ABSTRACT 
  
Title of Document: EXPLORING THE EFFECTS OF ADVERSE CHILDHOOD 
EXPERIENCES AND SOCIAL INTEGRATION ON 
ADOLESCENT SUICIDAL IDEATION: A SECONDARY 
ANALYSIS OF THE NATIONAL COMORBIDITY SURVEY 
REPLICATION ADOLESCENT SUPPLEMENT (NCS-A) 
  
Kathleen Elaine Washington, Doctor of Philosophy, 
2018 
  
Directed By: Associate Professor, Sharon M. Desmond 
Department of Behavioral and Community Health 
 
 
Exposure to adverse childhood experiences is common among youth. Numerous 
studies have identified cumulative exposures to adversity during childhood as a serious 
public health issue. Studies have found a significant dose-response association between 
cumulative adversities and risk for negative behavioral and mental health outcomes in 
adults, and developmental delays and internalizing and externalizing disorders in 
children. Efforts to expand existing knowledge about childhood adversities and the 
factors that are protective has been a focus of recent efforts. 
In this dissertation, two studies were conducted to assess the relationship of 
adverse childhood experiences and social integration on risk for adolescent suicidal 
ideation. Study 1 focused on assessing the predictability of a 20-item childhood adversity 
assessment versus a 9-item assessment for risk of suicidal ideation. Results from this 
study found that additional items in the expanded assessment were predictive of suicidal 
ideation and increasing exposures significantly increased risk for suicidal ideation. 
However, the 20-item assessment was only slightly more predictive of suicidal ideation 
compared to the 9-item assessment. 
Study 2 examined the protective effect of family, school, religious/spiritual, peer 
and teacher integration as a protective factor of the relationship of cumulative childhood 
adversity and adolescent suicidal ideation. Results indicate family, school and 
religious/spiritual integration reduce risk for suicidal ideation for adolescents reporting 1- 
3 adversities. Only family integration significantly reduced risk for suicidal ideation for 
youth reporting 4-6 adversities. None of the five social integration factors reduced 
suicidal ideation risk for youth reporting 7 or more adversities. Further, peer and teacher 
integration were not found to be a protective influence across any of the adverse 
childhood experiences risk groups. 
Future public health research must focus on identifying those childhood 
experiences that may increase risk for poor behavioral and mental health outcomes. 
Current priorities should also focus on youth who have experienced higher levels of 
adversity so that more appropriate prevention and intervention programs and policies can 
be developed. Finally, given the impact of childhood adversities on health and wellbeing 
over the life-course, more efforts are needed to reduce or eliminate childhood adversities. 
 
 
 
 
EXPLORING THE EFFECTS OF ADVERSE CHILDHOOD EXPERIENCES AND SOCIAL 
INTEGRATION ON ADOLESCENT SUICIDAL IDEATION: A SECONDARY ANALYSIS OF 
THE NATIONAL COMORBIDITY SURVEY REPLICATION ADOLESCENT SUPPLEMENT 
(NCS-A) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
by 
 
 
Kathleen Elaine Washington 
 
 
 
 
 
Dissertation submitted to the Faculty of the Graduate School of the 
University of Maryland, College Park, in partial fulfillment 
of the requirements for the degree of 
Doctor of Philosophy 
2018 
 
 
 
 
Advisory Committee 
Dr. Sharon M. Desmond, Chairperson 
Dr. Robert H. Feldman 
Dr. Kerry M. Green 
Dr. Gregory R. Hancock, Dean’s Representative 
Dr. Natalie Slopen 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
© Copyright by 
Kathleen Elaine Washington 
2018 
Dedication 
 
 
This dissertation is dedicated to the many girls and young women who I have had 
the honor to meet. Graciously you each have shared your lived experiences that too often 
have included tremendous amounts adversities. Your bravery, will to find the bright spot 
in each day, and sheer brilliance is my inspiration. 
 ii 
Acknowledgements 
 
The completion of my doctoral program and dissertation would not have been 
possible without the help of numerous faculty members, colleagues, friends and family – 
both extended and non-extended. 
I would like to first send a heartfelt thank you to Dr. Robert Feldman. As my 
academic advisor you saw me through some of the hardest moments academically and 
through some of my greatest challenges personally. Not only did you stand by me as my 
advisor but also as a friend. I cannot begin to express to you how much I appreciate all 
that you have done. I would not be here today if had not been for you. Thank you! 
To Dr. Sharon Desmond, my dissertation committee chair, thank you so much for 
all of your encouragement and for hanging in there with me through what turned into 
quite a long dissertation process. There were many moments when I thought I would not 
make it to this day, but you were always there with a smile, a well needed pep talk, and 
most importantly a belief in me and my passion for improving the lives of the next 
generation that brought me to this program so many year ago. Your sense of humor was 
invaluable. Those moments of sitting in your office and laughing about life and family 
helped to ground me throughout this process. It reminded me that it was okay to simply 
be human. I will miss those moments. I am forever grateful to you. 
I would like to thank Dr. Natalie Slopen and Dr. Kerry Green for your support, 
encouragement and enthusiasm for my project. You both are truly inspirational, and I 
aspire to be as brilliant and knowledgeable in the future. And Dr. Gregory Hancock, I am 
truly thankful for all that you have done over the past few years. You truly are amazing! I 
can honestly say that if it had not been for your guidance and dedication to the work and 
your students I would not be here today. I am forever grateful to you for all of your 
assistance and kindness through this process. 
To my super smart and amazing academic colleagues Dr. Delya Banda, Dr. 
Katrina Debman, and Dr. Chandria Jones – you ALL are the best! You kept me laughing 
through the early years of this process and through my struggles. A big THANK YOU to 
all of my work colleagues, especially Jeannette Pai-Espinosa! We’ve known each other 
for many years and you have been there as a friend during some of my toughest moments 
in life. Thank you for your dedication to the work, the next generation of girls and 
women, and for insisting that I take the time needed to complete this process - especially 
when I simply would not do it on my own. You all are the best! 
Finally, I would like to thank (in alphabetical order) my friends Kimberley Bush, 
Robin Campbell, Dr. Chris DeGannes, Marci Edwards, Koree Furnari, Lisa Joyner, 
Samantha Lopez, Dr. Nicole Mason, Malika Saada Saar, Janelle Smith, Joanne Smith, 
LaToya Stubbs, Lauren Tinsley, Kimberly Washington, (late) Dr. Monica Wells Kisura, 
Joy Zarembka, my very patient and amazing partner Deirdre Ottengren, and of course my 
parents, Barbara and Edward Washington, and my big sister Deidre Washington for 
always being in my corner through the good times and the bad. Love you all to pieces. 
 iii 
Table of Contents 
 
 
 
 
ABSTRACT ....................................................................................................................... 0 
Dedication .......................................................................................................................... ii 
Acknowledgements .......................................................................................................... iii 
Table of Contents ............................................................................................................. iv 
List of Tables .................................................................................................................... vi 
List of Figures .................................................................................................................. vii 
Chapter 1: Introduction .................................................................................................. 1 
Introduction ............................................................................................................................... 1 
Problem Statement .................................................................................................................... 2 
Measuring Youth Exposure to Adversity and Stressful Events ............................................... 2 
Prevalence of Adversity and Stressful Events Among Youth .................................................. 3 
The Impact of Exposure to Adversity and Stressful Experiences ............................................ 7 
Potential Protective Factors of Childhood Adversity and Adolescent Health.......................... 8 
Justification for Current Study ................................................................................................ 9 
Research Questions and Study Aims ..................................................................................... 11 
Definition of Terms .................................................................................................................. 12 
Chapter 2: Literature Review ........................................................................................ 14 
Introduction ............................................................................................................................. 14 
An Ecobiodevelopmental Framework to Understanding Stress ............................................ 16 
Defining Adverse Childhood Experiences and Stressors ...................................................... 20 
The Impact of Cumulative Adversity and Stress on Health ................................................ 23 
Adversity and Internalizing and Externalizing Behaviors Among Youth............................. 26 
Limitations of the Adverse Childhood Experience Study ..................................................... 27 
Other Potential Childhood Adversities and Stressors ........................................................... 30 
Potential Protective Factors of Childhood Adversity and Health ....................................... 33 
Durkheim’s Theory of Social Integration as a Protective Factor of Health .......................... 33 
Studies of Durkheim’s Theory of Social Integration and Health .......................................... 37 
Conclusion ................................................................................................................................ 38 
Chapter 3: ........................................................................................................................ 40 
Study #1: Adverse Childhood Experiences and Adolescent Suicidal Ideation .......... 40 
Abstract .................................................................................................................................... 40 
Introduction ............................................................................................................................. 42 
Methods .................................................................................................................................... 46 
Discussion ................................................................................................................................. 60 
Chapter 4 ......................................................................................................................... 65 
Study #2: Adolescent Social Integration as a Protective Factor of Cumulative 
Adverse Childhood Experiences and Adolescent Suicidal Ideation ........................... 65 
Abstract .................................................................................................................................... 65 
 iv 
Introduction ............................................................................................................................. 67 
Methods .................................................................................................................................... 74 
Discussion ................................................................................................................................. 92 
Chapter 5: Summary ................................................................................................... 100 
Appendices .................................................................................................................... 108 
Appendix A:  Methods .......................................................................................................... 109 
The National Comorbidity Survey Replication Adolescent Supplement Study .............. 109 
NCS-A Eligibility and Sampling Frame ............................................................................. 110 
NCS-A Adolescent and Parent Supplemental Questionnaires ............................................ 111 
NCS-A Limitations ............................................................................................................. 113 
Dissertation Study Design ..................................................................................................... 113 
Conceptual Model ............................................................................................................... 113 
Study Variables ................................................................................................................... 114 
Data Analysis Plan ................................................................................................................ 121 
Sampling Weights ............................................................................................................... 121 
Missing Data and Data Assumption .................................................................................... 121 
Scale Reliability, Descriptive and Bivariate Analyses ........................................................ 122 
Multivariate Logistic Regression and Path Analysis .......................................................... 122 
Power Analysis ................................................................................................................... 124 
Human Subjects Concerns ................................................................................................... 125 
Appendix B: Principal Components Analysis Results ....................................................... 126 
Religious Integration Scale Results .................................................................................... 127 
Peer Integration Scale Results............................................................................................. 127 
School Integration Scale Results ........................................................................................ 128 
Teacher Integration Scale Results ....................................................................................... 128 
Family Integration Scale Results......................................................................................... 129 
Appendix C: Institutional Review Board Approval Letters .............................................. 130 
References ..................................................................................................................... 134 
v 
 
List of Tables 
 
 
Table 1.1: Sample Questions and Percentage of Youth Reporting 
Experiencing an Exposure in the Past Year: 2013 Youth Risk 
Behavioral Survey ........................................................................................................6 
Table 3.1: Childhood Adversity Items and Related NCS-A Questions ...................... 51 
Table 3.2: Demographic Characteristics, Prevalence of Original 
ACEs by Group ........................................................................................................... 55 
Table 3.3: Prevalence and Risk of Suicidal Thoughts by ACE Item ........................... 57 
Table 3.4: Risk Estimate for Suicidality by Original and Expanded ACE items…….58 
Table 3.5: Estimated Risk for Suicidality by ACE Score ........................................... 60 
Table 4.1: Childhood Adversity Items and Related NCS-A Questions ...................... 79 
Table 4.2: Social Integration Items ............................................................................. 83 
Table 4.3: Participant Characteristics ......................................................................... 85 
Table 4.4: Prevalence of ACEs and Suicidal Ideation by ACE Group ....................... 87 
Table 4.5: Estimated Risk for Suicidal Ideation by ACE Group ................................ 87 
Table 4.6: Path Analysis Results of Social Integration on Suicidal Ideation by 
ACE Group ................................................................................................................. 91 
 vi 
List of Figures 
Figure 2.1: Ecobiodevelopmental Framework ............................................................ 19 
Figure 4.1: ACE Multigroup Path Analysis Results for Influence of 
Social Integration on Suicidal Ideation................................................................................... 90 
 vii 
Chapter 1: Introduction 
 
Introduction 
 
Exposure to cumulative interpersonal violence and other stress inducing adverse 
life events is a serious public health issue common to many children and adolescents 
living in the United States. This silent epidemic among youth often goes unidentified and 
untreated with devastating short and long-term consequences. Exposure to interpersonal 
violence and other stressful life events during childhood has been associated with 
numerous negative behavioral, mental and physical health, and social outcomes over the 
life course (Fang, Brown, Florence, & Mercy, 2012; Felitti et al., 1998). Further, 
exposures to interpersonal violence and other life stressors during childhood have been 
linked to financial costs to society as a whole. According to the Centers for Disease 
Control and Prevention (CDC), in 2012 the cost associated with just one year of 
confirmed cases of child maltreatment (n = 580,740) in the form of physical, sexual, 
psychological, or negligent abuse in the U.S. had an estimated total lifetime financial cost 
of approximately $124 billion (Fang et al., 2012). Cost estimates for the 1,740 fatal and 
579,000 non-fatal confirmed child maltreatment cases ranged between $8,000 to 
$1,200,000 in costs associated with childhood and adult medical care, productivity loss, 
child welfare, criminal justice, and special education services for this population (Fang et 
al., 2012). 
While much research has been conducted to better understand the prevalence 
(Child Trends, 2013; Finkelhor, Turner, Shattuck, & Hamby, 2015; U.S. Department of 
Health and Human Services, Office of Disease Prevention and Health Promotion, 2015) 
and effects of these experiences on behavior, health and social outcomes (Anda et al., 
 1 
2007, 2008; Bethell, Newacheck, Hawes, & Halfon, 2014; Chapman et al., 2004; Dube et 
al., 2003, 2006, 2009; Felitti et al., 1998; 2001; Kalmakis & Chandler, 2015; Kelly-irving 
et al., 2013), early identification and treatment of childhood exposure to these 
experiences among youth must continue to be a public health priority. As noted in the 
literature, there are numerous exposures that may produce stress responses among youth 
that must be more fully assessed, not all populations experience stressful life events at the 
same rate, and individual responses to these events vary depending on the presence of 
risk and protective factors such as family and peer social support and individual coping 
styles that may attenuate the effects of these exposures on youth outcomes (Rosenthal, 
Wilson, & Futch, 2009). 
This dissertation seeks to expand existing knowledge about the effects of 
adversity and stressful events during childhood on behavioral and mental health 
outcomes during adolescence. Specifically, in this secondary data analysis of the 
National Comorbidity Survey Replication Adolescent Supplement (NCS-A), the 
researcher examines a broader range of exposures as potential stressors during childhood, 
the role of social integration as a protective factor of suicidal ideation, and how these 
relationships vary across different levels of cumulative childhood adversity. 
Problem Statement 
 
Measuring Youth Exposure to Adversity and Stressful Events 
 
It is difficult to know the true scope of adversity experienced by youth. Estimates 
of prevalence rates of events such as interpersonal violence alone vary based on a host of 
factors such as how interpersonal violence is defined, the methods used to collect data, 
policies regarding reporting requirements, and how abuse, neglect and stressful life 
 2 
experiences are documented and classified (Voisin, 2007). For example, some 
researchers may assess exposure with a broadly worded question such as “Have you ever 
been physically abused?” while others may assess the same concept with more specific 
language such as “Have you ever been hit, slapped, or pushed by a boyfriend/girlfriend?” 
Further, nearly half of all victimizations among youth are not reported and therefore go 
undocumented, resulting in an underestimation of the pervasiveness of the problem 
(Office of Juvenile Justice and Delinquency Prevention, 2000). However, while it is 
difficult to estimate the prevalence of these types of exposures among our youth, a 
number of annual reports and nationally representative studies provide a glimpse into the 
nature of this epidemic in the U.S. (Finkelhor et al., 2015; Kann et al., 2014; U.S. 
Department of Health & Human Services, Administration for Children and Families, 
Administration on & Children, Youth and Families, Children’s Bureau, 2016). 
Prevalence of Adversity and Stressful Events Among Youth 
 
Every year, the U.S. Department of Health and Human Services (DHHS) 
publishes an annual national report on referrals and cases of child maltreatment (U.S. 
Department of Health & Human Services, Administration for Children and Families, 
Administration on & Children, Youth and Families, Children’s Bureau., 2016). 
According to the DHHS Office of Administration on Children and Families (2016), 
approximately 3 million referrals for child maltreatment are made annually to state Child 
Protective Service (CPS) agencies. Of these referrals, in 2014 approximately 702,000 
cases of child maltreatment (9.4 per 1,000) were substantiated (U.S. Department of 
Health & Human Services, Administration for Children and Families, Administration on 
& Children, Youth and Families, Children’s Bureau., 2016). The vast majority of these 
 3 
documented CPS cases were for neglect (75%) with the remaining cases being for 
physical abuse (17%), sexual abuse (8%), and other forms of maltreatment such as threats 
of abuse, exposure to parental drug and alcohol abuse, and lack of parental supervision 
(10%) (U.S. Department of Health & Human Services, Administration for Children and 
Families, Administration on & Children, Youth and Families, Children’s Bureau., 2016). 
DHHS also reports that boys and girls are almost evenly represented among 
documented CPS cases and exposure to child maltreatment is reported to occur among 
youth from the point of infancy throughout adolescence or until they reach the age of 18 
when they are no longer considered a child. Most often, the perpetrators identified in 
CPS cases are individuals most trusted and proximal to the child (i.e., one or both parents 
and acquaintances), and infants, under 1 year of age were reported to be the most 
vulnerable of all age groups (i.e., rate of 24.4 per 1000 compared to 12.3 or less per 1000 
for children 1 year of age and older). Finally, race/ethnic disparities exist among 
documented CPS cases. In 2014, African American children had the highest rates of 
victimization (15.3 per 1,000 children) compared to American Indian/Alaska Natives 
(13.4), Multiracial youth (10.6), Pacific Islanders (8.6), Hispanics (8.8), Whites (8.4), and 
Asian youth (1.7)(U.S. Department of Health & Human Services, Administration for 
Children and Families, Administration on & Children, Youth and Families, Children’s 
Bureau., 2016). 
The Youth Risk Behavior Surveillance (YRBS) study is an annual national survey 
of students in grades 9-12. This annual study provides a snapshot of adolescent health 
indicators nationally, and is one of the primary sources of data used to measure progress 
towards the DHHS Healthy People objectives among adolescents (Kann et al., 2014). 
 4 
Based on self-report information, this annual survey also provides a glimpse of the 
pervasive nature of violence and victimization among adolescents (Kann et al., 2014). 
As noted in Table 1.1 between 3.1% and 24.7% of study participants reported 
being exposed to some form of violence in the past year. For example, nearly one quarter 
of participants reported being in at least one physical fight (24.7%), nearly 20% reported 
being electronically bullied, and approximately 10% of all youth who were dating at the 
time of the survey stated they had either been physically or sexually assaulted by 
someone they were dating or involved with intimately (Kann et al., 2014). The YRBS 
also documented gender differences for certain categories of interpersonal violence 
exposure. Females were more likely than male students to report not going to school for 
safety reasons, being electronically bullied, being bullied on school property, being 
forced to have sexual intercourse, and experiencing physical and sexual dating violence 
(Kann et al., 2014). 
Additional studies have attempted to quantify the prevalence of interpersonal 
violence and abuse among youth. Among the most cited, is Finkelhor, Turner, Ormrod, 
& Hamby’s 2009 nationally representative study (N = 4,549), where they concluded 
approximately 60% of all youth experience at least one form of direct or indirect violence 
annually (e.g., physical assaults, robbery and other property crimes, child maltreatment, 
sexual victimization, or witnessing violence in their communities or families) (Finkelhor, 
Ormrod, & Turner, 2009; Turner, Finkelhor, & Ormrod, 2010). Further, Finkelhor and 
colleagues (2009) concluded victimization among youth often is not an isolated 
occurrence but may occur multiple times and in varying forms. For example, researchers 
found that nearly 40% of participating youth had been exposed to 2 or more direct forms 
 5 
of violence, slightly more than 10% had been exposed to 5 or more direct victimizations, 
and nearly 3% had experienced 10 or more exposures to violence during the prior year. 
 
Table 1.1: Sample Questions and Percentage of Youth Reporting 
Experiencing an Exposure in the Past Year: 2013 Youth Risk Behavioral 
Survey 
Item Percentage 
Threatened or injured with a weapon 6.9% 
Physical fight one or more times 24.7% 
Sustained an injury requiring medical care due to a physical 3.1% 
fight 
Electronically bullied via emails, chat rooms, instant 14.8% 
messaging, websites, or texting; 
Reported being bullied on school property 19.6% 
Reported ever being physically forced to have sexual 7.3% 
intercourse when they did not want to 
Dating Violence - reported being hit, slammed into something, 10.3% 
or injured with an object or weapon on purpose 
Dating violence -reported being kissed, touched, or physically 10.4% 
forced to have sexual intercourse when they did not want to 
 
 
Exposure to adversity or stressful events is not restricted to direct experiences of 
violence among youth. Indirect exposure (i.e., witnessing or hearing about violence) 
within the proximal and distal environments of children is also very common, and has 
been associated with numerous negative outcomes among youth (Stein, Jaycox, Kataoka, 
Rhodes, & Vestal, 2003). For example, a 2006 study estimated that more than 15 million 
children reside in families in which partner violence has occurred within the past year, 
with approximately 7 million children living in families in which severe partner violence 
is present (Herman-Smith, 2013; McDonald, Jouriles, Ramisetty-Mikler, Caetano, & 
Green, 2006). Another study of 120 African American male youth living in New York 
City found that 70% had been indirectly exposed to violence by hearing of a serious 
injury to someone close to them (Voisin, 2003). 
 6 
Further, there are a number of indirect exposures to stress inducing experiences 
that have been found to impact child wellbeing that are often overlooked in studies of 
cumulative childhood adversity. For example, the impact of homelessness, food scarcity, 
living in a single parent household, and being involved in the child welfare system 
through foster care are all potential stress inducing exposures that have been found to 
have negative impacts on child health (Alaimo, Briefel, Frongillo Jr, & Olson, 1998; 
Barth et al., 2007; Briggs et al., 2013; Cook et al., 2013; Edidin, Ganim, Hunter, & 
Karnik, 2012; Fawley-King & Snowden, 2012; Laraia, 2013; Leung, Epel, Willett, 
Rimm, & Laraia, 2015; Seligman, Laraia, & Kushel, 2010; Weinreb et al., 2002). Not 
only are these types of exposures widespread, they most often have a disproportionate 
impact on low-income and racial/ethnic minority youth (Voisin, 2007). 
The Impact of Exposure to Adversity and Stressful Experiences 
 
Recent advances in the fields of biological, behavioral, and social sciences have 
expanded our understanding of the impact of adversity and stress on short and long term 
health (Evans & Kim, 2013; S. B. Johnson, Riley, Granger, & Riis, 2013; Kolk, 
McFarlane, Weisaeth, & Greenberg, 1997; Middlebrooks & Audage, 2008; Shonkoff et 
al., 2012; van Der Kolk, McFarlane, & Weisæth, 1996; van Der Kolk & Saporta, 1991; 
Wadsworth, 2015). These advances have identified the physiological responses that act as 
protective mechanisms allowing humans to respond to and adjust to naturally occurring 
stressful events (Shonkoff et al., 2012). However, studies have also found that when 
stress becomes overwhelming, intense, and prolonged, the cumulative effect of exposures 
without appropriate supports can compromise the natural human coping response 
(Middlebrooks & Audage, 2008). What is typically considered “positive” or “tolerable” 
 7 
stress, becomes what has been termed by the National Scientific Council on the 
Developing Child as “toxic stress” (National Scientific Council on the Developing Child, 
2012). When left untreated, toxic stress has been found to disrupt normal brain 
development, compromise normal functioning of biological systems, and has been 
associated with a number of long-term physiological and mental health problems (Kolk et 
al., 1997; Middlebrooks & Audage, 2008; Shonkoff et al., 2012; van Der Kolk et al., 
1996; Wadsworth, 2015). 
Potential Protective Factors of Childhood Adversity and Adolescent Health 
 
While there is much evidence supporting the association of childhood adversity 
with poor behavioral and health outcomes, individual responses to these exposures often 
vary. While some individuals may demonstrate resiliency, others may exhibit a range of 
maladaptive responses to the same or similar experiences (Lazarus & Cohen, 1977). 
Using the “Ecobiodevelopmental (EBD) Framework” as a guide to understanding the 
impact of stressful events on child development, it is speculated that these differences 
may be a result of varying environmental and individual factors, and that access to 
necessary supports and resources as well as individual coping styles may provide a better 
understanding of these differences across individuals and populations (Folkman, 1984a; 
Lazarus, 1966; Lazarus & Cohen, 1977; Shonkoff et al., 2012). 
For example, social ties and relationships have been shown to act as “buffers” 
against negative behavior and health outcomes. It is postulated that these relationships or 
interactions provide a number of positive influences such as a sense of belonging and life 
purpose, access to necessary resources or information, and psychological support – all of 
which are believed to reduce maladaptive responses to stressful life situations (Cohen, 
 8 
1988; Cohen, Underwood, Gottlieb, & Fetzer Institute, 2000). Additionally, higher 
ratings of positive family, peer, school, and religious support or connectedness have all 
been associated with lower rates of suicidality, depression, and perceived stress, as well 
as higher scores on psychosocial wellbeing indicators and global life satisfaction scores 
when compared to adolescents with lower ratings on these four domains of social ties and 
relationships (Matlin, Molock, & Tebes, 2011; Rose, Joe, Shields, & Caldwell, 2014; 
Siddall, Huebner, & Jiang, 2013; Stadler, Feifel, Rohrmann, Vermeiren, & Poustka, 
2010; Tajima, Herrenkohl, Moylan, & Derr, 2011; Young, Berenson, Cohen, & Garcia, 
2005). Having a better understanding of how these relationships interact and influence 
behavioral and mental health outcomes of adolescents who have experienced adversity, 
and how these relationships vary across populations may provide necessary information 
for the development of appropriate prevention and treatment services. 
Justification for Current Study 
 
As we continue to strive to reduce the prevalence of adversity experienced by our 
youth, how best to interrupt the devastating effects of these adversities on morbidity and 
mortality across the life-course must remain a priority of public health research. Prior 
research has identified a number of adverse and stressful exposures associated with 
negative short and long-term health outcomes (Felitti et al., 1998). However, current 
research must continue to assess and expand our understanding of the types of events that 
may cause toxic stress among youth, as well as how these exposures may vary among 
diverse populations. As previously noted, there are a number of experiences such as 
homelessness and food scarcity, that disproportionately impact certain populations, yet 
 9 
have not been taken into consideration in prior studies of cumulative childhood adversity 
and child health. 
The role of social relationships and connectedness as a potential protective factor 
may be key in developing effective interventions for youth exposed to adversity and 
stress. Positive social relationships and ties with family, school, peers, and religious 
involvement or spiritual connections have been found to act as protective factors against 
negative behavioral and health outcomes among youth (Matlin et al., 2011; Rose et al., 
2014; Siddall et al., 2013; Stadler et al., 2010; Tajima et al., 2011; Young et al., 2005; 
Zhang, 2015). However, the role of family, peer, school, and religious connections as 
protective coping mechanisms among youth are most often examined individually and 
not as co-occurring protective factors. Further, these mechanisms as protective factors of 
the association of adverse childhood exposures and behavioral and mental health 
outcomes are rarely examined across varying levels of exposures, especially those with 
exposures reaching beyond four to five exposures. 
In conclusion, increased understanding of the pathways of cumulative exposures 
to adversity and other stressors during childhood has numerous public health 
implications. Expanded knowledge among public health professionals would provide the 
necessary groundwork to develop improved methods for identifying high risk populations 
and development of effective prevention and treatment programs that are 
developmentally, culturally, and trauma informed. Finally, further insight into these 
relationships would assist in the implementation of appropriate social and health policies 
that recognize and appropriately address the impact of cumulative violence and stressful 
events during childhood on the developmental, behavioral, and health trajectories of 
 10 
individuals exposed to these events during their lifetime (Dashiff, DiMicco, Myers, & 
Sheppard, 2009). 
Research Questions and Study Aims 
Study Aims and Hypotheses 
The specific aims and hypotheses of this study are as follows: 
 
Aim 1: To determine if an expanded assessment of adverse childhood experiences better 
 
predicts adolescent suicidal ideation. 
 
H.1.a. An expanded assessment of childhood adversity (i.e., 20-item assessment) 
 
will better predict variance in risk for suicidal ideation than a shorter assessment 
of childhood adversity (i.e., 9-item assessment) based on the original ACE 
assessment. 
H.1.b. Adolescents with higher cumulative childhood adversities will 
 
demonstrate greater risk for suicidal ideation compared to those with fewer 
cumulative childhood adversities. 
Aim 2: To determine if social integration (i.e., family, school, religious/spiritual, teacher, 
 
and peer integration) influences the impact of adverse childhood experiences on risk for 
suicidal ideation among adolescents. 
H.2.a Higher perceived family, school, religious/spiritual, teacher, and peer 
 
integration will each decrease risk for suicidal ideation among adolescents. 
 
