3675 0.111 9 0 .0125 - 0.5868 - 0.148 3 Goodkind et al., 2010 CBITS - 0.412 9 0.2172 0 .0 47 2 - 0.838 6 0 .0 12 9 Jaycox et al., 2010 CBITS - 0.5695 0.142 8 0.020 4 - 0.8494 - 0.289 7 Kataoka et al., 2003 CBITS - 0.332 8 0.1691 0.0 28 6 - 0.6642 - 0.00 13 Morsette et al., 2012 CBITS - 0.322 7 0.1642 0.0 2 70 - 0.644 6 - 0.000 8 Stein et al., 2003 CBITS - 0.377 9 0.187 1 0.0 350 - 0.744 6 - 0.0 11 2 Cohen et al., 2004a CBT - CTG - 0.5820 0.230 6 0.0 53 2 - 1.0339 - 0.130 2 Kjellgren et al., 2013 CPC - CBT - 0.7527 0.2362 0.0 55 8 - 1.2157 - 0.289 8 Runyon et al., 2009 CPC - CBT - 0.5125 0.2659 0.0707 - 1.033 7 0.00 8 7 Berger & Gelkopf, 2009 ERASE - Stress - 0.4309 0.1570 0.0 24 7 - 0.738 7 - 0.1231 Gelkopf & Berger, 2009 ERASE - Stress - 0.385 9 0.1958 0.0 383 - 0.769 7 - 0.00 2 1 Misurell et al., 2011 GB - CBT - 0.43 90 0.25 40 0.0 64 6 - 0.9367 0.0587 118 Table 11. Springer et al., 2012 GB - CBT - 1.03 0.309 3 0.095 7 - 1.636 2 - 0.4238 Berliner & Saunders, 1996 General CBT 0.104 2 0.2794 0.0 78 1 - 0.443 5 0.651 8 Brown et al., 2006 General CBT - 0.447 1 0.135 4 0.0 183 - 0.712 5 - 0.1817 de Roos et al., 2011 General CBT - 0.0 392 0.277 4 0.0 769 - 0.582 9 0.5044 Deblinger et al., 1990 General CBT - 0.6951 0.2556 0.0 653 - 1.196 2 - 0.1941 Deblinger et al., 1996 General CBT - 0.742 3 0.3117 0.0 97 2 - 1.353 2 - 0.1313 Deblinger et al, 1996 #2 General CBT - 0.6796 0.3035 0.0 921 - 1.274 6 - 0.084 7 Graham et al., 2017 General CBT - 0.3194 0.0 96 9 0.00 9 4 - 0.509 3 - 0.1295 Habigzang et al., 2013 General CBT - 0.392 9 0.148 3 0.0 2 20 - 0.683 5 - 0.102 3 Habigzang et al., 2016 General CBT - 0.333 9 0.1012 0.0 102 - 0.5323 - 0.1354 Ito et al., 2016 General CBT - 0.536 8 0.2280 0.0 520 - 0.9837 - 0.0 898 King et al., 2000 General CBT - 0.2937 0.4619 0.213 4 - 1.199 1 0.611 7 King et al., 2000 #2 General CBT - 0.3051 0.4621 0.213 6 - 1.2109 0.6006 Saltzman et al., 2001 General CBT - 0.35 50 0.202 2 0.0 40 9 - 0.751 3 0.0 413 Smith et al., 2007 General CBT - 0.73 20 0.421 7 0.1778 - 1.5585 0.0 945 119 Farkas, 2009 MASTR - EMDR - 0.9021 0.332 3 0.1104 - 1.5534 - 0.2508 March et al., 1998 MMTT - 0.973 8 0.294 5 0.0 867 - 1.5509 - 0.3966 Grefe, 2011 PARTNERS with Teens - 0.553 7 0.480 3 0.230 7 - 1.4949 0.3876 Salloum & Overstreet, 2008 Project LAST - 0.520 3 0.1588 0.0 252 - 0.831 6 - 0.2089 Hubel et al., 2014 Project SAFE - 0.4352 0.1318 0.0 17 4 - 0.693 6 - 0.176 9 Foa et al., 2013 PE - A 0.703 1 0.2639 0.0 696 0.185 9 1.2203 Gilboa - Schechtman et al., 2010 PE - A - 0.0 78 2 0.324 6 0.1053 - 0.7143 0.55 80 Danielson et al., 2012 RRFT 0.0536 0.3652 0.133 4 - 0.662 2 0.7694 Sinclair et al., 1995 SAY Group - 0.2933 0.1702 0.0 2 90 - 0.6269 0.0 40 3 Cohen & Mannarino, 1998 SAS - CBT - 0.609 2 0.299 6 0.0 89 8 - 1.196 4 - 0.0 220 Bicanic et al., 2014 STEPS - 0.554 6 0.167 8 0.0 281 - 0.883 4 - 0.225 8 Barron & Abdallah, 2017 TRT - 0.3833 0.160 6 0.0 25 8 - 0.6980 - 0.0 686 Barron et al., 2016 TRT - 0.012 5 0.170 2 0.0 2 90 - 0.34 60 0.321 1 Barron et al., 2017 TRT - 0.326 9 0.49 60 0.24 60 - 1.29 90 0.645 2 120 Chen et al., 2014 TRT 0.1126 0.4285 0.1836 - 0.727 3 0.952 5 Ehntholt et al., 2005 TRT - 0.2627 0.3986 0.1589 - 1.0440 0.518 6 Eloranta et al., 2017 TRT - 0.3010 0.100 1 0.0 10 1 - 0.4972 - 0.1048 Giannopoulou et al., 2006 TRT - 1.069 1 0.323 7 0.104 8 - 1.7034 - 0.434 7 Ooi et al., 2016 TRT - 0.0 25 1 0.2219 0.0 49 3 - 0.4600 0.4099 Qouta et al., 2012 TRT 0.0466 0.0 911 0.00 8 3 - 0.1319 0.2252 Sarkadi et al., 2018 TRT - 0.5794 0.1593 0.0 25 4 - 0.8917 - 0.2671 Allen & Hoskowitz, 2017 TF - CBT - 0.522 2 0.0 661 0.00 43 - 0.651 8 - 0.3926 Cohen et al., 2004 TF - CBT - 0.4033 0.1493 0.0 223 - 0.6960 - 0.1106 Cohen et al., 2005 TF - CBT - 0.5165 0.304 3 0.0 92 6 - 1.1128 0.0 798 Cohen et al., 2016 TF - CBT - 0.468 5 0.1862 0.034 7 - 0.8334 - 0.103 5 Costantino et al., 2014 TF - CBT 0.178 1 0.224 3 0.050 3 - 0.2615 0.6176 Damra et al., 2014 TF - CBT - 4.5083 0.8870 0.7868 - 6.246 9 - 2.769 8 Deblinger et al., 2006 TF - CBT - 0.408 9 0.1647 0.0271 - 0.731 8 - 0.0 860 Dorsey et al., 2014 TF - CBT - 0.559 4 0.215 1 0.0 46 3 - 0.9809 - 0.1378 121 Feather & Ronan, 2009 TF - CBT - 1.065 9 0.473 3 0.2240 - 1.993 5 - 0.1382 Hartman, 2011 TF - CBT 0.1256 0.2683 0.0 7 20 - 0.400 3 0.651 5 Jaycox et al., 2010 TF - CBT - 0.3661 0.276 1 0.0 762 - 0.9072 0.1749 Jensen et al., 2014 TF - CBT - 0.549 1 0.1884 0.0 35 5 - 0.918 4 - 0.179 8 Nixon et al., 2012 TF - CBT - 0.00 7 4 0.4369 0.1909 - 0.863 8 0.84 90 al., 2014 TF - CBT - 0.431 5 0.130 7 0.0 17 1 - 0.6876 - 0.1753 Ruiz, 2016 TF - CBT - 0.348 3 0.1608 0.0 25 9 - 0.6635 - 0.0 330 Scheeringa et al., 2011 TF - CBT - 0.647 5 0.38 80 0.1505 - 1.407 9 0.11 30 Stewart et al., 2017 TF - CBT - 0.97 60 0.3137 0.0 984 - 1.590 9 - 0.3611 122 Table 12 . Summary of Initial Meta - Analytic Results k St Diff in Means Standard Error Variance Confidence Interval Z - value p - value Q - Stat I 2 - Stat Fail - Safe N Random Effects PTS 94 - 0.567 4 0.0 450 0.00 20 - 0.6556 to - 0.4791 - 12.595 4 < 0.00 1 628.3455 *** 85.040 1 7078 Anxiety 38 - 0.399 9 0.0 57 6 0.00 33 - 0.5127 to - 0.2870 - 6.9421 < 0.00 1 124.599 1*** 69.502 2 1451 Depression 64 - 0.399 7 0.0 35 9 0.00 1 3 - 0.4699 to - 0.3294 - 11.1452 < 0.001 131.5867 *** 51.362 9 4411 Note . Fail - Safe N = the number of studies with an effect of 0 that would be necessary to lead to nonsignificant overall results; I 2 - Stat = the proportion of observed variance reflecting real differences in effect sizes; K = number of independent samples that contributed to an effect size ; PTS = p osttraumatic stress; Q - Stat = variability among effect sizes (the Q - statistic is tested for significant at the .05 level) *** p < .001 123 Table 13 . Moderator and Mediator Analysis Data Moderator / M ediator Q p Q - val R 2 p R 2 Category k d SE 95% CI p PTS Subtreatment 12.965 3 0.0 7 3 CBITS 5 - 0.534 2 0.1760 - 0.8792 to - 0.1891 0.002 CPC - CBT 3 - 1.227 3 0.2595 - 1.735 9 to - 0.7187 < 0.00 1 ERASE Stress 4 - 0.5112 0.1969 - 0.897 2 to - 0.1253 0.009 GB - CBT 3 - 0.377 3 0.2410 - 0.8496 to - 0.0951 0.117 PE - A 5 - 0.456 5 0.2001 - 0.848 7 to - 0.0643 0.023 TRT 12 - 0.320 9 0.1225 - 0.5610 to - 0.0807 0.009 TF - CBT 28 - 0.664 9 0.0 82 2 - 0.82 60 to - 0.5038 <0.001 Other 35 - 0.564 2 0.0 7 30 - 0.707 2 to - 0.4212 <0.001 Trauma Exposure 24.085 6 0.00 4 * Natural Disaster 10 - 0.565 7 0.1295 - 0.8195 to - 0.3118 <0.001 Physical Abuse 4 - 1.4215 0.2481 - 1.907 9 to - 0.9352 <0.001 Sexual Abuse/Assault 28 - 0.592 7 0.0 792 - 0.7479 to - 0.4375 <0.001 Single Incident Trauma 3 - 1.017 6 0.303 9 - 1.6131 to - 0.4220 0.001 Terrorism 3 - 0.2197 0.2201 - 0.6512 to 0.2117 0.318 124 Traumatic Grief 4 - 0.83 60 0.19 50 - 1.2181 to - 0.4538 <0.001 Various 13 - 0.6168 0.116 4 - 0.844 9 to - 0.3887 <0.001 Violence 3 - 0.578 9 0.22 20 - 1.013 9 to - 0.1438 0.009 War - Related Violence 20 - 0.380 4 0.0 873 - 0.5515 to - 0.2092 <0.001 Other 7 - 0.424 7 0.159 2 - 0.7366 to - 0.1127 0.008 Predominant Race in Sample 0.791 1 0.673 Black/African American 13 - 0.4719 0.139 2 - 0.744 7 to - 0.1992 0.001 Hispanic/Latinx 6 - 0.577 2 0.197 9 - 0.9650 to - 0.1894 0.004 White/European American 18 - 0.6299 0.110 7 - 0.846 9 to - 0.4129 <0.001 Gender 10.679 9 0.005 * Female Only 12 - 0.640 4 0.1309 - 0.89 70 to - 0.3838 <0.001 Male Only 4 - 1.356 2 0.2530 - 1.852 1 to - 0.8603 <0.001 Mixed 78 - 0.5280 0.0 48 8 - 0.623 7 to - 0.4324 <0.001 Study Design 2.211 3 0.331 Pre - Post 48 - 0.619 5 0.0 630 - 0.74 30 to - 0.4960 <0.001 Quasi - Experimental 8 - 0.6365 0.14 90 - 0.9285 to - 0.3446 <0.001 RCT 39 - 0.47 90 0.0 77 1 - 0.6300 to - 0.3280 <0.001 125 Treatment Setting 10.1135 0.0 72 Clinic 3 - 0.663 7 0.353 9 - 1.357 3 to 0.0299 0.061 Community 32 - 0.626 5 0.073 9 - 0.7713 to - 0.4817 <0.001 Hospital Outpatient 3 - 0.436 8 0.2408 - 0.908 8 to 0.0352 0.070 School 31 - 0.431 1 0.0 69 3 - 0.5669 to - 0.2952 <0.001 University 3 - 1.175 6 0.279 2 - 1.7227 to - 0.6284 <0.001 Other 22 - 0.632 2 0.0 90 5 - 0.8095 to - 0.4548 <0.001 Parental Involvement - 0.727 5 0.68 9 Parents Included 50 - 0.6022 0.0 63 2 - 0.726 1 to - 0.4784 <0.001 Parents Not Included 35 - 0.517 2 0.073 5 - 0.661 2 to - 0.3732 <0.001 Other Treatment Techniques 0.0 4532 0.831 Inclusion of Other Techniques 19 - 0.5512 0.0 989 - 0.745 1 to - 0.3574 <0.001 Other Techniques Not Included 7 6 - 0.5750 0.0 52 2 - 0.677 3 to - 0.4728 <0.001 Treatment Delivery 2.8242 0.24 4 Individual 48 - 0.603 5 0.0 664 - 0.733 7 to - 0.4732 <0.001 Group 41 - 0.5710 0.0 646 - 0.697 7 to - 0.4444 <0.001 Session Frequency 3.5893 0.309 Biweekly 5 - 0.1720 0.2244 - 0.6118 to 0.2678 0.443 Three Times a Week 4 - 0.4 700 0.2298 - 0.9204 to - 0.0195 0.041 Weekly 52 - 0.6007 0.0 623 - 0.722 9 to - 0.4786 <0.001 126 Other 34 - 0.5780 0.0 74 5 - 0.72 40 to - 0.4320 <0.001 Age - 0.0058 0.755 94 Session Length - 0.0007 0.770 64 Number of Sessions - 0.0104 0.366 83 Anxiety Trauma Exposure 4.485 8 0.344 Sexual Abuse/Assault 18 - 0.4 100 0.083 4 - 0.5734 to - 0.2465 <0.001 Single Incident Trauma 3 - 0.7630 0.26 90 - 1.2901 to - 0.2359 0.005 Various 3 - 0.5650 0.221 8 - 0.9997 to - 0.1303 0.011 War - Related Violence 6 - 0.2026 0.136 5 - 0.470 2 to 0.0649 0.138 Other 9 - 0.408 5 0.1167 - 0.6373 to - 0.1797 <0.001 Predominant Race in Sample 0.490 3 0.78 3 Black/African American 3 - 0.348 3 0.1540 - 0.6501 to - 0.0464 0.024 Hispanic/Latinx 4 - 0.313 7 0.1488 - 0.605 3 to - 0.0220 0.035 White/European American 12 - 0.4211 0.0 793 - 0.5766 to - 0.2656 <0.001 Gender 0.871 9 0.350 Female Only 4 - 0.5457 0.1647 - 0.8686 to - 0.2228 0.001 Mixed 34 - 0.380 6 0.0 643 - 0.5067 to - 0.2545 <0.001 Study Design 4.7295 0.0 9 4 Pre - Post 19 - 0.496 7 0.0 789 - 0.6513 to - 0.3420 <0.001 127 Quasi - Experimental 4 - 0.4709 0.1684 - 0.801 1 to 0.1408 0.005 RCT 16 - 0.2317 0.0 963 - 0.4205 to - 0.0429 0.016 Treatment Setting 3.923 1 0.14 1 Community 13 - 0.470 7 0.100 3 - 0.667 2 to - 0.2742 <0.001 School 11 - 0.237 4 0.0 996 - 0.432 7 to - 0.0421 0.017 Other 15 - 0.484 1 0.0 96 3 - 0.6727 to - 0.2954 <0.001 Parental Involvement 1.045 5 0.59 3 Parents Included 21 - 0.3883 0.0 83 8 - 0.5525 to - 0.2241 <0.001 Parents Not Included 14 - 0.3728 0.0 981 - 0.565 1 to - 0.1805 <0.001 Other Treatment Techniques 0.0 109 0.91 7 Inclusion of Other Techniques 7 - 0.414 6 0.131 2 - 0.671 7 to - 0.1574 0.002 Other Techniques Not Included 32 - 0.399 2 0.0 67 3 - 0.531 1 to - 0.2673 <0.001 Treatment Delivery 3.77 20 0.15 2 Individual 21 - 0.4439 0.0 903 - 0.62 10 to - 0.2668 <0.001 Group 15 - 0.444 4 0.0 927 - 0.626 1 to - 0.2627 <0.001 Age 0.0072 0.813 38 Session Length - 0.0020 0.601 27 Number of Sessions - 0.0009 0.962 33 Depre - ssion Subtreatment 3.700 2 0.29 6 CBITS 5 - 0.408 2 0.11 10 - 0.6257 to - 0.1906 <0.001 128 TRT 10 - 0.2487 0.0 861 - 0.417 6 to - 0.0799 0.004 TF - CBT 17 - 0.442 6 0.0 694 - 0.578 7 to - 0.3065 <0.001 Other 33 - 0.424 3 0.0500 - 0.522 2 to - 0.3264 <0.001 Trauma Exposure 14.186 4 0.11 6 Natural Disaster 7 - 0.4600 0.102 5 - 0.660 9 to - 0.2592 <0.001 Physical Abuse 3 - 0.8519 0.210 7 - 1.264 9 to - 0.4390 <0.001 Sexual Abuse/Assault 21 - 0.3690 0.0 61 4 - 0.489 3 to - 0.2487 <0.001 Single Incident Trauma 3 - 0.663 8 0.237 3 - 1.128 8 to - 0.1988 0.005 Terrorism 3 - 0.2645 0.1461 - 0.5509 to - 0.0219 0.070 Traumatic Grief 4 - 0.4685 0.1214 - 0.7065 to - 0.2305 <0.001 Various 8 - 0.5733 0.115 9 - 0.8004 to - 0.3463 <0.001 Violence 3 - 0.3667 0.1490 - 0.658 8 to - 0.0747 0.014 War - Related Violence 7 - 0.2143 0.089 7 - 0.390 1 to - 0.0386 0.017 Other 6 - 0.397 9 0.1248 - 0.642 5 to - 0.1533 0.001 Predominant Race in Sample 2.666 6 0.26 4 Black/African American 9 - 0.376 8 0.102 9 - 0.578 4 to - 0.1752 <0.001 Hispanic/Latinx 6 - 0.2663 0.1113 - 0.4845 to - 0.0481 0.017 129 White/European American 16 - 0.472 6 0.066 1 - 0.602 1 to - 0.3431 <0.001 Gender 0.627 8 0.428 Female Only 7 - 0.3305 0.3305 - 0.50 90 to - 0.1311 0.001 Mixed 55 - 0.4018 0.0 36 9 - 0.474 1 to - 0.3296 <0.001 Study Design 10.946 5 0.00 4 * Pre - Post 31 - 0.495 5 0.0 45 3 - 0.584 3 to - 0.4068 <0.001 Quasi - Experimental 6 - 0.354 4 0.0 91 7 - 0.534 2 to - 0.1746 <0.001 