LIFESTYLE BEHAVIORS AND CHRONIC DISEASE RISK IN EMERGING ADULTS 
WITH A HISTORY OF ADVERSE CHILDHOOD EXPERIENCES (ACES) 
by 
Alexander J. Capiaghi 
May, 2021 
Director of Thesis:  Matthew C. Whited, Ph.D. 
Major Department:  Psychology 
 The growing body of adverse childhood experiences (ACEs) literature over the past two 
decades provides strong evidence that there is a dose-response relationship between the number 
of ACEs experienced and the negative downstream effects on one’s health. However, due to an 
abundance of retrospective public health survey data, but absence of data across the lifespan, the 
mechanisms that contribute to worse health outcomes remain unclear. Recent findings indicate 
that ACEs are associated with higher chronic disease prevalence even among young adults. 
Therefore, the present study was designed to examine the relationships between ACEs and 
various aspects of physical and mental health in a healthy emerging adult population.  
Using a college student sample, participants completed an online questionnaire that 
included the Adverse Childhood Experience Questionnaire (ACE-Q) and measures of physical 
health, mental health, and lifestyle health behaviors. Correlation analysis revealed strong 
positive associations between ACEs and symptoms of PTSD, depression, anxiety, ADHD, 
emotion dysregulation, sleep difficulties, and somatic symptoms; mild positive associations 
were also observed between ACEs and perceived stress, disordered eating, illicit substance use, 
and e-cigarette use. Hierarchical linear regression analysis demonstrated that ACE-Q scores 
significantly predicted these factors when controlling for demographic covariates such as 
gender, race/ethnicity, and SES. 
 
 
The findings of this study help to fill an etiological knowledge gap by providing more 
information about the prevalent risk factors related to individuals with a history of high ACEs 
that could contribute to the development of chronic diseases. In addition to various mental 
health symptoms, sleep difficulties and somatic symptoms appear to be prominent health issues 
for emerging adults with a history of ACEs. Healthcare providers for this age group may want 
to increase screening, prevention, and intervention efforts to address these symptoms, as this 
may have a positive influence on reducing chronic disease risk.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
LIFESTYLE BEHAVIORS AND CHR0NIC DISEASE RISK IN EMERGING ADULTS WITH 
A HISTORY OF ADVERSE CHILDHOOD EXPERIENCES (ACES)   
 
 
 
 
 
 
 
A Thesis 
 
Presented to the Faculty of the Department of Psychology 
 
East Carolina University 
 
 
 
 
 
In Partial Fulfillment of the Requirements for the Degree 
 
Master of Arts in Psychology, Clinical Concentration 
 
 
 
 
 
by 
 
Alexander J. Capiaghi 
 
May, 2021 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
© Alexander J. Capiaghi, 2021 
 
 
 
 
 
 
 
 
 
 
 
LIFESTYLE BEHAVIORS AND CHRONIC DISEASE RISK IN EMERGING ADULTS 
WITH A HISTORY OF ADVERSE CHILDHOOD EXPERIENCES (ACES) 
by 
Alexander J. Capiaghi 
APPROVED BY: 
DIRECTOR OF THESIS: 
Matthew C. Whited, PhD 
COMMITTEE MEMBER: 
Christyn Dolbier, PhD 
COMMITTEE MEMBER: 
D. Erik Everhart, PhD 
 
CHAIR OF THE DEPARTMENT OF PSYCHOLOGY: 
 
Alan J. Christensen, PhD 
DEAN OF THE GRADUATE SCHOOL: 
 
Paul J. Gemperline, PhD 
 
 
 