H.2.b Higher perceived social integration, while attenuated, will decrease risk for 
 
suicidal ideation among youth with a greater number of cumulative childhood 
adversities. 
 11 
Definition of Terms 
 
Adolescence or adolescent: Adolescence is most often defined as the period 
 
between puberty and adulthood in human development and is most associated with the 
“teen years.” For the purpose of this dissertation, adolescence or adolescent will refer to 
youth between the ages of 13 and 18 years. 
Adverse Childhood Experiences: According to Kalmakis and colleagues (2014), 
 
adverse childhood experiences have five key characteristics. An experience is considered 
adverse if it is harmful, chronic, distressing, cumulative and varying in severity. Adverse 
childhood experiences are “operationally defined as childhood events, varying in severity 
and often chronic, occurring in a family or social environment and causing harm or 
distress” (Kalmakis & Chandler, 2014, pg. 1490). The concept is complex and includes 
numerous in-family and social-environmental sources (Kalmakis & Chandler, 2014). 
Complex trauma: According to Courtois & Ford (2009), complex psychological 
 
trauma “represents extreme forms of traumatic stressors due to their nature and timing. In 
addition to often being life-threatening or physically violating, terrifying, or horrifying, 
these experiences are typically chronic rather than one-time or limited, and they 
compromise the individual’s personality development and basic trust in primary 
relationships” (Courtois & Ford, 2009, pg. 14). 
Direct exposure or direct victimization: Direct exposure or direct victimization 
 
are considered “primary” exposures to violence where the child is the intended target of 
the exposure (Stein et al., 2003). 
Indirect exposure or indirect victimization: Indirect exposure or victimization is 
 
most commonly used to describe exposures where the child is witness to violence but is 
 12 
not the intended target. Such exposures include but are not limited to hearing violence 
transpire, hearing or learning about violence, knowing someone who has been victimized, 
and seeing violence on television or the movies (Stein et al., 2003) 
Resiliency: Resiliency is the successful adaptation to negative life events, trauma, 
 
stress, and other forms of risk (Fraser, Richman, & Galinsky, 1999). It is the ability of 
people to achieve a positive outcome in the face of adversity (McEwen, Gray, & Nasca, 
2015). 
Risk Factor: Risk factors are a combination of individual, relational, community, 
 
and societal factors that increase risk for maladaptive behaviors, physical or mental 
disorders, and negative social outcomes. 
Toxic Stress: Toxic stress is defined as the result of “strong, frequent, or 
 
prolonged activation of the body’s stress response systems in the absence of the buffering 
presence of protection of a supportive adult relationship” (Shonkoff et al., 2012, pg. 
e236). 
Protective Factors: Protective factors are a combination of individual, relational, 
 
community, and societal factors that decrease risk for maladaptive behaviors, physical or 
mental disorders, and negative social outcomes. 
Victim or Victimization: A victim is a person harmed, injured, or killed as a result 
 
of a crime, accident, or other event or action. Victimization is the act of making a person 
a victim. 
 13 
Chapter 2: Literature Review 
 
Introduction 
 
Exposure to traumatic events across the lifespan among adults and youth is not 
uncommon. In the United States, more than half of all adults report experiencing at least 
one traumatic event in their lifetime and 60% of youth under the age of 18 report 
experiencing at least one exposure to violence, either directly as a victim or indirectly as 
a witness to a violent act, annually (Briere & Scott, 2014; Finkelhor et al., 2009). Efforts 
to better understand the extent of childhood adversity and to reduce its occurrence 
remains a public health priority. This is evidenced in its inclusion as a Healthy People 
2020 (U.S. Department of Health and Human Services, Office of Disease Prevention and 
Health Promotion, 2015) priority and the inclusion of the original Adverse Childhood 
Experiences index in the Behavioral Risk Factor Surveillance System (BRFSS) annual 
assessment of population health in a growing number of U.S. states, in a number of 
replication studies internationally (i.e., Canada, China, Jordan, Norway, the Philippines, 
and the United Kingdom), and as an addendum to documents published by the World 
Health Organization (Center for Disease Control and Prevention (CDC), 2015b; Centers 
for Disease Control and Prevention (CDC), n.d.). Unfortunately, while our understanding 
of the extent of exposures has grown, efforts to reduce such events within the lives of 
youth remains virtually unchanged. 
In the 2013-14 National Survey of Children’s Exposure to Violence, researchers 
found that exposures to violence, crime, and abuse continue to remain high and have not 
changed significantly in most exposure areas when compared to earlier studies (Finkelhor 
D, Turner HA, Shattuck A, & Hamby SL, 2015). In this nationally representative study of 
 14 
4000 children under the age of 17, 37.3% of participants experienced a physical assault 
(e.g., physical intimidation, relational aggression, internet or cell phone harassment, gang 
and non-sibling peer assault, etc.), 5% experienced a sexual offense (e.g., sexual assault, 
attempted or completed rape, sexual harassment, flashing, internet sex talk, etc.), 15.2% 
experienced maltreatment from a parent or guardian (e.g., physical, sexual or negligent 
abuse, etc.), 27.1% had experienced a property crime (e.g., robbery, vandalism, theft, 
etc.), and 24.5% witnessed or were indirectly exposed to violence (i.e., family assault, 
physical abuse, assault in the community, shooting, etc.) in the prior 12 months to 
completing the survey. Progress towards significant reductions in violent exposures 
among adolescents were only noted for past-year exposure to dating violence and lifetime 
exposure to household theft (Finkelhor et al., 2015). 
While prevention efforts to reduce exposures to stress inducing events among 
children must remain a priority of public health programs and policy efforts, findings 
such as those observed in the National Survey of Children’s Exposure to Violence 
indicate that much work remains. Therefore, identifying how best to address the needs of 
children who are exposed to high levels of stress and trauma must remain a research 
priority. These efforts must include identifying the array of exposures that may cause 
stress and trauma among children, as well as expanding our understanding of the 
interpersonal and environmental factors most salient to increasing resiliency among 
children exposed to traumatic and stressful life events. Further, research efforts must 
continue to identify the extent to which adverse childhood experiences and the factors 
that influence the relationship between childhood trauma and health outcomes vary 
across populations. 
 15 
In this chapter a summary of what is currently known about childhood adversity 
and stress is presented, specifically: what is currently known about the impact of 
cumulative exposures to adverse childhood experiences and stress; how adverse 
childhood experiences are currently defined; and what is currently known about 
adolescent health and the role of social relationships and social ties, especially in relation 
to Durkheim’s theory of social integration. 
An Ecobiodevelopmental Framework to Understanding Stress 
 
The implementation of effective prevention and treatment programs to address the 
impact of childhood stressors requires a framework describing how adversity impacts the 
development of children both physically and mentally. While the study of childhood 
adversity and stress continues to be a growing field of inquiry, new advances in a broad 
spectrum of disciplines (i.e., neuroscience, molecular biology, genomics, developmental 
psychology, epidemiology, sociology, and economics) have expanded our understanding 
about the harms of stressful events during childhood on individual learning, behavior, and 
physical and mental health over the life course (Folkman, 1984b; Lazarus, 1966; Lazarus 
& Cohen, 1977; Shonkoff et al., 2012). 
Exposure to stressful experiences has been found to be both beneficial and 
harmful – with three distinct categories of human stress responses (Shonkoff et al., 2012). 
“Positive” stress responses are physiological states that are characterized as brief, mild to 
moderate in magnitude, and typically includes appropriate and adequate supports (i.e., a 
caring and responsive adult) to assist the child to return to a normal functional state. 
Examples of positive stress inducing events may include stressors such as receiving an 
 16 
immunization, normal frustrations, or anxiety resulting from events such as a child’s first 
day at school (Shonkoff et al., 2012). 
“Tolerated” stress responses occur when the individual is exposed to “non- 
normative” experiences such as a contentious divorce, death of a family member or 
friend, serious illness or injury, natural disasters, or an act of terrorism (Shonkoff et al., 
2012). Again, these experiences are considered tolerable to the extent that protective 
adult relationships facilitate the child’s adaptive coping and sense of control. Finally, 
there is the “toxic” stress response. This stress response is considered the most harmful 
and is characterized by strong, frequent, or prolonged activation of the body’s stress 
response systems. Unlike “positive” or “tolerable” stress responses “toxic stress” is 
characterized by the absence of protective factors such as adequate and appropriate 
support of a caring adult or social systems. Examples of toxic stressors include child 
abuse or neglect, parental substance abuse or mental illness, and witnessing violence 
towards another. 
The “Ecobiodevelopmental (EBD) Framework,” created by the Center on the 
Developing Child at Harvard University, provides a structure for understanding the 
etiology of childhood adversity and stress on health and development, and provides an 
outline for potential areas of intervention and treatment (see Figure 2.1: 
Ecobiodevelopmental Framework) (Shonkoff et al., 2012). The EBD framework suggests 
the consequences of adversity and stress during childhood are determined by interlocking 
relationships of three factors. These factors are biology (i.e., physiological adaptations 
and disruptions), ecology (i.e., social and physical environment), and health disposition 
and human development (i.e., learning, behavior, and physical and mental well-being) 
 17 
(Shonkoff et al., 2012). It is believed that environmental exposures (i.e., ecology) 
beginning as early as the prenatal period trigger and set the stage for current and future 
responses to adversity and stress (i.e., biology), which then influence future behavioral 
and psychological responses (i.e., health and development) throughout childhood and 
over the life-course (Committee on Psychosocial Aspects of Child and Family Health, 
2012; Shonkoff et al., 2012). 
It is currently thought the human body is biologically designed to respond to 
threatening or stress inducing experiences (i.e., “positive” and “tolerated” stress) through 
a series of hormonal responses that act as protective factors (Committee on Psychosocial 
Aspects of Child and Family Health, 2012; Compas, 2006; Shonkoff et al., 2012; 
Wadsworth, 2015). These protective responses to ecological exposures are centered 
within the brain and include a combination of immediate, unconscious biological 
reactions and less reactionary, conscious responses (i.e., appraisal of experiences) 
(Compas, 2006; McEwen, 1993; McEwen et al., 2015). In situations of adversity, threat 
or stress, these biological stress reactions set in motion the “fight or flight” response or a 
series of biological and psychological responses to the event or exposure (Compas, 
2006). 
For example, studies have demonstrated that biological reactions (i.e., the 
immediate and unconscious) to stressors include hormonal responses that activate the 
hypothalamic-pituitary-adrenocortical axis and the sympathetic-adrenomedullary system. 
When activated, these systems increase the levels of stress hormones produced by the 
body such as corticotropin-releasing hormones (CRH), cortisol, norepinephrine, 
 18 
adrenaline, along with a number of other physiological responses such as elevated 
inflammatory cytokines and activation of the para-sympathetic nervous system. 
 
 
Figure 2.1: Ecobiodevelopmental Framework 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
When threats or stressors are minimal and the social and physical environment 
provide necessary protections (e.g., social support), these physiological responses most 
often return to normal homeostasis balance (Shonkoff et al., 2012; Shonkoff, Boyce, & 
McEwen, 2009). As noted by McEwen et. al., (2015), these early childhood experiences 
allow for healthy brain development and adaption to life stressors. They provide the basis 
for “good self-esteem, and a locus of control for effective self-regulation, not only of 
behavior but also of the physiological responses to stressors that are regulated by the 
central and peripheral nervous systems” (McEwen et al., 2015, pg. 2). 
On the other hand, when individuals are exposed to repeated or chronic stress this 
process may become deregulated and may result in harmful physiological conditions 
known as “allostatic load or overloading” and maladaptive psychological coping skills 
 19 
(McEwen, 1993). Studies focused on the biological impact of allostasis have found that 
chronic stressors and activation of the stress system may actually alter the structure of the 
brain, and normal biological and psychological functioning (McEwen et al., 2015; 
McEwen & Wingfield, 2003). For example, studies have found childhood adversity to be 
a significant predictor of physiological changes such as elevated blood pressure, elevated 
hypothalamic pituitary adrenal axis activity, dysregulation of metabolic activity, 
compromised immune function, increases in inflammatory markers, and elevated 
allostatic load increase risk for morbidity and mortality across the lifespan (Araújo et al., 
2009; Berasain et al., 2009; Chen & Miller, 2007; Danese et al., 2008; Evans & Kim, 
2013; Felitti et al., 1998; Middlebrooks & Audage, 2008; Slopen, Fitzmaurice, Williams, 
& Gilman, 2012; Slopen, Kubzansky, McLaughlin, & Koenen, 2013; Slopen, 
McLaughlin, Dunn, & Koenen, 2013). Further, exposure to toxic stress during childhood 
may lead to increased risk for a number of maladaptive psychological and behavioral 
responses such as depression and anxiety, suicidality, avoidance and dissociation, 
aggression, early sexual initiation and risky sexual behaviors, impaired school function 
and memory, and alcohol and tobacco consumption (Chapman et al., 2004; Dube et al., 
2003; Felitti et al., 1998; Middlebrooks & Audage, 2008; Stein et al., 2003). 
Defining Adverse Childhood Experiences and Stressors 
 
Currently, the study of adverse childhood experiences lacks a consistent 
vocabulary and methodological approach to assessing childhood adversity (Kalmakis & 
Chandler, 2014). According to a recent literature review conducted by Kalmakis and 
colleagues, childhood adversity is generally defined by the specific exposures under 
investigation, and these exposures often vary in nature, severity, and frequency across 
 20 
studies. The operationalization of childhood adversity may include single exposures or a 
varying combination of exposures that may or may not include experiences such as 
childhood physical, sexual and emotional abuse; war or terrorist attacks; natural disasters; 
transportation and/or vehicle accidents; witnessing various forms of family and 
community violence; rape and sexual assault; stranger physical assault; intimate partner 
violence; sex trafficking; witnessing the death or suicide of another; loss of a parent or 
significant care giver; persistent poverty; and life threatening events and/or medical 
conditions and other experiences with the potential of creating stress or trauma (Briere & 
Scott, 2014; Courtois & Ford, 2009; Felitti et al., 1998; Kalmakis & Chandler, 2014). 
Further, assessment tools used to operationalize childhood adversity often vary 
across studies (Kalmakis & Chandler, 2014). These assessments differ in wording as well 
as how total exposure to adversity is scored. For example, the ACE index assesses ten 
potential adversities experienced prior to age 18 and produces a cumulative score of 
exposures ranging between 0 and 10 (Felitti et al., 1998). The ACE score indicates a 
lifetime exposure to an experience even if the experience occurred once or multiple times 
over an extended period of time. The Juvenile Victimization Questionnaire assesses a 
wide range of potential victimization exposures (i.e., child maltreatment, crime 
victimizations, and other exposures such as bullying and witnessing violence) using a 34- 
item assessment (Finkelhor, Hamby, Ormrod, & Turner, 2005). Unlike the ACE 
assessment, the JVC is designed to collect more than single lifetime exposures to 
victimization. The JVC collects information such as 12-month and lifetime exposure, 
frequency of the exposure, information on the perpetrator of violence, if an injury 
occurred, and the severity of the injury. The Conflict Tactics Scale-Parent Child Version 
 21 
on the other hand uses an assessment that may include upwards of 39-items and is 
specifically intended to measure psychological and physical maltreatment and neglect of 
children by parents and modes of discipline used within the family (Straus, 1979; Straus, 
Hamby, Finkelhor, Moore, & Runyan, 1998). Finally, the review by Kalmakis and 
Chandler (2014) finds that the concept of childhood adversity is often described using a 
variety of interchangeable terms such as childhood trauma, childhood abuse and 
maltreatment, childhood stressors, childhood adversity, and polyvictimization, further 
obscuring the meaning and characteristics of the concept of adverse childhood 
experiences. 
While a clear-cut definition of the construct “adverse childhood experiences” 
remains elusive, several key elements have been identified that distinguish this concept 
from other determinants of health. First, adverse childhood experiences are events that 
occur specifically during the formative developmental years of childhood, and have a 
unique impact on health outcomes when compared to individuals who are exposed to 
similar experiences during adulthood (Courtois & Ford, 2009; Kalmakis & Chandler, 
2014; Terr, 1991). Second, the sources of childhood adversity are located within the 
social environmental context of the child and are stressors that occur “outside” of the 
individual (Lazarus & Cohen, 1977; Terr, 1991). In other words, adverse childhood 
events are not a result of actions controlled by the child but are a result of events 
controlled by others or factors within the environment of the child. These exposures are 
stressors that may be traumatic in nature and may have the potential to alter ordinary 
coping and defensive mechanisms of the individual (Terr, 1991). Finally, adverse 
childhood experiences are events that can be distinguished as either singe-incident 
 22 
traumas (e.g., a sever accident, a natural disaster, a terrorist attack, a single incident of 
direct abuse or witnessing of violence, etc.) or as repetitive or cumulative exposures to 
traumatic or stressful events (e.g., recurring abuse, domestic violence, community 
violence and crime, etc.). (Courtois & Ford, 2009; Terr, 1991; van Der Kolk et al., 1996). 
While prior research has established single traumatic events as significant predictors of 
health, the study of adverse childhood experiences is most often associated with the study 
of traumatic events that are chronic and/or cumulative in nature (Courtois & Ford, 2009; 
Kalmakis & Chandler, 2014). 
The Impact of Cumulative Adversity and Stress on Health 
 
The cumulative effect of traumatic or stressful experiences during childhood and 
its impact on behavioral, physical, and mental health outcomes is most commonly 
associated with the Adverse Childhood Experiences (ACE) Study (Felitti et al., 1998). 
The ACE study was a collaborative effort of the Centers for Disease Control and 
Prevention (CDC) and Kaiser Permanente that began in the 1990s. The ACE study was 
designed to explore childhood antecedents of health risk behaviors, disability, disease, 
and premature mortality amongst a cohort of over 17,000 adult patients who were 
members of Kaiser Permanente. Using a retrospective assessment of ten stressful 
childhood events (i.e., emotional, physical, and sexual abuse; emotional and physical 
neglect; living in a household where a mother was treated violently; living in a household 
with a substance abuser; living with someone with a mental illness; exposure to the 
separation or divorce of parents; and living in a household where someone had been 
incarcerated) the association of ACEs to the incidence of diseases, use of prescription 
pharmaceuticals, healthcare costs, premature mortality, causes of death, and a host of 
 23 
other behavioral, social, and health outcomes was examined (Felitti et al., 1998; Larkin, 
Shields, & Anda, 2012). 
There were a number of significant findings associated with this original study 
(Larkin et al., 2012; Nurius, Russell, Herting, Hooven, & Thompson, 2009). First, Felitti 
and colleagues (1998) found exposure to adversity or stressful events during childhood 
was not uncommon and often co-occurring (Dong et al., 2004; Felitti et al., 1998). Of a 
sample of 9,508 adults who completed a standardized medical evaluation survey that 
included seven items from the ACE index (i.e., psychological, physical, or sexual abuse, 
violence against mother, or living with a household member who was a substance abuser, 
mental ill or suicidal, or ever imprisoned), more than half responded positively to at least 
one adverse experience and one-fourth reported at least two or more exposures during 
their childhood. 
Second, the literature produced from the original ACE study has consistently 
found a “dose-response” relationship between the number of reported exposures to 
adversity during childhood and increased risk for poor behavioral, mental and physical 
health outcomes, even after controlling for potential confounders (Larkin et al., 2012). 
Findings demonstrated that as the number of childhood exposures increased, so increased 
risk for conditions such as ischemic heart disease; lung cancer and chronic lung disease; 
smoking; alcohol and illicit drug use; obesity; mental health disorders such as depression 
and anxiety; memory disturbances; early initiation of and other sexual risk taking 
behaviors; unintended adult and teen pregnancy; poor-self rated health; headaches; 
impaired job performance; and premature mortality (Anda, 1999; Anda et al., 2004, 2008; 
Anda, Tietjen, Schulman, Felitti, & Croft, 2010; Brown et al., 2007; Chapman et al., 
 24 
2004, 2011; Dietz, 1999; Dong et al., 2004; Dube et al., 2001, 2003, 2006b, 2009b; 
 
Edwards, Anda, Felitti, & Dube, 2004; Edwards, Fivush, Anda, Felitti, & Nordenberg, 
2001; Edwards, Holden, Felitti, & Anda, 2003; Hillis et al., 2004; Hillis, Anda, Felitti, & 
Marchbanks, 2001; Hillis, Anda, Felitti, Nordenberg, & Marchbanks, 2000; Williamson, 
Thompson, Anda, Dietz, & Felitti, 2002). 
Third, findings from the original ACE study also demonstrated a consistent and 
significant increase in risk for poor behavioral, social, and health outcomes among 
individuals with ACE scores greater than or equal to four when compared to individuals 
with ACE scores equal to zero. Felitti and colleagues (1998), found a 4 to 12-fold 
increase in risk for alcoholism, drug abuse, depression, and suicide attempt; a 2 to 4-fold 
increase in smoking, poor self-rated health, fifty or more sexual partners, and sexually 
transmitted disease; and 1.4-1.6 fold increase in physical inactivity and severe obesity. 
Two studies conducted by Anda and colleagues (2007) found that individuals with ACE 
scores equal to or greater than five had a 3-fold increase in rates of psychotropic 
prescription drug use, a 2.6 times risk of prevalent chronic obstructive pulmonary disease 
(COPD), and a 2.0 time risk of COPD-related hospitalizations (Anda et al., 2008) 
compared to individuals with ACE scores equal to zero. Another study on depressive 
disorders found that individuals reporting five or more ACEs had a five-fold increased 
risk for a lifetime history of depressive disorders and a six or more greater risk for a 
recent depressive disorder or episode compared to individuals with ACE scores equal to 
zero (Chapman et al., 2004). 
 25 
Adversity and Internalizing and Externalizing Behaviors Among Youth 
 
While there is overlap, the behavioral and mental health manifestations of 
childhood adversity vary when comparing children to adults. For children, the effects of 
adversity and stress are most often associated with a number of disorders considered key 
indicators of maladaptive development that may impact health over the life course. These 
indicators are often categorized as externalizing and internalizing disorders. 
Externalizing disorders are a classification of problem behaviors that are 
manifested in behaviors where the child is acting out negatively on the external 
environment, and are characterized by problems in emotional and behavioral self-control 
(Liu, 2004). Examples of externalizing disorders include but are not limited to 
oppositional defiant disorder (ODD), conduct disorder (CD), attention- 
deficit/hyperactivity disorder (ADHD), substance use disorders, and a range of eating 
disorders (American Psychiatric Association, 2013). Internalizing disorders on the other 
hand most often refer to problems that more centrally affect the child’s internal 
psychological environment as opposed to their external environment (Liu, 2004). 
Examples of internalizing disorders include but also are not limited to depressive 
disorders and dysthymia, posttraumatic stress disorder (PTSD), and a range of anxiety 
disorders (i.e., separation, generalized, obsessive-compulsive disorders, phobias, and 
panic disorders) (American Psychiatric Association, 2013). It has also been noted, that 
while these internalizing and externalizing behaviors are typically diagnosed as distinct 
disorders, there is significant and substantial co-morbidity (Liu, 2004). 
Exposure to adversity and stress has been linked to a number of internalizing and 
externalizing behaviors among youth such as higher rates of posttraumatic stress 
 26 
disorders, depression, distress, aggression, dissociation, and anxiety disorders (Buka, 
Stichick, Birdthistle, & Earls, 2001; Clarke, 2006; Cloitre et al., 2009; Dunn, Gilman, 
Willett, Slopen, & Molnar, 2012; Gaylord-Harden, Cunningham, & Zelencik, 2011; 
Haller & Chassin, 2012; Hunt, Martens, & Belcher, 2011; Lilly & Valdez, 2012; Luthra 
et al., 2009; McLaughlin et al., 2013; Mohammad, Shapiro, Wainwright, & Carter, 2014; 
Nooner KB et al., 2012; Turner, Finkelhor, & Ormrod, 2006). For example, a 2009 study 
using the original ACE index, found similar graded-risk responses on health indicators 
such as emotional distress (i.e., depression, anxiety, hopelessness and anger), life stress 
(i.e., number of stressful events, effect of stress, and family distress), suicide risk (i.e., 
suicide ideation, suicide exposure, and lethality of prior attempts), and risk behaviors 
(i.e., alcohol, illicit drug use, high risk behaviors, and peer high risk behaviors) – all of 
which are key indicators of adolescent and child health (Nurius et al., 2009). 
Limitations of the Adverse Childhood Experience Study 
 
While much has been learned from the original ACE study, limitations of the 
original study design and assessment instrument have been noted in the literature 
(Cronholm et al., 2015; Finkelhor, Shattuck, Turner, & Hamby, 2013, 2015a; Wade et al., 
2015; Wade, Shea, Rubin, & Wood, 2014).  One such limitation is that the study relied 
on retrospective assessments of events that occurred prior to the age of 18 using a study 
sample of adults with more than 50% of participants over 50 years of age (Felitti et al., 
1998). While this original approach has been validated as a reliable method for assessing 
stressful events that occurred during childhood, it has been noted that assessing adversity 
during childhood may offer improvements to our understanding of the impact of 
adversity to individual health outcomes (Dube, Williamson, Thompson, Felitti, & Anda, 
 27 
2004). As noted by Finkelhor et. al. (2013), assessments conducted during childhood 
would provide a more accurate and comprehensive reporting of stressors due to improved 
recall of events. Further, assessments conducted during childhood would provide a better 
understanding of the causal sequencing of the relationship of these experiences and future 
health outcomes (Finkelhor et al., 2013). 
It has also been noted that the 10-item scale is limited in its assessment of 
adversities experienced during childhood (Cronholm et al., 2015; Finkelhor et al., 2013; 
Finkelhor, Shattuck, et al., 2015; Wade et al., 2015). In a 2013 study, the original ACE 
assessment was expanded using items from the Juvenile Victimization Questionnaire - an 
inventory of childhood victimization (Finkelhor et al., 2013). This study compared 
explained variance in childhood distress using the 10-item ACE assessment and an 
expanded 30-item adversity assessment. Three important findings were noted from this 
study. 
First, when examining the original 10-items only, results found that two 
childhood experiences (i.e., having in incarcerated household member and parental 
separation or divorce) were not significantly associated with increased childhood distress. 
Second, when the additional adversity items were included in the analysis, several items 
on the original ACE assessment were no longer significant predictors of childhood 
distress (i.e., household substance abuse, mother treated violently, and incarcerated 
household member) but 10 of the new childhood stressor items (i.e., peer victimization, 
parents arguing, property victimization, someone close having a bad accident or illness, 
exposure to community violence, no good friends, socioeconomic status, below-average 
grades, someone close died from illness/accident, parent lost their job) were significant 
 28 
predictors of childhood distress (Finkelhor et al., 2013). Finally, the enhanced childhood 
adversity assessment explained a significantly greater amount of the variance in the 
outcome measure of childhood distress (R2 = .34) compared to the original 10-item 
assessment (R2 = .21) (Finkelhor et al., 2013). 
Generalizability of the original ACE study findings has also been questioned 
given that the study population was a predominantly White, middle to upper income, and 
highly educated population (Cronholm et al., 2015). Recent qualitative and quantitative 
studies within predominantly Black, low-income young adult populations have found 
significant differences in types of adversity experienced compared to the original ACE 
study assessment (Cronholm et al., 2015; Wade et al., 2014). The qualitative study (n = 
119) found single-parent homes, family related violence, exposure to community 
violence such as shootings and other crimes, negative adult activities and criminal 
behavior, personal victimization such as child abuse and neglect, economic hardship and 
struggles of parents to make ends meet, and less cited hardships such as bullying by 
peers, loss of friends to death or incarceration, and discrimination were considered the 
most significant stressors or adversities among this study sample (Wade et al., 2014). 
Further, participants did not endorse divorce and separation or living with a caregiver that 
was mentally ill as an adverse experience during their childhood (Wade et al., 2014). 
A quantitative study (n = 1,784) of predominantly White (45.2%) and Black 
(43.6%) adults between the ages of 18-97 (mean age 48.6 years) was also conducted to 
assess the effectiveness of an expanded ACE assessment that included the original ACE 
items as well as items assessing exposure to experiencing racism or discrimination, 
witnessing violence, living in an unsafe neighborhood, experiencing bullying, and having 
 29 
a history of living in foster care (Cronholm et al., 2015). There were a number of 
significant findings from this study that further supports the need for an expanded 
assessment of the original ACE instrument. 
First, when examining outcomes of the original ACE assessment, participants in 
the study reported a greater number of ACE exposures compared to the original ACE 
study sample. Approximately, 21% of participants in this current study reported four or 
more exposures, compared to approximately 13% of the participants in the original 
Kaiser study. Second, half of the respondents in the study experienced at least one 
exposure that was assessed with the expanded ACE survey; 13.4% experienced three or 
more of the expanded ACE items; and 13.9% of the respondents only experienced items 
from the expanded assessment and no exposures from the traditional ACE items – 
indicating they would have gone unrecognized if only the traditional ACE index had been 
used as the primary assessment of childhood adversity (Cronholm et al., 2015). Finally, 
comparison of demographic associations with the original ACE assessment versus the 
expanded ACE assessment found that males, individuals identifying as Black/African 
American, Hispanic/Latino, Asian/Pacific Islander, or other; individuals who were 
divorced or separated; and those living below 150% of the poverty line were between 
1.51 and 5.93 times more likely to have been exposed to one or more of the expanded 
ACE items during childhood (Cronholm et al., 2015). 
Other Potential Childhood Adversities and Stressors 
 
Given these findings further research is warranted to identify other potential 
stressful events during childhood that may impact individual wellbeing and how these 
different stressors may vary across different populations. For example, there is evidence 
 30 
that exposure to marginal-to-severe hunger or food insecurity during the life course is 
significantly associated with higher levels of negative internalizing behaviors, physical 
health problems, poor diets, and chronic disease even after controlling for known 
correlates (Alaimo, Olson, Frongillo Jr, & Briefel, 2001; Cook et al., 2013; Laraia, 2013; 
Weinreb et al., 2002). Further, past studies have found that food scarcity is most 
prevalent among low-income families and may have specific relevance to child behavior 
and mental health among communities living in poverty (Alaimo et al., 1998). 
Individuals with a history of foster care placement have been identified as a high- 
risk population for a number of negative behavioral and mental health outcomes. 
According to the U.S. DHHS Administration on Children and Families, in 2012 
approximately 397,000 children were in foster care (Children’s Bureau, 2014). While 
involvement in the foster care system is most often a result of exposure to violence and 
neglect, other factors inherent to the foster care system have been found to be 
significantly associated with negative outcomes among system-involved children. 
Placement instability is one of many factors noted as having a negative impact on foster 
care youth. Researchers have found that nearly a fifth of children in foster care have had 
four or more placements within a 3-year period (Barth et al., 2007) and multiple 
placements are a significant predictor of increased mental health symptomology, use of 
outpatient mental health services, and future psychiatric hospitalization (Fawley-King & 
Snowden, 2012). 
Homelessness or housing instability has also been found to have negative health 
and social outcomes among youth. According to Child Trends, it was estimated that 
approximately 1.4 million children under the age of 18 who were registered as students 
 31 
during the 2013-14 school year were homeless (Child Trends, 2015). When compared to 
other low-income youth living in stable housing conditions, youth experiencing 
homelessness or housing instability were more likely to demonstrate acute and chronic 
health problems such as infections, respiratory disorders, gastrointestinal problems, and 
other injuries; more likely to encounter food insecurity and scarcity; less likely to access 
necessary medical care; and were less likely to attain a high school diploma or general 
equivalency diploma by age 18 (Briggs et al., 2013). Further, homeless children, 
especially those who are unaccompanied by an adult, were at increased risk for child 
sexual exploitation; to have psychiatric disorders; to engage in substance and alcohol use; 
and to demonstrate developmental delays compared to other youth living in stable 
housing conditions (Edidin et al., 2012). 
Family structure or more specifically single-parent households, has also been 
associated with poor behavioral and mental health outcomes for children. According to 
the U.S. Census Bureau, in 2013 there were approximately 12 million single-parent 
households in the U.S. (U.S. Census Bureau, 2013). Of these households, 83% were 
headed by single-parenting mothers and approximately 35% of all single-parent 
households were living below poverty (U.S. Census Bureau, 2013). A study examining 
lifetime drug use (i.e., marijuana, inhalants, and amphetamines) among adolescents in 8th, 
10th, and 12th grade (n = 37,507) found that children living in single-parent families (i.e., 
both male and female headed households) were significantly more likely to use drugs 
when compared to intact or dual-headed households (Hemovich & Crano, 2009). This 
study also found that girls living in father-only households were significantly more likely 
 32 
to use all three drugs compared to youth residing in mother-only and dual-parent 
households (Hemovich & Crano, 2009). 
Further, analyses of data from the National Survey of America’s Families study (n 
 
= 67,558) of children between the ages of 0 to 17, found significant differences in parent 
reported physical and mental health status of children based on family structure (Ziol- 
Guest & Dunifon, 2014). As with prior studies, children residing in dual-parent 
households were found to fair better than children living in a variety of family structures 
such as single-parent households (i.e., both mother and father), step-parents (i.e., both 
step father and mother), cohabiting adults (i.e., biological parents or only one biological 
parent), and custodial grandparents. Children living in foster families were found to have 
the worst mental and physical health outcomes compared to all other family structures 
studied (Ziol-Guest & Dunifon, 2014). 
Potential Protective Factors of Childhood Adversity and Health 
 
Durkheim’s Theory of Social Integration as a Protective Factor of Health 
 
As previously noted, healthy child development is dependent on adequate and 
appropriate supports and connections within children’s ecological environment. These 
social relationships and ties are believed to affect mental and physical health by having a 
positive influence on individual emotions, cognitions, and behaviors (Berkman, Glass, 
Brissette, & Seeman, 2000; Cohen, 2004; Seeman, 1996). Cohen (1988) proposed two 
models – the “main effect” and “stress-buffering” models - to explain how social ties and 
relationships influence health and behavioral outcomes (Cohen, 1988; Cohen et al., 
2000). 
 33 
The “Main Effect Model” theorizes that individuals engaged in a range of positive 
social relationships or ties are subject to social controls and pressures that assist the 
individual to engage in behaviors that are beneficial to their well-being (Cohen, 2004; 
Cohen et al., 2000). These controls and pressures are believed to influence health 
behaviors (e.g., exercising, eating healthy diets, not smoking, etc.), provide positive 
psychological states (e.g., generalized positive affect, sense of predictability and stability, 
life purpose, sense of belonging and security, and self-worth, etc.), provide multiple 
sources of information that may reduce risk for engaging in harmful behaviors or assist in 
accessing necessary care, treatment and services; and reduce a sense of isolation which 
may be both psychologically and physiologically harmful. 
The “Stress-buffer Model” on the other hand views social relationships or ties as 
a means of preventing harmful responses to stressful events by influencing individual 
perceptions of or actual availability of support when stressful events occur (Cohen, 1988; 
Cohen et al., 2000). Cohen proposes that the belief or perception that existing social 
relationships or ties will provide necessary resources and supports may in itself mitigate 
negative psychological and maladaptive behavioral responses to stressful events and may 
increase an individual’s perceived ability to cope. Further, Cohen proposes that receiving 
actual support in response to a stressful event may impact how one appraises a stressful 
event – this may lead to solutions, reduce the perceived importance of a situation, or 
provide a distraction from the presenting problem. 
Durkheim’s theory of social integration and regulation was one of the first 
theories to examine the impact of social relationships and social ties on human health and 
behavior, and is reflective of the two models proposed by Cohen (Berkman et al., 2000). 
 34 
Specifically, Durkheim examined variations in rates of suicide as a consequence of the 
quality of social relationships or connectedness (i.e., “social facts”) which he referred to 
as social integration and social regulation and risk for suicide (Durkheim, 1951). 
Durkheim theorized the loss of organized and supporting relationships (i.e., family, 
community, and work ties) or social integration and social constraints as a result of 
individuals migrating to more industrialized, less organized communities would have a 
negative impact on their psychological wellbeing, resulting in increased risk for 
suicidality (Cohen et al., 2000). 
According to Durkheim, suicide rates could be predicted based on levels of social 
integration (i.e., social support) and social regulation (i.e., social control). Social 
integration was theorized to provide an interchange of ideas and feelings that form the 
social bonds that attach individuals to common goals and shared social causes. Social 
regulation, expressed through religious, political and moral beliefs, were theorized to 
control individual behaviors that would prevent individuals from acting in opposition to 
the benefit of society. The quantity and intensity of these relationships determine how 
attached individuals are to each other, whether they have common goals, and the extent to 
which natural human desires are regulated through the use of socially accepted authority 
figures to limit individual desires and passions (Durkheim, 1951). 
Durkheim believed that rapid social change and turbulence at the macro-level 
leads to a weakening of social controls and norms, a deregulation of values, beliefs and 
general norms, and creates social conditions that prevent effective integration and 
regulation of the individual into society. He postulated that the more attached or 
integrated the individual is to the social group, the more the individual will conform to 
 35 
group norms in their behaviors, and the individual will strive to maintain the affective ties 
of important relationships or connections. Further, Durkheim’s theory of social 
integration suggests that effective social integration of the individual also provides a 
positive means of social supports and interactions that lead to better psychosocial 
outcomes for the individual, whereas ineffective integration leads to negative outcomes 
or in the case of Durkheim’s investigations, higher rates of suicide (Rose et al., 2014). 
Finally, it is suggested that social integration is the central construct of Durkheim’s 
theoretical framework, and that without strong social integration or bonds, social 
regulation would have limited or no impact on the individual (Thorlindsson & Bjarnason, 
1998). 
Criticisms of Durkheim’s theory of social integration and regulation on rates of 
suicide have primarily focused on the lack of clarification of the role of macro-level (i.e., 
social) and micro-level (i.e., interpersonal) social factors as measures of social integration 
and regulation and their impact on suicide (Thorlindsson & Bjarnason, 1998). Critics 
have sited that Durkheim’s approach often fluctuated unsystematically between social 
and interpersonal factors, and that while factors at the macro level may be anomic (i.e., 
socially unstable, alienated, and disorganized), individuals may be highly integrated and 
regulated at the interpersonal level - mitigating the effects of societal disintegration. 
Finally, critics of Durkheim’s theory highlight the lack of clarity regarding the concepts 
of integration and regulation. In other words, critics were unclear if Durkheim’s theory 
viewed social integration and social regulation as independent, uncorrelated constructs or 
as a series of correlated concepts that made up one construct of social factors impacting 
individual suicidality (Thorlindsson & Bjarnason, 1998). 
 36 
Studies of Durkheim’s Theory of Social Integration and Health 
 