RCT 28 - 0.258 9 0.0 56 4 - 0.3694 to - 0.1484 <0.001 Treatment Setting 10.975 3 0.00 4 * Community 22 - 0.5616 0.0 60 3 - 0.6797 to - 0.4435 <0.001 School 21 - 0.308 8 0.0 5 20 - 0.410 7 to - 0.2069 <0.001 Other 22 - 0.3473 0.0 62 4 - 0.469 6 to - 0.2251 <0.001 Parental Involvement 3.488 9 0.17 5 Parents Included 32 - 0.475 2 0.054 1 - 0.581 1 to - 0.3692 <0.001 Parents Not Included 28 - 0.3 400 0.0518 - 0.4415 to - 0.2384 <0.001 Other Treatment Techniques 1.3298 0.24 9 Inclusion of Other Techniques 13 - 0.4783 0.0 772 - 0.6296 to - 0.3270 <0.001 Other Techniques Not Included 52 - 0.3781 0.0 399 - 0.4563 to - 0.2999 <0.001 Treatment Delivery 1.7499 0.41 7 Individual 30 - 0.401 4 0.060 8 - 0.520 5 to - 0.2823 <0.001 Group 32 - 0.3812 0.0 474 - 0.474 2 to - 0.2883 <0.001 130 Session Frequency 2.214 1 0.33 1 Biweekly 3 - 0.170 2 0.176 7 - 0.5164 to - 0.1761 0.335 Weekly 40 - 0.387 2 0.0 44 4 - 0.474 3 to - 0.3000 <0.001 Other 22 - 0.4370 0.0 55 7 - 0.5461 to - 0.3279 <0.001 Age 0.0246 0.133 64 Session Length - 0.0015 0.524 40 Number of Sessions 0.0043 0.652 56 * p < .05, indicating statistical significance 131 Table 14 . Populations that Need Examining in Future Research Moderator Specific Populations Racial Identity Asian/Asian American youth Native American youth Youth of racial minority backgrounds in other countries since race is conceptualized differently outside the United States Trauma Type Racial Trauma Generational Trauma Neglect Socioeconomic Status Low Income Middle Income High Income Sexual Orientation Any youth who identify was lesbian, gay, bisexual, transgender, queer/questioning (LGBTQ) Gender Identity Transgender Gender nonconforming Comorbidity Youth with comorbid physical health concerns Youth with comorbid mental health concerns Setting Juvenile detention center Residential placement facility 132 Figure 1 . Conceptual Framework on Factors T hat Impact Traumatized Youth Mental Health Problems . Traumatic Event Exposure Biological Factors (e.g., heightened sympathetic nervous system activity, prolonged stress response, genetic traits, family psychiatric history) Psychological Factors (e.g., negative cognitive appraisals, low intelligence, poor cognitive flexibility, low self - regulation skills, low self - efficacy, low locus of control) Ecological Factors (e.g., family instability, social support, past trauma exposure, cultural factors like age, SES, gender) Trauma Severity Mental Health Dysfunction/ Poor Outcomes 133 Figure 2 . Cyclic Influence of Emotion, Cognition, and Behavior (Tolin, 2016). Emotions/ Feelings Cognitions Behaviors 134 Figure 3 . Flow c hart of Meta - Analysis Phases ( Moher et al., 2009 ) *Note . 94 studies that produced 97 effect sizes were included in the meta - analysi s Studies identified through database search (n = 22,334 ) Additional studies identified through other sources (n = 19 ) Studies after remo val of duplicates (n = 19,273 ) Studies screened (n = 19,273 ) Full - text articles assessed for eligibility (n = 176 ) Studies included in the meta - analysis (n = 94* ) Studies excluded (n = 19,097 ) Not a Study (n = 7414) Not CBT (n = 503) Not Involving Children (n = 5347) Not English (n = 73) Not Trauma Related (n = 3866) Not Treatment Study (n = 1894) Studies excluded due to missing outcome measures (n = 82 ) 135 Figure 4 . Funnel Plot to Determine Publication Bias for Posttraumatic Stress Symptom Data 136 Figure 5 . Funnel Plot to Determine Publication Bias for Posttraumatic Stress Symptom Data ( W ithout O utlier) 137 Figure 6 . Funnel Plot to Determine Publication Bias for Anxiety Symptom Data 138 Figure 7 . Funnel Plot to Determine Publication Bias for Depression Symptom Data 139 Figure 8 . Funnel Plot to Determine Publication Bias for Depression Symptom Data ( W ithout O utlier) 140 APPENDIX B CODING SHEET Coder Initials: Article Characteristics ---- How many authors? ---- order) ---- What is the year of publication/completion? ---- What type of report is this? 1. Journal 2. Book or book chapter 3. Dissertation 4. 5. Private report 6. Government report 7. Conference paper ---- Was this a peer - reviewed document (note if unknown)? ---- What kind of organization produced this report? 1. University 2. Government entity 3. Contract research firm ---- What is the name of the organization (note if unknown)? ---- Journal that published this article (note if this is not published in a journal) ---- Article title ---- Study/Sample Characteristics What was the sampling procedure? 1. Random 2. Purposive ---- Was the treatment group compared to a controlled group? 1. Yes, waitlist 2. Yes, another treatment (specify) 3. No ---- What type of setting was this study in? 1. Clinical 2. School 3. Other 141 What was the geographic location of the study? 1. United States (specify the state/region) 2. International (specify country) ---- What was the community setting? 1. Urban 2. Suburban 3. Rural 4. Mixed 5. Unknown ---- Demographic Characteristics (Moderating Variables) Number of youth involved in the study at the beginning after meeting inclusion criteria ---- Number of youth who completed treatment Number of youth who left the study without completing treatment (note if there were any significant differences between attrition and completion groups) ---- Gender of youth (number of female participants: number of male participants) Age range ---- Mean age ---- Grade (note if unknown) ---- SES range (note if unknown) Race of youth (note percentage breakdown) ---- Sexuality of youth (note if unknown) ---- Mental health diagnos is (note if unknown) ---- Type of trauma exposure 1. Interpersonal 2. Non - interpersonal 3. Mixed 4. Unknown ---- Type of trauma exposure 1. Acute 2. Chronic 3. Mixed 4. Unknown ---- Specify the specific types of trauma youth were exposed to including percentage breakdowns (e.g., sexual abuse, physical abuse, witnessing violence, neglect, car accident) Treatment Characteristics (Mediating Variables) Type of treatment ---- 142 1. TF - CBT 2. CBITS 3. Other Frequency of intervention 1. Twice a week 2. Weekly 3. Monthly 4. Other (specify) 5. Unknown ---- Number of sessions (specify) ---- Duration of intervention sessions (specify) ---- Were there any modifications to the treatment manual (if yes, specify) Was there a parent psychoeducation component (if yes, specify the number of sessions) ---- Was there a follow - up assessment? ---- If yes, after how many months? 1. 1 month 2. 3 months 3. 6 months 4. 12 months 5. Other (specify) ---- If yes, was the control group also assessed? ---- How was treatment delivered? 1. Group 2. Individual 3. Mixed 4. Unknown ---- Outcome Assessment Characteristics Type of assessment used (select all that apply) 1. 2. Child PTSD Symptom Scale (CPSS) 3. UCLA PTSD Reaction Index 4. SCARED 5. Child Behavior Checklist ( CBCL ) 6. K - SADS 7. Other (specify) ---- Validity and reliability information available 1. Internal consistency 2. Test - retest correlation 3. 4. Other (specify) 5. Unknown ---- 143 Analysis Characteristics Type of analysis 1. ANOVA 2. ANCOVA 3. MANOVA 4. MANCOVA 5. Mean Difference 6. Other (specify) 7. Unknown ---- Units used in the statistical analysis 1. Groups 2. Individuals ---- Type of study design 1. Randomized controlled design 2. Quasi - experimental 3. Pre - post 4. Other (specify) ---- 144 APPENDIX C NARRATIVE SUMMARIES Well - Established Treatment General Cognitive Behavioral Treatments . Overall, 18 journal articles examined general cognitive behavioral treatments. Ten of the 18 studies used pre - post design, seven used RCTs, and one was a quasi - experimental study. One study occurred in a sexual assault clinic, seven were in schools, four w ere in the community, two were in a clinic, and four did not report the study setting. Seven studies took place in the United States and ten occurred internationally. The international studies were in Australia (King et al., 2000), Brazil ( Habigzang et al. , 2014; Habigzang et al., 2016), Iran (Jaberghaderi et al., 2004), Israel (Wolmer et al., 2013); Japan (Ito et al., 2016), the Netherlands (de Roos et al., 2011), Rwanda (Sezibera et al., 2009) ,and Turkey (Gormez et al., 2017; Wolmer et al., 2003). Of the seven United States - with the predominant race in the samples being White /European American in four studies, Black /African American in one study, and Hispanic/ Latinx in one study. Most studies ( k = 13) had a mix of fe male and male participants while four studies (Deblinger et al., 1990; Habigzang et al., 2013; Habigzang et al., 2016; Jaberghaderi et al., 2004) examined female - only samples. The types of trauma examined in the general cogn itive behavioral treatments were sexual abuse/ violence ( k = 8 ), war - related violence/terrorism ( k = 4), natural disasters ( k = 3), a fireworks disaster ( k = 1), single incident trauma ( k = 1) and community violence ( k = 1). T wo of the studies used other tr eatment techniques. Specifically, and colleagues; (2013) treatment included integrated balanced exercise. Eight studies had parental involvement in treatment, seven did not have pa rental involvement, and three did not explicitly report if there was parental involvement. Group treatment occurred in eight of the studies, nine studies delivered treatment through individual sessions, and one study did not report treatment delivery. Trea tment typically occurred weekly ( k = 13) and session length ranged from 45 minutes to two hours. The number of treatment sessions provided to participants in the studies ranged from one to 20. Teaching Recovery Techniques (TRT) . Teaching Recovery Techni ques (TRT) was examined in 12 studies. Two studies used pre - post design, three used quasi - experimental design, and seven used RCTs. One study occurred in a secure facility, two were in the community setting, eight were in schools, and one study did not rep ort treatment setting. All twelve studies were conducted internationally in Australia (Ooi et al., 2016), China (Chen et al., 2014), Greece (Giannopoulou et al., 2006), Palestine (Barron & Abdallah, 2017; Barron et al., 2016; Eloranta et al., 2017; Kangas lampi et al., 2016; Qouta et al., 2012), Scotland (Barron et al., 2017), Sweden (Sarkadi et al., 2018), Thailand (Pityaratstian et al., 2015), and the United Kingdom (Ehntholt et al., 2005). All twelve studies had a mix of female and male participants. The types of trauma youth were exposed to were war - related violence ( k = 7), domestic trauma ( k = 1), traumatic grief ( k = 1), and 145 natural disasters ( k = 3). Two studies involved treatment techniques in addition to cognitive behavioral techniques. Specificall y, Chen and colleagues (2014) incorporated dual - attention therapy, and Sarkadi and colleagues (2018) incorporated dual attention tasks and dreamwork. Three studies included parental involvement in treatment, and nine studies did not involve parents in trea tment. All studies were conducted in a group setting. Treatment occurred weekly in three studies and biweekly in three studies. Treatment occurred for three days in one study, and treatment frequency was not reported in five studies. Session length ranged from 40 minutes to 120 minutes, with two studies not reporting session length. The number of treatment sessions provided to participants in the studies ranged from three to 16, with one study not reporting the number of sessions. Trauma - Focused C ognitive Behavioral Therapy (TF - CBT) . Trauma - focused cognitive behavioral therapy was the most common subtreatment examined in this meta - analysis ( k = 28). Sixteen of the 28 studies used pre - post design and 12 used RCTs. Sixteen occurred in a community set ting, one was telehealth, one was in a vocational training setting, one was in residential treatment, one was in a university hospital, one was in a hospital, three were in schools, one was in a clinical setting, and three did not report the study setting. Fifteen studies took place in the United States and 13 occurred internationally. The international studies were in Australia (Nixon et al., 2012), Canada (Bambrah et al., 2018; H é bert & Daignault, 2015; Thornback & Muller, 2015 ), DR Congo (McMullen et al. al., 2015), Jordan (Damra et al., 2014), New Zealand (Feather & Ronan, 2009), (Murray et al., 2013). All thirteen United States - based studies included race data, with the predominant race in the samples being White/ European American in eight studies, Black/African American in two studies, Hispanic/Latinx in four studies, and Biracial in one study. Most o f the studies ( k = 25) had a mix of female female - only sample, and two studies (Damra et al., 2014; McMullen et al., 2013) examined male - only samples. The types of trauma examined in the TF - CBT studies were sexual abuse/exploitation ( k = 9), physical abuse ( k = 1), maltreatment ( k = 1), witnessing violence ( k n = 1), traumatic grief ( k = 1), war - related violence/terrorism ( k = 4), natural disasters ( k = 1), single incident trauma ( k = 1), and various traumas ( k = 9). Two studies involved other treatment techniques. Twenty - three studies included parental involvement in treatment, two did not include parental involvement, and three did not report parental involvement information. Three studies conducted treatment in a group setting, 22 studies engaged in in dividual treatment, and three did not report treatment delivery. Treatment typically occurred weekly ( k = 11), but many studies did not report treatment frequency ( k = 13). Session length ranged from 30 minutes to two hours, and the number of treatment se ssions provided to participants in the studies ranged from nine to 21. 