TABLE OF CONTENTS 
LIST OF TABLES ......................................................................................................................... vi 
LIST OF FIGURES ...................................................................................................................... vii 
INTRODUCTION .......................................................................................................................... 1 
LITERATURE REVIEW ............................................................................................................... 8 
Adverse Childhood Experiences (ACEs) .................................................................................... 8 
Physiological Stress and Dysfunction ....................................................................................... 11 
PROPOSED STUDY .................................................................................................................... 19 
Overview ................................................................................................................................... 19 
Proposed Aims and Hypotheses ................................................................................................ 22 
Proposed Methods ..................................................................................................................... 25 
Proposed Statistical Analysis .................................................................................................... 40 
CURRENT STUDY...................................................................................................................... 42 
Aims and Hypotheses ................................................................................................................ 42 
METHODS ................................................................................................................................... 45 
Participants and Demographics ................................................................................................. 45 
Data Cleaning ............................................................................................................................ 45 
Descriptive Statistics ................................................................................................................. 46 
Aim 1 ......................................................................................................................................... 53 
Aim 2 ......................................................................................................................................... 59 
Aim 3 ......................................................................................................................................... 65 
DISCUSSION ............................................................................................................................... 69 
Mental Health Factors ............................................................................................................... 69 
Physical Health Factors ............................................................................................................. 71 
Lifestyle Health Behaviors ........................................................................................................ 72 
Key Differences by ACE-Q Subscale ....................................................................................... 74 
CONCLUSION ............................................................................................................................. 75 
Limitations ................................................................................................................................ 77 
REFERENCES ............................................................................................................................. 80 
APPENDIX A: AIM 1 HIERARCHICAL LINEAR REGRESSION RESULTS (SPSS) ........... 94 
APPENDIX B: AIM 2 HIERARCHICAL LINEAR REGRESSION RESULTS (SPSS) ......... 106 
APPENDIX C: AIM 3 HIERARCHICAL LINEAR REGRESSION RESULTS (SPSS) ......... 115 
 
 
 
APPENDIX D: IRB APPROVAL LETTER .............................................................................. 125 
APPENDIX E: INFORMED CONSENT FOR ONLINE PORTION OF STUDY ................... 127 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
List of Tables 
Table 1. Hypothesis 1 Correlation Matrix …………………………………………………….   53 
Table 2. Hypothesis 2 Correlation Matrix …………………………………………………….   56 
Table 3. Hypothesis 3 Correlation Matrix …………………………………………………….   60 
Table 4. Hypothesis 4 Correlation Matrix …………………………………………………….   63 
Table 5. Hypothesis 5 Correlation Matrix …………………………………………………….   66 
Table 6. Hypothesis 6 Correlation Matrix …………………………………………………….   67 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
List of Figures 
Figure 1. Adaptive versus maladaptive HRV responses ………………………………………………   24 
Figure 2. Frequency of Adverse Childhood Experiences ………………………………………………   47
 
Introduction 
 Over the last two decades, emerging research on adverse childhood experiences (ACEs) 
has demonstrated a clear association between early life adversity (e.g. abuse, neglect, household 
dysfunction) and various states of pathology. Although some recent literature has sought to 
explore the underlying mechanisms connecting ACEs with subsequent disease states, the 
pathophysiology remains largely unknown. This is mainly due to the high frequency of ACEs 
population data and relatively low implementation of experimental approaches to explore disease 
pathways associated with ACEs. Therefore, specific physiological mechanisms must be explored 
to inform interventions that target vulnerable youth, adolescents, and young adults with a history 
of ACEs, as these individuals are at high risk for developing chronic physical and mental health 
conditions (Felitti et al., 1998). 
For individuals with a history of ACEs, the pathway to chronic illness is a process that 
involves interaction between the body’s major systems – namely, the nervous, cardiovascular, 
endocrine, and immune systems. Cardiovascular disease remains the most prominent cause of 
death in the US, and is responsible for approximately 25% of all deaths (Benjamin et al., 2019). 
Accordingly, cardiovascular health is perhaps the most significant target for health interventions, 
and a large diversity of interventions have been developed to address the presence of biological, 
behavioral, social, and psychological risk factors at the cardiovascular level (Su, Jimenez, 
Roberts, & Loucks, 2015). Due to the interactions of these risk factors, the heightened risk for 
various forms of heart disease is well-documented among individuals with a history of ACEs 
(Felitti et al., 1998; Dong, et al., 2004; Campbell, Walker, & Egede, 2016; Hughes et al., 2017). 
In particular, chronic stress has been used to explain the effects of various biological, 
 