Prior studies focused on Durkheim’s theory of social integration have 
demonstrated a positive association between social integration and health outcomes 
(Berkman & Syme, 1979; Crittenden et al., 2014; Easton & Renner, 2013; House, 
Robbins, & Metzner, 1982; Kuramoto, Wilcox, & Latkin, 2013; Knoester & Haynie, 
2005; Rose et al., 2014; Schoenbach, Kaplan, Fredman, & Kleinbaum, 1986; Tenorio & 
Lo, 2011). One early, better known study examined levels of social integration (i.e., 
summary index reflecting ties with a spouse, close friends and relatives, and participation 
in church and other types of groups) was a 9-year probability study of mortality among a 
cohort of adults (ages 30 to 69 years) who had completed a health survey in 1965 in 
Alameda County (N = 4775) (Berkman & Syme, 1979). After controlling for 
sociodemographic variables, findings from this study showed that lower levels of social 
integration were associated with higher mortality. The findings from this study on 
mortality were replicated in follow up studies in Tecumseh, Michigan as well as Evans 
County, Georgia (House et al., 1982; Schoenbach et al., 1986) 
Thorlindsson and Bjarnason (1998) investigated Durkheim’s theory of social 
integration and regulation by examining “social facts” at the micro-level with a specific 
focus on the role of family integration and parental regulation on youth suicidality. 
Using a nationally representative sample of 4,314 high school students, the relationship 
of family social support (i.e., social integration), parental regulation (i.e., social 
regulation), social anomie, and social norms regarding suicide as expressed by knowing 
others who were thinking about, attempted or committed suicide and their association to 
individual suicidality were examined. In this study, analyses demonstrated that the 
 37 
concepts of social integration and social regulation were distinct constructs, with strong 
family integration or social support having a significant inverse effect on rates of 
suicidality among the sample population. Further, parental regulation was not found to be 
a significant mediator of suicidality amongst this sample population (Thorlindsson & 
Bjarnason, 1998). 
A more recent study examined Durkheim’s theory of social integration on mental 
health outcomes of a nationally representative sample of Black adolescents (n = 1,170) 
between the ages of 13-17 (Rose et al., 2014). Specifically, Rose and colleagues 
examined the relationship of family, school, and religious involvement and religious 
commitment to psychological wellbeing. The study finding supported Durkheim’s theory 
of social integration and concluded that family, school, and religious integration are 
significant predictors of better mental health for Black adolescents. Further, among this 
cohort, school integration emerged as the strongest predictor of both positive and 
negative psychological well-being, and religious involvement was mediated by individual 
religious commitment – indicating that self-directed choice to attend religious services 
and activities actually play a greater beneficial role in adolescent health than rote 
religious involvement (Rose et al., 2014). 
Conclusion 
 
There is much evidence supporting the association of cumulative childhood 
adversity and stress with poor behavioral and mental health outcomes of children. While 
a clear definition does not currently exist and the operationalization of “adversity” varies, 
its associate to internalizing and externalizing disorders is evident. However, further 
research is needed to determine the impact of other potential childhood stressors, and 
 38 
more importantly how these stressors vary across differing populations. As we continue 
to work towards reducing the prevalence of adversity and stress in the lives of youth, the 
development of effective interventions and treatments will require a better understanding 
of the environmental and individual factors that are protective against such exposures. 
 39 
Chapter 3: 
 
Study #1: Adverse Childhood Experiences and Adolescent Suicidal Ideation 
Abstract: 
Background: Exposure to adversity and stressful experiences during childhood is 
 
not uncommon. These experiences have been linked to numerous negative behavioral 
and mental health outcomes throughout the life course, and premature mortality. Suicide, 
which is the second leading cause of death among adolescents, has been associated with a 
variety of interpersonal adversities. While the original Adverse Childhood Experiences 
study has expanded our understanding about the influences of these early events, 
continued exploration of other potential adversities as influencers of adolescent health, 
especially as it relates to suicidal ideation, is needed. 
Methods: Using logistic regression analyses, a secondary analysis of the National 
 
Comorbidity Survey Replication Adolescent Supplement (NCS-A) study was conducted 
to investigate the association of twenty adverse childhood experiences and suicidal 
ideation among a nationally representative cohort of adolescents (n = 10,128) between 
the ages of 13 and 18. 
Results: Exposure to adverse childhood experiences was found to be high among 
 
this study population. Exposure to parental domestic violence, physical abuse, sexual 
abuse, physical neglect, parental mental health conditions, the death or adversities of 
another, having a serious or life-threatening illness, and exposures to other traumatic 
experiences not specified by the instrument were significantly associated with adolescent 
suicidality. Conclusion: Cumulative childhood adversities are significant predictors of 
suicidality among adolescent populations. Given these findings, public health 
 40 
interventions geared towards reducing suicidal behaviors among adolescents should focus 
on early screening and detection of interpersonal adversities. 
 41 
Introduction 
 
Adverse childhood experiences among youth under age 18 are not uncommon and 
considered a serious public health issue. In a 2009 national study investigators found 
nearly 60% of all youth reported exposure to at least one form of direct or indirect 
violence annually (Finkelhor, Ormrod, et al., 2009). In 2014, the Department of Health 
and Human Services, Office of Administration on Children and Families documented 
approximately 3 million referrals for child maltreatment to state Child Protective Services 
(CPS) agencies, with over 700,000 cases being substantiated (U.S. Department of Health 
& Human Services, Administration for Children and Families, Administration on & 
Children, Youth and Families, Children’s Bureau., 2016). Further, the 2014 Youth Risk 
Behavior Surveillance (YRBSS) survey of 9th – 12th graders found between 3.1% and 
24.7% of study participants reported experiencing some form of violence such as being in 
a physical fight, electronically bullied, or experiencing dating violence in the past year 
(Kann et al., 2014). 
While some level of adversity is considered normal and necessary for healthy 
development, many children experience toxic levels of stress and adversity that are often 
co-occurring or cumulative. These experiences often go unidentified and without 
appropriate supports and treatment with far reaching consequences over the life course 
(National Scientific Council on the Developing Child, 2012). Further, exposure to 
violence and stressors result in financial costs to society as a whole. According to the 
Centers for Disease Control and Prevention (CDC), in 2012 the cost associated with just 
one year of confirmed cases of child maltreatment in the form of physical, sexual, 
 42 
psychological, or negligent abuse in the U.S. had an estimated total lifetime financial cost 
of approximately $124 billion (Fang et al., 2012). 
One of the most well-known investigations of cumulative childhood adversity is 
the Adverse Childhood Experiences (ACE) study (Felitti et al., 1998). The ACE study 
was a collaborative effort between Kaiser Permanente and the CDC. The study examined 
the association of cumulative exposure to ten childhood adversities related to abuse, 
neglect, and household dysfunction and long-term behavioral and health outcomes in a 
cohort of more than 1700 adults (Felitti et al., 1998). Results from this study have 
consistently shown graded increased risk for negative behavioral, physical and mental 
health outcomes, and a significant and substantial increase in risk when comparing those 
reporting no experiences of adversity and those reporting four or more exposures. 
For example, outcomes from numerous studies have found a four to five-fold 
increased risk for negative physical health outcomes such as ischemic heart disease, lung 
cancer and chronic lung disease, headaches, and premature death; poor behavioral health 
outcomes such as smoking, alcohol and illicit drug use, obesity, early initiation of and 
other sexual risk taking behaviors, and unintended adult and teen pregnancy; and a 
variety of negative mental health disorders such as depression and anxiety, memory 
disturbances, poor-self rated health, and impaired job performance (Anda, 1999; Anda et 
al., 2004, 2008; Anda, Tietjen, Schulman, Felitti, & Croft, 2010; Brown et al., 2007; 
Chapman et al., 2004, 2011; Dietz, 1999; Dong et al., 2004; Dube et al., 2001, 2003, 
2006b, 2009b; Edwards, Anda, Felitti, & Dube, 2004; Edwards, Fivush, Anda, Felitti, & 
Nordenberg, 2001; Edwards, Holden, Felitti, & Anda, 2003; Hillis et al., 2004; Hillis, 
Anda, Felitti, & Marchbanks, 2001; Hillis, Anda, Felitti, Nordenberg, & Marchbanks, 
 43 
2000; Williamson, Thompson, Anda, Dietz, & Felitti, 2002). Similar studies among 
children exposed to extremely stressful or traumatic experiences have also shown an 
association between adversity and a variety of internalizing and externalizing disorders, 
that often are co-occurring (American Psychiatric Association, 2013; Liu, 2004). 
While much has been learned from the ACE study, limitations of the original 
study design and assessment tool have been noted (Cronholm et al., 2015; Finkelhor, 
Shattuck, Turner, & Hamby, 2013, 2015a; Wade et al., 2015; Wade, Shea, Rubin, & 
Wood, 2014). Lack of generalizability of the original study to other populations is one 
such limitation (Cronholm et al., 2015). Where the original study was retrospective and 
conducted among an adult predominantly non-Hispanic white, middle to upper income, 
and well educated sample, recent studies among predominantly Black, low-income 
populations have found differences in the types of childhood experiences reported as 
significant stressors when compared to the ten items assessed in the original ACE study 
population (Cronholm et al., 2015; Wade et al., 2014). For example, a qualitative study 
conducted in Philadelphia, Pennsylvania (n = 119) found living in a single-parent home, 
exposure to community violence such as shootings and other crimes, negative adult 
activities other than criminal behavior, economic hardship, and peer bullying, loss of 
friends to death or incarceration, and racial discrimination were also considered 
significant stressors or adversities (Wade et al., 2014). 
Others have noted the 10-item scale is limited, and that a more comprehensive 
assessment of childhood adversities may provide a more accurate understanding of the 
breadth of adversities experienced by youth that may influence child and adult health 
(Cronholm et al., 2015; Finkelhor et al., 2013; Finkelhor, Shattuck, et al., 2015; Wade et 
 44 
al., 2015). Finkelhor and colleagues (2013) expanded the original ACE assessment using 
items from the Juvenile Victimization Questionnaire. Findings demonstrated several of 
the original adversities were not significantly associated with reported “distress 
symptoms,” while ten of the newly added adversities were significantly associated with 
“distress symptoms.” Further, the enhanced assessment explained a significantly greater 
amount of the variance (34%) in the outcome measure when compared to the original 10- 
item ACE assessment (21%) (Finkelhor et al., 2013). 
Expanding our understanding about experiences that may be traumatic in nature 
or may cause toxic stress among children is a public health imperative. A more 
comprehensive understanding of such experiences will likely help inform the 
development of effective methods for assessing such childhood stressors early in life and 
may expand our understanding about which adversities have the most significant impact 
on child health and wellbeing. Further, improved understanding will likely contribute to 
the development and implementation of prevention and treatment programs and policies 
focused on reducing the prevalence of childhood exposures found to be most harmful to 
children. Not only would this expanded knowledge have a lasting impact on individual 
health, it would also likely reduce the social and financial burdens associated with these 
harmful experiences. 
This study specifically seeks to expand existing knowledge by examining the 
impact of additional adversities on risk for suicidal ideation (the process of thinking 
about, considering, or planning to commit suicide) among a nationally representative 
sample of adolescents. Suicidal ideation was specifically selected because suicide is the 
second leading cause of death among youth (Center for Disease Control and Prevention 
 45 
(CDC), 2015a). It is estimated 17% of students in grades 9-12 seriously consider 
attempting suicide and 13.6% make a plan about how they would commit suicide (Center 
for Disease Control and Prevention (CDC), 2015c). Further, while the impact of ACEs on 
a variety of youth behavioral and health outcomes have been conducted across a wide 
range of childhood adversities, few studies have specifically focused on suicidal ideation 
(Dube et al., 2001; Goldston et al., 2016; Hardt et al., 2008a; Perez, Jennings, Piquero, & 
Baglivio, 2016). 
Using data from the National Comorbidity Survey Replication Adolescent 
Supplement (NCS-A), this study examined the impact of nine (9) adversities studied in 
the original ACE study and the influence of eleven (11) additional adversities on risk for 
suicidal ideation among a nationally representative sample of adolescents. Using the 
twenty (20) item adversity assessment, overall ACE scores were examined to determine 
if there was a graded impact on suicidal ideation as seen in other studies focused on other 
behavioral and health outcomes. 
Methods 
 
This study analyzes of the National Comorbidity Survey Replication Adolescent 
Supplement (NCS-A) study (Kessler, 2013). Data for the NCS-A were collected between 
February 2001 and January 2004 by the Survey Research Center of the Institute for 
Social Research at the University of Michigan as an add-on to the National Comorbidity 
Survey Replication (NCS-R) study, an investigation of the prevalence and correlates of 
mental health disorders among adults in the US (Kessler, 2013). The NCS-A study was 
considered the first nationally representative study on the prevalence, correlates and 
patterns of service use for DSM-IV mental health disorders among U.S. adolescents 
 46 
(Kessler, Avenevoli, Costello, et al., 2009a). This seminal study was designed to provide 
the groundwork for follow-up studies of risk and protective factors, consequences, and 
early expressions of adult mental health disorders (Kessler et al., 2009a, 2009b; 
Merikangas et al., 2009). 
NCS-A Eligibility and Sampling Frame.  Participation in the NCS-A study was 
 
restricted to English-speaking, non-institutionalized adolescents between the ages of 13 
and 18 (n=10,148) living in the United States. A dual-framed, complex cluster sampling 
methodology was used to draw a sample of adolescents for the study. The first sampling- 
frame included adolescents (n=904) living in households where an adult had participated 
in the National Comorbidity Survey-Replication (NCS-R) study. To supplement the low 
number of interviews completed, a second sampling-frame of adolescents (n=9,244) were 
recruited to the NCS-A study from a population of students attending 320 schools located 
in the same counties in which the NCS-R was conducted (Kessler et al., 2009b). Eligible 
schools included all accredited private and public middle, junior high, and high schools 
within the 62 primary sampling units (PSUs) used in the NCS-R study (Kessler et al., 
2004, 2009b).  A probability sampling methodology proportional to the size of the 
student body was used to select schools. Forty to fifty students were randomly selected 
from each school to participate in the NCS-A study. The overall adolescent response rate 
(i.e., both household and student sample populations) was 75.6%. The majority of non- 
responders (21.3%) refused to participate; the remaining 3.2% were unable to participate 
due to “circumstances” or “non-contact” (Kessler et al., 2009b). Further details regarding 
the sampling frame for the NCS-A study can be found elsewhere (Kessler et al., 2004). 
 47 
NCS-A Adolescent Questionnaire. The NCS-A adolescent questionnaire includes 
 
data on a number of factors considered relevant to adolescent behavior and mental health. 
Areas assessed include individual level factors (e.g. demographics, developmental 
factors, cognitive and academic abilities-achievements, physical health, and stressful life 
events), family level factors (e.g. family structure, stability and adaptability, parenting 
behaviors, parental psychopathology, and family stress), and environmental level factors 
such as school and neighborhood characteristics (Merikangas et al., 2009). Further, a 
modified version of the fully structured Composite International Diagnostic Interview 
(CIDI) of behavioral and mental health disorders was used in the adolescent 
questionnaire to assess for fifteen DSM-IV disorders among participants such as mood, 
anxiety, behavior, and substance abuse disorders (Kessler et al., 2012). 
Procedures. In the presence of a parent or guardian, adolescent interviews were 
 
administered face-to-face in the home of each respondent using a laptop, computer- 
assisted personal interview (CAPI). Trained survey interviewers from the Survey 
Research Center of the Institute of Social Research at the University of Michigan 
administered each adolescent interview. On average, the NCS-A adolescent interviews 
took approximately two and a half hours to complete, with interviews ranging between 69 
to 347 minutes (Kessler et al., 2009b). This variation in interview length was primarily 
due to the number of lifetime disorders assessed for each adolescent (determined based 
on an initial screening tool administered at the beginning of each interview) (Kessler et 
al., 2009b; Merikangas et al., 2009). 
This secondary analysis includes all participants in the NCS-A study (N = 
10,123). Approval for this study was obtained from the University of Maryland, College 
 48 
Park institutional review board and the University of Michigan Inter-University 
Consortium of Political and Social Research (ICPSR) institutional review board. 
Measures 
Adversity Measures. Forty-six (46) variables in the NCS-A were used to create 
 
the twenty (20) childhood adversity items for this study. These items were derived from a 
number of standardized assessment tools selected by the original NCS-A research team 
(Kessler et al., 2009a; Merikangas et al., 2009). Table 3.1 provides a summary of the 
twenty (20) categories of adversity used for this study and the NCS-A questions used to 
assess exposure for each category. Emotional neglect, which was among the 10-items 
assessed in the original ACE study, was not included in this study. Appropriate proxies 
for this item could not be identified in the NCS-A survey instrument. Therefore, only 9 of 
the 10 original ACE items are represented in this study. 
A positive response to any question within an adversity category was coded as “1 
 
= Yes” meaning the participant had been exposed to the adversity category. If all of the 
items within an adversity category received a negative response, the respondent was 
coded as “0 = No” indicating the respondent had not been exposed to that particular 
adversity category. A conservative approach was used to score adversity variables 
measured using a Likert scale. A respondent was coded “1= Yes” if the participant 
reported an event occurred “often or sometimes” and was coded “0 = No” if the 
participant reported “not very often or never.” Further, missing responses to any adversity 
question was coded as “0 = No” for not being exposed to the experience. An overall 
ACEs score was generated for each participant by summing all twenty adversity items to 
create an aggregate score for each participant ranging from 0–20. However, due to small 
 49 
numbers, youth reporting seven or more adversities were coded as “7” to ensure better 
stability during the analyses. Therefore, the final ACE score for each participant ranged 
between 0-7 for this study. 
Dependent Variable. To assess suicidal ideation, participants were asked if they 
 
had ever had this experience - “You seriously thought about committing suicide.” This 
item was used to determine if a participant had at any time in her/his life seriously 
considered suicide. Possible response options were “1 = Yes” and “0 = No.” 
Control Variables. Several variables were examined as potential confounding 
 
variables. These variables included adolescent race/ethnicity (i.e., White, Black, 
Hispanic, or other), age at interview, gender (i.e., male or female), highest maternal or 
paternal education (i.e., less than high school, high school graduate, some college, college 
degree), poverty ratio (1 = “<=1.5”, 2 = “1.5 to 3”, 3 = “3 to 6” and 4 = “ >= 6” times the 
2000 poverty threshold), type of region of residency (i.e., metropolitan [large city], other 
urban area [suburbs, small city, town/village], and rural area), and U.S. citizen status (i.e., 
yes or no). Using a significance level of .05, univariate analysis of each item indicated a 
significant relationship to suicidality, therefore each item was controlled for in the 
analyses. 
 50 
 
Table 3.1: Childhood Adversity Items and Related NCS-A Questions 
Parental domestic violence – “Yes” = 1, “No” = 0; “Often” or Sometimes” = 1, “Not very often” or “Never” = 0 
• Ever witness serious physical fights at home, like when your father beat up your mother? 
• How often did (woman/man) who raised you insult, swear, shout, yell, scream, threaten to hit each other? 
• How often did (woman/man) grab, shove, throw something, slap or hit each other? 
• How often did (woman/man) kick, bit, hit with a fist, beat up, choke, burn or scald, threaten with a knife or gun each other? 
Emotional abuse – “Often” or “Sometimes = 1, “Not very often” or Never” = 0 
• How often did woman/man who raised you insult, swear, shout, yell or scream, or threatened to hit you? 
• How often did sibling insult, swear, shout, yell or scream, or threaten to hit you? 
• How often did someone you were dating insult, swear, shout, yell or scream, or threaten to hit you? 
Physical abuse – “Yes” = 1, “No” = 0; “Often” or “Sometimes” = 1, “Not very often” or “Never” = 0 
• Ever badly beaten up by your parents or people who raised you? 
• Ever badly beaten up by someone you were dating or with whom you were romantically involved? 
• Were you ever badly beaten up by anyone else (other than parent or someone you were dating)? 
• When growing up, how often did woman or man who raised you push, grab or shove, throw something, slap or hit you? 
Kick, bit or hit with a fist, beat up, choke, burn or scald, or threaten with a knife or gun? 
• When growing up, how often did sibling push, grab or shove, throw something, slap or hit you? Kick, bit or hit with a fist, 
beat up, choke, burn or scald, or threaten with a knife or gun? 
Parental divorce – “Yes” = 1, “No” = 0 
• “Biological parents were divorced”. 
Sexual abuse – “Yes” = 1, “No” = 0 
• Have you ever been raped? 
• Have you ever been sexually assaulted or molested (other than rape)? 
• Have you ever been stalked? 
Physical neglect – Responses = “Often” or “Sometimes” = 1, “Not very often” or “Never” = 0 
• How often made to do chores that were too difficult or dangerous for someone your age? 
• How often left alone or unsupervised when you were too young to be alone? 
• How often go without things like clothes, shoes, or school supplies because parent spent money on themselves? 
 51 
 
• How often did parents/caregivers make you go hungry or not prepare regular meals? 
• How often did parents/caregivers ignore or fail to get you medical treatment when sick or hurt? 
Parental mental health – Any question “Yes” = 1, “No” = 0 
• Was the (man/woman) who raised you ever sad or depressed lasting two weeks or more? 
• Was the (man/woman) who raised you ever nervous, edgy, or anxious more than a month? 
• Did the (man/woman) who raised you ever have anxiety attacks? 
• Did the (man/woman) who raised you ever attempt or commit suicide? 
Parent drug/alcohol abuse “Yes” = 1, “No” = 0 
• Did the (man/woman) who raised you ever have a problem with drugs? 
• Did the (man/woman) who raised you ever have a problem with drinking alcohol? 
Parent criminal/arrest history “Yes” = 1, “No” = 0 
• Was the (man/woman) who raised you ever arrested or sent to prison? 
• Was the (man/woman) who raised you ever involved in criminal activities like burglary or selling stolen property? 
Single parent household 
• Participant reported living mostly in a single parent household. 
Foster care involvement 
• Ever in foster care? 
Homelessness – “Yes” = 1, “No” =2 
• Have you ever been homeless? 
Food scarcity – “Yes” = 1, “No” =0 
• In the past 12 months, did you go hungry because there was not enough money for food? 
• In the past 12 months, did you eat less because there was not enough money for food? 
Major disaster – “Yes” = 1, “No” =0 
• Ever involved in a major disaster like a flood, hurricane, fire, bomb explosion, or earthquake? 
Community violence – “Yes” = 1, “No” =0 
• Were you ever mugged, held up, or threatened with a weapon? 
• Did you ever see someone being badly injured or killed, or unexpectedly see a dead body? 
Death or stress of other – “Yes” = 1, “No” =0 
 52 
 
• Anyone close to you die unexpectedly (e.g., killed in accident, murder, committed suicide, or fatal heart attack)? 
• Anyone close to you ever have a stressful or life-threatening experience, like kidnapping, torture or rape? 
War, revolution, political coup, refugee – “Yes” = 1, “No” =0 
• Ever in a place where there was war, revolution, military coup or ongoing terror? 
• Were you ever a refugee? 
Car or other life-threatening accident – “Yes” = 1, “No” =0 
• Ever involved in a very serious or life-threatening car accident? 
• Did you ever have any other very serious or life-threatening accident? 
Serious illness, poison or chemical – “Yes” = 1, “No” =0 
• Ever exposed to poisonous chemical or substance that could cause serious harm? 
• Did you ever have a very serious or life-threatening illness? 
Any other stressor or trauma not mentioned – “Yes” = 1, “No” =0 
• Ever experience other extremely upsetting or life-threatening event that I haven’t asked about yet? 
 53 
Analysis 
The complex sample module of the Statistical Package for the Social Sciences 
(SPSS) version 24 was used for the analyses. Missing data were assessed for each 
variable included in the study. For all predictor variables it was determined that less than 
5% of the data was missing, therefore, list-wise deletion was used for the study analyses. 
Multicollinearity of the study predictor variables was assessed using tolerance and 
variance inflation factors (VIF) statistics and were found to be acceptable. 
Several logistic regression models, controlling for covariates, were conducted to 
assess the relationship of exposure to adversity during childhood and the dependent 
variable “suicidal ideation.” Univariate analyses were conducted to obtain adjusted odds 
ratios (ORs) and 95% confidence intervals (CIs) for each of the twenty adversity items to 
determine if there was a significant relationship to the dependent variable. Next, two full 
regression models were conducted first with the 9-item ACE assessment and then the 20- 
item ACE assessment. Model fit and R2 statistics were examined to determine if 
including all 20 adversity items was a better predictor of risk for suicidal thoughts when 
compared to the model with the 9-item ACE assessment. Finally, logistic modeling was 
used to examine the graded cumulative effect of the 20-item ACEs assessment and 
suicidal ideation. Model results include mean and prevalence estimates, model statistics, 
overall classification, R2 estimates, and odds ratios with 95% confidence intervals. 
Results 
Characteristics of Study Population. Demographic characteristics and prevalence 
 
of the nine original ACEs of the study sample (weighted n = 10,123) is provided in table 
 
3.2. The study sample was nearly evenly divided between males (51.3%) and females 
(48.7%), with a mean age of 15.2 years (S.E. = .063). The majority were non-Hispanic 
 54 
Whites (65.6%) and citizens of the United States (95.1%). Approximately 45% of parents 
and/or guardians of participants had received a high school diploma or less and 35% had 
received a college degree. Slightly more than 84% of participants lived in either a 
metropolitan or other urban community, and more than 60% lived in households with 
incomes three times greater than the 2000 poverty level. For those who responded (n = 
9878) to the question “Have you ever seriously considered suicide?” 12.1% (S.E. = 0.6%) 
responded positively. Of those youth who reported they had thought seriously about 
suicide, 8.7% reported no exposures to adversity, 50.1% reported 1 – 3 exposures, 36.1% 
reported 4 – 6 exposures, and 5.0 % reported 7 – 9 exposures on the 9-item adversity 
scale. 
Table 3.2: Demographic Characteristics, Prevalence of Original ACEs by Group1 
   Original ACE Score by Group 
% N 0 1 – 3 4 – 6 7 – 9 
Age, mean = 15.2 (.063)       
 13 15.2% 1652 24.4% 60.9% 13.5% 1.2% 
 14 21.0% 2218 24.6% 60.1% 13.9% 1.4% 
 15 20.5% 1887 24.3% 59.3% 15.5% 0.9% 
 16 21.0% 2010 20.3% 59.5% 19.0% 1.2% 
 17 16.8% 1758 18.5% 59.7% 20.0% 1.8% 
 18 5.4% 598 22.4% 55.5% 20.1% 2.1% 
Gender       
 Female 48.7% 4953 21.7% 58.6% 17.9% 1.7% 
 Male 51.3% 5170 23.1% 60.5% 15.4% 1.0% 
Race/Ethnicity       
 White 65.6% 5634 24.6% 58.5% 15.7% 1.2% 
 Hispanic 15.1% 1953 16.2% 64.9% 17.2% 1.6% 
 Black 14.4% 1914 19.9% 57.9% 20.6% 1.6% 
 Other 5.0% 622 20.6% 62.1% 15.9% 1.3% 
Poverty Ratio (2000)       
 < 1.5 of threshold 17.1% 1716 20.4% 58.6% 19.5% 1.4% 
 <= 3 of threshold 19.3% 2021 21.2% 59.3% 18.1% 1.3% 
 <= 6 of threshold 31.0% 3104 22.4% 59.3% 17.3% 1.1% 
 > 6 of threshold 32.6% 3282 24.0%% 60.5% 13.9% 1.6% 
Urbanicity       
 Metropolitan 47.5% 4508 22.5% 59.4% 17.0% 1.1% 
 55 
 Other Urban 37.6% 3304 21.3% 60.1% 17.1% 1.6% 
 Rural 14.9% 2311 25.2% 59.0% 14.3% 1.6% 
Parent(s) Highest Edu.       
 < High School 15.5% 1684 20.6% 58.7% 18.9% 1.8% 
 High School grad. 29.7% 3081 18.1% 60.2% 19.8% 1.9% 
 Some college 19.4% 1998 21.3% 56.6% 20.2% 1.8% 
 College grad 35.3% 3360 27.5% 61.1% 10.9% 0.4% 
U.S. Citizen       
 No 4.9% 409 28.9% 59.4% 11.6% 0.0% 
 Yes 95.1% 9682 21.9% 59.8% 16.9% 1.4% 
Suicidality2       
 No 87.9% 8750 24.5% 60.7% 13.9% 0.9% 
 Yes 12.1% 1128 8.7% 50.1% 36.1% 5.0% 
1 Unweighted N = 10,123 
2Unweighted valid N is 9878 due to missing data. 
 
 
Prevalence of ACEs and Univariate Analyses. Table 3.3 provides the prevalence 
 
rates and univariate results for each of the 20 ACEs items. The prevalence rate for the 
adversity items ranged from a low of .3% (war) and a high of 41.7% (physical abuse). 
The five highest reported exposure categories were physical abuse (41.7%), emotional 
abuse (35.1%), community violence (33.5%), parent(s) with a mental health condition 
(28.7%), and exposure to parental domestic violence (26.2%). 
Examination of individual ACE scores showed a mean ACE score of 1.97 (S.E. 
 
= .041) for the 9-item ACEs assessment and a mean ACE score of 3.03 (S.E. = .058) for 
the 20-item assessment. Compared to the original 9 ACE items, the expanded 20-item 
adversity assessment identified fewer participants reporting no exposures (22.4% vs. 
12.4%) or 1-3 exposures (59.6% vs. 51.8%) to adversity and more participants reporting 
 
4 or more adversities (18.0% vs. 35.9%). 
 