146 Probably Efficacious Treatment Cognitive Behavioral Intervention for Trauma in Schools (CBITS). Overall, five journal articles examined Cognitive Behavioral Intervention for Trauma i n Schools (CBITS). Three of the five studies used pre - post design and two used RCTs. All five studies occurred in schools, and all five took place in the United States. Only one study (Stein et al., 2003) did not report data about the n the other four studies, the predominant race in the samples were Native American ( k = 2), Hispanic/Latinx ( k = 1), and White/European American ( k = 1). All five studies examined a mix of female and male participants. The types of trauma examined in the C BITS studies were violence ( k = 3), community violence ( k = 1), and natural disasters ( k = 1). One study used other treatment techniques. Specifically, Morsette and colleagues (2012) incorporated traditional cultural and healing practices. All but one stud y (Stein et al., 2003) included parental involvement in treatment. Group treatment occurred in all five studies. Treatment typically occurred weekly ( k = 3) and the number of treatment sessions ranged from eight to ten sessions. No studies reported session length. Enhancing Resiliency Amongst Students Experiencing Stress (ERASE - Stress). Overall, four journal articles examined Enhancing Resiliency Amongst Students Experiencing Stress (ERASE - Stress). Three of the four studies used quasi - experimental design and one used pre - post design. All four studies occurred in schools internationally. Two studies occurred in Israel (Berger et al., 2012; Gelkopf & Berger, 2009), one study occurred in Sri Lanka (Berger & Gelkopf, 2009), and one st udy occurred in Palestine (Shaheen & Oppenheim, 2016). Three studies examined a mix of female and male participants while one study (Gelkopf & Berger, 2009) examined a male - only sample. The types of trauma examined in ERASE - Stress studies were natural disa sters ( k = 1) and terrorism/war - related violence ( k = 3). Two studies used other treatment techniques in addition to cognitive behavioral techniques. Specifically, the treatment in Berger and - oriented exercise, and expressive religious and spiritual practices in addition to meditative practices. One study (Shaheen & Oppenheim, 2016) included parental involvement in treatment. Treat ment typically occurred weekly ( k = 3) in a group setting ( k = 3). Session length was 90 minutes, and the number of treatment sessions provided to participants in the studies ranged from 12 to 16. Narrative Exposure Therapy for Children ( KIDNET). Narrat ive Exposure Therapy for Children (KIDNET) was examined in two studies (Catani et al., 2009; Ruf et al., 2010). Both studies used RCT designs. One study occurred in a refugee camp and one occurred in an outpatient clinic. Both studies occurred internationa lly, with one study occurring in Germany with youth exposed to war - related violence (Ruf et al., 2010) and one study occurring in Sri Lanka with youth exposed to a natural disaster (Catani et al., 2009). Both studies examined a mix of female and male parti cipants. Both studies used only cognitive behavioral techniques, and neither study included parental involvement. Catani and - minute individual treatment three times a week 147 provided 90 - minute individual treatment weekly for eight sessions. Prolonged Exposure Therapy for Adolescents (PE - A) . Prolonged Exposure Therapy for Adolescents (PE - A) was examined in five studies. All five studies used an RCT design, with two studies oc curring in a crisis center and three occurring in a community setting. Four studies were in the United States, with the predominant races in the samples being Black/African American (Brownlow et al., 2016; Capaldi et al., 2016; Foa et al., 2013; Kaczkurkin et al., 2016). Gilboa - Schechtman and colleagues (2010) did not report study location. One study examined a mix of female and male participants exposed to various traumatic events (Gilboa - Schechtman et al., 2010). Female - only samples exposed to sexual abus e/assault were treated in the other four studies. Only cognitive behavioral techniques were used in all five studies. Parents were not involved in three of the treatment studies (Brownlow et al., 2016; Capaldi et al., 2016; Gilboa - Schechtman et al., 2010), and two studies did not report parental involvement. Sixty - minute individual sessions were provided weekly in four studies. Capaldi and colleagues (2016) did not report session frequency, but they also provided 60 - minute individual treatment sessions. The number of treatment sessions provided to participants ranged from 12 to 14. Possibly Efficacious Treatment Classroom - Based Intervention (CBI). One study examined Classroom - Based sed an RCT design in a school setting. The study occurred in Nepal, and it examined a mix of female and male participants. Jordans and colleagues (2012) examined youth exposed to war - related violence. Play therapy, creative - expressive therapy, and experien tial therapy techniques were used in addition to cognitive behavioral techniques. Parents were not involved in treatment. Treatment was conducted three times a week for 60 minutes in a group setting. Overall, each group was exposed to 15 treatment sessions . Motivation - Adaptive Skills - Trauma Resolution Eye Movement Desensitization and Reprocessing (MASTR - EMDR) . One study examined Motivation - Adaptive Skills - Trauma Resolution Eye Movement Desensitization and Reprocessing (MASTR - EMDR ; Farkas, 2009 ). community setting. The study occurred in Canada, and it examined a mix of female and male participants exposed to various traumatic events. EMDR and motivational interviewing were used in addition to cognitive behavioral techniques. Parental involvement in treatment was not reported. Treatment was conducted weekly for 90 minutes. Overall, individual treatment occurred for 12 sessions. Overshadowing the Threat of Terrorism (OTT) . One study examined Overshadowing the Threat of Terrorism (OTT; Berger et al., 2007) using a quasi - experimental design in a school setting. The study occurred in Israel, and it examined a mix of female and male participants exposed to war - related violence . Mindfulness, body - oriented exercise, and expressive therapy techniques were used in addition to cognitive behavioral techniques. Parents were involved in treatment. 148 - minute group treatment weekly for eight sessions. Risk Reduction Through Family Therapy (RRFT) (2012) study examined Risk Reduction Through Family Therapy (RRFT) through an RCT design in a community setting. The study occurred in the United States, and the predomin ant race in the sample was Black/African American. The sample involved a mix of female and male participants exposed to sexual assault. Multisystemic therapy was used in conjunction with cognitive behavioral techniques. Parents were involved in the weekly 60 - minute sessions. Treatment was provided individually for 23 sessions. Sexual Abuse - Specific Cognitive Behavioral Therapy (SAS - CBT). Cohen and - Specific Cognitive Behavioral Therapy (SAS - CBT) through an RC T design in an outpatient clinic. The study occurred in the United States, with the predominant race in the sample being White/European American. The sample was a mix of female and male participants exposed to sexual abuse. Parents were involved in this tr eatment that used only cognitive behavioral techniques. Treatment was provided in 45 - minute individual sessions weekly for 12 sessions. Stress Inoculation Training (SIT) examined Stress Inoculation Training (SIT) thr ough an RCT design in a school setting. The study occurred in Israel, and the sample was a mix of female and male participants exposed to war - related violence. Parent involvement in treatment was not reported in SIT, a treatment that used only cognitive be havioral techniques. Treatment was provided in 45 - minute weekly sessions for 14 sessions. Treatment delivery method (i.e., group versus individual treatment) was not described. Experimental Treatment Cognitive Behavioral Therapy for Childhood Traumatic Grief (CBT - CTG). One study examined Cognitive Behavioral Therapy for Childhood Traumatic Grief (CBT - - post design in an outpatient cl inic. The study occurred in the United States, and the predominant race in the sample was White/ European American. The authors examined a mix of female and male youth exposed to traumatic grief. No other treatment techniques outside of cognitive behaviora l techniques were used. Parents were involved in treatment. Treatment was conducted individually in weekly, 60 - minute sessions for 16 sessions. Combined Parent - Child Cognitive Behavioral Therapy (CPC - CBT). Overall, three studies examined Combined Parent - Child Cognitive Behavioral Therapy (CPC - CBT). All three studies used pre - post design. One study was in a community and two were in university settings. Two studies took place in the United States and one study took place in Sweden (Kjellgren et al., 2013) . Only one United States - based study included complete race data. The predominant race in that sample was Black/African American. All five studies examined a mix of female and male participants. The types of trauma examined in the three studies were physic al abuse. No studies used other treatment techniques, but all three studies included parental involvement in treatment. Two studies conducted treatment individually while one study conducted treatment in a group setting. In 149 all three studies, treatment was provided in 120 - minute weekly sessions for 16 sessions. Game - Based Cognitive - Behavioral Therapy (GB - CBT). Three studies examined Game - Based Cognitive Behavioral Therapy (GB - CBT; Misurell et al., 2011; Misurell et al., 2014; Springer et al., 2012). All three studies used pre - post treatment group only designs. Two studies occurred in hospitals and one occurred in a community setting. All three studies occurred in the United States, with the predominant race in the three samples being Black/African America n. All three studies examined a mix of female and male participants who had been exposed to sexual abuse. All three studies used play therapy techniques in addition to cognitive behavioral techniques. Only Misurell and colleagues (2014) included parental i nvolvement in treatment. Only one study (Springer et al., 2012) reported treatment frequency; specifically, treatment occurred weekly. Misurell and colleagues (2014) conducted