 
 
psychological, and environmental determinants on both cardiovascular and mental health (Su et 
al., 2015).  
In addition to increased vulnerability for heart disease, individuals with a history of 
childhood adversity also have a heightened risk for various forms of psychopathology. Those 
with elevated ACE scores have been consistently found to have higher prevalence of depression, 
suicide attempts, alcohol and other substance use, interpersonal violence and overall poor mental 
health (Felitti et al., 1998; Hughes et al., 2017). Combined with this, there is extensive data 
supporting the increased prevalence mental health conditions among heart disease patients, and 
the prevalence of depression is about three times higher than in the general population (Chaddha, 
Robinson, Kline-Rogers, Alexandris-Souphis, & Rubenfire, 2016). While the bidirectionality of 
this association has been widely accepted – that is, that mental health conditions contribute to 
heart disease and heart disease contributes to mental health conditions – the exact mechanisms of 
this relationship are complex, and still largely unknown. Accordingly, it is fitting for this link 
between psychological and cardiovascular pathology to be simultaneously studied among the 
ACEs population, a group with heightened vulnerability for both conditions. 
Autonomic nervous system (ANS) function is particularly useful in the study of chronic 
stress due to its sensitivity to both physiological and psychological health influences and the 
ability to monitor ANS functioning in a laboratory setting. For instance, multiple studies have 
found that participants with ACE scores  ? 4 had dysregulated cortisol levels, connecting ACE 
prevalence to stress sensitization and dysregulation of the HPA axis (Bunea, Szentágotai-T?tar, 
& Miu, 2017; Kalmakis, Meyer, Chiodo, & Leung, 2015). While autonomic dysfunction is 
speculated to play a role in the high rates of heart disease among adults with a history of multiple 
ACEs (Su et al., 2015), the current literature has not yet thoroughly explored this relationship. In 
2 
 
 
 
a recent meta-analysis of ACEs and indicators of poor health, only two studies explored 
autonomic function, and both concluded significant relationships between adversity and 
physiological dysregulation (Rigterink, Katz, & Hessler, 2010; Shonkoff et al., 2012). This 
highlights the need for additional research to examine the role of autonomic dysregulation as a 
potential mechanism in the progression from childhood adversity to subsequent cardiovascular 
disease. 
Through the use of electrocardiogram (ECG) monitoring, heart rate variability (HRV) is 
one of the most widely used measurements of ANS functioning, and has been conceptualized as 
a transdiagnostic biomarker of both physical and mental illness (Beauchaine & Thayer, 2015). 
Among the different frequencies of HRV (e.g. low frequency, very low frequency), high 
frequency HRV (HF HRV) is often utilized due to the ability to detect vagal control – that is, 
one’s ability to self-regulate heart rate via the vagus nerve. Accordingly, vagal control has been 
conceptualized as the proper adaptation to the stressors in one’s environment, which allows for 
the conservation of energy, and thereby prevents unnecessary burden on the cardiovascular 
system (Thayer, Ahs, Fredrikson, Sollers, & Wager, 2012). Favorable health outcomes have been 
associated with high HF HRV, representing one’s ability to engage in adaptive regulatory 
processes carried out by sympathetic and parasympathetic activity. In comparison, low HF HRV 
is associated with poorer health outcomes, and indicates problems with self-regulation and 
cognitive control (Thayer, Hansen, Saus-Rose, & Johnsen, 2009).  
In particular, the behavioral inhibition and activation systems (BIS/BAS) have been used 
as a conceptual framework to interpret HRV analysis (Brenner, Beauchaine, & Sylvers, 2005). 
These constructs were initially developed to explain how motivational systems can predict the 
relationship between affect and behavior (Gray, 1981). As such, social engagement is considered 
3 
 