Univariate logistic regression analysis for each of the 20 adversity items indicated 
eighteen (18) exposures were significantly associated with increased risk for suicidal 
ideation. The adjusted odds ratios ranged between 1.25 to 4.72 increased risk. The 
 56 
adversities with the five highest risk were sexual abuse (AOR = 4.72), foster care 
involvement (AOR = 4.61), parental mental health issues (AOR = 3.17), other 
stressors/traumas (AOR = 2.97), and parental domestic violence (AOR = 2.95). Living in 
a single parent household and war related exposures were not significantly associated 
with suicidal ideation. 
Table 3.3: Prevalence and Univariate Risk of Suicidal Thoughts by ACE Item 
 Prevalence, % AOR (95%, CI)1 
Original ACEs2 Yes Thought about Suicide?3 
Parent domestic violence 26.2% 2.95 (2.46 – 3.54) 
Emotional abuse 35.1% 1.56 (1.27 - 1.91) 
Physical abuse 41.7% 1.89 (1.59 - 2.25) 
Parental divorce 17.2% 1.25 (1.00 - 1.58) 
Sexual Abuse 9.3% 4.72 (3.48 – 6.39) 
Physically neglected 10.4% 2.44 (1.88 – 3.17) 
Parent mental illness 28.7% 3.17 (2.56 – 3.93) 
Parent drug/alcohol 17.7% 2.32 (1.86 - 2.89) 
Parent criminal/arrest 12.4% 2.02 (1.62 – 2.51) 
Additional ACEs   
Single parent household 11.6% 1.24 (.87 - 1.77) 
War 0.3% .89 (.56 - 1.43) 
Foster care involvement 1.3% 4.61 (1.72 – 12.35) 
Homeless 12.9% 2.63 (1.33 – 5.20) 
Disaster 3.1% 1.44 (1.04 – 2.01) 
Food scarcity 7.6% 2.76 (1.83 - 4.18) 
Community violence 33.5% 2.82 (1.98 – 4.01) 
Death/stress of other 2.6% 1.83 (1.50 – 2.24) 
Car/other accidents 15.2% 1.70 (1.22 - 2.35) 
Serious illness, poison 8.3% 2.06 (1.57 - 2.72) 
Other stressors/trauma 9.2% 2.97 (2.47 - 3.57) 
1 Model odds ratios adjusted for age, gender, race/ethnicity, poverty ratio, 
parent(s) highest educational status, urbanicity, and U.S. citizenship. 
2 Referent for each item is “No” exposure to the adversity. 
3 Referent is “No” the participant has not thought about suicide. 
 
 
Multivariate Analysis. Table 3.4 shows the adjusted odds ratios and 95% 
 
confidence intervals for each adversity item for the two regression models that included 
all 9-ACEs items from the original assessment and then a separate model including all 
 57 
twenty (20) adversity items. The overall model with the original nine ACE items was 
statistically significant (Wald (X2) = 1064.82, df = 23, p < .001). The model had an 
overall model classification of 88.5% and a Nagelkerke’s R2 of .190, indicating an 
estimated prediction in variance among risk groups of approximately 19%. 
Table 3.4: Risk Estimate for Suicidality by Original and Expanded ACE items. 
 Original ACE Items Expanded ACE Items 
(n = 9451)3 (n = 9418)4 
Predictor Variables1 AOR2 95% CI AOR2 95% CI 
Original ACEs     
Parent domestic violence 1.99 1.62 – 2.45 1.93 1.57 – 2.37 
Emotional abuse .99 .73 – 1.36 1.02 .74 – 1.40 
Physical abuse 1.53 1.23 – 1.89 1.46 1.17 – 1.82 
Parental divorce 1.09 .85 – 1.40 1.01 .78 – 1.32 
Sexual Abuse 3.22 2.31 – 4.51 2.67 1.89 – 3.79 
Physical neglect 1.48 1.08 – 2.04 1.40 1.04 – 1.91 
Parent mental health 2.20 1.78 – 2.72 2.04 1.66 – 2.51 
Parent drug/alcohol 1.33 .95 – 1.85 1.30 .94 – 1.80 
Parent criminal/arrest 1.01 .73 – 1.38 .96 .69 – 1.32 
Additional ACEsa     
Single parent household  1.15 .78 – 1.69 
War  .57 .35 – .93 
Foster care involvement  1.07 .32 – 3.58 
Homeless  .75 .30 – 1.88 
Disaster  1.02 .73 – 1.42 
Food scarcity  1.23 .80 – 1.90 
Community violence  1.45 .91 – 2.32 
Death/stressful of other  1.31 1.06 – 1.62 
Car/other accidents  1.04 .72 – 1.49 
Illness/poisoning  1.45 1.09 – 1.94 
Other stressor/trauma  1.72 1.32 – 2.23 
Note: Model odds ratios adjusted for age, gender, race/ethnicity, poverty ratio, 
parent(s) highest educational status, urbanicity, and U.S. citizenship. 
1 Referent for each item is “No” exposure to the adversity. 
2 Referent is “No” the participant has not thought about suicide. 
3 Overall model fit: Wald (X2) = 1064.82 (23), p < .001; Overall classification = 
88.5%; Cox and Snell = .098, Negelkerke R2 = .190, McFadden = .142 
4 Over all model fit: Wald (X2) = 6154.64 (34), p < .001; Overall classification = 
88.5%; Cox and Snell = .106, Negelkerke R2 = .206, and McFadden = .155 
 58 
In this model sexual abuse (AOR = 3.22), parental mental health (AOR = 2.20), parental 
domestic violence (AOR = 1.99), physical abuse (AOR = 1.53), and physical neglect of 
the child (AOR = 1.48) were statistically significantly associated with increased risk for 
suicidal ideation. 
The second model, which included all twenty (20) adversities was statistically 
significant (Wald (X2) = 6154.64, df = 34, p < .001). The model had an overall model 
classification of 88.5%, and a Nagelkerke’s R2 of .206 indicating an estimated prediction 
in variance among risk groups of approximately 21%. While slightly attenuated, the same 
five adversities - sexual abuse (AOR = 2.67), parental mental health (AOR = 2.04), 
parental domestic violence (AOR = 1.93), physical abuse (AOR = 1.46), and physical 
neglect of the child (AOR = 1.40) - were significantly associated with increased risk for 
suicidal thoughts. Of the additional 11 adversity items, other stressors and/or traumatic 
events (AOR = 1.72), having a serious illness and/or poisoning (AOR = 1.45), and the 
death and/or stressful situation of another (AOR = 1.31) were statistically significantly 
associated with increased risk for suicidal ideation. 
Table 3.5 shows the estimated risk for suicidal ideation by ACE score. This model 
was statistically significant (Wald (X2) = 1020.58, df = 21, p < .001), the overall model 
classification was 88.7%, and the Nagelkerke’s R2 was .170, indicating an estimated 
prediction in variance of approximately 17%. While those reporting only one exposure to 
adversity was not statistically significant, there was a graded effect on risk for suicidal 
ideation as individual ACE scores increased and significant risk increasing substantially 
for those reporting 4 or more exposures. 
 59 
Table 3.5: Estimated Risk for Suicidality by ACE Score (N=9850)1 
 Model2 
AOR 95% CI 
20-item ACE Score   
0 Referent Referent 
1 1.23 .71 – 2.13 
2 2.17 1.22 – 3.88 
3 2.71 1.47 – 5.01 
4 5.27 3.07 – 9.05 
5 6.69 4.02 – 11.14 
6 7.49 4.01 – 13.99 
7 or more 14.76 8.94 – 24.39 
1Note: Model odds ratios adjusted for age, gender, race/ethnicity, poverty ratio, 
parent(s) highest educational status, urbanicity, and U.S. citizenship. 
2Overall model fit: Wald (X2) = 1020.58 (21), p < .001; Overall classification = 
88.7%; Cox and Snell = .088, Negelkerke R2 = .170, McFadden = .126 
 
 
Discussion 
 
The Adverse Childhood Experiences study by Felitti and colleagues (1998) was a 
seminal study that expanded our understanding of the long and short-term impact of 
traumatic experiences during childhood. Since that time, researchers have sought to better 
understand how these experiences influence behavioral, psychological, and physiological 
responses to such exposures (Cronholm et al., 2015; Finkelhor et al., 2013; McLaughlin 
et al., 2013; National Scientific Council on the Developing Child, 2012; Wade et al., 
2015, 2014). This study similarly seeks to advance our current understanding of the 
prevalence of various adversities children may be exposed to during their formative 
developmental years and the association of such exposures with adolescent suicidal 
ideation. 
This study confirms adversity during childhood is not uncommon and may in fact 
be more prevalent than noted in the original ACE study. The original Felitti (1998) study 
found 36% of adult participants reported no exposures to childhood adversities using a 
 60 
10-item childhood adversity assessment. In the current study 22% of participants reported 
no exposures based on a 9-item assessment of similar adversities assessed in the original 
ACE study and 12% reported no exposures when using an expanded 20-item adversity 
assessment. Finally, several adversities included in the expanded assessment were 
reported by a substantial number of study participants. Of the eleven additional 
adversities included in this study, exposure to community violence, car and other 
potentially fatal or serious accidents, homelessness, living in a single parent household, 
and exposure to other stressors or traumas were among the five highest reported 
exposures in the expanded assessment. These findings support the need to expand 
existing assessments to include a broader range of experiences that may cause stress and 
harm during important developmental stages. 
This study further confirms that interpersonal adversities are significant predictors 
of adolescent suicidal ideation and perhaps other suicide related behaviors (Johnson et al., 
2002). Regression models using both the 9-item assessment and the 20-item assessment 
found a significant association of exposure to parental domestic violence, physical abuse, 
sexual abuse, physical neglect, parental mental health conditions, the death or adversities 
of another, and having a serious or life-threatening illness as significant predictors of risk 
for suicidal ideation. Perhaps most notable is that the variable “other stressors and 
traumas” not assessed by the NCS-A was statistically significant in the fully regressed 
20-item model. This finding suggests other potential stressors not identified in the 
original ACE assessment as well as the expanded assessment for this study may play a 
significant role in suicidal ideation. As demonstrated in prior studies experiences such as 
exposure to discrimination, living in unsafe neighborhoods, school failure, and being 
 61 
bullied may be significant adversities not included in many studies focused on childhood 
adversities and health outcomes, and may be specific to populations that are more 
racially, ethnically, and economically diverse compared to the original ACE study 
population (Cronholm et al., 2015; Finkelhor et al., 2013; Johnson et al., 2002). Hence, 
expanding our knowledge about the influence of other exposures among diverse 
populations is warranted. 
There were several unexpected outcomes of this study. First, unlike other 
childhood adversities, exposure to “war” related experiences were found to reduce risk 
for suicidal thoughts when included in the full 20-item regression model. A possible 
explanation, is a variety of other factors associated with children who have survived 
experiences of war may be playing a role in children’s reduced risk for suicidal thoughts 
and sense of resilience among this sample population. For example, it has been noted that 
ecological factors such as meaning making of the experience, family functioning and 
other social supports and networks unique to those who have experienced war related 
exposures may increase resiliency and reduce risk for future negative behavioral and 
mental health outcomes including suicidal ideation as seen in this study population 
(Betancourt & Khan, 2008; Slone & Mann, 2016). 
Second, while the original hypothesis suggested that additional adversities would 
be additive to the model, the full 20-item model in this study only slightly better 
predicted risk for suicidal ideation among adolescents, with three additional adversities 
found to be significantly associated with the outcome. It is theorized this result may be 
related to the specific nature of the risk factors associated with suicidal behaviors which 
are most often interpersonal in nature (i.e., witnessing parental domestic violence, 
 62 
physical and sexual abuse, exposure to serious illness, and death of a significant person) 
versus other forms of stressful events that may be attributed to factors not as personal in 
nature and may be more socially driven (Johnson et al., 2002). For example, while prior 
research has supported the theory that exposures to experiences such as living in a single 
parent household, being involved in foster care, and exposure to food scarcity may have 
short and long-term impacts on overall health and wellbeing (Cook et al., 2013; 
Courtney, Piliavin, Grogan-Kaylor, & Nesmith, 2001; Cronholm et al., 2015; 
McLaughlin et al., 2012; Weinreb et al., 2002), these findings were not confirmed in 
multivariate analyses in this study on suicidal ideation. Another possible reason for these 
findings may be due in part to rarity of exposures such as foster care involvement and 
exposure to disasters or how well the specific questions captured the adversity in this 
study. Further research on these items as potential adversities that may influence suicidal 
ideation is warranted. 
Limitations 
 
There are several limitations that should be considered when interpreting the 
results of this study. First, due to the cross-sectional observational design, temporal 
precedence and causation cannot be determined. It is unclear whether the adversities were 
a precursor to the outcome measure of interest or if the suicidal ideation preceded the 
reported adversities. However, these results are consistent with current theory and other 
studies assessing cumulative childhood exposures to adversity and health related 
outcomes including suicidal ideation and behaviors (Dube et al., 2001; Felitti et al., 
1998). A second limitation is possible recall bias due to the retrospective reporting of 
exposures to various stressors and adversities. While this potential bias is present with 
 63 
any retrospective assessment, exposures for this study were assessed during adolescence 
rather than during adulthood, providing a shorter time span between actual events and 
reporting. Further, studies have indicated that retrospective assessments of traumatic or 
stressful events is an appropriate approach (Dube, Williamson, Thompson, Felitti, & 
Anda, 2004). Finally, the rarity of several adversities included in this study may have 
prevented an accurate assessment of the association between these exposures and suicidal 
ideation. 
Conclusion 
 
This study confirms adversity during childhood is common and may include 
additional exposures not assessed in the original Adverse Childhood Experiences Study. 
These experiences may have dramatic effects on child and adult behavioral and health 
outcomes, and in this case, suicidal ideation among children during their adolescence. 
Therefore, prevention and treatment interventions among youth geared towards reducing 
youth suicide must consider the multiple factors that may act as drivers to the second 
leading cause of death among adolescents and young adults. 
 64 
Chapter 4: 
 
Study #2: Adolescent Social Integration as a Protective Factor of Cumulative 
Adverse Childhood Experiences and Adolescent Suicidal Ideation. 
Abstract: 
 
Background: Adverse Childhood Experiences have been associated with 
 
numerous negative behavioral and mental health outcomes over the life-course. While 
much has been learned about the physiological impact of cumulative childhood 
adversities and trauma, research is needed to better understand the factors that protect 
those exposed to potentially toxic levels of adversity and stress. 
Methods: Guided by Durkheim’s theory of social integration and regulation, an 
 
analysis of the National Comorbidity Survey Replication Adolescent Supplement (NCS- 
 
A) study was conducted to investigate the influence of adolescent social integration on 
the relationship of childhood adversity and suicidal ideation among a nationally 
representative sample of adolescents (n = 10,128) between the ages of 13 and 18. Family, 
school, peer, teacher, and religious/spiritual integration or connectedness were assessed 
as protective factors for suicidal ideation among those with childhood adversities using 
multigroup path analysis. 
Results: Social integration was found to reduce risk for suicidal ideation. 
 
However, only family, school and religious integration or connectedness were found to 
be significant protective factors in this study. Further, the effect of these three integration 
factors was only demonstrated among youth reporting 1-3 adversities, and only family 
connectedness reduced risk for suicidal ideation for youth reporting 4 – 6 exposures. 
 65 
Social integration did not reduce risk for suicidal ideation for youth reporting 7 or more 
adversities. 
Conclusion: Family, school, and religious/spiritual integration and connectedness 
 
reduce risk for suicidal ideation among youth reporting low to medium levels of 
cumulative adverse experiences. However, the findings of this study suggest that the 
impact of higher exposures to adversity (i.e., 7 or more exposures) mitigate the positive 
influence of social integration. Further research is warranted to address the specific needs 
of youth exposed to higher levels of stress and adversity with a specific focus on how 
best to reduce or eliminate childhood adversities and trauma. 
 66 
Introduction 
 
Adverse childhood experiences among youth under age 18 are common and 
considered a serious public health issue. In 2009, a national study found nearly 60% of all 
youth reported exposure to at least one form of direct or indirect violence annually 
(Finkelhor, Ormrod, et al., 2009). While exposure to adversity is considered normal and 
necessary for healthy development, many children experience toxic levels of stress and 
adversity that are often co-occurring or cumulative. These experiences often go 
unidentified and without appropriate supports and treatment with far reaching 
consequences over the life course (National Scientific Council on the Developing Child, 
2012). 
Suicidal ideation (the process of thinking about, considering, or planning to 
commit suicide), which often precedes suicide related behaviors is one of many factors 
that have been associated with cumulative childhood adversities (Beautrais, 2000; Choi et 
al., 2017; Dube et al., 2001; Hardt et al., 2008b; Johnson et al., 2002). Led only by 
accidental injuries, suicide is the second leading cause of death among youth and young 
adults between the ages of 15 and 25 (Center for Disease Control and Prevention (CDC), 
2015a). It is estimated that 17% of students in grades 9-12 seriously consider attempting 
suicide and 13.6% make a plan about how they would commit suicide (Center for 
Disease Control and Prevention (CDC), 2015c). While estimates are not available 
specifically for youth under the age of 18, estimated national cost of suicides and suicide 
attempts range between $58.4 to $93.5 billion annually (Shepard, Gurewich, Lwin, Reed, 
& Silverman, 2016). Given our understanding of the connection between childhood 
adversity and the devastating impact of suicide among youth and young adults, 
 67 
identifying the factors that increase resiliency among at risk populations is a public health 
imperative. 
One of the most well-known investigations of cumulative childhood adversity is 
the Adverse Childhood Experiences (ACE) study (Felitti et al., 1998). The ACE study 
was a collaborative effort between Kaiser Permanente and the Centers for Disease 
Control and Prevention (CDC) that examined the association of cumulative exposure to 
ten childhood adversities related to abuse, neglect, and household dysfunction and long- 
term behavioral and health outcomes in a cohort of more than 1700 adults (Felitti et al., 
1998). Results from this study and many other studies focused on childhood adversity 
have consistently shown a graded increased risk for negative behavioral, physical and 
mental health outcomes among adults and children (Anda, 1999; Anda et al., 2004, 2008; 
Anda, Tietjen, Schulman, Felitti, & Croft, 2010; Brown et al., 2007; Chapman et al., 
2004, 2011; Dietz, 1999; Dong et al., 2004; Dube et al., 2001, 2003, 2006b, 2009b; 
Edwards, Anda, Felitti, & Dube, 2004; Edwards, Fivush, Anda, Felitti, & Nordenberg, 
2001; Edwards, Holden, Felitti, & Anda, 2003; Hillis et al., 2004; Hillis, Anda, Felitti, & 
Marchbanks, 2001; Hillis, Anda, Felitti, Nordenberg, & Marchbanks, 2000; Williamson, 
Thompson, Anda, Dietz, & Felitti, 2002). Since the ACE study, advances in a broad 
spectrum of disciplines (i.e., neuroscience, molecular biology, genomics, developmental 
psychology, epidemiology, sociology, and economics) have expanded our understanding 
about the influence of stressful events on child development (Folkman, 1984b; Lazarus, 
1966; Lazarus & Cohen, 1977; Shonkoff et al., 2012). 
The “Ecobiodevelopment (EBD) Framework,” created by the Center on the 
Developing Child at Harvard University, provides a structure for understanding the 
 68 
etiology of childhood adversity and stress on health and development, and provides an 
outline for potential areas of intervention and treatment (see Figure 1: 
Ecobiodevelopmental Framework) (Shonkoff et al., 2012). The EBD framework suggests 
the consequences of adversity and stress during childhood are determined by interlocking 
relationships of biology (i.e., physiological adaptations and disruptions), ecology (i.e., 
social and physical environment), and health disposition and human development (i.e., 
learning, behavior, and physical and mental well-being) (Shonkoff et al., 2012). It is 
thought that environmental exposures (i.e., ecology) beginning as early as the prenatal 
period trigger and set the stage for current and future responses to adversity and stress 
(i.e., biology), which then influence future behavioral and psychological responses (i.e., 
health and development) throughout childhood and over the life-course (Committee on 
Psychosocial Aspects of Child and Family Health, 2012; Shonkoff et al., 2012). 
At the biological level, it is understood that the human body is designed to 
respond to stress inducing experiences through a series of hormonal responses that act as 
protective factors (Committee on Psychosocial Aspects of Child and Family Health, 
2012; Compas, 2006; Shonkoff et al., 2012; Wadsworth, 2015). These protective 
responses to stressful ecological exposures are centered within the brain and include a 
combination of immediate, unconscious biological reactions and less reactionary, 
conscious responses (i.e., appraisal of experiences) (Compas, 2006; McEwen, 1993; 
McEwen et al., 2015). In situations of adversity, threat or stress, these biological stress 
reactions set in motion the “fight or flight” response or a series of biological and 
psychological responses to the event or exposure (Shonkoff et al., 2009). Evidence has 
suggested these responses can either facilitate healthy development through “positive” or 
 69 
“tolerable” stressful events or disrupt normal development in children when exposures 
become “toxic” (Shonkoff et al., 2012). 
“Toxic” stress, which is the most harmful to healthy development, is 
characterized by strong, frequent, or prolonged activation of the body’s stress response 
systems in the absence of protective factors (Shonkoff et al., 2012). When individuals are 
exposed to repeated or chronic stress the human stress response may become deregulated 
and may result in harmful physiological conditions known as “allostatic load or 
overloading” and maladaptive psychological coping skills (McEwen, 1993). For example, 
studies have found toxic levels of adversity to be a significant predictor of physiological 
changes such as elevated blood pressure, elevated hypothalamic pituitary adrenal axis 
activity, dysregulation of metabolic activity, compromised immune function, increases in 
inflammatory markers, and elevated allostatic load - all of which increases risk for 
morbidity and mortality (Araújo et al., 2009; Berasain et al., 2009; Chen & Miller, 2007; 
Danese et al., 2008; Evans & Kim, 2013; Felitti et al., 1998; Middlebrooks & Audage, 
2008; Slopen et al., 2012; Slopen, Kubzansky, et al., 2013; Slopen, McLaughlin, et al., 
2013). In children, toxic stress has been associated with increased risk for a number of 
maladaptive psychological and behavioral responses such as depression and anxiety, 
avoidance and dissociation, aggression, early sexual initiation and risky sexual behaviors, 
impaired school function and memory, alcohol and tobacco consumption, and suicidality 
(Chapman et al., 2004; Dube et al., 2003; Felitti et al., 1998; Middlebrooks & Audage, 
2008; Stein et al., 2003). 
It is understood that healthy child development in the face of adversity and trauma 
is dependent on adequate and appropriate supports and connections within the ecological 
 70 
environment. These social relationships and ties are thought to affect mental and physical 
health by having a positive influence on individual emotions, cognitions, and behaviors 
(Berkman et al., 2000; Cohen, 2004; Seeman, 1996). Durkheim was one of the first to 
examine the impact of social relationships and social ties on human health and behavior, 
and more specifically suicide, through his theory of social integration and regulation 
(Berkman et al., 2000). Durkheim theorized that variations in rates of suicide were a 
consequence of the quality of social relationships or connectedness (i.e., “social facts”) 
(Durkheim, 1951). Further, the loss of organized and supporting relationships (i.e., 
family, community, and work ties) and lack of social constraints as a result of individuals 
migrating to more industrialized, less organized communities would have a negative 
impact on their psychological wellbeing, resulting in increased risk for suicidality (Cohen 
et al., 2000). 
Social integration, which many consider to be the central construct of Durkheim’s 
theoretical framework, was posited to provide an interchange of ideas and feelings that 
form the social bonds that attach individuals to common goals and shared social causes 
(Thorlindsson & Bjarnason, 1998). Effective social integration of the individual is also 
believed to provide a positive means of social supports and interactions leading to better 
psychosocial outcomes for the individual, whereas ineffective integration leads to 
negative outcomes or higher rates of suicide (Rose et al., 2014). Social regulation, 
expressed through religious, political and moral beliefs, were theorized to control 
individual behaviors that would prevent individuals from acting in opposition to the 
benefit of society. The quantity and intensity of these relationships determine how 
attached individuals are to each other, whether they have common goals, and the extent to 
 71 
which natural human desires are regulated through the use of socially accepted authority 
figures to limit individual desires and passions (Durkheim, 1951). 
Prior studies focused on Durkheim’s theory of social integration have 
demonstrated a positive association between social integration and health outcomes 
(Berkman & Syme, 1979; Crittenden et al., 2014; Easton & Renner, 2013; House, 
Robbins, & Metzner, 1982; Kuramoto, Wilcox, & Latkin, 2013; Knoester & Haynie, 
2005; Rose et al., 2014; Schoenbach, Kaplan, Fredman, & Kleinbaum, 1986; Tenorio & 
Lo, 2011). For example, a 9-year probability study of mortality among a cohort of adults 
ages 30 to 69 years (N = 4775), found that lower levels of social integration (i.e., 
summary index reflecting ties with a spouse, close friends and relatives, and participation 
in church and other types of groups) were associated with higher mortality (Berkman & 
Syme, 1979). 
Thorlindsson and Bjarnason (1998) investigated Durkheim’s theory of social 
integration and regulation by examining “social facts” at the micro-level with a specific 
focus on the role of family integration and parental regulation on youth suicidality. 
Using a nationally representative sample of 4,314 high school students, the relationship 
of family social support (i.e., social integration), parental regulation (i.e., social 
regulation), social anomie (i.e., a condition of instability resulting from a breakdown of 
standards and values or lack of purpose or ideas), and social norms regarding suicide (i.e., 
knowing others who were thinking about, attempted or committed suicide) and their 
association to individual suicidality were examined. Results from their study 
demonstrated the concepts of social integration and social regulation were distinct 
constructs, with strong family integration or social support having a significant inverse 
 72 
effect on rates of suicidality among the sample population. Further, parental regulation 
was not found to be a significant predictor of suicidality amongst this sample population 
(Thorlindsson & Bjarnason, 1998). 
Finally, Rose and colleagues (2014) examined Durkheim’s theory of social 
integration (i.e., family, school, and religious involvement and religious commitment) on 
mental health outcomes of a nationally representative sample of Black adolescents (n = 
1,170) between the ages of 13-17. The findings from this study supported Durkheim’s 
theory of social integration and concluded that family, school, and religious integration 
are significant predictors of better mental health for Black adolescents, with school 
integration emerging as the strongest predictor of positive psychological well-being 
(Rose et al., 2014). 
Using the Ecobiodevelopmental (EBD) Framework as a guide, this study focused 
on expanding our understanding of the role of social integration or connectedness as a 
protective factor for adolescent suicidal ideation. Unlike prior studies, this investigation 
used an expanded assessment of childhood adversity to assess the prevalence and 
influence of higher levels of adversity on suicidal ideation. This investigation also 
expanded on the Rose and colleagues (2014) study by examining the co-occurrence of 
family, school and religious connectedness with two additional factors (i.e., peer and 
teacher integration) commonly viewed as important positive influences on early child and 
adolescent developmental on suicidal ideation. Based on Durkheim’s theory of social 
integration, it was hypothesized that each of the five social integration factors would have 
a significant protect effect on risk for suicidal ideation across each childhood adversity 
 73 
group, and while attenuated, this protective relationship would remain significant across 
each adversity group. 
Methods 
 
This study is an analysis of the National Comorbidity Survey Replication 
Adolescent Supplement (NCS-A) study (Kessler, 2013). Data for the NCS-A were 
collected between February 2001 and January 2004 by the Survey Research Center of the 
Institute for Social Research at the University of Michigan and was an add-on to the 
National Comorbidity Survey Replication (NCS-R) study, an investigation of the 
prevalence and correlates of mental health disorders among adults in the US (Kessler, 
2013). The NCS-A study was considered the first nationally representative study on the 
prevalence, correlates and patterns of service use for DSM-IV mental health disorders 
among U.S. adolescents (Kessler, Avenevoli, Costello, et al., 2009a). This seminal study 
was designed to provide the groundwork for follow-up studies of risk and protective 
factors, consequences, and early expressions of adult mental health disorders (Kessler et 
al., 2009a, 2009b; Merikangas, Avenevoli, Costello, Koretz, & Kessler, 2009). 
NCS-A Eligibility and Sampling Frame. Participation in the NCS-A study was 
 
restricted to English-speaking, non-institutionalized adolescents between the ages of 13 
and 18 (n=10,148) living in the United States. A dual-framed, complex cluster sampling 
methodology was used to draw a sample of adolescents for the study. The first sampling- 
frame included adolescents (n=904) living in households where an adult had participated 
in the National Comorbidity Survey-Replication (NCS-R) study. To supplement the low 
number of interviews completed, a second sampling-frame of adolescents (n=9244) were 
recruited to the NCS-A study from a population of students attending 320 schools located 
 74 
in the same counties in which the NCS-R was conducted (Kessler et al., 2009b). Eligible 
schools included all accredited private and public middle schools, junior high schools and 
high schools within the 62 primary sampling units (PSUs) used in the NCS-R study 
(Kessler et al., 2004, 2009b). A probability sampling methodology proportional to the 
size of the student body was used to select schools. Forty to fifty students were randomly 
selected from each school to participate in the NCS-A study. The overall adolescent 
response rate (i.e., both household and student sample populations) was 75.6%. The 
majority of non-responders (21.3%) refused to participate; the remaining 3.2% were 
unable to participate due to “circumstances” or “non-contact” (Kessler et al., 2009b). 
Further details regarding the sampling frame for the NCS-A study can be found 
elsewhere (Kessler et al., 2004). 
NCS-A Adolescent Questionnaire. The NCS-A adolescent questionnaire includes 
 
data on a number of factors considered relevant to adolescent behavior and mental health. 
Areas assessed include individual level factors (e.g. demographics, developmental 
factors, cognitive and academic abilities-achievements, physical health, and stressful life 
events), family level factors (e.g. family structure, stability and adaptability, parenting 
behaviors, parental psychopathology, and family stress), and environmental level factors 
such as school and neighborhood characteristics (Merikangas et al., 2009). Further, a 
modified version of the fully structured Composite International Diagnostic Interview 
(CIDI) of behavioral and mental health disorders was used in the adolescent 
questionnaire to assess for fifteen DSM-IV disorders among participants such as mood, 
anxiety, behavior, and substance abuse disorders (Kessler et al., 2012). 
 75 
Procedures. In the presence of a parent or guardian, adolescent interviews were 
 
administered face-to-face in the home of each respondent using a laptop, computer- 
assisted personal interview (CAPI). Trained survey interviewers from the Survey 
Research Center of the Institute of Social Research at the University of Michigan 
administered each adolescent interview. On average, the NCS-A adolescent interviews 
took approximately two and a half hours to complete, with interviews ranging between 69 
to 347 minutes (Kessler et al., 2009b). This variation in interview length was primarily 
due to the number of lifetime disorders assessed for each adolescent that was determined 
based on an initial screening tool administered at the beginning of each interview 
(Kessler et al., 2009b; Merikangas et al., 2009). 
This analysis includes all participants in the NCS-A study. Approval for this study 
was obtained from the University of Maryland, College Park institutional review board 
and the University of Michigan Inter-University Consortium of Political and Social 
Research (ICPSR) institutional review board. 
Measures 
 
Adversity Measures. Forty-six (46) questions in the NCS-A were used to create 
 
the twenty (20) childhood adversity items for this study. These items were derived from a 
number of standardized assessment tools selected by the original NCS-A research team 
(Kessler et al., 2009a; Merikangas et al., 2009)(Kessler et al., 2009a; Merikangas et al., 
2009). Table 4.1 provides a summary of the twenty (20) categories of adversity used and 
the NCS-A questions used to assess exposure for each category. 
A positive response to any question within an adversity category was coded as “1 
 
= Yes” meaning the participant had been exposed to the adversity category. If all of the 
 76 
items within an adversity category received a negative response, the respondent was 
coded as ‘0 = No” indicating he/she had not been exposed to that particular adversity 
category. A conservative approach was used to score adversity variables measured using 
a Likert scale in the NCS-A. A respondent was coded “1= Yes” if the participant reported 
an event occurred “often or sometimes” and was coded “0= No” if the participant 
reported “not very often or never.” Further, missing responses to any adversity question 
was coded as “0=No” for not being exposed to the experience. 
An overall adverse childhood adversity (ACE) score was generated for each 
participant by summing all twenty adversity items to create an aggregate score for each 
participant ranging from 0–20. However, due to small numbers, youth reporting seven or 
more adversities were coded as “7” to ensure better stability during the analyses. 
Therefore, the final ACE score for each participant ranged between 0-7 for this study. 
Dependent Variable. To assess suicidal ideation participants were asked if they 
had ever had the experience - “You seriously thought about committing suicide.” For this 
study, this item was used to determine if a participant had at any time in her/his life 
seriously considered suicide. Possible response options were “1=Yes” and “0=No.” 
Social Integration Variables. Table 4.2 summarizes items used from the NCS-A 
 
used to generate each social integration variable. To construct the new social integration 
variables, first reverse coding was conducted to align negatively worded items or 
response options for directional consistency across each social integration item. The 
range of responses for each item was 0 to 4, with higher scores representing higher levels 
of perceived social integration or connectedness. Next, principal components analysis 
(using direct oblimin rotations) and scale reliability were used to create the five social 
 77 
integration variables for the final analysis (Nunnally, 1975). While several items used in 
the NCS-A appeared to be consistent with a specific social integration construct, only 
those items with a component loading greater than .30 were included in each social 
integration scale (Kline, 2010). Finally, scale reliability was assessed for each of the 
newly constructed social integration variables to examine internal consistency of each 
integration scale and mean scores for each of the five social integration items was 
generated for each participant. 
The religious/spiritual integration variable was derived from four items assessing 
religious engagement, importance, and emotional support. It included items such as 
“How often do you usually attend religious services” and “In general, how important are 
religious or spiritual beliefs in your daily life?” (Mean = 2.50, S.E. = .02, Cronbach’s 
Alpha = .882). The family integration item was derived from twelve items used in the 
NCS-A study and assessed family emotional support, engagement, and decision-making. 
It included items such as “How often did family members feel very close to each other” 
and “How often did the whole family do things together” (Mean = 2.86, S.E. = .01, 
Cronbach’s Alpha = .855). 
 78 
 