treatment individually while Misurell and colleagues (2011) and Springer and col leagues (2012) conducted treatment in a group setting. Session length was 90 minutes, and the number of treatment sessions provided to participants in the studies ranged from 11.5 to 12. Mein Weg. One study examined Mein Weg (Pfeiffer & Goldbeck). The s tudy used a pre - post design in child welfare agencies. The study occurred in Germany, and the authors examined males exposed to war - related violence. Group processing principles were used in addition to cognitive behavioral techniques. Parents were not inv olved in the weekly group treatments. Treatment occurred for 90 minutes for six sessions. Multi - Modality Trauma Treatment (MMTT). One study examined Multi - Modality Trauma Treatment (MMTT; March et al., 1998). The study used a pre - post design in a school. United States, with the predominant race in the sample being White/European American. The sample involved a mix of female and male participants who had been exposed to a single incident stressor. Only cog nitive behavioral techniques were used without parental involvement in treatment. Session length was not reported, and 18 weekly group treatment sessions were provided to participants. PARTNERS with Teens. One study examined PARTNERS with Teens (Grefe, 2011). The study used a pre - post design in a university setting. The study occurred in the United States, with the predominant race in the sample being Black/African American. The sample involved only female participants exposed to various traumas. Dialect ical Behavior Therapy (DBT) and motivational interviewing were used in addition to cognitive behavioral techniques. Parents were included in Project Loss and Survival Team (Project LAST) . One study examined Project Loss and Survival Team (Project LAST; Salloum & Overstreet, 2008). The study used a pre - post design in the community. The study occurred in the United States, with the predominant race in the sample being Bla ck/African American. The sample involved a mix of female and male participants exposed to traumatic grief. Narrative therapy was used in addition to cognitive behavioral techniques. Parents udy provided 60 - minute group treatment weekly for ten sessions. 150 Project Sexual Abuse Family Education (Project SAFE) (2014) study examined Project Sexual Abuse Family Education (Project SAFE). The study used a pre - post design in a child advocacy center. The study occurred in the United States, with the predominant race in the sample being White/European American. The sample involved a mix of female and male participants exposed to sexual abuse. Parents were included in Project SAF E, which used only cognitive behavioral techniques. The study provided 90 - minute group treatment weekly for 12 sessions. Real Life Heroes. Heroes. The study used a pre - post design in both residential treatment and in an outpatient mental health clinic. The study occurred in the United States, with the predominant race in the sample being White/European American. The sample involved a mix of female and male participants exposed to various traumatic events. Parents were included in Real Life Heroes, which used only cognitive individual sessions, but data on treatment frequency, session length, and session numbers were not provided. SAY Group study used a pre - post design in a group home. The study occurred in the United States, with the predominant race in the sample being White/European American. The sample included only females exposed to sexual abuse. Parents were not included in SAY Group, a treatment that used only cognitive behavioral techniques. The study p rovided 20 weekly sessions in a group setting. Session length was not reported. STEPS pre - post design in a community setting. The study occurred in the Netherlands with a female - only sample exposed to rape. Parents were included in this treatment that used only cognitive behavioral techniques. Treatment was provided in 120 - minute weekly sessions in a group setting for nine sessions. 151 REFERENCES 152 REFERENCES Achenbach, T.M. (1991). Manual for the youth self - report and 1991 profile. Burlington, VT: University of Vermont Department of Psychiatry Ahmad, A., Sundelin - Wahlsten, V., Sofi, M. A., Qahar, J. A., & Knorring, V. (2000). Reliability and vali dity of a child - specific cross - cultural instruments for assessing posttraumatic stress disorder . European Child & Adolescent Psychiatry, 9, 285 294. Alisic , E., Zalta, A. K., van Wesel, F., Larsen, S. E., Hafstad, G. S., . . . & Smid, G. E. (2014). Rates of post - traumatic stress disorder in trauma - exposed children and adolescents: Meta - analysis. The British Journal of Psychiatry, 204 , 335 - 340. doi: 10.1192/b jp.bp.113.131227 *Allen, B., & Hoskowitz, N. A. (2017). Structured trauma - focused CBT and unstructured play/experiential techniques in the treatment of sexually abused children: A field study with practicing clinicians. Child Maltreatment, 22, 112 - 120. do i: 10.1177/1077559516681866 American Academy of Child and Adolescent Psychiatry (2010). Practice parameter for the assessment and treatment of children and adolescents with posttraumatic stress disorder. Journal of the American Academy of Child & Adolesce nt Psychiatry, 49, 414 - 430. American Psychiatric Association. ( 1987 ). Diagnostic and statist ical manual of mental disorders : DSM - III - R . Arlington, VA: American Psychiatric Association. American Psychiatric Association. ( 2003 ). Diagnostic and statist ical manual of mental disorders : DSM - IV . Arlington, VA: American Psychiatric Association. American Psychiatric Association. (2013). Diagnostic and statist ical manual of mental disorders : DSM - 5 ( 5th ed.). Arlington, VA: American Psychiatric Associat ion. American Psychological Association (2008). Children and trauma: Update for mental health professionals. Retrieved from http://www.apa.org/pi/families/resources/children - trauma - update.aspx American Psychological Association (2017). Multicultural guidelines: An ecological approach to context, identity, and intersectionality. Retrieved from http://www.apa.org/about/policy/multicultural - guidelines.pdf Angold, A., Costello, E. J., Messer, S. C. , Pickles, A., Winder, F., & Silver, D. (1995). Development of a short questionnaire for use in epidemiological studies of depression in children and adolescents. International Journal of Methods in P sychiatric Research, 5 , 237 249. 153 APA Task Force on SES. (2007). Report of the APA Task force on SES. Washington, D.C.: APA. Atdjian, S., & Vega, W. A. (2005). Disparities in mental health treatment in U.S. racial and ethnic minority groups: Implications for psychiatrists. Psychiatric Services, 56, 1600 - 1602. doi: 10.1176/appi.ps.56.12.1600 *Bambrah, V., Mastorakos, T., Cordeiro, K. M., Thornback, K., & Muller, R. T. (2018). Parent - child discordance and child trauma symptomatology throughout therapy: Correlates and treatment response. Journal of Family Violence, 33 , 281 - 295. doi: 10.1007/s108 96 - 017 - 9948 - x Bandura, A. (1977). Social Learning Theory . Englewood Cliffs, NJ: Prentice - Hall. *Barron, I., & Abdallah, G. (2017). Field trial of a complicated grief psychosocial program for adolescents in occupied Palestine. Journal of Aggression, Malt reatment, & Trauma, 26 , 372 - 390. doi : 10.1080/10926771.2017.1290000 *Barron, I., Abdallah, G., & Heltne, U. (2016). Randomized control trial of Teaching Recovery Techniques in rural occupied Palestine: Effect on adolescent dissociation. Journal of Aggression, Maltreatment, & Trauma, 25, 955 - 973. doi: 10.1080/10926771.2016. 1231149 *Barron, I., Mitchell, D., & Yule, W. (2017). Pilot study of a group - based psychosocial trauma recovery program in secure accommodation in Scotland . Journal of Family Violence, 32 , 595 - 606. doi: 10.1007/s10896 - 017 - 9921 - 8 *Bartlett, J. D., Griffin, J. L., Spinazzola, J., Fraser, J. G., Noro ñ a, C. R., . . . & Barto, B. (2018). The impact of a statewide trauma - informed care initiative in child welfare on the well - being of children and youth with complex trauma. Children and Youth Services Revi ew, 84 , 110 - 117. doi: 10.1016/j.childyouth.2017.11.015 Beck, A.T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961) . An inventory for measuring depression. Archives of General Psychiatry, 4, 561 - 571. Benjamin, C. L., Puleo, C. M., Settipani, C. A., Brodman, D. M., Edmunds, J. M., . . . & Kendall, P. C. (2011). History of cognitive - behavioral therapy in youth. Child & Adolescent Psychiatric Clinics of North America, 20 , 179 - 189. doi: 10.1016/j.chc.2011.01.011 *Berger, R., & Gelkopf, M. (2009). S chool - based intervention for the treatment of tsunami - related distress in children: A quasi - randomized controlled trial. Psychotherapy and Psychosomatics, 78 , 364 - 371. doi: 10.1159/000235976 *Berger, R., Gelkopf, M., & Heineberg, Y. (2012). A teacher - delivered intervention for adolescents exposed to ongoing and intense traumatic war - related stress: A quasi - 154 randomized controlled study. Journal of Adolescent Health, 51 , 453 - 461. doi: 10.1016/j.j adohealth.2012.02.011 *Berger, R., Pat - Horenczyk, R., & Gelkopf, M. (2007). School - based intervention for prevention and treatment of elementary - - related distress in Israel: A quasi - randomized controlled trial. Journal of Traumatic Stress, 20 , 541 - 551. doi : 1 0.1002/jts.20225 *Berliner, L., & Saunders, B. E. (1996). Treating fear and anxiety in sexually abused children: Results of a controlled 2 - year follow - up study. Child Maltreatment, 1 , 294 - 309. doi: 10.1177/1077559596001004002 Biaggio, A. & Spielberger, C. D. (1983). Inventário de Ansiedade Traço - Estado - Idate - C Manual. Rio de Janeiro: CEPA. *Bicanic , I., de Roos, C., van Wesel, F., Sinnema, G., & van de Putte, E. (2014). Rape - related symptoms in adolescents: Short - and long - term outcome after cognitive behavior group therapy. European Journal of Psychotraumatology, 5 , 1 - 8. doi: 10.3402/ejpt.v5.22969 Birleson, P. (1981). The validity of depressive disorder in childhood and the development of a self - rating scale: A research report. Journal of Child Psychology and Psychiatry, 22 , 73 - 88 . doi: 10.1111/j.1469 - 7610.1981.tb00533.x Birmaher, B., Brent, D. A., Chiappetta, L., Bridge, J., Monga, S., & Baugher, M. (1999). Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED): A replication study. Journal of the American Academy of Child & Adolescen t Psychiatry, 38 , 1230 - 1236. doi: 10.1097/00004583 - 199910000 - 00011 Blueprints for Healthy Youth Development (2018). Blueprints programs. Retrieved from http://www.blueprintsprograms.com/programs Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2005). Comprehensive meta - analysis (Version 2) . Englewood, NJ: Biostat. Borenstein, M. , Rothstein, H. R., Hedges, L. V., & Higgins, J. P. T. (2009). Introduction to meta - analysis. West Sussex, United Kingdom: John Wiley & Sons, Ltd. Borntrager, C., Chorpita, B. F., Higa - McMillan, C. K., Daleiden, E. L., & Starace, N. (2013). Usual care for trauma - exposed youth: Are clinician - reported therapy techniques evidence - based? Children and You th Services Review, 35, 133 - 141. doi: 10.1016/j.childyouth.2012.09.018 Brand, T., & Jungmann, T. (2014). Participant characteristics and process variables predict attrition from a home - based early intervention program. Early Childhood Research Quarterly, 29 , 155 - 167. doi: 10.1016/j.ecresq.2013.12.001 155 Brent, D. A., Holder, D., & Kolko, D. (1997). A clinical psychotherapy trial for adolescent depression comparing cognitive, family, and supportive therapy. Archives of General Psychiatry, 54 , 877 - 88 5. Briere, J. (1996). Trauma symptom checklist for children (TSCC). Odessa, FL: Psychological Assessment Resources, Inc. Briere, J. (2005). Trauma symptom checklist for young children: Professional manual. Lutz, FL: Psychological Assessment Resources. * Brown, E. J., McQuaid, J., Farina, L., Ali, R., & Winnick - Gelles, A. (2006). Matching school - based trauma intervention program. Education and Treatment of Children, 2 9 , 257 - 286. Brown, R. C., Witt, A., Fegert, J. M., Keller, F., Rassenhofer, M., & Plener, P. L. (2017). Psychosocial interventions for children and adolescents after man - made and natural disasters: A meta - analysis and systematic review. Psychological Med icine, 47 , 1893 - 1905. doi : 10.1017/S0033291717000496 *Brownlow, J. A., McLean, C. P., Gehrman, P. R., Harb, G. C., Ross, R. J., & Foa, E. B. (2016). Influence of sleep disturbance on global functioning after posttraumatic stress disorder treatment. Journa l of Traumatic Stress, 29 , 515 - 521. doi: 1 0.1002/jts.22139 California Evidence - Based Clearinghouse for Child Welfare (2018). Trauma treatment Client - level interventions (child & adolescent). Retrieved from http://www.cebc4cw.org/topic/trauma - treatment - c lient - level - interventions - child - adolescent/ *Capaldi, S., Asnaani, A., Zandberg, L. J., Carpenter, J. K., & Foa, E. B. (2016). Therapeutic alliance during prolonged exposure versus client - centered therapy for adolescent posttraumatic stress disorder. Journal of Clinical Psychology, 72 , 1026 - 1036. doi: 10.1002/jclp.22303 Depression Inventory: Construct and discriminant validity across clinical and nonreferred (control) populations. Journal of Consulting and Clinical Psychology, 55 , 755 - 761. doi: 10.1037/0022 - 006X.55.5.755 Carrion, V. G., & Hull, K. (2009). Treatment manual for trauma - exposed youth: Case studies. Clinical Child Psychology and Psychiatry, 15 , 27 - 38. doi: 10.1177/1359104509338150 Carrion, V. G., Wong, S. S., & Kletter, H. (2013). Update on neuroimaging and c ognitive functioning in maltreatment - related pediatric PTSD: Treatment implications. Journal of Family Violence, 28 , 53 - 61. doi: 10.1007/s10896 - 012 - 9489 - 2 156 Carter, R. T. (2007). Racism and psychological and emotional injury: Recognizing and assessing race - b ased traumatic stress. The Counseling Psychologist, 35 , 13 - 105. doi: 10.1177/0011000006292033 Cartledge, G., Kea, C., & Simmons - Reed, E. (2002). Serving culturally diverse children with serious emotional disturbance and their families. Journal of Child and Family Studies, 11 (1), 113 - 126. Cary, C. E., & McMillen, J. C. (2012). The data behind the dissemination: A systematic review of trauma - focused cognitive behavioral therapy for use with children and youth. Children and Youth Services Review, 34 , 748 - 757. doi: 10.1016/j.childyouth.2012.01.003 *Catani, C., Kohiladevy, M., Ruf, M., Schauer, E., Elbert, T., & Neuner, F. (2009). Treating children traumatized by war and tsunami: A comparison between exposure therapy and meditation - relaxation in North - East Sri Lanka. BMC Psychiatry, 9 , 1 - 11. doi: 1 0.1186/1471 - 244X - 9 - 22 Chambless, D. L., & Hollon, S. D. (1998). Defining empirically supported therapies. Journal of Consulting and Clinical Psychology, 66 , 7 - 18. *Chen, Y., Shen, W. W., Gao, K., Lam, C. S., Chang, W. C., & Deng, H. (2014). Effectiveness RCT of a CBT intervention for youths who lost parents in the Sichuan, China, earthquake. Psychiatric Services, 65 , 259 - 262. doi: 10.1176/appi.ps.201200470. Cheung, M. W. L., Ho, R. C. M., Lim, Y., & Mak, A. (2012). Conducting a meta - analysis: Basics and good practice. International Journal of Rheumatic Disease, 15 , 129 - 135. Chorpita, B. F., Daleiden, E. L., & Collins, K. S. (2014). Managing and adapting practice: A system for applying evidence in clinical care with youth and families. Clinical Social Work Journal, 42 , 134 - 142. doi: 10.1007/s10615 - 013 - 0460 - 3 Chorpita, B. F., Daleiden, E., & Weisz, J. R. (2005). Modularity in the design and application of therapeutic interventions. Applied and Preventive Psychology , 11 , 141 - 156 . Chorpita, B. F., & Regan, J. (2009). Dissemination of effective mental health treatment procedures: Maximizing the return on a significant investment. Behavior Research and Therapy , 11 , 990 - 993. doi: 10.1016/j.brat.2009.07.002 Cohen, J. A., Deblinger, E., & Mannarino, A. P. (2018). Trauma - focused cognitive behavioral therapy for children and families. Psychotherapy Research, 28 , 47 - 57. doi: 10.1080/10503307.2016.1208375 *Cohen, J. A., Deblinger, E., Mannarino, A. P., & Steer, R. A. (2004). A multisite, randomized controlled trial for children with sexual abuse - related PTSD symptoms. Journal of the American Academy of Child and Adolescent Psychiatry, 43 , 393 - 402. doi: 10.1 097/00004583 - 200404000 - 00005 157 Cohen, J. A., & Mannarino, A. P. (1996). A treatment outcome study for sexually abused preschool children: Initial findings. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 42 - 50. *Cohen, J. A., & Mannarino, A. P. (1998). Interventions for sexually abused children: Initial treatment outcome findings. Child Maltreatment, 3 , 17 - 26. doi: 10.1177/1077559598003001002 Cohen, J. A., Mannarino, A. P., & Deblinger, E. (2006). Treating traum a and traumatic grief in children and adolescents. New York, NY: The Guilford Press. Cohen, J. A., Mannarino, A. P, & Iyengar, S. (2011). Community treatment of posttraumatic stress disorder for children exposed to intimate partner violence: A randomized controlled trial. Archives of Pediatric & Adolescent Medicine, 165, 16 - 21. doi: 10.1001/archpediatrics.2010.247 *Cohen, J. A., Mannarino, A. P., Jankowski, K., Rosenberg, S., Kodya, S., & Wolford II, G. L. (2016). A randomized implementation study of trauma - focused cognitive behavioral therapy for adjudicated teens in residential treatment facilities. Child Maltreat ment, 21 , 156 - 167. doi: 1 0.1177/1077559515624775 *Cohen, J. A., Mannarino, A. P., & Knudsen, K. (2004a). Treating childhood traumatic grief: A pilot study. Journal of the American Academy of Child and Adolescent Psychiatry, 43 , 1225 - 1233. *Cohen, J. A., Mannarino, A.P., & Knudsen, K. (2005). Treating sexually abused children: 1 year follow - up of a randomized controlled trial. Child Abuse & Neglect, 29 , 135 - 145. doi: 10.1016/j.chiabu.2004.12.005 Connor, D. F., Ford, J. D., Arnsten, A. F. T. & Greene, C. A. (2015). An update on posttraumatic stress disorder in children and adolescents. Clinical Pediatrics, 54, 517 - 528. doi: 10.1177/0009922814540793 *Costantino, G., Primavera, L. H., Malgady, R. G., & Costantino, E. (2014). Culturally oriented trauma treat ments for Latino children post 9/11. Journal of Child & Adolescent Trauma, 7 , 247 - 255. doi: 10.1007/s40653 - 014 - 0031 - y Cuijpers, P., Huibers, M., Ebert, D. D., Koole, S. L., & Andersson, G. (2013). How much psychotherapy is needed to treat depression? A metaregression analysis. Journal of Affective Disorders , 149 , 1 - 13. doi: 10.1016/j.jad.2013.02.030 *Damra, J. K. M., Nas sar, Y. H., & Ghabri, T. M. F. (2014). Trauma - focused cognitive behavioral therapy: Cultural adaptations for application in Jordanian culture. Counselling Psychology Quarterly, 27 , 308 - 323. doi: 10.1080/09515070.2014.918534 158 *Danielson, C. K., McCart, M. R ., Walsh, K., de Arellano, M. A., White, D., & Resnick, H. S. (2012). Reducing substance use risk and mental health problems among sexually assaulted adolescents: A pilot randomized controlled trial. Journal of Family Psychology, 26 , 628 - 635. doi: 10.1037/ a0028862 *de Roos, C., Greenwald, R., den Hollander - Gijsman, M., Noorthoorn, E., van Buuren, S., & de Jongh, A. (2011). A randomised comparison of cognitive behavioural therapy (CBT) and eye movement desensitisation and reprocessing (EMDR) in disaster - exp osed children. European Journal of Psychotraumatology, 2 , 1 - 11. doi : 10.3402/ejpt.v2i0.5694 *Deblinger, E., Lippman, J., & Steer, R. (1996). Sexually abused children suffering posttraumatic stress symptoms: Initial treatment outcome findings. Child Maltre atment, 1 , 310 - 321. doi : 10.1177/1077559596001004003 *Deblinger, E., Mannarino, A. P., Cohen, J. A., & Steer, R. A. (2006). A follow - up study of a multisite, randomized, controlled trial for children with sexual abuse - related PTSD symptoms. Journal of the American Academy of Child & Adolescent Psychiatry, 45 , 1474 - 1484. doi: 10.1097/01.chi.0000240839.56114.bb *Deblinger, E., McLeer, S. V., & Henry, D. (1990). Cognitive behavioral treatment for sexually abused children suffering post - traumatic stress: Preliminary findings. American Academy of Child and Adolescent Psychiatry, 29 , 747 - 752. doi: 1 0.1097/00004583 - 19 9009000 - 00012 *Deblinger, E., Pollio, E., Runyon, M. K., & Steer, R. A. (2017). Improvements in personal resiliency among youth who have completed trauma - focused cognitive behavioral therapy: A preliminary examination. Child Abuse & Neglect, 65, 132 - 139. doi: 10.1016/j.chiabu.2016.12.014 *Deblinger, E., Stauffer, L. B., & Steer, R. A. (2001). Comparative efficacies of supportive and cognitive behavioral group therapies for young children who have been sexually abused and their nonoffending mother s. Child Maltreatment, 6 , 332 - 343. doi: 10.1177/1077559501006004006 De Young, A. C., Kenardy, J. A., & Cobham, V. E. (2011). Trauma in early childhood: A neglected population. Clinical Child and Family Psychology Review, 14 , 231 - 250. doi: 10.1007/s10567 - 0 11 - 0094 - 3 DePrince, A. P., Weinzierl, K., M., & Combs, M. D. (2009). Executive function performance and trauma exposure in a community sample of children. Child Abuse & Neglect, 33 , 353 - 361. doi:10.1016/j.chiabu.2008.08.002 Dorsey, S., McLaughlin, K. A., Kerns, S. E. U., Harrison, J. P., Lambert, H. K., Briggs, E. C., . . . & Amaya - Jackson, L. (2017). Evidence base update for psychosocial treatments for children and adolescents exposed to traumatic events. Journal of Clinical Child & Adolescent Psychology, 46 , 303 - 330. do i: 10.1080/15374416.2016.1220309 159 *Dorsey, S., Pullmann, M. D., Berliner, L., Koschmann, E., McKay, M., & Deblinger, E. (2014). En gaging foster parents in treatment: A randomized trial of supplementing Trauma - focused Cognitive Behavioral Therapy with evidence - based engagement strategies. Child Abuse & Neglect, 38 , 1508 - 1520. doi : 10.1016/j.chiabu.2014.03.020 Egger, H.L., Erkanli, A., Keeler, G., Potts, E., Walter, B.K., & Angold, A. (2006). Test retest reliability of the Preschool Age Psychiatric Assessment (PAPA). Journal of the American Academy of Child and Adolescent Psychiatry, 45 , 538 549. *Ehntholt, K. A., Smith, P. A., & Yule, W. (2005). School - based cognitive - behavioural therapy group intervention for refugee children who have experienced war - related trauma. Clinical Child Psychology and Psychiatry, 10 , 235 - 250. doi: 10.1177/1359104505051214 *Eloranta , S. J., Peltonen, K., Palosaari, E., Qouta, S. R., & Punam äki , R. (2017). The role of attachment and emotion regulation in the psychosocial intervention among war - affected children. Journal of Child & Adolescent Trauma, 10 , 301 - 314. doi: 10.1007/s40653 - 01 6 - 0115 - y *Farkas, L. (2009). The effects of Motivation - Adaptive Skills - Trauma Resolution (MASTR) Eye Movement Desensitization and Reprocessing (EMDR) on traumatized adolescents with conduct problems. (Doctoral dissertation). Universite de Montreal, Montreal. *Feather, J. S., & Ronan, K. R. (2009). Trauma - focused CBT with maltreated children: A clinic - based evaluation of a new treatment manual. Australian Psychologist, 44 , 174 - 194. doi: 10.1080/00050 060903147083 national indicators of well - being, 2010. Washington, U.S. Government Printing Office. Retrieved from http://childstats.gov . (n.d). Finkelhor, D., Ormro d, R. K., & Turner, H. A. (2007). Polyvictimization and trauma in a national longitudinal cohort. Development and Psychopathology, 19 , 149 - 166. doi:10.1017/S0954579407070083 Foa, E. B., Johnson, K. M., Feeny, N. C., & Treadwell, K. R. H. (2001). The Child PTSD Symptom Scale: A preliminary examination of its psychometric properties. Journal of Clinical Child & Adolescent Psychology, 30 , 376 - 384. doi: 10.1207/ S15374424JCCP3003_ 9 *Foa, E. B., McLean, C. P., Capaldi, S., & Rosenfield, D. (2013). Prolonged exposure vs supportive counseling for sexual abuse - related PTSD in adolescent girls: A randomized clinical trial. Journal of the American Medical Association, 310, 2650 - 2657. do i: 10.1001/jama.2013.282829. 