 
 
to be an adaptive response to the stressors in one’s environment, and would be expected to 
accompany HRV reactivity that is appropriately responsive to experiences of stress. In contrast, 
various forms of avoidance are viewed as maladaptive, and would theoretically be found 
associated with HRV responses that are inadequately lower or excessively higher than what 
would be expected in relation to the presenting stressor. Accordingly, self-report measures of 
behavioral inhibition and behavioral activation (Carver & White, 1994) have been used to 
describe how affective and emotion regulation differences between individuals that may 
contribute to physiological dysfunction (Brenner et al., 2005). 
Significantly, issues of emotion regulation have been identified as a prevalent risk factor 
among individuals with high ACE scores. The ACEs literature has consistently shown that 
exposure to early adversity is associated with higher prevalence of mental health issues in 
adulthood, including depression, suicide, interpersonal violence, and alcohol and other substance 
use (Felitti et al., 1998; Hughes et al., 2017). A recent assessment of mental health issues in an 
adolescent sample (ages 13-17) found that elevated ACE scores were associated with heightened 
symptoms of ADHD (impulsivity, inattention, and combined types), generalized anxiety disorder 
(GAD), major depressive disorder (MDD), and dysthymic disorder (Horn, Leve, Levitt, & 
Fisher, 2019). Given the application of HRV analysis to transdiagnostic mental health issues, it is 
likely to observe an association between ACEs and ANS dysregulation in laboratory samples. 
Additionally, some statistical models have concluded that emotion and cognitive processes could 
mediate the relationship between childhood adversity and physiological changes (Jin, Kim, Kim, 
Hyun, & Lee, 2018). 
Given these factors, the ecobiodevelopmental (EBD) theory is the leading framework for 
understanding the interaction of early life adversity, chronic stress, and subsequent poor health 
4 
 
 
 
outcomes (Shonkoff et al., 2012). In a recent policy proposal, the American Academy of 
Pediatrics voiced their adoption of this framework to discuss the various influences of chronic, or 
toxic, stress throughout the lifespan. The concept of toxic stress has been used to contrast the 
experiences of positive stress and tolerable stress, as toxic stress involves prolonged activation of 
the body’s stress response in a way that alters brain development and associated functioning 
(Franke, 2014). For instance, neurodevelopmental changes to the amygdala, hippocampus, and 
prefrontal cortex due to toxic stress have been shown to contribute to deficits in learning, 
memory, executive functioning and anxiety (Shonkoff et al., 2012).  
Past research has shown that these neurodevelopmental changes also contribute to 
autonomic dysregulation. The results of these modified frontolimbic functions have been 
identified as altered cognition, reduced stress reactivity, and emotion dysregulation, which may 
lead to impulsivity and a focus on short-term goal attainment, and consequently, may increase 
one’s risk for poor health behaviors (e.g. addiction/substance use, unhealthy eating, physical 
inactivity) (Lovallo, 2013). Therefore, these changes are likely to influence reactivity and vagal 
control during a stressor task. In past research, various stress tasks have been implemented 
during HRV monitoring to generate stress responses that require participants to engage in 
emotion regulation and cognitive control. This methodology may provide insight into the ways 
that toxic stress and associated neurodevelopmental alterations can also lead to changes in 
autonomic dysregulation and cardiovascular reactivity. 
While the growing ACEs literature has recently emphasized neurodevelopmental changes 
within the EBD framework, very few studies have looked at alterations to cardiovascular 
activity. This is surprising, since research has indicated an increased risk to various forms of 
heart disease – such as myocardial infarction, coronary artery disease, stroke, and diabetes – for 
5 
 