Table 4.1: Childhood Adversity Items and Related NCS-A Questions 
Parental domestic violence – “Yes” = 1, “No” = 0; “Often” or Sometimes” = 1, “Not very often” or “Never” = 0 
• Ever witness serious physical fights at home, like when your father beat up your mother? 
• How often did (woman/man) who raised you insult, swear, shout, yell, scream, threaten to hit each other? 
• How often did (woman/man) grab, shove, throw something, slap or hit each other? 
• How often did (woman/man) kick, bit, hit with a fist, beat up, choke, burn or scald, threaten with a knife or gun each other? 
Emotional abuse – “Often” or “Sometimes = 1, “Not very often” or Never” = 0 
• How often did woman/man who raised you insult, swear, shout, yell or scream, or threatened to hit you? 
• How often did sibling insult, swear, shout, yell or scream, or threaten to hit you? 
• How often did someone you were dating insult, swear, shout, yell or scream, or threaten to hit you? 
Physical abuse – “Yes” = 1, “No” = 0; “Often” or “Sometimes” = 1, “Not very often” or “Never” = 0 
• Ever badly beaten up by your parents or people who raised you? 
• Ever badly beaten up by someone you were dating or with whom you were romantically involved? 
• Were you ever badly beaten up by anyone else (other than parent or someone you were dating)? 
• When growing up, how often did woman or man who raised you push, grab or shove, throw something, slap or hit you? 
Kick, bit or hit with a fist, beat up, choke, burn or scald, or threaten with a knife or gun? 
• When growing up, how often did sibling push, grab or shove, throw something, slap or hit you? Kick, bit or hit with a fist, 
beat up, choke, burn or scald, or threaten with a knife or gun? 
Parental divorce – “Yes” = 1, “No” = 0 
• “Biological parents were divorced”. 
Sexual abuse – “Yes” = 1, “No” = 0 
• Have you ever been raped? 
• Have you ever been sexually assaulted or molested (other than rape)? 
• Have you ever been stalked? 
Physical neglect – Responses = “Often” or “Sometimes” = 1, “Not very often” or “Never” = 0 
• How often made to do chores that were too difficult or dangerous for someone your age? 
• How often left alone or unsupervised when you were too young to be alone? 
• How often go without things like clothes, shoes, or school supplies because parent spent money on themselves? 
 79 
 
• How often did parents/caregivers make you go hungry or not prepare regular meals? 
• How often did parents/caregivers ignore or fail to get you medical treatment when sick or hurt? 
Parental mental health – Any question “Yes” = 1, “No” = 0 
• Was the (man/woman) who raised you ever sad or depressed lasting two weeks or more? 
• Was the (man/woman) who raised you ever nervous, edgy, or anxious more than a month? 
• Did the (man/woman) who raised you ever have anxiety attacks? 
• Did the (man/woman) who raised you ever attempt or commit suicide? 
Parent drug/alcohol abuse “Yes” = 1, “No” = 0 
• Did the (man/woman) who raised you ever have a problem with drugs? 
• Did the (man/woman) who raised you ever have a problem with drinking alcohol? 
Parent criminal/arrest history “Yes” = 1, “No” = 0 
• Was the (man/woman) who raised you ever arrested or sent to prison? 
• Was the (man/woman) who raised you ever involved in criminal activities like burglary or selling stolen property? 
Single parent household 
• Participant reported living mostly in a single parent household. 
Foster care involvement 
• Ever in foster care? 
Homelessness – “Yes” = 1, “No” =2 
• Have you ever been homeless? 
Food scarcity – “Yes” = 1, “No” =0 
• In the past 12 months, did you go hungry because there was not enough money for food? 
• In the past 12 months, did you eat less because there was not enough money for food? 
Major disaster – “Yes” = 1, “No” =0 
• Ever involved in a major disaster like a flood, hurricane, fire, bomb explosion, or earthquake? 
Community violence – “Yes” = 1, “No” =0 
• Were you ever mugged, held up, or threatened with a weapon? 
• Did you ever see someone being badly injured or killed, or unexpectedly see a dead body? 
Death or stress of other – “Yes” = 1, “No” =0 
 80 
 
• Anyone close to you die unexpectedly (e.g., killed in accident, murder, committed suicide, or fatal heart attack)? 
• Anyone close to you ever have a stressful or life-threatening experience, like kidnapping, torture or rape? 
War, revolution, political coup, refugee – “Yes” = 1, “No” =0 
• Ever in a place where there was war, revolution, military coup or ongoing terror? 
• Were you ever a refugee? 
Car or other life-threatening accident – “Yes” = 1, “No” =0 
• Ever involved in a very serious or life-threatening car accident? 
• Did you ever have any other very serious or life-threatening accident? 
Serious illness, poison or chemical – “Yes” = 1, “No” =0 
• Ever exposed to poisonous chemical or substance that could cause serious harm? 
• Did you ever have a very serious or life-threatening illness? 
Any other stressor or trauma not mentioned – “Yes” = 1, “No” =0 
• Ever experience other extremely upsetting or life-threatening event that I haven’t asked about yet? 
 81 
 
 
Principal component analysis found two distinct components among the education 
related observed variables - school and teacher connectedness. The school integration 
item was derived from seven observed items and included questions such as “I like 
school” and “Getting good grades is important to me” (Mean = 3.11, S.E. = .01, 
Cronbach’s Alpha = .748). The teacher integration item was derived from three observed 
variables focused on the participants connection with their teachers and included items 
such as “Most of my teachers treat me fairly” and “I like my teachers” (Mean = 3.27, S.E. 
= .01, Cronbach’s Alpha = .665). Finally, the peer integration item was derived from 
three observed variables that included items such as “How much can you rely on friends 
when you have a serious problem” and “How often do you let your friends know about 
your problems and worries” (Mean = 3.06, S.E. = .01, Cronbach’s Alpha = .632). 
Control Variables. Several variables were examined as potential confounding 
 
variables. These variables included adolescent race and ethnicity (i.e., White/other, 
 
Black, and Hispanic), age at interview, gender (i.e., male or female), highest maternal or 
paternal education (i.e., less than high school, high school graduate, some college, college 
degree), poverty ratio (1 = “<=1.5”, 2 = “1.5 to 3”, 3 = “3 to 6” and 4 = “ >= 6” times the 
poverty threshold), type of region of residency (i.e., metropolitan [large city], other urban 
area [suburbs, small city, town/village], and rural area), and U.S. citizen status (i.e., yes 
or no). 
 82 
Table 4.2: Social Integration Items 
Religious/Spiritual Integration 
• How often do you usually attend religious services? 
• In general, how important are religious/spiritual beliefs in your daily life? 
• When you have problems/difficulties in your family, work, etc. how often do you 
seek comfort through religious or spiritual means such as prayer, meditating, 
attending service, or talking to religious/spiritual advisor? 
• When you have decisions to make in your daily life, how often do you think about 
what your religious or spiritual beliefs suggest you should do? 
Family Integration 
• How often did family members feel very close to each other? (R) 
• How often could family members talk to each other about their feelings? (R) 
• How often did family members let each other know when they were sad or 
worried? (R) 
• How often did family members keep their feelings to themselves? 
• How often did the whole family do things together? (R) 
• How often did family members share interests/hobbies with each other? (R) 
• How often did family members avoid each other? 
• How often did family members go along w/ what the family decided to do? (R) 
• How often did each family member have input in major family decisions? (R) 
• How often did children have a say in their discipline? (R) 
• How often did everyone compromise when there were disagreements? (R) 
• How often was it difficult to get everyone to agree on decisions? 
School Integration 
• What sort of grades did you get in your last years at school? 
• I like(d) school 
• Getting good grades (is/was) important to me. 
• Homework (is/was) a waste of time. (R) 
• I try(ied) hard at school. 
• I Feel/felt as if I don’t/didn’t belong at school (R). 
• Most of the things I learn(ed) in school are unimportant. (R). 
Teacher Integration 
• Most of my teachers treat(ed) me fairly. 
• I care(d) a lot about what my teacher thinks/thought about me. 
• I like(d) my teachers. 
Peer Integration 
• How often do you talk on the phone/hangout/get together with your friend(s)? 
• How much can you rely on your friend(s) for help if you have a serious problem? 
• How much can you open up to your friend(s) if you need to talk about your 
worries? 
 83 
The number of DSM-IV disorders were also controlled for in this study. This 
variable was generated based on the fifteen lifetime DSM-IV disorders (i.e., distress, fear, 
behavioral, substance abuse, and bipolar disorders) assessed and coded for each 
participant by the original NCS-A research team. The range of cumulative disorders was 
0-5 in this study, indicating a participant had anywhere from 0 to 5 DSM IV disorder 
diagnoses. Given the small number of participants with five DSM IV disorders (n = 59), 
those with 4 and 5 disorders were combined into one category of 4 or more disorders (n = 
313). Therefore, the range of DSM IV disorders for this study was between 0 to 4 
disorders. 
Analysis 
 
The Statistical Package for the Social Sciences (SPSS) version 24 was used to test 
for assumptions, develop the social integration items, and prepare the dataset for further 
analyses. Missing data were assessed for each variable included in the study and it was 
determined that less than 5% of the data was missing. Next, multicollinearity was 
assessed using Tolerance (R2smc < .10) and Variance Inflation Factor (1/1-R2smc > 10) 
statistics (Kline, 2010) and were found to be acceptable. 
Mplus version 7.4 was used to test the hypothesis using multi-group path analyses 
(Muthén & Muthén, 2018). The full sample was retained for this analysis and full 
information maximum likelihood (FIML) method was used to handle missing data. 
Weighted least squares (WLSMV) was used to estimate the path coefficients given the 
complex sampling methodology and the inclusion of both continuous and categorical 
variables. As recommended by Hu and Bentler (1999), data to model fit was examined 
using comparative fit index (CFI > .95) and the root mean square error of approximation 
 84 
(RMSEA < .06) indices. Parameter estimates were examined and statistically significant 
path coefficients and R-square values for each multi-group analysis are reported (Kline, 
2010). 
Results 
 
Characteristics of Study Population. Table 4.3 provides a description of the 
 
demographic characteristics of the study sample (weighted n = 10,123). The study sample 
was nearly evenly divided between males (51.3%) and females (48.7%), with a mean age 
of 15.2 years (range = 13 to 18 years). The majority were non-Hispanic Whites (65.6%) 
and citizens of the United States (95.1%). Approximately 45% of parents and/or 
guardians of participants had received a high school diploma or less and 35% had 
received a college degree. Slightly more than 84% of participants lived in either a 
metropolitan or other urban community, and more than 60% lived in households with 
incomes three times greater than the 2000 poverty level. 
Prevalence rates of DSM-IV disorders in this study population showed that 51.0% 
of participants did not meet the diagnostic criteria for any of the fifteen behavioral and 
mental health disorders assessed. Of the remaining participants, 26.7% had one disorder, 
13.1% had two disorders, 6.0% had three disorders, and 3.2% had four or five DSM 
disorders. For those who responded (n = 9878) to the question “Have you ever seriously 
considered suicide?” 12.1% (S.E. = 0.6%) responded positively. 
Table 4.3: Participant Characteristics1 
 Column 
Age (mean = 15.2 years) % N 
 13 15.2% 1652 
 14 21.0% 2218 
 15 20.5% 1887 
 16 21.0% 2010 
 17 16.8% 1758 
 85 
 18 5.4% 598 
Gender   
 Female 48.7% 4953 
 Male 51.3% 5170 
Race/Ethnicity   
 White 65.6% 5634 
 Hispanic 15.1% 1953 
 Black 14.4% 1914 
 Other 5.0% 622 
Poverty Ratio   
 < 1.5 17.1% 1716 
 <= 3 19.3% 2021 
 <= 6 31.0% 3104 
 > 6 32.6% 3282 
Urbanicity   
 Metropolitan 47.5% 4508 
 Other Urban 37.6% 3304 
 Rural 14.9% 2311 
Parent(s) Highest Education   
 < High school 15.5% 1684 
 High school graduate 29.7% 3081 
 Some college 19.4% 1998 
 College graduate 35.3% 3360 
U.S. Citizen   
 No 4.9% 409 
 Yes 95.1% 9682 
Any DSM-IV Disorder   
 No 51.0% 5186 
 Yes 49.0% 4937 
1 Unweighted N = 10,123 
 
 
Using the expanded childhood adversity assessment, 12.5% of participants 
reported no exposures to adversity (Table 4.4). Just over half of the participants (51.7%) 
of participants reported 1–3 exposures, 27.1% reported 4–6 exposures, and 8.8% of 
participants reported 7 or more exposures to childhood adversities (Table 4.4). Finally, as 
depicted in Table 4, 3.8% of participants reporting no adversity exposures indicated they 
had seriously considered suicide, 7.1% of participants reporting 1–3 adversities seriously 
 86 
considered suicide, 18.1% of those reporting 4–6 adversities serious considered suicide, 
and 34.6% of those reporting 7 or more adversities seriously considered suicide. 
Table 4.4: Prevalence of ACEs and Suicidal Ideation by ACE Group 
ACE Group  Ever seriously consider suicide?1  
Prevalence No Yes Totals 
 Est. S.E. Est. S.E. Est. S.E. Count2 
0 12.5% 0.7% 96.2% 0.8% 3.8% 0.8% 1185 
1-3 51.7% 0.9% 92.9% 0.5% 7.1% 0.5% 5149 
4-6 27.1% 0.7% 81.9% 1.1% 18.1% 1.1% 2739 
7 => 8.8% 0.7% 65.4% 2.8% 34.6% 2.8% 805 
Totals 100% 0.0% 87.9% 0.6% 12.1% 0.6%  
1 Unweighted valid N is 9878 for the sample due to missing data. 
2 All counts are unweighted valid Ns. 
 
 
Multivariate Analysis. To assess the association of cumulative adverse childhood 
 
experiences to suicidal ideation, adjusted logistic regression modeling was used to 
examine this relationship (Table 4.5). 
Table 4.5: Estimated Risk for Suicidal Ideation by ACE Group 
ACE Group Model1 
AOR 95% C.I. 
0 ACEs group Referent Referent 
1 – 3 ACES group 1.39 .811, 2.38 
4 – 6 ACEs group 2.87 1.70, 4.84 
7 or more ACEs group 4.11 2.56, 6.61 
Note: Model odds ratios adjusted for age, gender, race/ethnicity (i.e., Black, 
Hispanic and Referent = white and other), poverty ratio, parent(s) highest 
educational status, urbanicity (i.e., metropolitan, other urban and Referent = rural), 
U.S. citizenship, and number of DSM-IV disorders. 
 
1 Overall model fit: Wald (X2) = 1005.46, df = 20, p < .001; Overall classification = 
88.9%; Cox and Snell = .144, Negelkerke R2 = .277, McFadden = .212. 
 
 
This overall model fit was also statistically significant (Wald (X2) = 1005.46, df = 20, p 
 
< .001). The overall model classification was 88.9%, and the Nagelkerke’s R2 was .277, 
indicating an estimated prediction in variance of approximately 28%. Risk for suicidal 
ideation for those reporting 1–3 childhood adversities was not statistically significantly 
 87 
different from those reporting no adversities (AOR=1.39, 95% C.I. [.81, 2.38]) in this 
model. Those reporting 4–6 childhood adversities (AOR=2.87, 95% C.I. [1.70, 4.84]) and 
7 or more childhood adversities (AOR=4.11, 95% C.I. [2.56, 6.61]) were statistically 
significantly at greater risk for suicidal ideation compared to those reporting no 
exposures. 
Multi-Group Path Analysis Results. Results of the multi-group path analysis 
 
examining the influence of cumulative adversities on the relationship of social integration 
and suicidal ideation are presented in Figure 4.1 and Table 4.6. Controlling for all 
covariates, model fit statistics determined that the multi-group model was an acceptable 
fit to the data, CFI/TFI = 1.00 and RMSEA = .000, 90% C.I. [.000, .000]. Results for 
group 1 (i.e., adolescents reporting no adverse childhood experiences) found no 
statistically significant effects of the five social integration factors to risk for suicidal 
ideation. Increasing age (Est.=.140, S.E.=.083, p=.092) was mildly associated with 
increased risk for suicidal ideation. Results for group 2 (i.e., adolescents reporting 1 to 3 
adverse childhood experiences) indicated a statistically significant effect of three of the 
five social integration factors. Religious integration (Est.=- .089, S.E.=.049, p=.066) was 
mildly associated with decreased risk for suicidal ideation. Family integration 
(Est.=- .249, S.E.=.070, p < .001) and school integration (Est.=- .275, S.E.=.076, p 
 
< .000) were both statistically significantly associated with decreased risk for suicidal 
ideation. Examination of the covariates among this group found that age (Est.=.014, S.E. 
=.020, p = .041), higher parental education (Est.=.099, S.E =.035, p = .004), and DSM IV 
disorders (Est =.443, S.E.=.043, p < .001) were statistically significantly associated with 
 88 
increased risk for suicidal ideation. Females (Est.=.189, S.E.=.111, p = .088) were mildly 
associated with increased risk among this group. 
Group 3 (i.e., adolescents reporting 4 to 6 adverse childhood experiences) results 
showed fewer social integration items as protective factors of suicidal ideation. Only 
family integration (Est.=- .330, S.E. =.096, p = .001) was associated with a statistically 
significant reduction in risk for suicidal ideation among this group. Females (Est.=.377, 
S.E.=.080, p < .000) and DSM IV disorders (Est.=.361, S.E.=.060, p < .001) remained 
statistically associated with increased risk for suicidal ideation. U.S. citizenship 
(Est.=.604, S.E.=.353, p = .087) was also mildly associated with increased risk for 
suicidal ideation among this group. Finally, results for group 4 (i.e., adolescents reporting 
7 or more adverse childhood experiences) indicated none of the five social integration 
factors were associated with a reduction in adolescent risk for suicidal ideation. Both 
females (Est.=.591, S.E.=.178, p = .001) and DSM IV disorders (Est.=.341, S.E.=.102, p 
= .001) remained statistically significantly associated with increased risk for suicidal 
ideation among this high adversity exposure group. 
Given these results, additional analyses were conducted to assess the moderating 
effect of ACE categories on the relationship of each social integration item and 
adolescent suicidal ideation. Pairwise comparisons of parameter estimates were found to 
be insignificant, indicating that a statistically significant moderating effect of cumulative 
adversities was not present in this study. 
 89 
 
 
Figure 4.1: ACE Multigroup Path Analysis Results for Influence of Social Integration on Suicidal Ideation  
 
 
 
 
Religion 
ACE0 = ns 
ACE1 = - .089† 
ACE2 = ns 
Social ACE3 = ns 
Integration Family Suicidal 
Variables Ideation ACE0 = ns 
*** 
ACE1 = - .249
*** 
ACE2 = - .330
ACE3 = ns 
School 
ACE0 = ns 
*** 
ACE1 = - .275
ACE2 = ns 
ACE3 = ns 
 
Note: ns = not significant 
Note: ACE0 = no adversities, ACE1 = 1-3 adversities, ACE2 = 4-6 adversities, ACE3 = 7 or more adversities 
Note: Unstandardized coefficients are presented 
† * ** *** 
= .10; = .05; = .01; < .001 
 90 
 
 
 
 
Table 4.6: Path Analysis Results of Social Integration on Suicidal Ideation by ACE Group1 
 Participant ACE Category 
Parameter Estimates 0 1 – 3 4 – 6 7 or more 
Est. (S.E.) Est. (S.E.) Est. (S.E.) Est. (S.E.) 
Religion → Suicidal Ideation .071 (.717) - .089 (.049) † - .010 (.125) .011 (.100) 
Family → Suicidal Ideation - .178 (.926) - .249 (.070) *** - .330 (.096) *** - .199 (.203) 
Friend → Suicidal Ideation - .305 (.510) - .026 (.058) - .006 (.120) - .112 (.196) 
Teacher → Suicidal Ideation - .456 (.996) - .109 (.082) .000 (.161) .028 (.195) 
School → Suicidal Ideation - .248 (2.243) - .275 (.076) *** - .084 (.240) - .243 (.153) 
R-Square .253 (.407) .218 (.025) *** .311 (.101) ** .340 (.078) *** 
† = .10; * = .05; ** = .01; *** < .001 
1 All estimates are unstandardized coefficients (S.E.). 
 91 
Discussion 
 
The Adverse Childhood Experiences study by Felitti and colleagues (1998) was a 
seminal study that expanded our understanding of the long and short-term impact of 
traumatic experiences during childhood. Since that time, researchers have sought to better 
understand how these experiences influence behavioral, psychological, and physiological 
responses to such exposures, and the factors that are protective to such exposures. This 
study specifically examined the effects of cumulative childhood adversity on the 
relationship of social integration or connectedness to adolescent suicidal ideation. 
This study confirms several findings in the existing literature. It confirms 
adversity during childhood is not uncommon (Felitti et al., 1998; Finkelhor et al., 2009; 
Finkelhor, Shattuck, Turner, & Hamby, 2015; Turner, Finkelhor, & Ormrod, 2010). 
Using an expanded childhood adversity assessment, eighty-eight percent of adolescents 
in this study reported experiencing at least one traumatic or stressful event in their 
lifetime, with another thirty-five percent reporting four or more adversities. Higher levels 
of childhood adversity were associated with increased risk for suicidal ideation. Risk for 
suicidal ideation among youth reporting four to six adversities were nearly three times the 
risk of those reporting no adversities, and more than four times the risk for youth 
reporting seven or more adversities compared to youth reporting no exposures to 
adversity. Finally, this study found social integration or connectedness as a protective 
factor against suicidal ideation was partially supported. In other words, the protective 
benefits of higher perceived social integration were not consistent across varying 
 92 
adversity risk groups and only three of the five protective factors significantly reduced 
risk for suicidal ideation. 
While this study supports the findings that family and school connectedness or 
integration act as protective factors for suicidal ideation (Czyz, Liu, & King, 2012; 
Eisenberg, Ackard, & Resnick, 2007; Eisenberg & Resnick, 2006; Kaminski et al., 2010; 
Matlin et al., 2011) this study indicates that these influences may vary as exposure to 
adversities increase. Results indicated individuals experiencing lower levels of adversity 
(i.e., 1 to 3 adversities) demonstrated a significant reduction in risk for suicidal ideation 
as perceived family, school and to a lesser extent, religious social integration increased. 
However, the benefits of these protective factors were not maintained for youth reporting 
higher levels of cumulative childhood adversities. Youth reporting 4 to 6 adversities 
experienced a statistically significant reduction in suicidal ideation from higher perceived 
family integration only, while youth reporting 7 or more adversities experienced no 
benefits from higher perceived family or school integration. These findings suggest as 
substantial negative childhood exposures accumulate the positive influences of school, 
family and other forms of social integration or connectedness may be mitigated by 
uncontrolled biological and psychological responses often associated with higher levels 
of stress. Such responses may overwhelm individual systems and coping skills and may 
diminish and interrupt healthy interactions with existing protective factors. 
Finally, findings from this study did not support the hypothesized effect of higher 
perceived teacher or peer integration. Across all childhood adversity groups, higher 
perceived teacher and peer integration were not found to be statistically significant 
 93 
protective factors against suicidal ideation. However, these results may be due in part to 
how these two factors were operationalized in this study. Further, prior studies focused 
on these factors have also demonstrated inconsistent findings (Eisenberg et al., 2007; 
McNeely & Falci, 2009). 
For example, McNeely and colleagues (2009), conducted a longitudinal study to 
assess teacher support and social belonging as protective factors for several health risk 
behaviors among adolescents. While they found teacher support was a protective factor 
for initiation of sexual intercourse, weapon related violence, initiation of cigarette, 
drinking and marijuana use, and suicide attempts, it was not a protective factor for 
suicidal ideation. However, another study focused specifically on the protective influence 
of family, school safety, adult caring, and teacher caring for suicidal ideation and suicide 
attempts among youth with a history of sexual abuse did find “teacher caring” to be a 
protective factor (Eisenberg et al., 2007). 
Similarly, studies on peer integration or connectedness as a protective factor to 
suicidal ideation has also found mixed results (Whitlock, Wyman, & Moore, 2014). A 
search for recent studies found four studies that examined peer support as a protective 
factor for suicidal ideation. Of these studies, two determined peer support was not a 
protective factor for suicidal ideation (Miller, Esposito-Smythers, & Leichtweis, 2015; 
Rojas et al., 2017), one study found peer connectedness to be a statistically significant 
protective factor (Kaminski et al., 2010), and the final study determined peer 
connectedness was a complex relationship that may be contingent on other factors and 
context (Czyz et al., 2012). For example, the longitudinal study of suicidal adolescents by 
 94 
Czyz and colleagues (2012) found the influence of peer connectedness as a protective 
factor for suicidal ideation varied across gender, severity of baseline suicidal ideation, 
baseline status of multiple suicide attempts, and changes in perceived family and peer 
connectedness over time. Given the existing, yet limited evidence in the literature, further 
investigations on the influence of teacher and peer connectedness on suicidal ideation are 
warranted. 
The findings of this study are important given the current focus of public health 
prevention and treatment efforts on the role of social connectedness as a protective factor 
for poor health outcomes resulting from exposures to adversity during childhood. 
Organizations such as the Center on the Developing Child at Harvard University have 
identified and endorsed the presence of a caring supportive adult and relationships with 
other caring individuals (i.e., immediate and extended family members, early care 
providers, teachers, and community members) as the most salient factor in supporting 
positive and healthy child development in the face of adversity (Center on the 
Developing Child, 2007; National Scientific Council on the Developing Child, 2007). Yet 
the findings of this study suggest the positive effects of various forms of social support 
and connectedness may be greatly attenuated by the harmful effects of high levels of 
adversity. 
In this study, approximately 27% of participants reported experiencing four to six 
adversities in their lifetime, with another 9% reporting seven or more exposures. 
Research, program and policy efforts designed to address the harmful influence of 
childhood adversities, often times do not consider the social and environmental 
 95 
conditions of children and adolescents exposed to untenable adversities. Many prevention 
and treatment programs identify the specific behavior or health condition as the focus of 
their efforts without considering the breath of contributing factors or root causes. For a 
significant number of youth, exposures to childhood adversity not only include many of 
the adversities assessed in this study, but other co-occurring adversities such as 
intractable intergenerational poverty, various types and intersecting forms of 
discrimination and oppression, persistent lack of access to quality education and housing, 
inadequate health related systems and care, inconsistent and inadequate access to 
transportation and other social systems, and living in isolated communities faced with a 
number of issues such as food deserts and high rates of unemployment and violence. 
The impact of high levels of cumulative adversity can also be seen among some 
of our most socially disconnected youth – youth disproportionately impacted by school 
disciplinary policies and youth involved in the juvenile justice system – both of which are 
important predictors of lifelong health and well-being for children. A recently released 
report focused on school discipline disparities among Black youth by the Government 
Accountability Office (GAO) found that high rates of trauma related behaviors (i.e. 
increased anxiety, problems with self-regulation, aggressive and self-destructive 
behaviors, and a number of mental health problems such as depression) were reported by 
school administrators as a growing problem among school populations, and important 
drivers of school suspensions and expulsions (Office of U. S. Government 
Accountability, 2018). Recent studies among juvenile justice involved youth, who often 
come from poor communities of color, have also found exceptionally high rates of 
 96 
childhood adversities among both boys and girls (Baglivio et al., 2014; Cannon, 2016). 
For example, an assessment of 220 incarcerated youth found 86% had experienced four 
or more adversities during their childhood – a figure that is seven times higher than that 
found in the original Adverse Childhood Experiences study (Baglivio et al., 2014). 
Future efforts to address childhood adversity must be focused on out most 
vulnerable populations it our society is to make significant progress towards reducing 
adversity and trauma related morbidity and mortality. This will require substantial 
investments in research focused on uncovering the root cause of childhood adversity, and 
more importantly, in finding solutions that not only reduce but eliminate toxic levels of 
trauma and stress. Further this will require a fundamental shift in our beliefs. In other 
words, such a shift will require that our primary focus will move from increasing 
resiliency to investing in healthy viable communities that provide the social context 
necessary for health childhood development. 
Limitations 
 
There are several limitations that must be considered when interpreting the results 
of this study. First, this study is a cross-sectional design. This type of design limits our 
ability to determine temporal precedence and causation. However, these results are 
consistent with other studies assessing the influence of childhood exposures to adversity 
and the role of social integration or connectedness as a protective factor against negative 
health related outcomes including suicidal ideation and behaviors (Dube et al., 2001; 
Eisenberg et al., 2007; Felitti MD et al., 1998; Hardt et al., 2008a; Johnson et al., 2002; 
Rojas et al., 2017; Waldrop et al., 2007). 
 97 
A second limitation is possible recall bias due to the retrospective reporting of 
exposures to various stressors and adversities. While this potential bias is present with 
any retrospective assessment, exposures in this study were assessed during adolescence 
rather than adulthood, providing a shorter time span between actual events and reporting. 
Further, studies have indicated retrospective assessments of traumatic or stressful events 
is an appropriate approach (Dube et al., 2004). Along with recall bias, this study relied on 
self-report data that was not confirmed with other sources. Given the nature of many of 
the questions used for this study, under reporting of adversities and negative social 
relationships are a possibility. 
Instrument bias is also a serious consideration for this study. Since this study was 
a secondary analysis, the assessment of the social integration constructs was restricted to 
items included on the NCS-A questionnaire. Both teacher and peer integration scale items 
were derived from only three questions. Further, the three peer integration items did not 
reflect the influence of deviant peer relationships which could have a negative influence 
on thoughts of suicidal ideation. However, methods used to construct these items have 
been used previously (Rose et al., 2014). However, future studies with instruments that 
have been validated is recommended. 
Conclusion 
 
Much research has been focused on identifying factors that increase individual 
resilience of children exposed to adversity. Yet the findings of this study suggest the 
positive effects of various forms of social support and the harmful effects of untenable 
adversity may significantly attenuate connectedness. While individual and family 
 98 
resilience should remain a priority, such approaches place the primary burden of 
responsibility on those most impacted (i.e., the children) by toxic levels of cumulative 
adversities and traumas – many of which may be a product of existing social conditions. 
Given the findings of this study and what is currently known about toxic stress on 
child development, future research must focus on addressing the ecological factors that 
are the drivers of adversity and trauma during childhood. Early identification and 
intervention as a public health program and policy approach should be practiced as a 
strategy to reduce the lifelong harms of childhood adversity, especially among those 
working in early childhood health fields (Johnson, Riley, Granger, & Riis, 2013). Further, 
future public health investments must continue to link children and families who face 
significant adversities to supportive services such as trauma informed programs and 
systems such as the Nurse-Family Partnership program and trauma-focused cognitive 
behavioral therapeutic models. Each of these modalities have been shown to be 
promising or evidence-based prevention and treatment programs (Garner, 2013; Olds et 
al., 2013; Salloum, Scheeringa, Cohen, & Storch, 2014). Finally, and most importantly, 
there must be a paradigm shift in our thinking as a society where it is no longer 
acceptable for our youth to experience multiple childhood adversities. We must make the 
necessary social, financial and policy commitments to eliminate these exposures that so 
seriously harm our youth and adults. 
 99 
Chapter 5: Summary 
 
Using a secondary analysis of the National Comorbidity Survey Replication 
Adolescent Supplement (NCS-A), this dissertation sought to expand the existing 
literature focused on the impact of adverse childhood experiences and the role of social 
integration or connectedness as a protective factor to adolescent suicidal ideation. This 
chapter is a summary of the results of the two studies presented in this dissertation, the 
limitations, and implications for future research. 
Study #1 Aim: To determine if an expanded assessment of adverse childhood experiences 
 
better predicts adolescent suicidal ideation. 
 