160 Ford, J. (2015). An affective cognitive neuroscience - based approach to PTSD psychotherapy: The TARGET model. Journal of Cognitive Psychotherapy: An International Quarterly, 29 , 68 - 91. doi: 10.1891/0889 - 8391.29.1.68 Ford, J. D., Elhai, J. D., Connor, D. F., & Frueh, B. C. (2010). Poly - victimization and risk of posttraumatic, depressive, and substance use disorders and involvement in delinquency in a national sample of adolescents. Journal of Adolescent Health, 46, 545 552. doi:10.1016/j.jadohealth.2009.11.212 Fraynt, R., Ross, L., Baker, B. L., Rystad, I., Lee, J., & Briggs, E. C. (2014). Predictors of treatment engagement in ethnically diverse, urban children receiving treatment for trauma exposure. Journal of Traumatic Stress, 27, 66 - 73. doi: 10.1002/jts.21889 Garland, A. F., Bickman, L., & Chorpita, B. F. (2010). Change what? Identifying quality improvement targets by investigating usual mental health care. Administration and Policy in Me ntal Health and Mental Health Services Research, 37 , 15 - 26. doi: 1 0.1007/s10488 - 010 - 0279 - y *Gelkopf, M., & Berger, R. (2009). A school - based, teacher - mediated prevention program (ERASE - Stress) for reducing terror - related tsunami reactions in Israeli youth : A quasi - randomized controlled trial. The Journal of Child Psychology and Psychiatry, 50 , 962 - 971. doi: 10.1111/j.1469 - 7610.2008.02021.x *Giannopoulou, I., Dikaiakou, A., & Yule, W. (2006). Cognitive - behavioural group intervention for PTSD symptoms in children following the Athens 1999 earthquake: A pilot study. Clinical Child Psychology and Psychiatry, 11 , 543 - 553. doi: 10.1177/13591045060 67876 *Gilboa - Schechtman, E., Foa, E. B., Shafran, N., Aderka, I. M., Powers, M. B., . . . & Apter, A. (2010). Prolonged exposure versus dynamic therapy for adolescent PTSD: A pilot randomized controlled trial . Journal of the American Academy of Child & A dolescent Psychiatry, 49 , 1034 - 1042. doi: 10.1016/j.jaac.2010.07.014 the treatment of post - traumatic stress disorder in children and adolescents. Cochrane Da tabase of Systematic Reviews, 12 , CD006726. doi: 10.1002/14651858 Goenjian, A. K., Molina, L., Steinberg, A. M., Fairbanks, L. A., Alvarez, M. L., Goenjian, H. A., et al. (2001). Posttraumatic stress and depressive reactions among Nicaraguan adolescents after hurricane Mitch. The American Journal of Psychiatry, 158 , 788 794. doi: 10.1176/appi.ajp.158.5.788 *Goodkind, J. R., LaNoue, M. D., & Milford, J. (2010). Adaptation and implementation of cognitive behavioral intervention for trauma in schools with A merican Indian youth. Journal of Clinical Child & Adolescent Psychology, 39 , 858 - 872. doi: 10.1080/15374416.2010.517166 161 * Evaluation of a school - based, teacher - delivered psychological intervention group program for trauma - affected Syrian refugee children in Istanbul, Turkey. Psych iatry and Clinical Psychopharmacology, 27 , 125 - 131. doi: 10.1080/24750573.2017.1304748 Gosch, E. A., Flannery - Schroeder, E., Mauro, C. F., & Compton, S. N. (2006). Principles of cognitive - behavioral therapy for anxiety disorders in children. Journal of Co gnitive Psychotherapy, 20 , 247 - 262. *Graham, R. A., Osofsky, J. D., Osofsky, H. J., & Hansel, T. C. (2017). School based post disaster mental health services: Decreased trauma symptoms in youth with multiple traumas. Advances in School Mental Health Prom otion, 10 , 161 - 175. doi: 10.1080/1754730X.2017.1311798 - traumatic symptoms: Development and preliminary validation of parent and child scales. Research on Social Work Practice , 9 , 61 75. *Grefe , C. N. (2011). Treating adolescent victims of physical and sexual abuse: A multiple baseline study. Gutermann, J., Schreiber, F., Matulis, S., Schwartzkopff, L., Deppe, J., & Steil, R. (2016). Psychological treatments for symptoms of posttraumatic stress disorder in children, adolescents, and young adults: A meta - analysis. Clinical Child and Family Psych ology Review, 19, 77 - 93. doi : 10.1007/s10567 - 016 - 0202 - 5 *Habigzang, L. F., Damásio, B. F., & Koller, S. H. (2013). Impact evaluation of a cognitive behavioral group therapy model in Brazilian sexually abused girls. Journal of Child Sexual Abuse, 22, 173 - 1 90. doi : 10.1080/10538712.2013.737445 *Habigzang, L. F., Pizarro de Freitas, C. P., Von Hohendorff, J., & Koller, S. H. (2016). Cognitive - behavioral group therapy for girls victims of sexual violence in Brazil: Are there differences in effectiveness when applied by different groups of psychologists? Effectiveness of group therapy for girls victims of sexual violence. Anales de Psicologia, 32 , 433 - 440. doi: 10.6018/analesps.32.2.213041 *Hartman, S. (2010). From efficacy to effectiveness: A look at trauma - f ocused cognitive behavioral therapy in a community setting. (Doctoral dissertation). Pace University, New York. Harvey, S. T., & Taylor, J. E. (2010). A meta - analysis of the effects of psychotherapy with sexually abused children and adolescents. Clinical Psychology Review, 30 , 517 - 535. doi: 10.1016/j.cpr.2010.03.006 162 * H é bert , M., & Daignault, I. V. (2015). Challenges in treatment of sexually abused preschoolers: A pilot study of TF - CBT in Quebec. Sexologies, 24 , e21 - e27. doi : 10.1016/j.sexol.2014.09.003 Henggeler, S. W., Schoenwald, S. K., & Pickrel, S. G. (1995). Multisys temic therapy: Bridging the gap between university - and community - based treatment. Journal of Consulting and Clinical Psychology, 63 , 709 - 717. Efforts to dec rease LGBTQ bullying and harassment in Seattle public schools. Journal of School Health, 84, 1 - 9. Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child - clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65 , 599 - 610. *Hubel, G. S., Campbell, C., West, T., Friedenberg, S., Schreier, A., . . . & Hansen, D. J. (2014). Child advocacy center based group treatment for child s exual abuse. Journal of Child Sexual Abuse, 23 , 304 - 325. doi: 10.1080/10538712.2014.888121 Ingraham, C. L., & Oka, E. R. (2006). Multicultural issues in evidence - based interventions. Journal of Applied School Psychology, 22 , 127 - 149. doi: 10.1300/J370v22n02_07 *Ito, D. , Koseki, S., & Ohtani, T. (2016). A brief school - based cognitive - behavioral intervention for Japanese adolescents with severe posttraumatic stress. Journal of Traumatic Stress, 29 , 577 - 580. doi: 10.1002/jts.22145 *Jaberghaderi, N., Greenwald, R., Rubin, A., Zand, S. O., & Dolatabadi, S. (2004). A comparison of CBT and EMDR for sexually - abused Iranian girls. Clinical Psychology and Psychotherapy, 11 , 358 - 368. doi: 10.1002/cpp.395 *Jaycox, L. H., Cohen, J. A., Mannarino, A. P., Walker, D. W., Langley, A. K ., . . . & Schonlau, trauma - focused psychotherapies. Journal of Traumatic Stress, 23 , 223 - 231. doi: 10.1002/jts.20518 Jaycox, L. H., Kataoka, S. H., Stein. B. D., Langl ey, A. K., & Wong, M. (2012). Cognitive behavioral intervention for trauma in schools. Journal of Applied School Psychology, 28 , 239 - 255. doi: 10.1080/15377903.2012.695766 Jaycox, L. H., Langley, A. K., Stein, B. D., Wong, M., Sharma, P. Scott, M. . . . S chonlau, M. (2009). Support for students exposed to trauma: A pilot study. School Mental Health, 1 , 49 - 60. doi: 10.1007/s12310 - 009 - 9007 - 8 *Jensen, T. K., Holt, T., Ormhaug, S. M., Egeland, K., Granly, L., . . . & Wentzel - Larsen, T. (2014). A randomized effectiveness study comparing trauma - focused cognitive 163 behavioral therapy with therapy as usual for youth. Journal of Clinical Child & Adoles cent Psychology, 43 , 356 - 369. doi: 10.1080/15374416.2013.822307 Johnson, M. E. (2018). The effects of traumatic experiences on academic relationships and expectations in justice - involved children. Psychology in the Schools, 55, 240 - 249. doi: 10.1002/pits. 22102 *Jordans, M. J. D., Komproe, I. H., Tol, W. A., Kohrt, B. A., Luitel, N. P., . . . & de Jong, J. T. V. M. (2010). Evaluation of a classroom - based psychosocial intervention in conflict - affected Nepal: A cluster randomized controlled trial. Journal of Child Psychology and Psychiatry, 51 , 818 - 826. doi: 10.1111/j.1469 - 7610.2010.02209.x *Kaczkurkin, A. N., Asnaani, A., Zhong, J., & Foa, E. B. (2016). The moderating effect of state anger on treatment outcome in female adolescents with PTSD. Journal of Tra umatic Stress, 29 , 325 - 331. doi: 10.1002/jts.22116 *Kagan, R., Douglas, A. N., Hornik, J., & Kratz, S. L. (2008). Real life heroes pilot study: Evaluation of a treatment model for children with traumatic stress. Journal of Child & Adolescent Trauma, 1, 5 - 22. doi: 10.1080/19361520801929845 *Kameoka, S., Ya gi, J., Arai, Y., Nosaka, S., Saito, A., . . . & Asukai, N. (2015). Feasibility of trauma - focused cognitive behavioral therapy for traumatized children in Japan: A pilot study. International Journal of Mental Health Systems, 9 , 1 - 5. doi : 10.1186/s13033 - 015 - 0021 - y *Kangaslampi, S., Punam äki , R, Qouta, S., Diab, M., & Peltonen, K. (2016). Psychosocial group intervention among war - affected children: An analysis of changes in posttraumatic cognitions. Journal of Traumatic Stress, 29 , 546 - 555. doi: 10.1002/jts. 22149 *Kataoka, S. H., Stein, B. D., Jaycox, L. H., Wong, M., Escudero, P., . . . & Fink, A. (2003). A school - based mental health program for traumatized Latino immigrant children. Journal of the American Academy of Child & Adolescent Psychiatry, 42 , 311 - 3318. doi: 10.1097/00004583 - 200303000 - 00011 Kaufman, J., Birmaher, B., Brent, D., Rao, U., Flynn, C., Moreci, P., . . . & Ryan, N. (1997). Schedule for Affective Disorders and Schizophrenia for School - Age Children - Present and Lifetime Version (K - SADS - PL): Initial reliability and validity data. Journ al of the American Academy of Child & Adolescent Psychiatry, 36 , 980 - 988. doi: 10.1097/00004583 - 199707000 - 00021 Kaufman, N. K., Rohde, P., Seeley, J. R., Clarke, G. N., & Stice, E. (2005). Potential mediators of cognitive - behavioral therapy for adolescents with comorbid major depression and conduct disorder. Journal of Consulting and Clinical Psychology, 73, 48 - 46. doi: 10.1037/0022 - 006X.73.1.38 164 Kazdin, A. E. (2008). Evidence - based treatment and practice: New opportunities to bridge clinical research and practice, enhance the knowledge base, and improve patient care. American Psychologist, 63 , 146 - 159. doi : 1 0.1037/0 003 - 066X.63.3.146 Kazdin, A. E., & Nock, M. K. (2003). Delineating mechanisms of change in child and adolescent therapy: Methodological issues and research recommendations. Journal of Child Psychology and Psychiatry, 44 , 1116 - 1128. Kazdin, A. E., & Weis z, J. R. (1998). Identifying and developing empirically supported child and adolescent treatments. Journal of Consulting and Clinical Psychology, 66, 19 - 36. Kendall, P. C., & Braswell, L. (1982). Cognitive - behavioral self - control therapy for children: A components analysis. Journal of Consulting and Clinical Psychology, 50 , 672 - 689. Kendall, P. C., & Choudhury, M. S. (2003). Children and adolescents in cognitive - behavioral therapy: Some past efforts and current advances, and the challenges in our future . Cognitive Therapy and Research, 27 , 89 - 104. Kendall, P. C., & Southam - Gerow, M. A. (1996). Long - term follow - up of a cognitive - behavioral therapy for anxiety - disordered youth. Journal of Consulting and Clinical Psychology, 64 , 724 - 730. Kessler, R. C., Sonnega, A., Bromet, E., & Hughes, M. (1995). Posttraumatic stress disorder in the National Comorbidity Survey. Archives of General Psychiatry, 52 , 1048 - 1060. *King, N. J., Tonge, B. J., Mullen, P., Myerson, N., Heyne, D., . . . & Ollendic k, T. H. (2000). Treating sexually abused children with posttraumatic stress symptoms: A randomized clinical trial. Journal of the American Academy of Child & Adolescent Psychiatry, 39 , 1347 - 1355. doi: 10.1097/00004583 - 200011000 - 00008 *Kjellgren, C., Sved in, C. G., & Nilsson, D. (2013). Child