The first study examined the influence of additional adversities on suicidal 
ideation by constructing a 20-item adversity assessment that included nine of the original 
ACE adversities and eleven additional adversities assessed in the NCS-A study. The nine 
items assessed in this study representing the original ACE adversities were physical, 
emotional, and sexual abuse; physical neglect; parental divorce; parental mental health 
disorders; parental drug and alcohol abuse; parental domestic violence; and parental 
criminal and arrest history. Additional exposures examined in the analyses included 
community violence, death of someone close, living in a single parent household, 
homelessness, foster care involvement, food scarcity and other items such as war related 
adversities, natural disasters, and life-threatening accidents and illness. Both individual 
items and cumulative adversity were examined as risk factors for suicidal ideation. 
H.1.a. An expanded assessment of childhood adversity (i.e., the 20-item 
 
assessment) will better predict variance in risk for suicidal ideation than a shorter 
 100 
assessment of childhood adversity (i.e., the 9-item assessment based on the original ACE 
assessment). 
This hypothesis was partially supported by the findings in this study. Based on 
prior research, it was expected that the expanded adversity assessment would be a better 
predictor of suicidal ideation compared to the nine-item assessment. For example, 
Finkelhor and colleagues (2013) found that an expanded childhood adversity assessment 
was a better predictor of variance in childhood distress (34%) compared to a less 
exhaustive adversity assessment (21%). However, adjusted multivariate analyses 
determined the nine-item assessment predicted approximately 19% of variance in suicidal 
ideation, while the twenty-item assessment predicted approximately 21% of variance. 
H.1.b. Adolescents with higher cumulative childhood adversities will 
 
demonstrate greater risk for suicidal ideation compared to those with fewer cumulative 
childhood adversities. 
This hypothesis was supported by the findings in this study. As expected, 
exposure to childhood adversities was high among this cohort of adolescents. Using the 
twenty-item adversity assessment eighty-eight percent of study participants indicated 
they had been exposed to at least one adversity assessed in the NCS-A study. Thirty-five 
percent had been exposed to four or more adversities. Physical abuse by a parent or 
guardian, sibling, or intimate partner were extremely high in this cohort, along with 
exposure to community violence, emotional abuse, domestic violence between parents or 
guardian, and parental mental health disorders. 
 101 
Univariate analysis determined that all of the adversities, except for living in a 
single parent household and exposure to war related adversities, were significantly 
associated with suicidal ideation. A Multivariate analysis that included all twenty 
adversity items found that five of the original childhood adversity items (i.e., parental 
domestic violence, physical abuse, sexual abuse, parental mental health disorders, and 
physical neglect) and three items from the expanded assessment (i.e., death or stressful or 
life-threatening experience, like kidnapping, torture or rape of another significant person, 
a life-threatening or serious illness or injury, and “other stressors or traumas” not 
assessed by the NCS-A study) were each significant predictors of suicidal ideation. 
Cumulative adversities also had the expected dose-response relationship to the 
outcome variable. As cumulative exposures to adversities increased, so increased 
individual risk for suicidal ideation. However, it is noted that participants reporting seven 
or more adversities were at substantially higher risk for suicidal ideation compared to 
those reporting lower cumulative adversities. In the first full regression model adjusting 
for all covariates, the risk for suicidal ideation for those reporting seven or more 
adversities was nearly twice the risk of those reporting six or fewer exposures to 
adversity. It is also noted that participants reporting four or more cumulative adversities 
were at greater risk for suicidal ideation compared to any single adversity item assessed 
in this study, indicating that cumulative exposures to adversity may be a greater risk than 
single exposures to adversity. 
 102 
Study #2 Aim: To determine if social integration (i.e., family, school, 
 
religious/spiritual, teacher, and peer integration) influences the impact of adverse 
childhood experiences on risk for suicidal ideation among adolescents. 
The second study was guided by Durkheim’s theory of social integration and 
regulation, with a focus on social integration as a protective factor of suicidal ideation. 
Principal components analysis was used to generate five new social integration items 
believed to be significant influences in healthy child development. These new social 
integration variables were family, school, religious/spiritual, peer, and teacher integration 
or connectedness. Mplus path analysis was used to examine the hypotheses regarding the 
protective effect of the five social integration items on the relationship of cumulative 
childhood adversities and suicidal ideation. 
H.2.a Higher perceived family, school, religious/spiritual, teacher, and peer 
 
integration will each decrease risk for suicidal ideation among adolescents. 
 
This hypothesis was partially confirmed by this study. While each social 
integration factor was expected to significantly reduce risk for suicidal ideation, only 
three of the five social integration items were found to be significant in this study. Higher 
perceived family, school and to a lesser extent, religious/spiritual integration were found 
to be significant protective factors for suicidal ideation. In other words, as perceived 
integration or connectedness increased for these factors, risk for suicidal ideation 
decreased. Finally, peer and teacher connectedness were not found to be positive 
protective factors for suicidal ideation regardless of the number of cumulative childhood 
adversity exposures. While this finding was not predicted for this study, it is in line with 
 103 
the literature. The limited extant literature on the effects of peer and teacher integration or 
connectedness as a protective factor of suicidal ideation is inconclusive, meriting further 
research. 
H.2.b Higher perceived social integration, while attenuated, will decrease risk for 
 
suicidal ideation among youth with a greater number of cumulative childhood adversities. 
 
This hypothesis is partially confirmed by the findings in this study. Family, 
school, and religious/spiritual integration or connectedness were protective factors of the 
relationship between cumulative childhood adversities and suicidal ideation. As 
exposures to childhood adversities increased, this positive relationship of the three social 
integration factors were attenuated. However, we found it surprising that none of the 
social integration factors provided positive benefit among those reporting higher levels of 
adversity. The findings of this study demonstrated that family, school, and 
religious/spiritual integration decreased risk for suicidal ideation among youth reporting 
fewer than six adversities but did not reduce risk for youth reporting seven or more 
adversities. 
Research Implications 
 
Over the past few decades, there has been a dramatic growth in our 
understanding about the influences of childhood adversity on behavioral and mental 
health outcomes, including suicidal ideation. There is greater understanding about 
childhood adversities’ influence on physiological and psychological development of 
children during their formative developmental years, and much research has been 
conducted to better understand the factors that build resiliency in children faced with 
 104 
adversity or traumatic events. Prevention and treatment efforts have and continue to focus 
on the influence of healthy relationships and connectedness as the key factor to healthy 
coping skills for youth facing adversity. However, the findings of this study call into 
question this approach as the panacea to negative ecologic factors that have damaging 
impacts on health and wellbeing across the life-course. 
Future research efforts therefore must continue to expand on our understanding 
about the ecological factors that may cause trauma during children’s formative 
developmental years. Such research efforts require new approaches to identifying what 
individuals perceive to be stressful or traumatic experiences during childhood. The use of 
mixed methods study designs (i.e., the use of both quantitative and qualitative methods) 
is a suggested approach to providing a deeper understanding of various traumatic 
exposures, as well as a better understanding of those factors considered most supportive 
to prevention and recovery. 
Investment in longitudinal studies must also be a focus of future research efforts. 
 
Many studies focused on the harms of childhood adversity are often cross-sectional, 
which limits our ability to determine causal relationships, and are subject to numerous 
limitations. Further, future studies must focus on examining risk and protective factors 
across diverse communities (i.e., race/ethnicity, gender, sexual orientation, rural versus 
urban communities, etc.). This will ensure that prevention and treatment efforts are 
appropriately designed to meet the specific needs of diverse communities and address the 
specific environments in which people live. 
 105 
Research must also focus on developing standardized tools and assessments that 
can be used by systems and professionals (i.e., pediatricians and other medical staff, 
education administrators, therapist, etc.) working with youth to better assess the full 
breadth of adversities that place children at increased risk for poor behavioral and health 
outcomes. This will ensure appropriate interventions and treatments can be initiated in 
the early developmental years. 
Finally, the previously mentioned research efforts should be multi-disciplinary in 
nature and must cross multiple systems of care and services. Children and families at 
greatest risk for multiple traumas and adversities, often times move through numerous 
public systems that provide limited solutions to the multiple problems they face. 
Coordinated efforts across systems would not only put us in a better position to address 
possible gaps in needed services, they would also be more efficient and cost effective. 
Limitations 
There are a few limitations to the findings presented in this dissertation. The 
cross-sectional nature of the study limits our ability to determine causal relationships 
between adverse childhood experiences and suicidal ideation, as well as other risk and 
protective factors examined in this study. However, these results are consistent with other 
studies assessing cumulative childhood exposures to adversity and health related 
outcomes including suicidal ideation and behaviors (Dube et al., 2001; Felitti et al., 
1998). 
This study also primarily relied on self-report data that was assessed 
retrospectively. This increases risk for poor recall of prior events, and under and/or over 
 106 
reporting of certain behaviors, experiences, and perceptions on certain variables used for 
this study. However, this study was conducted during adolescence rather than during 
adulthood, providing a shorter time span between actual events and reporting. Further, 
studies have indicated that retrospective assessments of traumatic or stressful events is an 
appropriate approach (Dube et al., 2004) 
Conclusion 
 
The role of cumulative childhood adversities as a driver of many behavioral and 
mental health conditions and social issues must be recognized as a public health 
epidemic. Many of the adversities and traumas faced by children are often socially driven 
and avoidable. While our current goal of building resiliency is a worthy endeavor, 
reducing and/or eliminating childhood adversities must become a social and policy 
priority if we truly are committed to making a significant and permanent shift in trauma 
related morbidity and mortality. However, such an endeavor will require a clear 
commitment from community leaders, practitioners, and policymakers across disciplines 
and social systems to begin addressing the issues that drive many of the factors that are so 
harmful to our children, families and ultimately, our communities. 
 107 
Appendices 
 108 
Appendix A: Methods 
 
The purpose of this dissertation was to expand existing knowledge on factors that 
may influence adolescent health. Using a nationally representative sample of youth 
between the ages of 13 and 18, an expanded 20-item adverse childhood experiences index 
was generated to explore the role of childhood adversities as risk factors for suicidal 
ideation. Using Durkheim’s theory of social integration and social regulation as a guide, 
this dissertation also examined the role of five key ecological influences in the lives of 
adolescents – family, school, religious, teacher and peer integration - as potential 
protective factors to the relationships of childhood adversities and adolescent suicidal 
ideation. The following paragraphs provide a brief description of the NCS-A study and 
the methods used for the current dissertation (i.e., the conceptual model, study sample, 
and methods of analyses). 
The National Comorbidity Survey Replication Adolescent Supplement Study 
 
This dissertation involves secondary analysis of the National Comorbidity Survey 
Replication Adolescent Supplement (NCS-A) study (Kessler, 2013). Data for the NCS-A 
were collected between February 2001 and January 2004 by the Survey Research Center 
of the Institute for Social Research at the University of Michigan and was an add-on to 
the National Comorbidity Survey Replication (NCS-R) study, an investigation of the 
prevalence and correlates of mental health disorders among adults in the US (Kessler, 
2013). The NCS-A study was considered the first nationally representative study on the 
prevalence, correlates and patterns of service use for DSM-IV mental health disorders 
among U.S. adolescents (Kessler, Avenevoli, Costello, et al., 2009a). Further, this 
 109 
seminal study was designed to provide the groundwork for follow-up studies of risk and 
protective factors, consequences, and early expressions of adult mental disorders (Kessler 
et al., 2009a, 2009b; Merikangas, Avenevoli, Costello, Koretz, & Kessler, 2009). 
NCS-A Eligibility and Sampling Frame. 
 
Participation in the NCS-A study was restricted to English-speaking, non- 
institutionalized adolescents between the ages of 13-18 (n=10,148) living in the United 
States. A dual-framed, complex cluster sampling methodology was used to draw a 
sample of adolescents for the study. The first sampling-frame included adolescents 
(n=904) living in households where an adult had participated in the National Comorbidity 
Survey-Replication (NCS-R) study. Respondents to the NCS-R were selected from a 
four-stage area probability sample of the non-institutionalized civilian population using 
small area 2000 census data collected from the US Bureau of the Census. A probability 
sample of 84 primary sampling units (PSUs) and pseudo-PSUs representative of the US 
population were selected. Each PSU included all counties in a census-defined 
metropolitan statistical area (MSA) or individual counties representing themselves. 
Individual housing units were then identified and entered into a computer system. A 
random sample of housing units was then selected and all residents 18 years of age and 
older were identified in each housing unit. Probability sampling was then used to select 
one or two individuals within a housing unit to participate in the NCS-R study. Students 
living in campus group housing were also eligible to participate in the study if their 
permanent housing address was a housing unit selected for the study (Kessler et al., 
2004). Further details regarding the sampling frame for the NCS-R study can be found in 
 110 
Kessler’s article on the design and field procedures used for the NCS-R study (Kessler et 
al., 2004). 
To supplement the low number of interviews completed by adolescents living in 
households of adult participants in the NCS-R study, a second sampling-frame of 
adolescents (n=9244) were recruited to the NCS-A study from a population of students 
attending 320 schools located in the same counties in which the NCS-R was conducted 
(Kessler et al., 2009b). Eligible schools included all accredited private and public middle 
schools, junior high schools and high schools within the NCS-R counties. A probability 
sampling methodology proportional to the size of the student body in the classes relevant 
to the target population was used to select schools. This sample of schools included both 
unblinded schools (i.e., schools in which student rosters were provided to investigators to 
select and recruit adolescents and their families to participate in the study) and blinded 
schools (i.e., student identities were unknown until a signed consent was obtained from 
the school principals). Forty to fifty students were randomly selected from each school to 
participate in the NCS-A study. The overall adolescent response rate (i.e., both household 
and student sample populations) was 75.6%; the majority of non-responders (21.3%) 
were refusals to participate in the study with the remaining 3.2% unable to participate due 
to “circumstances” or “non-contact” (Kessler et al., 2009b). 
NCS-A Adolescent and Parent Supplemental Questionnaires 
 
The adolescent questionnaire included assessments of the known correlates to 
mental health of children and the fully structured Composite International Diagnostic 
Interview (CIDI) personifying fifteen DSM-IV disorders. The specific disorders assessed 
 111 
via the CIDI include mood disorders (major depressive disorder and dysthymia, bipolar I- 
II disorder and subthreshold bipolar disorder), anxiety disorders (panic disorder with or 
without agoraphobia, agoraphobia without panic disorder, social phobia, specific phobia, 
generalized anxiety disorder (GAD), post-traumatic stress disorder, separation anxiety 
disorder), behavior disorders (attention deficit hyperactivity disorder, oppositional defiant 
disorder, conduct disorder, eating disorders) and substance abuse disorders (alcohol and 
drug abuse with and without dependency) (Kessler et al., 2012). All diagnoses were 
made using the DSM-IV distress and impairment criteria and organic exclusion rules 
(i.e., if a symptomatic episode occurs as a result of factors such as physical illness, injury, 
or the use of medication, drugs, alcohol, then the participant is not coded as having a 
disorder even if they meet all other criteria) (Kessler et al., 2012). Further, it is noted that 
diagnostic hierarchy rules were not used due to an interest in studying co-morbidity 
among hierarchy-free disorders (i.e., all disorders were coded as present even if they only 
occurred in the presence of another disorder) (Kessler et al., 2012). 
In the presence of a parent or guardian, adolescent interviews were administered 
face-to-face in the home of each respondent using a laptop, computer-assisted personal 
interview (CAPI). Trained survey interviewers from the Survey Research Center of the 
Institute of Social Research at the University of Michigan administered each adolescent 
interview. Survey data was collected on a number of factors considered relevant to 
adolescent behavior and mental health including individual level factors (e.g. 
demographics, developmental factors, cognitive and academic abilities-achievements, 
physical health, and stressful life events), family level factors as reported by the 
 112 
adolescent (e.g. family structure, stability and adaptability, parenting behaviors, parental 
psychopathology, and family stress), and finally, environmental level factors such as 
school and neighborhood characteristics (Merikangas et al., 2009). 
On average, the NCS-A student interviews took approximately two and a half 
hours to complete, with interviews ranging between 69 minutes to 347 minutes (Kessler 
et al., 2009b). This variation in interview length was primarily due to the number of 
lifetime disorders assessed for each adolescent that was determined based on an initial 
screening tool administered at the beginning of each interview. Adolescent participants 
were each provided a $50 incentive for their participation (Kessler et al., 2009b; 
Merikangas et al., 2009). 
NCS-A Limitations 
 
Limitations of the NCS-A as noted by the Merkangas and colleagues (2009) 
included: 1) exclusion of children 12 years of age and under, which restricts the 
generalizability of findings; 2) the cross-sectional design of the survey prevented the 
assessment of temporal precedence of risk and protective factors and their associations to 
outcomes of interest; and 3) parental/guardian information was collected from only one 
parent/guardian using a self-administered questionnaire, which did not allow for 
clarification of responses and may have introduced bias. 
Dissertation Study Design 
Conceptual Model 
Durkheim’s Theory of Social Integration was used to guide the analyses used in 
 
this dissertation. This theory posits that involvement in formal and informal organizations 
 113 
provide necessary social supports and networks for positive wellbeing and coping 
(Durkheim, 1951). Individuals who are socially integrated into their social and 
environmental context are less likely to feel disconnected from their surroundings and 
have healthier dispositions and outcomes compared to those who are less socially 
integrated (Durkheim, 1951). In this study, it is proposed adolescents who experience 
Figure 2: Conceptual Model Based on Durkheim’s Theory of Social Integration1 
 
Social Integration Health Status 
 
Family School Religion 
Cumulative 
Adverse Childhood 
Experiences Suicidal 
 Ideation 
0 ACEs 
1-3 ACEs 
4-6 ACEs 
7 >= ACEs Peer Teacher 
 
 
1 Control Variables: age, gender, race/ethnicity, poverty ratio, parent(s) highest education status, urbanicity, 
US Citizen, comorbid DSM IV disorder 
 
stressful events in the form of adverse experiences who are more highly integrated in 
their social context – families, school, teacher, religious/spiritual connections, and peer 
networks – will be at lower risk of suicidal ideation compared to adolescents who are less 
well integrated in their social context (Figure 2). This study also proposes that while 
social integration will continue to act as a protective factor, higher exposures to adversity 
will significantly attenuate this relationship. 
Study Variables 
Similar to prior studies, variables representing the constructs of interest for this 
study were constructed using items from the NCS-A questionnaire (Rose et al., 2014). 
 114 
These new variables included five social integration variables representing the constructs 
of family integration, school integration, religious/spiritual integration, teacher 
integration, and peer integration. 
Adverse Experiences. The “Adverse Experiences” variables used for this study 
 
were constructed using 46 observed exposures to adversity assessed in the NCS-A 
questionnaire (Table 2). Similar to the coding scheme in prior studies, each potential 
exposure was coded as a dichotomous variable (i.e., 0 = no exposure, 1 = exposure) and 
summed (Felitti et al., 1998). Higher ACE scores represented more exposures to adverse 
experiences and the range of ACE scores was 0 to 20. However, due to the small cell 
sizes of those reporting higher levels of adversity, all participants report 8 or more 
adversities were recoded into the category of 7 or more adversities. Therefore, for this 
dissertation ACE scores ranged between 0 to 7 or more adversities. 
Social Integration Variables. Similar to prior studies, this dissertation examined 
 
three social contexts found to play a significant role in positive adolescent development 
(i.e., family, school, and religious/spiritual integration) (Rose et al., 2014). This 
dissertation expanded on Rose and colleague’s findings by also examining the role of 
positive peer and teacher integration on risk for suicidal ideation. The new social 
integration items for family, school, religious/spiritual, teacher and peer integration were 
derived from observed items assessed in the NCS-A questionnaire. 
Family Integration. The “Family Integration” scaled variable included observed 
variables representing perceived constructs such as family emotional support, 
engagement, and decision-making (Table 3: Family Integration). Response options for 
 115 
each item were measured on a 5-point Likert scale (i.e., 1 = “All of the time”, 2 = “Most 
of the time”, 3 = “Some of the time”, 4 = “A little bit of the time”, and 5 = “Never”). 
Items were reverse coded for directional consistency with the other integration variables. 
Higher scores represent higher levels of positive “family integration.” 
Table 3: Family Integration 
How often did family members feel very close to each other? (reverse) 
How often could family members talk to each other about their feelings? (reverse) 
How often did family members let each other know when they were sad or worried? 
(reverse) 
How often did family members keep their feelings to themselves? 
How often did the whole family do things together? (reverse) 
How often did family members share interests and hobbies with each other? (reverse) 
How often did family members avoid each other? 
How often did family members go along with what the family decided to do? (reverse) 
How often did family members find it easy to express their opinions to each other? 
(reverse) 
How often did each family member have input in major family decisions? (reverse) 
How often did children have a say in their discipline? (reverse) 
How often did everyone compromise when there were disagreements? (reverse) 
How often was it difficult to get everyone to agree on decisions? 
 
 
School Integration. The variable “School Integration” was constructed using 
observed variables representing constructs such as academic achievement and school 
bonding (Table 4: School Integration). One observed questionnaire item – “What sort of 
grades did you get in your last year of school?” - will be used to measure academic 
achievement. Response options for “That sort of grades did you get in your last years at 
school?” was recoded to match a 4-point response option scale and were 1.33 = below 
average, 2.66 = average, and 3.99 = above average. The remaining items had response 
 116 
options of 1 = very, 2 = somewhat, 3 = not very, and 4 = not at all. Where indicated, 
items were reverse coded so that higher scores represented higher levels of positive 
“school integration.” 
Table 4: School Integration 
What sort of grades did you get in your last years at school? 
I like(d) school. (reverse coded) 
Getting good grades (is/was) important to me. (reverse coded) 
Homework (is/was) a waste of time. 
I try(ied) hard at school. (reverse coded) 
I Feel/felt as if I don’t/didn’t belong at school. 
Most of the things I learn(ed) in school are unimportant. 
 
 
Teacher Integration. The variable “Teacher Integration” was constructed using 
observed variables representing how connected the participant felt towards their teach 
(Table 5: Teacher Integration). Three observed questionnaire items were used to measure 
teacher integration. Response options for these items included 1 = very, 2 = somewhat, 3 
= not very, and 4 = not at all. Where indicated, response options were reverse coded so 
that higher scores represented higher levels of positive “teacher integration.” 
Table 5: Teacher Integration 
Most of my teachers treat(ed) me fairly (reverse) 
I care(d) a lot about what my teachers (think/thought) about me. (reverse) 
I like(d) my teachers. (reverse) 
 
 
Peer Integration. The variable “Peer Integration” was constructed using observed 
variables representing constructs such as peer emotional support items (Table 6). 
Response options for the first two items were 1 = a lot, 2 = some, 3 = a little, 4 = not at 
all. The last item was recoded to match a 4-option response pattern and included .8 = 
always, 1.6 = most of the time, 2.4 = sometimes, 3.2 = not very often, and 4 = never. 
 117 
Where indicated, response options were reverse coded so that higher scores represented 
higher levels of positive “peer integration.” 
Table 6: Peer Integration 
How much can you rely on your friend(s) for help if you have a serious problem? 
(reverse) 
How much can you open up to your friend(s) if you need to talk about your worries? 
(reverse) 
When you have a problem or worry, how often do you let your friend(s) know about it? 
(reverse) 
 
 
Religious/Spiritual Integration. The variable “Religious/Spiritual Integration” 
was constructed using observed variables representing constructs such as 
religious/spiritual engagement, importance, and emotional support (Table 7: 
Religious/Spiritual Integration). Religious/spiritual engagement was derived from the 
item – “How often do you usually attend religious services?” Response options for this 
variable were recoded to match a 4-point scale and were .8 = less than once a week, 1.6 = 
about once a week, 2.4 = 1 to 3 times a month, 3.2 = less than once a month, and 4 = 
never. The response options for the remaining items were 1 = very important, 2 = 
somewhat important, 3 = not very important, and 4 = not at all important. Where 
indicated, response options were reverse coded so that higher scores indicated more 
positive religious/spiritual integration. 
Table 7: Religious/Spiritual Integration 
How often do you usually attend religious services? 
In general, how important are religious or spiritual beliefs in your daily life? 
When you have problems or difficulties in your family, work, etc. how often do you 
seek comfort through religious or spiritual means such as prayer, meditating, attending 
service, or talking to religious/spiritual advisor? 
When you have decisions to make in your daily life, how often do you think about 
what your religious or spiritual beliefs suggest you should do? 
 118 
 
 
 
Control Variables. Several variables were examined as potential confounding 
 
variables. These variables were adolescent race/ethnicity, age at interview, adolescent 
educational level, highest parental/guardian education, U.S. citizenship (yes/no), and type 
of region of residency (i.e., metropolitan [large city], other urban area [suburbs, small 
city, town/village], and rural area). Family household income as a ration to the 2000 
census poverty threshold was also examined as a potential confounding variable. In the 
NCS-A study, family household income was reported on the parent questionnaire and the 
appropriate annual national poverty threshold based on family size was created by the 
NCS-A research team. A categorical variable called “poverty” was generated by the 
original study investigators that included four income-to-poverty ratio response options 
(i.e., 1=“Low income” or less than 1.5 times the poverty line, 2=“Low to average 
income” or 1.5 to 3 times the poverty line, 3=“Average to high income” or 3 to 6 times 
the poverty line, and 4=“High income” or more than 6 times the poverty line) (Kessler et 
al., 2012). The “poverty” variable was examined as a potential confounder. 
Finally, a modified version of the World Health Organization (WHO) Composite 
International Diagnostic Interview (CIDI) and a supplemental parent questionnaire were 
used to assess symptomology for fifteen DSM-IV disorders among adolescents 
participating in the NCS-A study. The questionnaires assessed symptoms occurring 
within the prior 30-days, 12-months, and ever during the participants’ lifetime. This 
dissertation focused on disorder outcomes over the lifetime. All diagnoses were made by 
NCS-A investigators at the completion of each interview using the DSM-IV distress and 
 119 
impairment criteria and organic exclusion rules (Kessler et al., 2012). Final diagnostic 
outcomes were coded as “1” if the adolescent was confirmed as meeting DSM-IV criteria 
for a disorder and “5” if the adolescent did not meet the criteria for a disorder. For this 
dissertation, response options were recoded to “1” confirming a diagnosis and “0” for 
those not confirmed as meeting diagnostic criteria. A final cumulative DSM IV disorder 
variable was created for each participant indicating the total number of disorders 
diagnosed by the original NCS-A research team. Due to the small number of participants 
with five disorders, those with 4 and 5 disorders were collapsed into one category (i.e., 4 
or more disorders) for statistical purposes. The range for the final cumulative DSM 
disorder variable was 0 to 4. 
Table 8: Lifetime Prevalence of Estimated DSM-IV Disorder Types/Subtypes (%, SE) 
Fear Disorders (26.1%, 1.0) 
 Specific phobia (19.9%, 1.0) 
 Agoraphobia (with or without panic disorder) (2.6%, .04) 
 Social phobia (8.5%, 0.6) 
 Panic disorder (with or without agoraphobia) (2.4%, .02) 
Distress Disorders (25.4%, 0.9) 
 Separation anxiety disorder (7.6%, 0.5) 
 Post-traumatic stress disorder (4.7%, 0.4) 
 Major depressive episode/dysthymia (18.6%, 1.1) 
 Generalized anxiety disorder (2.2%, 0.4) 
Behavior Disorders (22.7%, 1.3) 
 Attention deficit hyperactivity disorder (ADHD) (8.1%, 0.6) 
 Oppositional defiant disorder (12.6%, 0.9) 
 Conduct disorder (6.8%, 0.9) 
 Eating disorders (5.1%, 0.4) 
(anorexia nervosa, bulimia nervosa and binge-eating disorder) 
Other Disorders (6.2%, 0.4) 
 Bipolar disorder (bipolar I, II and subthreshold) (6.2%, 0.4) 
Substance Disorders (11.4%, 0.9) 
 Alcohol abuse with or without dependency (6.1%, 0.5) 
 Drug abuse with or without dependency (8.9%, 0.8) 
 120 
Using a significance level of .05, confounding variables found to be significantly 
related to one or more of the three outcome variables of interest (i.e., behavior disorders, 
distress disorders, and suicidality) were controlled for in the final analyses. 
Data Analysis Plan 
Sampling Weights 
To address the dual-framed, clustered sampling methodology used to select 
participants in the study, the NCS-A investigators generated sample weights to account 
for the complex nature of the dataset (Kessler et al., 2009a). This study used the sample 
weight generated for the full adolescent sample (i.e., sample of adolescents from both the 
NCS-R household sample and the student sample). The full sample weight was generated 
from the student sample weight and the household sample weight which were generated 
using 2000 census data and select indicators to adjust each sample to be more 
representative of the general population. A final full sample weight 
(FINAL_UNBLENDED) was generated by combining the student and household weights 
and adjusting for discrepancies between the final student and household sample weights. 
Finally, to adjust for clustering effects, the investigators generated a stratum variable 
(STR) and cluster variable (SECU) that will also be used in the analyses (Kessler et al., 
2009a). 
Missing Data and Data Assumption 
 
Missing data for each independent and dependent variable were assessed at the 
initiation of the analyses. It was determined that less than 5% of the data were missing, 
and therefore it was assumed missing data were “missing at random” (MAR) (i.e., when 
 121 
missing observations on a variable X differ from the observed scores on that variable 
only by chance) or “missing completely at random” (MCAR) (i.e., when the presence 
versus absence of data on X is unrelated to any other variable in the data set) (Kline, 
2010). Multicollinearity (Tolerance: R2smc < .10 and Variance Inflation Factor: 1/1-R2smc 
> 10) was examined prior to initiation of the analyses (Kline, 2010). Age and grade level 
were found to be highly correlated with high tolerance and variance inflation factor 
scores. Given these results, grade level was removed from the analyses. 
Scale Reliability, Descriptive and Bivariate Analyses 
 
The complex sample module of the Statistical Package for the Social Sciences 
(SPSS) 22.0 was used for descriptive and bivariate analyses. Prior to initiation of the 
analyses, all social integration scaled items were examined using principal components 
analysis (i.e., direct oblimin) and scale reliability. Next, bivariate associations were 
analyzed to determine if variables identified as potential confounders should be 
controlled for in the multivariate analyses, and to determine if the main predictors of 
interest were associated with suicidal ideation. Pearson’s correlations coefficients and 
Chi-square were used to assess associations of continuous and categorical variables. 
Statistical significance was assessed at the p ≤ 0.05 level. 
Multivariate Logistic Regression and Path Analysis 
For paper #1, the complex sample module of the Statistical Package for the Social 
Sciences (SPSS) version 24 was used for the analyses (“IBM Release notes - IBM SPSS 
Statistics 24.0 - United States,” 2016). Several logistic regression models, controlling for 
covariates, were conducted to assess the relationship of exposure to adversity during 
 122 
childhood and the dependent variable “suicidal ideation.” Univariate analyses were 
conducted to obtain adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for 
each of the twenty adversity items to determine if there was a significant relationship to 
the dependent variable. Next, two full regression models were conducted first with the 9- 
item ACE assessment and then the 20-item ACE assessment. Model fit and R2 statistics 
were examined to determine if including all 20 adversity items was a better predictor of 
risk for suicidal thoughts when compared to the model with the 9-item ACE assessment. 
Finally, logistic modeling was used to examine the graded cumulative effect of the 20- 
item ACEs assessment on risk for suicidal ideation. Model results include mean and 
prevalence estimates, model statistics, overall classification, R2 estimates, and odds ratios 
with 95% confidence intervals. 
For paper #2, Mplus version 7.4 was used to test the hypothesis using multi-group 
path analyses (Muthén & Muthén, 2018). The full sample was retained for this analysis 
and full information maximum likelihood (FIML) method was used to handle missing 
data. Weighted least squares (WLSMV) was used to estimate the path coefficients given 
the complex sampling methodology and the inclusion of both continuous and categorical 
variables. As recommended by Hu and Bentler (1999), data to model fit was examining 
using comparative fit index (CFI > .95) and the root mean square error of approximation 
(RMSEA < .06) indices (Hu & Bentler, 1999). Parameter estimates were examined and 
statistically significant path coefficients and R-square values for each multi-group 
analysis are reported (Kline, 2010). 
 123 
Power Analysis 
 
The online power calculator “Free Statistics Calculators version 4.0 – Calculator: 
A-Priori Sample Size for Structural Equation Models” 
(http://www.danielsoper.com/statcalc/calculator.aspx?id=89) by Daniel Sopel was used 
to estimate the necessary sample size for the proposed analyses. Required information for 
the estimated sample size includes the 1) anticipated effect size, 2) desired statistical 
power level, 3) number of latent variables, 4) number of observed variables, and 5) the 
probability level. Estimates were provided for the minimum sample size to detect the 
desired effect size, the minimum sample size for the model structure, and the 
recommended minimum sample size for the analyses. 
To determine the largest sample size necessary for the proposed analyses, sample 
size estimates were calculated for the hypothesis (hypothesis 2b) with the largest number 
of observed variables (i.e., adversity = 21, family integration = 13, school integration = 
16, peer integration = 7, religious/spiritual integration = 4, internal and external locus of 
control = 10, behavioral and distress disorders = 7, suicidality =1, and socioeconomic 
status (i.e., family income) = 1) and largest number of latent variables (10). Estimates 
were calculated for small, medium and large effect sizes (see Table 9). A sample size of 
703 is recommended to detect a small, medium and large effect size with a statistical 
power level of .80 and probability level of .05. If potential control variables (i.e., age, 
education, gender, maternal education, paternal education, urbanicity, and two additional 
disorders) are included in the calculation the estimated sample size increases to a 
recommended minimum sample size of 1,012 (see Table 10). 
 124 
Table 9: Estimated Sample Size Calculations (without control variables) 
 Small Medium Large 
Anticipated effect size 0.1 0.3 0.5 
Desired statistical power level 0.80 0.80 0.80 
Number of latent variables 10 10 10 
Number of observed variables 80 80 80 
Probability level 0.05 0.05 0.05 
Minimum sample to detect effect 703 67 18 
Minimum sample size for model structure 700 700 700 
Recommended minimum sample size 703 700 700 
 
 
Table 10: Estimated Sample Size Calculations (including control variables) 
 Small Medium Large 
Anticipated effect size 0.1 0.3 0.5 
Desired statistical power level 0.80 0.80 0.80 
Number of latent variables 10 10 10 
Number of observed variables 88 88 88 
Probability level 0.05 0.05 0.05 
Minimum sample to detect effect 703 67 18 
Minimum sample size for model structure 1,012 1,012 1,012 
Recommended minimum sample size 1,012 1,012 1,012 
 
 
Human Subjects Concerns 
 
This dissertation is a secondary data analysis of the National Comorbidity Survey 
Replication Adolescent Supplement study (NCS-A) (Kessler, 2013). The NCS-A is 
maintained by the Inter-university Consortium for Political and Social Research (ICPSR) 
at the University of Michigan, Institute for Social Research (Kessler, 2013). To limit the 
risk of participant disclosure, data maintained by ICPSR undergo a confidentiality review 
and are altered when necessary. Prior to conducting this study approval was obtained 
from the ICPSR Institutional Review Board and from the University of Maryland, 
College Park Institutional Review Board (IRB). 
 125 
Appendix B: Principal Components Analysis Results 
 
 
This study theorized that adolescent social integration included a number of social 
factors that could potentially provide the support and connection deemed necessary for 
healthy child development. These five factors of adolescent social integration were 
family integration, school integration, religious/spiritual integration, teacher integration, 
and peer integration. While the NCS-A questionnaire did not include instruments 
designed to assess adolescent social integration in these five areas of interest, several 
items within the NCS-A were used as proxy measures to construct the five scaled 
integration variables. 
The following tables and summaries are the final results of a principal 
components analysis (PCA) conducted for each of the five integration variables. Direct 
oblimin rotation was used for these analyses. A summary of each PCA, Kaiser-Meyer- 
Olkin (KMO) statistics (i.e., measure of sampling adequacy), Bartlett’s Test of Sphericity 
(i.e., test of the null hypothesis that the scale variables are uncorrelated), and the 
component and/or rotated component matrix are presented (when applicable). The 
Cronbach’s Alpha for each scaled integration item is also presented. 
 126 
Religious Integration Scale Results 
 
Four items were identified as representing religious/spiritual integration. The 
Kaiser-Meyer-Olkin measure of sampling adequacy was .811 and the Bartlett’s Test of 
Sphericity was statistically significant (X2 = 19368.90, df = 6, p < .001). The total 
variance explained was 71.70% and all of the items loaded on one component. 
Cronbach’s Alpha = .882 
 
 
 
 
Component Matrix 
Items Component 
1 
During difficult times, seek comfort in religion .895 
Important of religion in daily life .885 
Decision making guided by religious beliefs .891 
Freq. attend religious services .699 
 
 
Peer Integration Scale Results 
 
Three items were identified as representing peer integration. The Kaiser-Meyer- 
Olkin measure of sampling adequacy was .606 and the Bartlett’s Test of Sphericity was 
statistically significant (X2 = 8761.40, df = 3, p < .001). The total variance explained was 
66.01% and all of the items loaded on one component. 
Cronbach’s Alpha = .632 
 
Component Matrix 
Items Component 
1 
How much can rely on friends when have serious problem .866 
How much can you open up to friends/talk about worries .891 
How often call/hang out/get together socially with friends .661 
 127 
School Integration Scale Results 
 
Twelve items were identified as representing family integration. The Kaiser- 
Meyer-Olkin measure of sampling adequacy was .884 and the Bartlett’s Test of 
Sphericity was statistically significant (X2 = 62893.21, df = 21, p < .001). The total 
variance explained was 68.02% and all of the items loaded on one component factors. 
Cronbach’s Alpha = .748 
Rotated Component Matrix 
Items Component 
1 
Grade in last or current year at school .551 
I like school .915 
Getting good grades is important to me .926 
Homework is a waste of time .766 
I try hard at school .943 
I feel as if I do not belong at school .760 
Most of the things I learn in school are unimportant .842 
 
Teacher Integration Scale Results 
 
Three items were identified as representing teacher integration. The Kaiser- 
Meyer-Olkin measure of sampling adequacy was .771 and the Bartlett’s Test of 
Sphericity was statistically significant (X2 = 29247.07, df = 3, p < .001). The total 
variance explained was 90.31% and all of the items loaded on one component. 
Cronbach’s Alpha = .665 
 
Component Matrix 
Items Component 
1 
I like my teachers .955 
Most of my teachers treat me fairly .942 
I care a lot about what my teachers think about me .953 
 128 
Family Integration Scale Results 
 
Twelve items were identified as representing family integration. The Kaiser- 
Meyer-Olkin measure of sampling adequacy was .958 and the Bartlett’s Test of 
Sphericity was statistically significant (X2 = 77832.10, df = 66, p < .001). The total 
variance explained was 57.71% and all of the items loaded on one component. 
Cronbach’s Alpha = .855 
 
 
 
 
Component Matrix 
Items Component 
1 
How is relationship with parents/guardian .623 
How often family members felt close to each other .803 
How often family members did things together .849 
How often family members avoided each other at home .780 
How often family members willing did what family decided .816 
How often family members shared interests and hobbies .857 
How often family members easily expressed opinions .841 
How often family members each had input on major decisions .741 
How often family members compromised .699 
How often family members talked about feelings .810 
How often family members talk when sad/worried .822 
How often family members kept feelings to themselves .671 
 129 
Appendix C: Institutional Review Board Approval Letters 
 130 
 1204 Marie Mount Hall 
College Park, MD 20742-5125 
TEL 301.405.4212 
FAX 301.314.1475 
irb@umd.edu 
 www.umresearch.umd.edu/IRB 
 
 
INSTITUTIONAL REVIEW BOARD 
 
 
 
DATE: June 2, 2017 
 
TO: Kathleen Washington 
FROM: University of Maryland College Park 
(UMCP) IRB 
 
 
 
PROJECT TITLE: [909882-2] Exploring the Effects of Poverty, Social Integration, and 
Locus of Control on Childhood Adversity and Other Stress Inducing 
Events 
REFERENCE #: 
SUBMISSION TYPE: Continuing Review/Progress Report 
 
 
ACTION: 
APPROVE 
D APPROVAL DATE: June 2, 
2017 
EXPIRATION DATE: June 22, 2018 
REVIEW TYPE: Expedited Review 
 
REVIEW CATEGORY: Expedited review category # 5 
 
 
Thank you for your submission of Continuing Review/Progress Report materials for this project. 
The University of Maryland College Park (UMCP) IRB has APPROVED your submission. This 
approval is based on an appropriate risk/benefit ratio and a project design wherein the risks have 
been minimized. All research must be conducted in accordance with this approved submission. 
 
Prior to submission to the IRB Office, this project received scientific review from the 
departmental IRB Liaison. 
 
This submission has received Expedited Review based on the applicable federal regulations. 
 
 
131 
 
This project has been determined to be a Minimal Risk project. Based on the risks, this project 
requires continuing review by this committee on an annual basis. Please use the appropriate forms 
for this procedure. Your documentation for continuing review must be received with sufficient 
time for review and continued approval before the expiration date of June 22, 2018. 
 
Please remember that informed consent is a process beginning with a description of the project 
and insurance of participant understanding followed by a signed consent form. Informed 
consent must continue throughout the project via a dialogue between the researcher and 
research participant. Unless a consent waiver or alteration has been approved, Federal 
regulations require that each participant receives a copy of the consent document. 
 
Please note that any revision to previously approved materials must be approved by this 
committee prior to initiation. Please use the appropriate revision forms for this procedure. 
 
All UNANTICIPATED PROBLEMS involving risks to subjects or others (UPIRSOs) and 
SERIOUS and UNEXPECTED adverse events must be reported promptly to this office. Please 
use the appropriate reporting forms for this procedure. All FDA and sponsor reporting 
requirements should also be followed. 
 
All NON-COMPLIANCE issues or COMPLAINTS regarding this project must be reported 
promptly to this office. 
Please note that all research records must be retained for a minimum of seven years after 
the completion of the project. 
 
If you have any questions, please contact the IRB Office at 301-405-4212 or 
irb@umd.edu. Please include your project title and reference number in all 
correspondence with this committee. 
 
 
 
 
 
This letter has been electronically signed in accordance with all applicable regulations, and a copy is retained within University of 
Maryland College Park (UMCP) IRB's records. 
 132 
 
Shakira Washington 
<shakira@nationalcrittenton.org> 
 
 
[ICPSR Access Request:26816] Exploring the Effects of Poverty, Social 
Integration, and Locus of Control on Childhood Adversity and Other 
Stress Inducing Events Modifications Approved 
2 messages 
petrinko@umich.edu <petrinko@umich.edu> Fri, Aug 
11, 2017 at 4:26 PM To: desmond@umd.edu 
Cc: shakira@nationalcrittenton.org, 
petrinko@umich.edu Dear Sharon 
M. Desmond; 
 
Your data access request for Exploring the Effects of Poverty, Social Integration, 
and Locus of Control on Childhood Adversity and Other Stress Inducing Events has 
been approved, as modified. If you requested additional datasets as part of this 
modification, we will contact you shortly with instructions for accessing these data. 
Please contact Shelly Petrinko (petrinko@umich.edu) if you have any 
difficulty with this process. Thank you, 
Shelly Petrinko 
Inter-university Consortium for 
Political and Social Research 
University of Michigan 
P.O. Box 1248 
Ann Arbor, MI 48106 
 133 
 
References 
 
 
Alaimo, K., Briefel, R. R., Frongillo Jr, E. A., & Olson, C. M. (1998). Food Insufficiency 
Exists in the United States: Results from the Third National Health and 
Nutrition Examination Survey (NHANES III). American Journal of Public 
Health, 88(3), 419. 
Alaimo, K., Olson, C. M., Frongillo Jr, E. A., & Briefel, R. R. (2001). Articles - Food 
Insufficiency, Family Income, and Health in US Preschool and School-Aged 
Children. American Journal of Public Health, 91(5), 781. 
American Psychiatric Association. (2013). Diagnostic and Statistical Manual of 
Mental Disorders, 5th Edition: DSM-5 (5 edition). Washington, D.C: American 
Psychiatric Publishing. 
Anda, R. F. (1999). Adverse Childhood Experiences and Smoking During 
Adolescence and Adulthood. JAMA, 282(17), 1652. 
https://doi.org/10.1001/jama.282.17.1652 
Anda, R. F., Brown, D. W., Dube, S. R., Bremner, J. D., Felitti, V. J., & Giles, W. H. (2008). 
Adverse Childhood Experiences and Chronic Obstructive Pulmonary Disease 
in Adults. American Journal of Preventive Medicine, 34(5), 396–403. 
https://doi.org/10.1016/j.amepre.2008.02.002 
Anda, R. F., Brown, D. W., Felitti, V. J., Bremner, J. D., Dube, S. R., & Giles, W. H. (2007). 
Adverse Childhood Experiences and Prescribed Psychotropic Medications in 
Adults. American Journal of Preventive Medicine, 32(5), 389–394. 
https://doi.org/10.1016/j.amepre.2007.01.005 
 134 
Anda, R. F., Fleisher, V. I., Felitti, V. J., Edwards, V. J., Whitfield, C. L., Dube, S. R., & 
Williamson, D. F. (2004). Childhood Abuse, Household Dysfunction, and 
Indicators of Impaired Adult Worker Performance. The Permanente Journal, 
8(1), 30–38. 
Anda, R., Tietjen, G., Schulman, E., Felitti, V., & Croft, J. (2010). Adverse Childhood 
Experiences and Frequent Headaches in Adults: October 2010. Headache: The 
Journal of Head and Face Pain, 50(9), 1473–1481. 
https://doi.org/10.1111/j.1526-4610.2010.01756.x 
Araújo, J. P., Lourenço, P., Azevedo, A., Friões, F., Rocha-Gonçalves, F., Ferreira, A., & 
Bettencourt, P. (2009). Prognostic Value of High-Sensitivity C-Reactive 
Protein in Heart Failure: A Systematic Review. Journal of Cardiac Failure, 
15(3), 256–266. https://doi.org/10.1016/j.cardfail.2008.10.030 
Baglivio, M. T., Epps, N., Swartz, K., Huq, M. S., Sheer, A., & Hardt, N. S. (2014). The 
prevalence of adverse childhood experiences (ACE) in the lives of juvenile 
offenders. Journal of Juvenile Justice, 3(2), 1. 
Baiden, P., Stewart, S. L., & Fallon, B. (2017). The mediating effect of depressive 
symptoms on the relationship between bullying victimization and non- 
suicidal self-injury among adolescents: Findings from community and 
inpatient mental health settings in Ontario, Canada. Psychiatry Research, 255, 
238–247. https://doi.org/10.1016/j.psychres.2017.05.018 
Barth, R. P., Lloyd, E. C., Green, R. L., James, S., Leslie, L. K., & Landsverk, J. (2007). 
 
Predictors of Placement Moves Among Children With and Without Emotional 
 135 
and Behavioral Disorders. Journal of Emotional and Behavioral Disorders, 
15(1), 46–55. https://doi.org/10.1177/10634266070150010501 
Beautrais, A. L. (2000). Risk Factors for Suicide and Attempted Suicide among Young 
People. Australian & New Zealand Journal of Psychiatry, 34(3), 420–436. 
https://doi.org/10.1080/j.1440-1614.2000.00691.x 
Beautrais, A. L. (2016). Risk Factors for Suicide and Attempted Suicide among Young 
People. Australian & New Zealand Journal of Psychiatry, 34(3), 420–436. 
Berasain, C., Castillo, J., Perugorria, M. J., Latasa, M. U., Prieto, J., & Avila, M. A. (2009). 
Inflammation and Liver Cancer: New Molecular Links. Annals of the New York 
Academy of Sciences, 1155(1), 206–221. https://doi.org/10.1111/j.1749- 
6632.2009.03704.x 
Berkman, L. F., & Syme, S. L. (1979). Social networks, host resistance, and mortality: 
a nine-year follow-up study of Alameda County residents. American Journal of 
Epidemiology, 109(2), 186–204. 
Berkman, Lisa F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social 
integration to health: Durkheim in the new millennium☆. Social Science & 
Medicine, 51(6), 843–857. https://doi.org/10.1016/S0277-9536(00)00065-4 
Betancourt, T. S., & Khan, K. T. (2008). The mental health of children affected by 
armed conflict: Protective processes and pathways to resilience. 
International Review of Psychiatry, 20(3), 317–328. 
 
Bethell, C. D., Newacheck, P., Hawes, E., & Halfon, N. (2014). Adverse childhood 
experiences: Assessing the impact on health and school engagement and the 
 136 
mitigating role of resilience. Health Affairs, 33(12), 2106–2115. 
https://doi.org/10.1377/hlthaff.2014.0914 
Briere, J. N., & Scott, C. (2014). Principles of Trauma Therapy: A Guide to Symptoms, 
Evaluation, and Treatment. SAGE Publications. 
Briggs, M. A., Granado-Villar, D. C., Gitterman, B. A., Brown, J. M., Chilton, L. A., Cotton, 
 
W. H., … Zind, B. (2013). Providing Care for Children and Adolescents Facing 
Homelessness and Housing Insecurity. Pediatrics, 131(6), 1206–1210. 
https://doi.org/10.1542/peds.2013-0645 
Brown, D. W., Anda, R. F., Edwards, V. J., Felitti, V. J., Dube, S. R., & Giles, W. H. (2007). 
 
Adverse childhood experiences and childhood autobiographical memory 
disturbance. Child Abuse & Neglect, 31(9), 961–969. 
https://doi.org/10.1016/j.chiabu.2007.02.011 
Buka, S. L., Stichick, T. L., Birdthistle, I., & Earls, F. J. (2001). Youth exposure to 
violence: Prevalence, risks, and consequences. American Journal of 
Orthopsychiatry, 71(3), 298–310. https://doi.org/10.1037/0002- 
9432.71.3.298 
 
Burke, N. J., Hellman, J. L., Scott, B. G., Weems, C. F., & Carrion, V. G. (2011). The 
impact of adverse childhood experiences on an urban pediatric population. 
Child Abuse & Neglect, 35(6), 408–413. 
https://doi.org/10.1016/j.chiabu.2011.02.006 
Cannon, Y. (2016). Adverse Childhood Experiences in the New Mexico Juvenile Justice 
Population (p. 12). Retrieved from 
 137 
https://nmsc.unm.edu/reports/2016/adverse-childhood-experiences-in- 
the-new-mexico-juvenile-justice-population.pdf 
Center for Disease Control and Prevention (CDC). (2015a). 10 Leading Causes of 
Death by Age Group, United States - 2015. Retrieved from 
https://www.cdc.gov/injury/wisqars/pdf/leading_causes_of_death_by_age_g 
roup_2015-a.pdf 
Center for Disease Control and Prevention (CDC). (2015b). Behavioral Risk Factor 
Surveillance System Survey Questionnaire. Atlanta, Georgia: U.S. Department 
of Health and Human Services, Centers for Disease Control and Prevention. 
Retrieved November 17, 2015, from 
http://www.cdc.gov/brfss/questionnaires/index.htm 
Center for Disease Control and Prevention (CDC). (2015c). Suicide - Facts at a 
Glance. Retrieved from 
https://www.cdc.gov/violenceprevention/pdf/suicide-datasheet-a.pdf 
Center on the Developing Child. (2007). The Impact of Early Adversity on Children’s 
Development (InBrief). Retrieved April 1, 2018, from 
https://developingchild.harvard.edu/resources/inbrief-the-impact-of-early- 
adversity-on-childrens-development/ 
Centers for Disease Control and Prevention (CDC). (n.d.). About the Study. Child 
Maltreatment|Violence Prevention|Injury Center|CDC. Retrieved January 19, 
2016, from http://www.cdc.gov/violenceprevention/acestudy/about.html 
 138 
Chapman, D. P., Wheaton, A. G., Anda, R. F., Croft, J. B., Edwards, V. J., Liu, Y., … Perry,  
 
G. S. (2011). Adverse childhood experiences and sleep disturbances in adults. 
 
Sleep Medicine, 12(8), 773–779. https://doi.org/10.1016/j.sleep.2011.03.013 
Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., & Anda, R. F. 
(2004). Adverse childhood experiences and the risk of depressive disorders 
in adulthood. Journal of Affective Disorders, 82(2), 217–225. 
https://doi.org/10.1016/j.jad.2003.12.013 
Chen, E., & Miller, G. E. (2007). Stress and inflammation in exacerbations of asthma. 
 
Brain, Behavior, and Immunity, 21(8), 993–999. 
https://doi.org/10.1016/j.bbi.2007.03.009 
Child Trends. (2013). Adverse Experiences: Indicators on Children and Youth. 
 
Retrieved from http://www.childtrends.org/?indicators=adverse- 
experiences 
Child Trends. (2015). Homeless Children and Youth. Retrieved from 
http://www.childtrends.org/wp- 
content/uploads/2015/01/112_Homeless_Children_and_Youth.pdf 
Children’s Bureau. (2014). Child Welfare Outcomes 2009-2012. Report to Congress. 
Safety. Permanency. Well-Being. U.S. Department of Health & Human Services, 
Administration on Children and Families. Retrieved from 
http://www.acf.hhs.gov/sites/default/files/cb/cwo09_12.pdf 
Choi, N. G., DiNitto, D. M., Marti, C. N., & Segal, S. P. (2017). Adverse childhood 
experiences and suicide attempts among those with mental and substance 
 139 
use disorders. Child Abuse & Neglect, 69, 252–262. 
https://doi.org/10.1016/j.chiabu.2017.04.024 
Clarke, A. T. (2006). Coping with Interpersonal Stress and Psychosocial Health 
Among Children and Adolescents: A Meta-Analysis. Journal of Youth and 
Adolescence, 35(1), 10–23. http://dx.doi.org/10.1007/s10964-005-9001-x 
Cloitre, M., Stolbach, B. C., Herman, J. L., Kolk, B. van der, Pynoos, R., Wang, J., & 
Petkova, E. (2009). A developmental approach to complex PTSD: Childhood 
and adult cumulative trauma as predictors of symptom complexity. Journal of 
Traumatic Stress, 22(5), 399–408. https://doi.org/10.1002/jts.20444 
Cohen, S. (1988). Psychosocial models of the role of social support in the etiology of 
physical disease. Health Psychology, 7(3), 269–297. 
https://doi.org/10.1037/0278-6133.7.3.269 
Cohen, S. (2004). Social Relationships and Health. American Psychologist, 59(8), 
676–684. 
Cohen, S., Underwood, L. G., Gottlieb, B. H., & Fetzer Institute. (2000). Social support 
measurement and intervention: a guide for health and social scientists. Oxford; 
New York: Oxford University Press. 
Committee on Psychosocial Aspects of Child and Family Health, G. A. (2012). Early 
childhood adversity, toxic stress, and the role of the pediatrician: translating 
developmental science into lifelong health. Early Childhood Adversity, Toxic 
Stress, and the Role of the Pediatrician: Translating Developmental Science into 
Lifelong Health., 129(1), e224-31. https://doi.org/10.1542/peds.2011-2662 
 140 
Compas, B. E. (2006). Psychobiological Processes of Stress and Coping: Implications 
for Resilience in Children and Adolescents--Comments on the Papers of 
Romeo &amp; McEwen and Fisher et al. Annals of the New York Academy of 
Sciences, 1094(1), 226–234. https://doi.org/10.1196/annals.1376.024 
Cook, J. T., Black, M., Chilton, M., Cutts, D., Ettinger de Cuba, S., Heeren, T. C., … Frank, 
 
D. A. (2013). Are Food Insecurity’s Health Impacts Underestimated in the U.S. 
Population? Marginal Food Security Also Predicts Adverse Health Outcomes 
in Young U.S. Children and Mothers. Advances in Nutrition: An International 
Review Journal, 4(1), 51–61. 
Courtney, M. E., Piliavin, I., Grogan-Kaylor, A., & Nesmith, A. (2001). Foster youth 
transitions to adulthood: a longitudinal view of youth leaving care. Child 
Welfare, 80(6), 685–717. 
Courtois, C. A., & Ford, J. D. (2009). Treating complex traumatic stress disorders: an 
evidence-based guide. New York: Guilford Press. Retrieved from Table of 
contents http://catdir.loc.gov/catdir/toc/ecip0823/2008030409.html 
Crittenden, C. N., Pressman, S. D., Cohen, S., Janicki-Deverts, D., Smith, B. W., & 
Seeman, T. E. (2014). Social integration and pulmonary function in the 
elderly. Health Psychology, 33(6), 535–543. 
https://doi.org/10.1037/hea0000029 
Cronholm, P., Forke, C. M., Wade, R., Bair-Merritt MH, Davis M, Harkins-Schwarz M, 
 
… Fein JA. (2015). Adverse Childhood Experiences: Expanding the Concept of  
 
Adversity. American Journal of Preventive Medicine, 49(3), 354–361. 
 141 
Czyz, E. K., Liu, Z., & King, C. A. (2012). Social connectedness and one-year 
trajectories among suicidal adolescents following psychiatric hospitalization. 
Journal of Clinical Child and Adolescent Psychology: The Official Journal for the 
Society of Clinical Child and Adolescent Psychology, American Psychological 
Association, Division 53, 41(2), 214–226. 
https://doi.org/10.1080/15374416.2012.651998 
Danese, A., Moffitt, T. E., Pariante, C., Ambler, A., Poulton, R., & Caspi, A. (2008). 26. 
Elevated inflammation levels in depressed adults with a history of childhood 
maltreatment. Brain, Behavior, and Immunity, 22(4, Supplement), 7. 
https://doi.org/10.1016/j.bbi.2008.04.028 
Dashiff, C., DiMicco, W., Myers, B., & Sheppard, K. (2009). Poverty and Adolescent 
Mental Health. Journal of Child and Adolescent Psychiatric Nursing, 22(1), 23– 
32. https://doi.org/10.1111/j.1744-6171.2008.00166.x 
Dietz, P. M. (1999). Unintended Pregnancy Among Adult Women Exposed to Abuse 
or Household Dysfunction During Their Childhood. JAMA, 282(14), 1359. 
https://doi.org/10.1001/jama.282.14.1359 
Dong, M. (2004). Insights Into Causal Pathways for Ischemic Heart Disease: Adverse 
Childhood Experiences Study. Circulation, 110(13), 1761–1766. 
https://doi.org/10.1161/01.CIR.0000143074.54995.7F 
Dong, Maxia, Anda, R. F., Felitti, V. J., Dube, S. R., Williamson, D. F., Thompson, T. J., … 
Giles, W. H. (2004). The interrelatedness of multiple forms of childhood 
abuse, neglect, and household dysfunction. Child Abuse & Neglect, 28(7), 771– 
784. 
 142 
Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P., Williamson, D. F., & Giles, W. H. 
(2001). Childhood Abuse, Household Dysfunction, and the Risk of Attempted 
Suicide Throughout the Life Span: Findings From the Adverse Childhood 
Experiences Study. JAMA, 286(24), 3089. 
https://doi.org/10.1001/jama.286.24.3089 
Dube, S. R., Fairweather, D., Pearson, W. S., Felitti, V. J., Anda, R. F., & Croft, J. B. 
(2009). Cumulative childhood stress and autoimmune diseases in adults. 
Psychosomatic Medicine, 71(2), 243–250. 
https://doi.org/10.1097/PSY.0b013e3181907888 
Dube, S. R., Felitti, V. J., Dong, M., Chapman, D. P., Giles, W. H., & Anda, R. F. (2003). 
Childhood Abuse, Neglect, and Household Dysfunction and the Risk of Illicit 
Drug Use: The Adverse Childhood Experiences Study. Pediatrics, 111(3), 564– 
572. https://doi.org/10.1542/peds.111.3.564 
Dube, S. R., Miller, J. W., Brown, D. W., Giles, W. H., Felitti, V. J., Dong, M., & Anda, R. F. 
(2006). Adverse childhood experiences and the association with ever using 
alcohol and initiating alcohol use during adolescence. Journal of Adolescent 
Health, 38(4), 444.e1-444.e10. 
https://doi.org/10.1016/j.jadohealth.2005.06.006 
 
Dube, S. R., Williamson, D. F., Thompson, T., Felitti, V. J., & Anda, R. F. (2004). 
Assessing the reliability of retrospective reports of adverse childhood 
experiences among adult HMO members attending a primary care clinic. 
Child Abuse & Neglect, 28(7), 729–737. 
 143 
Dunn, E. C., Gilman, S. E., Willett, J. B., Slopen, N. B., & Molnar, B. E. (2012). The 
Impact of Exposure to Interpersonal Violence on Gender Differences in 
Adolescent-Onset Major Depression: Results from the National Comobidity 
Survey Replication (NCS-R). Depression and Anxiety, 29(5), 392–399. 
https://doi.org/10.1002/da.21916 
Durkheim, É. (1951). Suicide, a study in sociology: Glencoe, Ill.,: Glencoe, Ill.,. 
 
Easton, S. D., & Renner, L. M. (2013). Factors from Durkheim’s Family Integration 
Related to Suicidal Ideation among Men with Histories of Child Sexual Abuse. 
Suicide and Life-Threatening Behavior, 43(3), 336–346. 
https://doi.org/10.1111/sltb.12020 
Edidin, J., Ganim, Z., Hunter, S., & Karnik, N. (2012). The Mental and Physical Health 
of Homeless Youth: A Literature Review. Child Psychiatry & Human 
Development, 43(3), 354–375. https://doi.org/10.1007/s10578-011-0270-1 
Edwards, V. J., Anda, R. F., Felitti, V. J., & Dube, S. R. (2004). Adverse childhood 
experiences and health-related quality of life as an adult. In K. A. Kendall- 
Tackett (Ed.), Health consequences of abuse in the family: A clinical guide for 
evidence-based practice. (pp. 81–94). Washington: American Psychological 
Association. Retrieved from http://content.apa.org/books/10674-005 
Edwards, V. J., Fivush, R., Anda, R. F., Felitti, V. J., & Nordenberg, D. F. (2001). 
 
Autobiographical Memory Disturbances in Childhood Abuse Survivors. 
 
Journal of Aggression, Maltreatment & Trauma, 4(2), 247–263. 
 