physical abuse experiences of combined treatment for children and their parents: A pilot study. Child Care in Practice, 19 , 275 - 290. doi: 10.1080/13575279.2013.785934 Kliethermes, M., Schacht, M., & Drewry, K. (201 4). Complex trauma. Child & Adolescent Psychiatric Clinics of North America, 23 , 339 - 361. doi: 10.1016/j.chc.2013.12.009 Multi - Health Systems. Kowalik, J., Weller, J. , Venter, J., & Drachman, D. (2011). Cognitive behavioral therapy for the treatment of pediatric posttraumatic stress disorder: A review and meta - analysis. Journal of Behavior Therapy and Experimental Psychiatry, 42 , 405 - 413. doi: 10.1016/j.jbtep.2011.02.0 02 165 (2012). Practice - based evidence for children and adolescents: Advancing the research agenda in schools. School Psychology Review, 41, 215 - 235. Kratochwill, T. R., & Shernoff, E. S. (2003). Evidence - based practice: Promoting evidence - based interventions in school psychology. School Psychology Quarterly, 18 , 389 - 408. doi : 10.1521/scpq.18.4.389.27000 Lang, P. J. (1977). Imagery in therapy: An information processing analysis of fear. Behavior Therapy, 8 , 862 - 886. Lilly, M. M., Valdez, C. E., & Graham - Bermann, S. (2011). The mediating effect of world assumptions on th e relationship between trauma exposure and depression. Journal of Interpersonal Violence, 26 , 2499 - 2516. doi:10.1177/0886260510383033 Lindebø Knutsen, M ., Sachser, C., Holt, T., Goldbeck, L., & Jensen, T. K. (2020). Trajectories and possible predictors of treatment outcome for youth receiving trauma - focused cognitive behavioral therapy. Psychological Trauma: Theory, Research, Practice, and Policy, 12 , 336 - 3 46. doi: 10.1037/tra0000482 Lipp, M. E., & Lucarelli, M. D. M. (1998). Escala de stress infantil ESI: Manual [Child Stress Scale (CSS): Manual]. São Paulo: Casa do Psicólogo. Lipsey, M. W., & Wilson, D. B. (2000). Practical meta - analysis. Thousand Oaks, CA: SAGE Publications. Littell, J. H., Corcoran, J., & Pillai, V. (2008). Systematic reviews and meta - analysis . New York, NY: Oxford University Press, Inc. Little, S. G., & Akin - Little, A. (2013). Trauma in child ren: A call to action in school psychology. Journal of Applied School Psychology, 29 , 375 - 388 . doi: 10.1080/15377903.2012.695769 Macartney, S., Bishaw, A., & Fontenot, K. (2013). Poverty rates for selected detailed race and Hispanic groups by state and pl ace: 2007 - 2011. Retrieved from https://www.census.gov/prod/2013pubs/acsbr11 - 17.pdf . *March, J. S., Amaya - Jackson, L., Murray, M. C., & Schulte, A. (1998). Cognitive - behavioral psychotherapy for children and adolescents with posttraumatic stress disorder after a single - incident stressor. Journal of the American Academy of Child & Adolescen t Psychiatry, 37 , 585 - 593. doi : 10.1097/ 00004583 - 199806000 - 00008 March, J. S., Parker, J. D., Sullivan, K., Stallings, P., & Conners, C. K. (1997). The Multidimensional Anxiety Scale for Children (MASC): Factor structure, reliability, and validity . Journa l of the American Academy of Child and Adolescent Psychiatry, 36 , 554 - 565. 166 Masten, A. S., & Narayan, A. J. (2012). Child development in the context of disaster , war, and terrorism: Pathways of risk and resilience. Annual Review of Psychology, 63 , 227 - 257. doi: 10.1146/annurev - psych - 120710 - 100356 McKeever, V. M., & Huff, M. E. (2003). A diathesis - stress model of posttraumatic stress disorder: Ecological, biologi cal, and residual stress pathways. Review of General Psychology, 7, 237 - 250. doi: 10.1037/1089 - 2680.7.3.237 McLaughlin, K. A., Koenen, K. C., Hill, E. D., Petukhova, M., Sampson, N. A., Zaslavsky, A. M., & Kessler, R. C. (2013). Traumatic event exposure and posttraumatic stress disorder in a national sample of adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 52 , 780 783. - focus ed cognitive - behavioural therapy with former child soldiers and other war - affected boys in the DR Congo: A randomised controlled trial. Journal of Child Psychology and Psychiatry, 54 , 1231 - 1241. doi: 10 .1111/jcpp.12094 *Misurell, J., Springer, C., Acosta, L., Liotta, L., & Kranzler, A. (2014). Game - based cognitive - behavioral therapy individual model (GB - CBT - IM) for child sexual abuse: A preliminary outcome study. Psychological Trauma: Theory, Research, Practice, and Policy, 6 , 250 - 258 . doi: 10.1037/a003341 1 *Misurell, J. R., Springer, C., & Tryon, W. W. (2011). Game - based cognitive - behavioral therapy (GB - CBT) group program for children who have experienced sexual abuse: A preliminary investigation. Journal of Child Sexual Abuse, 20 , 14 - 36. doi: 10.1080/105387 12.2011.540000 Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & The PRISMA Group (2009). Preferred reporting items for systematic reviews and meta - analyses: The PRISMA statement. PLoS Med, 6 , e 1 000097. doi:10.1371/journal.pmed1000097 Morina, N., Koerssen, R., & Pollet, T. V. (2016). Interventions for children and adolescents with posttraumatic stress disorder: A meta - analysis of comparative outcome studies. Clinical Psychology Review, 47 , 41 - 54. doi : 10.1016/j.cpr.2016.05.006 Morina, N., Malek, M., Nickerson, A., & Bryant, R. A. (2017). Psychological interventions for post - traumatic stress disorder and depression in young survivors of mass violence in low - and middle - income countries: A meta - analysis. The British Journal of Psychiatry, 210, 247 - 254. doi: 10.1192/bjp.bp.115.180265 * Morsette, A., van den Pol, R., Schuldberg, D., Swaney, G., & Stolle, D. (2012). Cognitive behavioral treatment for trauma symptoms in American Indian youth: Preliminary findings and issues in evidence - based p ractice and reservation culture. Advances in School Mental Health Promotion, 5 , 51 - 62. doi: 10.1080/1754730X.2012.664865 167 Motta, R. W. (2015). Trauma, PTSD, and secondary trauma in children and adolescents. In R. Flanagan, K. Allen, & E. Levine (Eds.), Cogn itive and behavioral interventions in the schools: Integrating theory and research into practice (67 - 84). New York, NY: Springer. *Murray, L. K., Familiar, I., Skavenski, S., Jere, E., Cohen, J., . . . & Bolton, P. (2013). An evaluation of trauma focused cognitive behavioral therapy for children in Zambia. Child Abuse & Neglect, 37 , 1175 - 1185. doi: 10.1016/j.chiabu.2013.04.017 Myschailyszyn, M. P. (2015) Transporting cognitive behavior interventions to the school setting. In R. Flanagan, K. Allen, & E. Le vine (Eds.), Cognitive and behavioral interventions in the schools: Integrating theory and research into practice (287 - 305). New York, NY: Springer. Nader , K. , Blake , D. , Kriegler , J. , & Pynoos , R . (1994). C linician Administered PTSD Scale for Children (CAPS - C), Current and Lifetime Diagnosis Version, and instruction manual . Los Angeles , CA : UCLA Neuropsychiatric Institute and National Center for PTSD Nader , K , & Fairbanks , L. (1994). The suppression of reexperiencing: I mpulse control and somat ic symptoms in children following traumatic exposure. Anxiety Stress Coping, 7 , 229 - 2 39. doi: 1080/10615809408249348. National Child Traumatic Stress Network (2003). The defining trauma and child traumatic stress. Retrieved from http://www.nctsnet.org/sit es/default/files/assets/pdfs/what_is_child_traumatic_stress_0.p df . National Child Traumatic Stress Network (2018). Empirically supported treatments and promising practices. Retrieved from http://nctsnet.org/resources/topics/treatments - that - work/promising - practices#q4 *Nixon, R. D. V., Sterk, J., & Pearce, A. (2012). A randomized trial of cognitive behavior therapy and cognitive therapy for children with posttraumatic stress disorder following single - incident trauma. Journal of Abnormal Child Psychology, 40 , 327 - 337. do i: 10.1007/s10802 - 011 - 9566 - 7 Nock, M. K. (2003). Progress review of the psychosocial treatment of child conduct problems. Clinical Psychology, 10 , 1 - 28. doi: 10.1093/clipsy.10.1.1 ring a trauma focused and non trauma focused intervention with war affected Congolese youth: A preliminary randomised trial. Intervention, 13 , 28 - 44. doi: 10.1097/ WTF.0000000000000054 (2013). A randomized controlled trial of trauma - focused cognitive behavioral therapy for sexually exploited, 168 war - affected Congolese girls. Journal of the American Academy of Child & Adolescent Psychiatry, 52 , 359 - 369. doi: 10.1016/j.jaac.2013.01.013 K. (2014). Treating unresolved grief and posttraumatic stress symptoms in orphaned children in Tanzania: Group - based trauma - focused cognitive behavioral therapy. Journal of Traumatic Stress, 27 , 664 - 671. doi: 10.1002/jts.21970 *Ooi, C. S., Rooney, R. M., Roberts, C., Kane, R. T., Wright, B., & Chatzisarantis, N. (2016). The efficacy of a group cognitive behavioral therapy for war - aff ected young migrants living in Australia: A cluster randomized controlled trial. Front Psychol, 7 , 1 - 14. How to use an overview. Evidence - Based Medicine Work Group. JAMA, 272, 1367 - 1371. Perfect, M. M., Turley, M. R., Carlson, J. S., Yohannan, J., & Pfenninger Saint Gilles, M. (2016). School - related outcomes of traumatic event exposure and traumatic stress symptoms in students: A systematic review of research from 1990 to 2015. *Pfeiffer , E., & Goldbeck, L. (2017). Evaluation of a trauma - focused group intervention for unaccompanied young refugees: A pilot study. Journal of Traumatic Stress, 30 , 531 - 536. doi: 10.1002/jts.22218 *Pityaratstian, N., Piyasil, V., Ketumarn, P., Sitdhiraksa, N. , Ularntinon, S., & Pariwatcharakul, P. (2015). Randomized controlled trial of group cognitive behavioural therapy for post - traumatic stress disorder in children and adolescents exposed to tsunami in Thailand. Behavioural and Cognitive Psychotherapy, 43 , 5 49 - 561. doi: 10.1017/S1352465813001197 Pynoos, R. S., Frederick, C., Nader, K., Arroyo, W., Steinberg, A., . . . & Fairbanks, L. (1987). Life threat and posttraumatic stress in school - age children. Archives of General Psychiatry, 44 , 1057 - 1063. *Qouta, S. R., Palosaari, E., Diab, M., & Punam äki , R. (2012). Intervention effectiveness among war - affected children: A cluster randomized controlled trial on improving mental health. Journal of Traumatic Stress, 25 , 288 - 298. doi: 10.1002/jts.21707 Radloff, L. S. (1977). The CES - D scale: A self report depression scale for research in the general population. Applied Psychological Measurements, 1, 385 - 401. Ramirez de Arellano, M. A., Lyman, D. R., Jobe - Shields, L., George, P., Doug herty, R. H., Daniels, A. S., . . . & Delphin - Rittmon, M. E. (2014). Trauma - focused cognitive - behavioral therapy for children and adolescents: Assessing the evidence. Psychiatric Services, 65 , 591 - 602. 