Edwards, V. J., Holden, G. W., Felitti, V. J., & Anda, R. F. (2003). Relationship between 
multiple forms of childhood maltreatment and adult mental health in 
 144 
community respondents: results from the adverse childhood experiences 
study. American Journal of Psychiatry, 160(8), 1453–1460. 
Eisenberg, M. E., Ackard, D. M., & Resnick, M. D. (2007). Protective Factors and 
Suicide Risk in Adolescents with a History of Sexual Abuse. The Journal of 
Pediatrics, 151(5), 482–487. https://doi.org/10.1016/j.jpeds.2007.04.033 
Eisenberg, M. E. S. D., M. P. H., & Resnick, M. D. P. D. (2006). Suicidality among Gay, 
Lesbian and Bisexual Youth: The Role of Protective Factors. Journal of 
Adolescent Health, 39(5), 662–668. 
Evans, G. W., & Kim, P. (2013). Childhood Poverty, Chronic Stress, Self-Regulation, 
and Coping. Child Development Perspectives, 7(1), 43–48. 
https://doi.org/10.1111/cdep.12013 
Fang, X., Brown, D. S., Florence, C. S., & Mercy, J. A. (2012). The economic burden of 
child maltreatment in the United States and implications for prevention. 
Child Abuse & Neglect, 36(2), 156–165. 
https://doi.org/10.1016/j.chiabu.2011.10.006 
Fawley-King, K., & Snowden, L. R. (2012). Relationship between placement change 
during foster care and utilization of emergency mental health services. 
Children and Youth Services Review, 34(2), 348–353. 
https://doi.org/10.1016/j.childyouth.2011.11.002 
Felitti MD, F., Vincent J., Anda MD, M., Robert F., Nordenberg MD, D., Williamson MS, 
P., David F., Spitz MS, M., Alison M., Edwards BA, V., … Marks MD, M., James S. 
(1998). Relationship of Childhood Abuse and Household Dysfunction to 
Many of the Leading Causes of Death in Adults: The Adverse Childhood 
 145 
Experiences (ACE) Study. American Journal of Preventive Medicine, 14(4), 
245–258. https://doi.org/10.1016/S0749-3797(98)00017-8 
Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., … 
Marks, J. S. (1998). Relationship of childhood abuse and household 
dysfunction to many of the leading causes of death in adults: The Adverse 
Childhood Experiences (ACE) Study. American Journal of Preventive Medicine, 
14(4), 245–258. 
Finkelhor, D., Hamby, S. L., Ormrod, R., & Turner, H. (2005). The Juvenile 
Victimization Questionnaire: Reliability, validity, and national norms. Child 
Abuse & Neglect, 29(4), 383–412. 
https://doi.org/10.1016/j.chiabu.2004.11.001 
Finkelhor, D., Ormrod, R. K., & Turner, H. A. (2009). Lifetime assessment of poly- 
victimization in a national sample of children and youth. Child Abuse & 
Neglect, 33(7), 403–411. https://doi.org/10.1016/j.chiabu.2008.09.012 
Finkelhor, D., Shattuck, A., Turner, H., & Hamby, S. (2013). Improving the Adverse 
Childhood Experiences Study Scale. JAMA Pediatrics, 167(1), 70–75. 
https://doi.org/10.1001/jamapediatrics.2013.420 
Finkelhor, D., Shattuck, A., Turner, H., & Hamby, S. (2015). A revised inventory of 
Adverse Childhood Experiences. Child Abuse & Neglect, 48, 13–21. 
Finkelhor, D., Turner, H. A., Shattuck, A., & Hamby, S. L. (2015). Prevalence of 
Childhood Exposure to Violence, Crime, and Abuse: Results From the 
National Survey of Children’s Exposure to Violence. JAMA Pediatrics, 169(8), 
746–754. https://doi.org/10.1001/jamapediatrics.2015.0676 
 146 
Finkelhor, D., Turner, H., Ormrod, R., & Hamby, S. L. (2009). Violence, Abuse, and 
Crime Exposure in a National Sample of Children and Youth. Pediatrics, 
124(5), 1411–1423. https://doi.org/10.1542/peds.2009-0467 
Finkelhor D, Turner HA, Shattuck A, & Hamby SL. (2015). Prevalence of childhood 
exposure to violence, crime, and abuse: Results from the national survey of 
children’s exposure to violence. JAMA Pediatrics, 169(8), 746–754. 
https://doi.org/10.1001/jamapediatrics.2015.0676 
Fitzpatrick, K. M. (1997). Aggression and environmental risk among low-income 
African-American Youth. Journal of Adolescent Health, 21(3), 172–178. 
https://doi.org/10.1016/S1054-139X(97)00047-5 
Folkman, S. (1984a). Personal control and stress and coping processes: A theoretical 
analysis. Journal of Personality and Social Psychology, 46(4), 839–852. 
https://doi.org/10.1037/0022-3514.46.4.839 
Folkman, S. (1984b). Personal control and stress and coping processes: A theoretical 
analysis. Journal of Personality and Social Psychology, 46(4), 839–852. 
https://doi.org/10.1037/0022-3514.46.4.839 
Fraser, M. W., Richman, J. M., & Galinsky, M. J. (1999). Risk, Protection, and 
Resilience: Toward a Conceptual Framework for Social Work Practice. Social 
Work Research, 23(3), 131–143. 
Garner, A. S. (2013). Home Visiting and the Biology of Toxic Stress: Opportunities to 
Address Early Childhood Adversity. Pediatrics, 132(Supplement 2), S65–S73. 
https://doi.org/10.1542/peds.2013-1021D 
 147 
Gaylord-Harden, N. K., Cunningham, J. A., & Zelencik, B. (2011). Effects of Exposure 
to Community Violence on Internalizing Symptoms: Does Desensitization to 
Violence Occur in African American Youth? Journal of Abnormal Child 
Psychology, 39(5), 711–719. https://doi.org/10.1007/s10802-011-9510-x 
Goldston, D. B., Erkanli, A., Daniel, S. S., Heilbron, N., Weller, B. E., & Doyle, O. U. of I., 
Chicago. (2016). Developmental Trajectories of Suicidal Thoughts and 
Behaviors From Adolescence Through Adulthood. Journal of the American 
Academy of Child & Adolescent Psychiatry, 55(5), 400-407.e1. 
Gordon KH, Simonich H, Wonderlich SA, Dhankikar S, Crosby RD, Cao L, … Engel SG. 
(2016). Emotion Dysregulation and Affective Intensity Mediate the 
Relationship Between Childhood Abuse and Suicide-Related Behaviors 
Among Women with Bulimia Nervosa. Suicide & Life-Threatening Behavior, 
46(1), 79–87. 
Haller, M., & Chassin, L. (2012). A Test of Adolescent Internalizing and Externalizing 
Symptoms as Prospective Predictors of Type of Trauma Exposure and 
Posttraumatic Stress Disorder: A Test of Adolescent Symptoms as Predictors. 
Journal of Traumatic Stress, 25(6), 691–699. 
https://doi.org/10.1002/jts.21751 
Hardt, J., Sidor, A., Nickel, R., Kappis, B., Petrak, P., & Egle, U. T. (2008a). Childhood 
Adversities and Suicide Attempts: A Retrospective Study. Journal of Family 
Violence, 23(8), 713–718. https://doi.org/10.1007/s10896-008-9196-1 
 148 
Hardt, J., Sidor, A., Nickel, R., Kappis, B., Petrak, P., & Egle, U. T. (2008b). Childhood 
Adversities and Suicide Attempts: A Retrospective Study. Journal of Family 
Violence, 23(8), 713–718. https://doi.org/10.1007/s10896-008-9196-1 
Hemovich, V., & Crano, W. D. (2009). Family Structure and Adolescent Drug Use: An 
Exploration of Single-Parent Families. Substance Use & Misuse, 44(14), 2099– 
2113. 
Herman-Smith, R. (2013). Intimate Partner Violence Exposure in Early Childhood: 
An Ecobiodevelopmental Perspective. Health & Social Work, 38(4), 231–239. 
https://doi.org/10.1093/hsw/hlt018 
Hillis, S. D., Anda, R. F., Dube, S. R., Felitti, V. J., Marchbanks, P. A., & Marks, J. S. 
(2004). The Association Between Adverse Childhood Experiences and 
Adolescent Pregnancy, Long-Term Psychosocial Consequences, and Fetal 
Death. PEDIATRICS, 113(2), 320–327. 
https://doi.org/10.1542/peds.113.2.320 
 
Hillis SD. (2001). Adverse childhood experiences and sexual risk behaviors in 
women: a retrospective cohort study. Adverse Childhood Experiences and 
Sexual Risk Behaviors in Women: A Retrospective Cohort Study., 33(5), 206– 
211. 
Hillis SD, Anda RF, Felitti VJ, & Marchbanks PA. (2001). Adverse childhood 
experiences and sexual risk behaviors in women: a retrospective cohort 
study. Family Planning Perspectives, 33(5), 206-211 6p. 
 149 
Hillis SD, Anda RF, Felitti VJ, Nordenberg D, & Marchbanks PA. (2000). Adverse 
childhood experiences and sexually transmitted diseases in men and women: 
a retrospective study. Pediatrics, 106(1), E11. 
House, J. S., Robbins, C., & Metzner, H. L. (1982). The Association of Social 
Relationships and Activities with Mortality: Prospective Evidence from the 
Tecumseh Community Health Study. American Journal of Epidemiology, 
116(1), 123–140. 
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure 
analysis: Conventional criteria versus new alternatives. Structural Equation 
Modeling: A Multidisciplinary Journal, 6(1), 1–55. 
https://doi.org/10.1080/10705519909540118 
Hunt, K. L., Martens, P. M., & Belcher, H. M. E. (2011). Risky business: Trauma 
exposure and rate of posttraumatic stress disorder in African American 
children and adolescents. Journal of Traumatic Stress, 24(3), 365–369. 
https://doi.org/10.1002/jts.20648 
IBM Release notes - IBM SPSS Statistics 24.0 - United States. (2016, March 15). 
[CT754]. Retrieved January 20, 2018, from http://www.ibm.com/support, 
//www.ibm.com/support/docview.wss?uid=swg27047057 
 
Janet Kuramoto, S., Wilcox, H. C., & Latkin, C. A. (2013). Social Integration and 
Suicide-Related Ideation from a Social Network Perspective: A Longitudinal 
Study among Inner-City African Americans. Suicide and Life-Threatening 
Behavior, 43(4), 366–378. https://doi.org/10.1111/sltb.12023 
 150 
Johnson, J. G., Cohen, P., Gould, M. S., Kasen, S., Brown, J., & Brook, J. S. (2002). 
Childhood Adversities, Interpersonal Difficulties, and Risk for Suicide 
Attempts During Late Adolescence and Early Adulthood. Archives of General 
Psychiatry, 59(8), 741. https://doi.org/10.1001/archpsyc.59.8.741 
Johnson, S. B., Riley, A. W., Granger, D. A., & Riis, J. (2013). The Science of Early Life 
Toxic Stress for Pediatric Practice and Advocacy. Pediatrics, 131(2), 319–327. 
https://doi.org/10.1542/peds.2012-0469 
Kalmakis, K. A., & Chandler, G. E. (2014). Adverse childhood experiences: towards a 
clear conceptual meaning. Journal of Advanced Nursing, 70(7), 1489–1501. 
https://doi.org/10.1111/jan.12329 
Kalmakis, K. A., & Chandler, G. E. (2015). Health consequences of adverse childhood 
experiences: A systematic review. Journal of the American Association of 
Nurse Practitioners, 27(8), 457–465. https://doi.org/10.1002/2327- 
6924.12215 
 
Kalmakis KA, & Chandler GE. (2014). Adverse childhood experiences: towards a 
clear conceptual meaning. Journal of Advanced Nursing, 70(7), 1489–1501. 
Kaminski, J. W., Puddy, R. W., Hall, D. M., Cashman, S. Y., Crosby, A. E., & Ortega, L. A. 
 
G. (2010). The Relative Influence of Different Domains of Social 
Connectedness on Self-Directed Violence in Adolescence. Journal of Youth and 
Adolescence, 39(5), 460–473. https://doi.org/10.1007/s10964-009-9472-2 
Kann, L., Kinchen, S., Shanklin, S., Flint, K., Hawkins, J., Harris, W., … Zaza, S. (2014). 
MMWR 2014. Youth Risk Behavior Surveillance - United States, 2013. Division 
of Adolescent and School Health, National Center for HIV/AIDS, Viral 
 151 
Hepatitis, STD, and TB Prevention, CDC, ICF International, Rockville, 
Maryland, Westat, Rockville, Maryland. Retrieved from 
http://www.cdc.gov/mmwr/pdf/ss/ss6304.pdf 
Kelly-irving, M., Lepage, B., Dedieu, D., Bartley, M., Blane, D., Grosclaude, P., … 
Delpierre, C. (2013). Adverse childhood experiences and premature all-cause 
mortality. European Journal of Epidemiology, 28(9), 721–734. 
http://dx.doi.org/10.1007/s10654-013-9832-9 
Kessler, R. C., Avenevoli, S., McLaughlin, K. A., Green, J. G., Lakoma, M. D., Petukhova, 
M., … Merikangas, K. R. (2012). Lifetime co-morbidity of DSM-IV disorders in 
the US National Comorbidity Survey Replication Adolescent Supplement 
(NCS-A). Psychological Medicine, 42(09), 1997–2010. 
https://doi.org/10.1017/S0033291712000025 
Kessler, Ronald C. (2013). National Comorbidity Survey: Adolescent Supplement 
(NCS-A), 2001-2004, ICPSR28581-v5. Ann Arbor, MI: INter-university 
Consortium for Political and Social Research [distributer]. Retrieved from 
http://doi.org/10.3886/ICPSR28581.v5 
Kessler, Ronald C., Avenevoli, S., Costello, E. J., Green, J. G., Gruber, M. J., Heeringa, S., 
 
… Zaslavsky, A. M. (2009a). Design and field procedures in the US National  
 
Comorbidity Survey Replication Adolescent Supplement (NCS-A). 
International Journal of Methods in Psychiatric Research, 18(2), 69–83. 
https://doi.org/10.1002/mpr.279 
Kessler, Ronald C., Avenevoli, S., Costello, E. J., Green, J. G., Gruber, M. J., Heeringa, S., 
 
… Zaslavsky, A. M. (2009b). National Comorbidity Survey Replication  
 152 
Adolescent Supplement (NCS-A): II. Overview and Design. Journal of the 
American Academy of Child & Adolescent Psychiatry, 48(4), 380–385. 
https://doi.org/10.1097/CHI.0b013e3181999705 
Kessler, Ronald C., Berglund, P., Chiu, W. T., Demler, O., Heeringa, S., Hiripi, E., … 
Zheng, H. (2004). The US National Comorbidity Survey Replication (NCS-R): 
design and field procedures. International Journal of Methods in Psychiatric 
Research, 13(2), 69–92. https://doi.org/10.1002/mpr.167 
Kline, R. B. (2010). Principles and Practice of Structural Equation Modeling, Third 
Edition (3rd edition). New York: The Guilford Press. 
Knoester, C., & Haynie, D. L. (2005). Community Context, Social Integration into 
Family, and Youth Violence. Journal of Marriage and Family, 67(3), 767–780. 
Kolk, B. A. van der, McFarlane, A. C., Weisaeth, L., & Greenberg, M. A. (1997). 
 
Traumatic Stress: The Effects of Overwhelming Experience on Mind, Body, 
and Society. Journal of Nervous and Mental Disease, 185(11), 707. 
Laraia, B. A. (2013). Food Insecurity and Chronic Disease. Advances in Nutrition: An 
International Review Journal, 4(2), 203–212. 
https://doi.org/10.3945/an.112.003277 
Larkin, H., Shields, J. J., & Anda, R. F. (2012). The Health and Social Consequences of 
Adverse Childhood Experiences (ACE) Across the Lifespan: An Introduction 
to Prevention and Intervention in the Community. Journal of Prevention & 
Intervention in the Community. Retrieved from http://www-tandfonline- 
com.proxy- 
um.researchport.umd.edu/doi/abs/10.1080/10852352.2012.707439 
 153 
Lazarus, R. S. (1966). Psychological stress and the coping process. New York: 
McGraw-Hill. 
Lazarus, R. S., & Cohen, J. B. (1977). Environmental Stress. In I. Altman & J. F. 
Wohlwill (Eds.), Human Behavior and Environment (pp. 89–127). Springer 
US. Retrieved from http://link.springer.com.proxy- 
um.researchport.umd.edu/chapter/10.1007/978-1-4684-0808-9_3 
Leung, C. W., Epel, E. S., Willett, W. C., Rimm, E. B., & Laraia, B. A. (2015). Household 
Food Insecurity Is Positively Associated with Depression among Low-Income 
Supplemental Nutrition Assistance Program Participants and Income-Eligible 
Nonparticipants. The Journal of Nutrition, 145(3), 622–627. 
https://doi.org/10.3945/jn.114.199414 
Lilly, M. M., & Valdez, C. E. (2012). Interpersonal trauma and PTSD: The roles of 
gender and a lifespan perspective in predicting risk. Psychological Trauma: 
Theory, Research, Practice, and Policy, 4(1), 140–144. 
https://doi.org/10.1037/a0022947 
Liu, J. (2004). Childhood Externalizing Behavior: Theory and Implications. Journal of 
Child and Adolescent Psychiatric Nursing, 17(3), 93–103. 
https://doi.org/10.1111/j.1744-6171.2004.tb00003.x 
Luthra, R., Abramovitz, R., Greenberg, R., Schoor, A., Newcorn, J., Schmeidler, J., … 
Chemtob, C. M. (2009). Relationship Between Type of Trauma Exposure and 
Posttraumatic Stress Disorder Among Urban Children and Adolescents. 
Journal of Interpersonal Violence, 24(11), 1919–1927. 
https://doi.org/10.1177/0886260508325494 
 154 
Matlin, S. L., Molock, S. D., & Tebes, J. K. (2011). Suicidality and Depression Among 
African American Adolescents: The Role of Family and Peer Support and 
Community Connectedness. American Journal of Orthopsychiatry, 81(1), 108– 
117. https://doi.org/10.1111/j.1939-0025.2010.01078.x 
McDonald, R., Jouriles, E. N., Ramisetty-Mikler, S., Caetano, R., & Green, C. E. (2006). 
 
Estimating the number of American children living in partner-violent 
families. Journal of Family Psychology: JFP: Journal of the Division of Family 
Psychology of the American Psychological Association (Division 43), 20(1), 
137–142. https://doi.org/10.1037/0893-3200.20.1.137 
McEwen, B. S. (1993). Stress and the Individual: Mechanisms Leading to Disease. 
 
Archives of Internal Medicine, 153(18), 2093. 
https://doi.org/10.1001/archinte.1993.00410180039004 
McEwen, B. S., Gray, J. D., & Nasca, C. (2015). Recognizing resilience: Learning from 
the effects of stress on the brain. Neurobiology of Stress, 1, 1–11. 
https://doi.org/10.1016/j.ynstr.2014.09.001 
McEwen, B. S., & Wingfield, J. C. (2003). The concept of allostasis in biology and 
biomedicine. Hormones and Behavior, 43(1), 2–15. 
https://doi.org/10.1016/S0018-506X(02)00024-7 
McLaughlin, K. A., Green, J. G., Alegría, M., Jane Costello, E., Gruber, M. J., Sampson, N. 
A., & Kessler, R. C. (2012). Food Insecurity and Mental Disorders in a National 
Sample of U.S. Adolescents. Journal of the American Academy of Child & 
Adolescent Psychiatry, 51(12), 1293–1303. 
https://doi.org/10.1016/j.jaac.2012.09.009 
 155 
McLaughlin, K. A., Koenen, K. C., Hill, E. D., Petukhova, M., Sampson, N. A., Zaslavsky, 
 
A. M., & Kessler, R. C. (2013). Trauma Exposure and Posttraumatic Stress 
Disorder in a National Sample of Adolescents. Journal of the American 
Academy of Child & Adolescent Psychiatry, 52(8), 815-830.e14. 
https://doi.org/10.1016/j.jaac.2013.05.011 
McNeely Clea, & Falci Christina. (2009). School Connectedness and the Transition 
Into and Out of Health‐Risk Behavior Among Adolescents: A Comparison of 
Social Belonging and Teacher Support. Journal of School Health, 74(7), 284– 
292. https://doi.org/10.1111/j.1746-1561.2004.tb08285.x 
Merikangas, K. R., Avenevoli, S., Costello, E. J., Koretz, D., & Kessler, R. C. (2009). 
 
National Comorbidity Survey Replication Adolescent Supplement (NCS-A): I. 
Background and Measures. Journal of the American Academy of Child & 
Adolescent Psychiatry, 48(4), 367–379. 
https://doi.org/10.1097/CHI.0b013e31819996f1 
Middlebrooks, J. S., & Audage, N. C. (2008). The Effects of Childhood Stress on Health 
Across the Lifespan. Atlanta (GA): Centers for Disease Control and Prevention, 
National Center for Injury Prevention and Control. 
Miller, A. B. M. A., Esposito-Smythers, C. P. D., & Leichtweis, R. N. P. D. (2015). Role of 
Social Support in Adolescent Suicidal Ideation and Suicide Attempts. Journal 
of Adolescent Health, 56(3), 286–292. 
Mohammad, E. T., Shapiro, E. R., Wainwright, L. D., & Carter, A. S. (2014). Impacts of 
Family and Community Violence Exposure on Child Coping and Mental 
 156 
Health. Journal of Abnormal Child Psychology, 43(2), 203–215. 
https://doi.org/10.1007/s10802-014-9889-2 
Muthén & Muthén. (2018). Mplus. Retrieved April 1, 2018, from 
http://www.statmodel.com/ 
National Scientific Council on the Developing Child. (2007). The Science of Early 
Childhood Development: Closing the Gap Between What We Know and What 
We Do. Retrieved April 1, 2018, from 
https://developingchild.harvard.edu/resources/inbrief-the-science-of- 
resilience/ 
National Scientific Council on the Developing Child. (2012). The Science of Neglect: 
The Persistent Absense of Responsive Care Disrupts the Developing Brain: 
Working Paper 12. Retrieved from http://www.developingchild.harvard.edu 
Nooner KB, Linares LO, Batinjane J, Kramer RA, Silva R, & Cloitre M. (2012). Factors 
related to posttraumatic stress disorder in adolescence. Trauma, Violence & 
Abuse, 13(3), 153–166. 
Nunnally, J. C. (1975). Psychometric Theory— 25 Years Ago and Now. Educational 
Researcher, 4(10), 7–21. https://doi.org/10.3102/0013189X004010007 
Nurius, P. S., Russell, P. L., Herting, J. R., Hooven, C., & Thompson, E. A. (2009). Risk 
and Protective Profiles Among Never Exposed, Single Form, and Multiple 
Form Violence Exposed Youth. Journal of Child & Adolescent Trauma, 2(2), 
106–123. https://doi.org/10.1080/19361520902880798 
 157 
Office of U. S. Government Accountability. (2018). K-12 Education: Discipline 
Disparities for Black Students, Boys, and Students with Disabilities, (GAO-18- 
258). Retrieved from https://www.gao.gov/products/GAO-18-258 
Olds, D., Donelan-McCall, N., O’Brien, R., MacMillan, H., Jack, S., Jenkins, T., … Beeber,  
 
L. (2013). Improving the Nurse–Family Partnership in Community Practice. 
 
Pediatrics, 132(Supplement 2), S110–S117. 
https://doi.org/10.1542/peds.2013-1021I 
Perez, N. M., Jennings, W. G., Piquero, A. R., & Baglivio, M. T. (2016). Adverse 
Childhood Experiences and Suicide Attempts: The Mediating Influence of 
Personality Development and Problem Behaviors. Journal of Youth and 
Adolescence, 45(8), 1527–1545. https://doi.org/10.1007/s10964-016-0519- 
x 
Ports, K. A., Merrick, M. T., Stone, D. M., Wilkins, N. J., Reed, J., Ebin, J., & Ford, D. C. E. 
 
D. C., Inc. .. Suicide Prevention Resource Center, Waltham, Massachusetts. 
(2017). Adverse Childhood Experiences and Suicide Risk: Toward 
Comprehensive Prevention. American Journal of Preventive Medicine, 53(3), 
400–403. 
Rojas S.M., Bilsky S.A., Dutton C., Feldner M.T., Leen-Feldner E.W., & Badour C.L. 
(2017). Lifetime histories of PTSD, suicidal ideation, and suicide attempts in 
a nationally representative sample of adolescents: Examining indirect effects 
via the roles of family and peer social support. Journal of Anxiety Disorders, 
49, 95–103. 
 158 
Rose, T., Joe, S., Shields, J., & Caldwell, C. H. (2014). Social Integration and the Mental 
Health of Black Adolescents. Child Development, 85(3), 1003–1018. 
https://doi.org/10.1111/cdev.12182 
Rosenthal, B., Wilson, W., & Futch, V. (2009). Trauma, protection, and distress in late 
adolescence: a multi-determinant approach. Adolescence, 44(176), 693–703. 
Salloum, A., Scheeringa, M. S., Cohen, J. A., & Storch, E. A. (2014). Development of 
Stepped Care Trauma-Focused Cognitive-Behavioral Therapy for Young 
Children. Cognitive and Behavioral Practice, 21(1), 97–108. 
Schoenbach, V. J., Kaplan, B. H., Fredman, L., & Kleinbaum, D. G. (1986). Social Ties 
and Mortality in Evans County, Georgia. American Journal of Epidemiology, 
123(4), 577–591. 
Seeman, T. E. F. the A. G. C., University of Southern California, Los Angeles, CA. .. USA. 
(1996). Social ties and health: The benefits of social integration. Annals of 
Epidemiology, 6(5), 442–451. 
Seligman, H. K., Laraia, B. A., & Kushel, M. B. (2010). Food Insecurity Is Associated 
with Chronic Disease among Low-Income NHANES Participants. The Journal 
of Nutrition, 140(2), 304–310. https://doi.org/10.3945/jn.109.112573 
Shepard DS, Gurewich D, Lwin AK, Reed GA Jr, & Silverman MM. (2016). Suicide and 
Suicidal Attempts in the United States: Costs and Policy Implications. Suicide 
& Life-Threatening Behavior, 46(3), 352–362. 
Shonkoff, J. P., Boyce, W. T., & McEwen, B. S. (2009). Neuroscience, Molecular 
Biology, and the Childhood Roots of Health Disparities: Building a New 
 159 
Framework for Health Promotion and Disease Prevention. JAMA, 301(21), 
2252. https://doi.org/10.1001/jama.2009.754 
Shonkoff, J. P., Garner, A. S., Health, T. C. on P. A. of C. and F., Childhood, C. on E., 
Adoption, Care, A. D., … Wood, D. L. (2012). The Lifelong Effects of Early 
Childhood Adversity and Toxic Stress. Pediatrics, 129(1), e232–e246. 
https://doi.org/10.1542/peds.2011-2663 
Siddall, J., Huebner, E. S., & Jiang, X. (2013). A Prospective Study of Differential 
Sources of School-Related Social Support and Adolescent Global Life 
Satisfaction. American Journal of Orthopsychiatry, 83(1), 107–114. 
https://doi.org/10.1111/ajop.12006 
Slone, M., & Mann, S. (2016). Effects of War, Terrorism and Armed Conflict on Young 
Children: A Systematic Review. Child Psychiatry & Human Development, 
47(6), 950–965. 
Slopen, N., Fitzmaurice, G. M., Williams, D. R., & Gilman, S. E. (2012). Common 
patterns of violence experiences and depression and anxiety among 
adolescents. Social Psychiatry and Psychiatric Epidemiology, 47(10), 1591– 
1605. https://doi.org/10.1007/s00127-011-0466-5 
Slopen, N., Kubzansky, L. D., McLaughlin, K. A., & Koenen, K. C. (2013). Childhood 
adversity and inflammatory processes in youth: A prospective study. 
Psychoneuroendocrinology, 38(2), 188–200. 
https://doi.org/10.1016/j.psyneuen.2012.05.013 
Slopen, N., McLaughlin, K. A., Dunn, E. C., & Koenen, K. C. (2013). Childhood adversity 
and cell-mediated immunity in young adulthood: Does type and timing 
 160 
matter? Brain, Behavior, and Immunity, 28, 63–71. 
https://doi.org/10.1016/j.bbi.2012.10.018 
Smalley, N., Scourfield, J., & Greenland, K. (2005). Young People, Gender and Suicide: 
A Review of Research on the Social Context. Journal of Social Work, 5(2), 133– 
154. https://doi.org/10.1177/1468017305054953 
Stadler, C., Feifel, J., Rohrmann, S., Vermeiren, R., & Poustka, F. (2010). Peer- 
Victimization and Mental Health Problems in Adolescents: Are Parental and 
School Support Protective? Child Psychiatry and Human Development, 41(4), 
371–386. http://dx.doi.org/10.1007/s10578-010-0174-5 
Stein, B. D., Jaycox, L. H., Kataoka, S., Rhodes, H. J., & Vestal, K. D. (2003). Prevalence 
of Child and Adolescent Exposure to Community Violence. Clinical Child and 
Family Psychology Review, 6(4), 247–264. 
https://doi.org/10.1023/B:CCFP.0000006292.61072.d2 
Straus, M. A. (1979). Measuring Intrafamily Conflict and Violence: The Conflict 
Tactics (CT) Scales. Journal of Marriage and the Family, 41(1), 75. 
Straus, M. A., Hamby, S. L., Finkelhor, D., Moore, D. W., & Runyan, D. (1998). 
 
Identification of Child Maltreatment With the Parent-Child Conflict Tactics 
Scales: Development and Psychometric Data for a National Sample of 
American Parents. Child Abuse & Neglect, 22(4), 249–270. 
https://doi.org/10.1016/S0145-2134(97)00174-9 
Tajima, E. A., Herrenkohl, T. I., Moylan, C. A., & Derr, A. S. (2011). Moderating the 
Effects of Childhood Exposure to Intimate Partner Violence: The Roles of 
Parenting Characteristics and Adolescent Peer Support. Journal of Research 
 161 
on Adolescence, 21(2), 376–394. https://doi.org/10.1111/j.1532- 
7795.2010.00676.x 
Tenorio, K. A., & Lo, C. C. (2011). Social Location, Social Integration, and the Co- 
occurrence of Substance Abuse and Psychological Distress. The American 
Journal of Drug and Alcohol Abuse, 37(4), 218–223. 
https://doi.org/10.3109/00952990.2011.568079 
Terr LC. (1991). Childhood traumas: an outline and overview. The American Journal 
of Psychiatry, 148(1), 10–20. 
Thorlindsson, T., & Bjarnason, T. (1998). Modeling Durkheim on the micro level: A 
study of youth suicidality. American Sociological Review, 63(1), 94–110. 
Turner, H. A., Finkelhor, D., & Ormrod, R. (2006). The effect of lifetime victimization 
on the mental health of children and adolescents. Social Science & Medicine, 
62(1), 13–27. https://doi.org/10.1016/j.socscimed.2005.05.030 
Turner, H. A., Finkelhor, D., & Ormrod, R. (2010). Poly-Victimization in a National 
Sample of Children and Youth. American Journal of Preventive Medicine, 
38(3), 323–330. https://doi.org/10.1016/j.amepre.2009.11.012 
U.S. Census Bureau. (2013). America’s Families and Living Arrangements: 2013: 
Family Groups (FG table series-Table FG5). U.S. Census Bureau, Current 
Population Survey, 2013 Annual Social and Economic Suppliment. Retrieved 
from https://www.census.gov/hhes/families/data/cps2013FG.html 
U.S. Department of Health & Human Services, Administration for Children and 
Families, Administration on, & Children, Youth and Families, Children’s 
Bureau. (2016). Child maltreatment 2014. Retrieved from 
 162 
http://www.acf.hhs.gov/programs/cb/research-data-technology/statistics- 
research/child-maltreatment 
U.S. Department of Health and Human Services, Office of Disease Prevention and 
Health Promotion. (2015). Adolescent Health: Healthy People 2020. 
Retrieved November 17, 2015, from 
http://www.healthypeople.gov/2020/topics-objectives/topic/Adolescent- 
Health/objectives 
U.S. Dept. of Health & Human Services, Office of the Assistant Secretary for Planning 
and Evaluation. (2014). ASPE Issue Brief: Information on Poverty and Income 
Statistics: A Summary of 2014 Current Population Survey Data. Retrieved from 
https://aspe.hhs.gov/sites/default/files/pdf/33766/ib_poverty2014.pdf 
van Der Kolk, B. A., McFarlane, A. C., & Weisæth, L. (1996). Traumatic stress: the 
effects of overwhelming experience on mind, body, and society. New York: 
Guilford Press. Retrieved from Sample text 
http://catdir.loc.gov/catdir/enhancements/fy0651/96010818-s.html 
van Der Kolk, B. A., & Saporta, J. (1991). The biological response to psychic trauma: 
Mechanisms and treatment of intrusion and numbing. Anxiety, Stress & 
Coping, 4(3), 199–212. 
Voisin, D. R. (2003). Victims of Community Violence and HIV Sexual Risk Behaviors 
Among African American Adolescent Males. Journal of HIV/AIDS Prevention & 
Education for Adolescents & Children, 5(3/4), 87–110. 
 163 
Voisin, D. R. (2007). The Effects of Family and Community Violence Exposure Among 
Youth: Recommendations for Practice and Policy. Journal of Social Work 
Education, 43(1), 51–66. 
Wade, R., Cronholm, P. F., Fein, J. A., Forke, C. M., Davis, M. B., Harkins-Schwarz, M., … 
Bair-Merritt, M. H. (2015). Household and community-level Adverse 
Childhood Experiences and adult health outcomes in a diverse urban 
population. Child Abuse & Neglect. 
Wade, R., Shea, J. A., Rubin, D., & Wood, J. (2014). Adverse Childhood Experiences of 
Low-Income Urban Youth. Pediatrics, 134(1), e13–e20. 
https://doi.org/10.1542/peds.2013-2475 
Wadsworth, M. E. (2015). Development of Maladaptive Coping: A Functional 
Adaptation to Chronic, Uncontrollable Stress. Child Development Perspectives, 
9(2), 96–100. https://doi.org/10.1111/cdep.12112 
Waldrop, A. E., Hanson, R. F., Resnick, H. S., Kilpatrick, D. G., Naugle, A. E., & 
Saunders, B. E. (2007). Risk factors for suicidal behavior among a national 
sample of adolescents: Implications for prevention. Journal of Traumatic 
Stress, 20(5), 869–879. https://doi.org/10.1002/jts.20291 
Weinreb, L., Wehler, C., Perloff, J., Scott, R., Hosmer, D., Sagor, L., & Gundersen, C. 
 
(2002). Hunger: Its Impact on Children’s Health and Mental Health.  
 
Pediatrics, 110(4), e41–e41. https://doi.org/10.1542/peds.110.4.e41 
Whitlock Janis, Wyman Peter A., & Moore Sarah R. (2014). Connectedness and 
Suicide Prevention in Adolescents: Pathways and Implications. Suicide and 
 164 
Life-Threatening Behavior, 44(3), 246–272. 
https://doi.org/10.1111/sltb.12071 
Williamson, D. F., Thompson, T. J., Anda, R. F., Dietz, W. H., & Felitti, V. (2002). Body 
weight and obesity in adults and self-reported abuse in childhood. 
International Journal of Obesity, 26(8), 1075–1082. 
https://doi.org/10.1038/sj.ijo.0802038 
Young, J. F., Berenson, K., Cohen, P., & Garcia, J. (2005). The Role of Parent and Peer 
Support in Predicting Adolescent Depression: A Longitudinal Community 
Study. Journal of Research on Adolescence, 15(4), 407–423. 
https://doi.org/10.1111/j.1532-7795.2005.00105.x 
Zhang, B. (2015). The Relationship Between Perceived Stress and Adolescent 
Depression: The Roles of Social Support and Gender. The Relationship 
Between Perceived Stress and Adolescent Depression: The Roles of Social 
Support and Gender, 123(2), 501–518. https://doi.org/10.1007/s11205-014- 
0739-y 
Ziol-Guest, K. M., & Dunifon, R. E. (2014). Complex Living Arrangements and Child 
Health: Examining Family Structure Linkages with Children’s Health 
Outcomes. Family Relations, 63(3), 424–437. 
https://doi.org/10.1111/fare.12071 
 165