169 Reynolds, C . R., & Kamphaus, R . W. (2004). BASC - 2: Behavior assessment system for children: manual . Circle Pines, MN: AGS Pub. Reynolds, W. M. (1987). Reynolds adolescent depression scale. Washington, D. C.: Psychological Assessment Resources. Reynolds, C. R., & Richmond, B. O. (1985). . Los Angeles: Western Psychological Services. Roberts, A. L., Gilman, S. E., Breslau, J., Breslau, N., & Koenen, K. C. (2011). Race/ethnic differences in exposure to traumatic events, development of post - traumatic stress disorder, and treatment - seeking for post - traumatic stress disorder in the United States. Psy chological Medicine, 41 , 71 - 83. doi : 10.1017/S0033291710000401 Roberts, N. P., Roberts, P. A., Jones, N., & Bisson, J. I. (2015). Psychological interventions for post - traumatic stress disorder and comorbid substance use disorder: A systematic review and m eta - analysis. Clinical Psychology Review, 38 , 25 - 38. doi: 10.1016/j.cpr.2015.02.007 Roberts, A. L., Rosario, M., Corliss, H. L., Koenen, K. C., & Austin, S. B. (2012). Childhood gender nonconformity: A risk indicator for childhood abuse and posttraumatic stress in youth. Pediatrics, 129, 410 - 417. doi: 10.1542/peds.2011 - 1804 *Ruf, M., Schauer, M., Neuner, F., Catani, C., Schauer, E., & Elbert, T. (2010). Narrative exposure therapy for 7 - to 16 - year - olds: A randomized controlled trial with traumatized refugee children. Journal of Traumatic Stress, 23 , 437 - 445. doi: 10.1002/jts. 20548 *Ruiz, E. (2016). Trauma symptoms in a diverse population of sexually abused children. Psychological Trauma: Theory, Research, Practice, and Policy, 8 , 680 - 687. doi: 10.1037/tra0000160 *Runyon, M. K., Deblinger, E., & Schroeder, C. M. (2009). Pilot evaluation of outcomes of combined parent - child cognitive - behavioral group therapy for families at risk for child physical abuse. Cognitive and Behavioral Practice, 16 , 101 - 118. doi: 10.1016/j.cbpra.2008.09.006 *Runyon, M. K., Deblinger, E., & Steer, R. A. (2010). Group cognitive behavioral treatment for parents and children at - risk for physical abuse: An initial study. Child & Family Behavior Therapy, 32 , 196 - 218. doi: 10.1080/07317107.2010.500515 Sachser, C., Berliner, L., Holt, T., Jensen, T. K., Jungbluth, N., . . . & Goldbeck, L. (2017). International development and psychometric properties of the Child and Adolescent Trauma Screen (CATS). Journal of Affective Disorders , 210 , 189 195. doi: 10.1 016/j.jad.2016.12.040 170 *Salloum, A ., & Overstreet, S. (2008). Evaluation of individual and group grief and trauma interventions for children post disaster. Journal of Clinical Child and Adolescent Psychology, 37 (3), 495 507. doi: 10.1080/15374410802148194 *Saltzman, W. R., Pynoos, R. S., Layne, C. M., Steinberg, A. M., & Aisenberg, E. (2001). Trauma - and grief - focused intervention for adolescents exposed to community violence: Results of a school - based screening and group treatment protocol. Group Dynamics : Theory, Research, and Practice, 5 , 291 - 303. doi: 10.1037//1089 - 2699.5.4.291 * Sarkadi, A., Ådahl, K, Stenvall, E., Ssegonja, R., Batti, H., . . . & Salari, R. (2018). Teaching recovery techniques: Evaluation of a group intervention for unaccompanied refu gee minors with symptoms of PTSD in Sweden. European Child & Adolescent Psychiatry, 27 , 467 - 479. doi: 10.1007/s00787 - 017 - 1093 - 9 Scheeringa, M. S. (2010). Young Child PTSD Checklist (YCPC). New Orleans, LA: Tulane University. *Scheeringa, M. S., Weems, C. F., Cohen, J. A., Amaya - Jackson, L., & Guthrie, D. (2011). Trauma - focused cognitive - behavioral therapy for posttraumatic stress disorder in three - through six year - old children: A randomized clinical trial. Journal of Child Psy chology and Psychiatry, 52, 853 - 860. doi: 10.1111/j.1469 - 7610.2010.02354.x *Schottelkorb, A. A., Doumas, D. M., & Garcia, R. (2012). Treatment for childhood refugee trauma: A randomized, controlled trial. International Journal of Play Therapy, 21 , 57 - 73. doi: 10.1037/a0027430 *Sezibera, V., Van Broeck, N., & Philippot, P. (2009). Intervening on persistent posttraumatic stress disorder; Rumination - focused cognitive and behavioral therapy in a population of young survivors of the 1994 genocide in Rwanda. Jo urnal of Cognitive Psychotherapy: An international Quarterly, 23 , 107 - 113. doi: 10.1891/0889 - 8391.23.2.107 Shaffer, D., Fisher P., Lucas, C. P., Dulcan, M. K., & Schwab - Stone, M. E. (2000). NIMH Diagnostic Interview Schedule for Children Version IV (NIMH DISC - IV): Description, differences from previous versions, and reliability of some common diagnoses. Journal of the American Academy of Child and Adolescent Psychiatry, 39 , 28 - 38. doi: 10.1097/00004583 - 200001000 - 00014 *Shaheen, M. M. A., & Oppenheim, S. (2016). Youth resilience makes a difference in mitigating stress: Teacher mediated school intervention in Bet hlehem. Intervention, 14, 305 - 319. doi: 10. 1097/WTF.0000000000000134 Silverman , W . K . , & Albano , A . M . (199 6) . Anxiety Disorders Interview Schedule for DSM - IV, Child and Parent Versions. San Antonio, TX: Psychological Corporation 171 Silverman , W. K., & Hinsha w, S. P. (2008). The second special issue of evidence - based psychosocial treatments for children and adolescents: A 10 - year update. Journal of Clinical Child & Adolescent Psychology, 37, 1 - 7. doi: 10.1080/15374410701817725 Silverman, W. K., Ortiz, C. D., Chockalingham, V., Burns, B. J., Kolko, D. J., . . . & Amaya - Jackson, L. (2008). Evidence - based psychosocial treatments for children and adolescents exposed to traumatic events. Journal of Clinical Child & Adolescent Psychiatry, 37 , 156 - 183. doi: 10.1080/1 5374410701818293 *Sinclair, J. J., Larzelere, R. E., Paine, M., Jones, P., Graham, K., & Jones, M. (1995). Outcome of group treatment for sexually abused adolescent females living in a group home setting. Journal of Interpersonal Violence, 10 , 533 - 542. doi: 10.1177/0886260 95010004011 Singh, N. N., Ellis, C. R., Oswald, D. P., Wechsler, H. A., & Curtis, W. J. (1997). Value and address diversity. Journal of Emotional and Behavioral Disorders, 5 (1), 24 - 35. Skinner, B. F. (1969). Contingencies of reinforcement: A theoretical analysis. New York: Appleton. Slade, M. K., & Warne, R. T. (2016). A meta - analysis of the effectiveness of trauma - focused cognitive - behavioral therapy and play therapy for child victims of abuse. Journal of Young Investigators, 30 , 36 - 43. Smith, P., P errin, S., Dyregrov, A., & Yule, W. (2003). Principal components analysis of the impact of event scale with children in war. Personality and Individual Differences, 34, 315 322. doi:10.1016/S0191 - 8869(02)00047 - 8 *Smith, P., Yule, W., Perrin, S., Tranah , T ., Dalgleish, T., & Clark, D. M. (2007). Cognitive - behavioral therapy for PTSD in children and adolescents: A preliminary randomized controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry, 46 , 1051 - 1061. Smucker, M. R., Craighead, W. E., Wilcoxon Craighead, L., & Green, B. J. (1986). Normative Journal of Abnormal Child Psychology, 14 , 25 - 39. Southam - Gerow, M. A., & Prinstein, M. J. (2014). Evi dence base updates: The evolution of the evaluation of psychological treatments for children and adolescents. Journal of Clinical Child & Adolescent Psychology, 43 , 1 - 6. doi: 10.1080/15374416.2013.855128 Spence, S. H. (1998). A measure of anxiety symptoms among children. Behaviour Research and Therapy, 36 , 545 - 566. *Springer, C., Misurell, J. R., & Hiller, A. (2012). Game - based cognitive - behavioral therapy (GB - CBT) group program for children who have experienced sexual abuse: A three - 172 month follow - up inve stigation. Journal of Child Sexual Abuse, 21 , 646 - 664. doi: 10.1080/10538712.2012.722592 * Stein, B. D., Jaycox, L. H., Kataoka, S. H., Wong, M., Tu, W., Elliott, M. N., & Fink, A. (2003). A mental health intervention for schoolchildren exposed to violence . The Journal of the American Medical Association, 290 , 603 - 611. Steinberg, A. M., Brymer, M. J., Decker, K. B., & Pynoos, R. S. (2004). The University of California at Los Angeles post - traumatic stress disorder reaction index. Current Psychiatry Reports , 6 , 96 - 100. *Stewart, R. W., Orengo - Aguayo, R. E., Cohen, J. A., Mannarino, A. P., & de Arellano, M. A. (2017). A pilot study of trauma - focused cognitive - behavioral therapy delivered via telehealth technology. Child Maltreatment, 22 , 324 - 333. doi: 10.11 77/1077559517725403 Strawn, J. R., Keeshin, B. R., DelBello, M. P., Geracioti, T. D., & Putnam, F. W. (2010). Psychopharmacologic treatment of posttraumatic stress disorder in children and adolescents: A review. Journal of Clinical Psychiatry, 71, 932 - 941. Substance Abuse and Mental Health Services Administration (2014) Trauma and Guidance for a Trauma - Informed Approach (HHS Publication No. 14 - 4884). Retrieved from http://store.samhsa.gov/sh in/content/SMA14 - 4884/SMA14 - 4884.pdf . Svanborg, P., & Asberg, M. (1994). A new self - rating scale for depression and anxiety states based on the Comprehensive Psychopathological Rating Scale. Acta Psychiatrica Scandinavica, 89, 21 - 28. Taylor, S., & Asmun dson, G. J. G. (2008). Posttraumatic stress disorder: Current concepts and controversies. Psychological Injury and Law, 1, 59 - 74. doi: 10.1007/s12207 - 008 - 9009 - 0 *Thornback, K., & Muller, R. T. (2015). Relationships among emotion regulation and symptoms during trauma - focused CBT for school - aged children. Child Abuse & Neglect, 50 , 182 - 192. doi: 10.1016/j.chiabu.2015.09.011 Tolan, P. H., Guerra, N. G., & Kendall, P . C. (1995). A developmental - ecological perspective on antisocial behavior in children and adolescents: Toward a unified risk and intervention framework. Journal of Consulting and Clinical Psychology, 63 , 579 - 584. Tolin, D. F. (2016). Doing CBT: A comprehensive guide to working with behaviors, thoughts, and emotions . New York, NY: The Guilford Press. United States Department of Health and Human Services, Administration for Children and Families. (2013). Child malt reatment. Retrieved from http://www.acf.hhs.gov/sites/default/files/cb/cm2013.pdf#page=31 . 173 Viechtbauer, W. (2010). Conducting meta - analyses in R with the metafor package. Journal of Statistical Software, 36 , 1 - 48. Walker, J. L., Carey, P. D., Mohr, N., Stein, D. J., & Seedat, S. (2004). Gender differences in the prevalence of childhood sexual abuse and in the development of pediatric PTSD. 111 - 121. doi: 10.1007/s00737 - 003 - 0039 - z Wamser - Nanney, R. & Steinzor, C. E. (2017). Factors related to attrition from trauma - focused cognitive behavioral therapy. Child Abuse & Neglect, 66 , 73 - 83. doi: 10.1016/j.chiabu.2016.11.031 Watson, J. B., & Rayner, R. (1920). Conditioned emotional reac tions. Journal of Experimental Psychology, 3, 1 - 14. Weersing, V. R., Jeffreys, M., Do, M. T., Schwartz, K. T. G., & Bolano, C. (2017). Evidence base update of psychosocial treatments for child and adolescent depression. Journal of Clinical Child & Adoles cent Psychology, 46 , 11 - 43. doi: 10.1080/15374416.2016.1220310 Weersing, V. R., & Weisz, J. R. (2002). Mechanisms of action in youth psychotherapy. Journal of Child Psychology and Psychiatry, 43 , 3 - 29. Weiss, D. S. (2004). The Impact of Event Scale - Revi sed. In J. P. Wilson & T. M. Keane (Eds.), Assessing psychological trauma and PTSD (2nd ed., pp. 168 189). New York, NY: Guilford Press. doi:10.1007/978 - 0 - 387 - 70990 - 1_10 Wolfe, V. V., Gentile, C., Michienzi, T., Sas, L., & Wolfe, D. A. (1991). The childre of traumatic events scale: A measure of post - sexual - abuse PTSD symptoms. Behavioral Assessment , 13 , 359 383. *Wolmer, L., disaster with teacher - based intervention: A controlled study. Journal of the American Academy of Child & Adolescent Psychiatry, 50 , 340 - 348. doi: 10.1016/j.jaac.2011.01.002 * Wolmer, L., Hamiel, D., Slone, M., Faians, M., Picker, M., Adiv, T., & Laor, N. (2013). Post - traumatic reaction of Israeli Jewish and Arab children exposed to rocket attacks before and after teacher - delivered intervention. Israel Journal of Psychiatry and Related Sciences, 50 , 165 - 172. *Wolmer, L., Laor, N., & Yazgan, Y. (2003). School reactivation programs after disaster: Could teachers serve as clinical mediators? Child and Adolescent Psychiatric Clinics of North America, 12 , 363 - 381. doi: 10.1016/S1056 - 4993(02)00104 - 9 Woodbridge, M. W., Sumi, W. C., Thornton, S. P., Fabrikant, N., Rouspil, K. M., . . . & Kataoka, S. H. (2016). Screening for trauma in early adolescence: Findings from a 174 diverse school district. School Mental Health, 8 , 89 - 105. doi: 10.1007/s12310 - 015 - 9169 - 5 Woods - Jaeger, B. A., Kava, C. M., Akiba, C. F., Lucid, L., Dorsey, S. (2017). The art and skill of delivering culturally responsive trauma - focused cognitive behavioral therapy in Tanzania and Kenya. Psychological Trauma: Theory, Research, Practice, and Policy, 9, 230 - 238. doi: 10.1037/tra0000170 Yasui, M., & Dishion, T. J. (2007). The ethnic context of child and adolescent problem behavior: Implications for child and family interventions. Clinical Child and Family Psychology, 10 (2), 137 - 179. doi: 10.1007/s10567 - 007 - 0021 - 9 Zinzow, H. M., Ruggiero, K. J., 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. Journal of Child Psychology and Psychiatry and Allied Disciplines, 50 , 441 - 450. doi: 10.111 1/j.1469 - 7610.2008.02004.x