 
 
individuals with a history of ACEs (Gilbert et al., 2015). Additionally, individuals with a higher 
prevalence of ACEs have been consistently shown to have an increased association with some of 
the main lifestyle and behavioral contributors to heart disease, such as heavy alcohol use, 
smoking, risky sex behaviors, physical inactivity, obesity, and problematic drug use (Campbell, 
Walker, & Egede, 2016; Hughes et al., 2017). These results are consistent with the sample used 
in the seminal ACE Study (Felitti et al., 1998), which showed a dose-response relationship 
between ACEs and ischemic heart disease. Further, a follow-up study found that ischemic heart 
disease was mediated by psychological risk factors commonly associated with childhood 
adversity (e.g. depression, anger, negative affect) rather than traditional risk factors of ischemia 
(Dong, Giles et al., 2004). Taken together, this suggests that psychological risk factors may pose 
a unique risk for developing CVD in individuals in with a history of ACEs, and that this may 
create vulnerability above and beyond that of unhealthy lifestyle factors (e.g. smoking, physical 
inactivity, substance misuse, unhealthy eating).  
Due to the predominant healthcare burden of heart disease, there remains a pressing need 
for interventions informed by a deeper understanding of the pathophysiology that contributes to 
heightened CVD rates in the ACEs population. Since cardiovascular diseases are typically not 
diagnosed in emerging adulthood, early psychophysiological indicators of CVD may allow for 
the implementation of early prevention and intervention strategies. Thus, HRV analysis in 
emerging adults may be able to identify potential mechanisms (e.g. poor vagal control, emotion 
dysregulation, cognitive dysfunction) for individuals with a history of ACEs. Further, while low 
HF HRV has been considered a biomarker of disease, self-regulatory vagal control has also been 
explored as a modifiable risk factor, and improvements may directly reduce one’s risk for 
disease (Thayer & Lane, 2007). Moreover, recent studies have identified that psychotherapeutic 
6 
 
 
 
interventions can improve physiological dysfunction during stress tasks, suggesting that adaptive 
physiological responses can be learned and practiced (Steffen, Fidalgo, Schmuck, Tsui, & 
Brown, 2014). In summary, HRV analysis in individuals with a history of ACEs may be able to 
inform effective prevention and intervention strategies related to self-regulation and stress 
reactivity, and as a result, reduce one’s heightened risk for cardiovascular disease.
7 
 
Literature Review 
Adverse Childhood Experiences (ACEs) 
Since the publication of the ACEs Study (Felitti et al., 1998), adverse childhood 
experiences (ACEs) have been generally conceptualized as experiences of abuse and household 
dysfunction prior to age 18. The construct of ACEs was developed as a way to quantify chronic 
negative early life experiences and examine the aggregate effects on subsequent physical and 
mental health. The operationalization of ACEs varies slightly between different measurement 
tools, but the most commonly assessed items include parental incarceration, domestic violence, 
household mental illness/suicide, parental separation/divorce, household substance abuse, 
physical and emotional neglect, and physical, sexual and emotional abuse; less common items 
include experiences of neighborhood/community violence, economic hardship, serious car 
accidents, bullying victimization, discrimination, and parental death (Bethell et al., 2017). The 
Adverse Childhood Experiences Study Questionnaire (ACE-Q) is both the original and most 
widely used measurement tool, and quantifies an ACE score based on the total number of 
affirmative questionnaire responses ranging from zero to ten (Felitti et al., 1998). 
The prevalence of childhood adversity has been derived from various sources of 
population data, all of which provide similar estimates. One sample that included the ACE-Q as 
part of the larger 2010 Behavioral Risk Factor Surveillance Survey administered across 11 
regions of the US found that 40.6% of respondents reported zero ACEs, 44.1% reported one to 
three ACEs, 12.7% reported four to six ACEs, and 2.6% reported seven to nine ACEs (Gilbert et 
al., 2015). Prevalence data were similar using the ACE questionnaire (Felitti et al., 1998) with 
undergraduate students at East Carolina University, which found that 40.5% of participants 
 
 
 
reported no history of ACEs, 47% reported 1-3 ACEs, and 12.5% reported 4 or more ACEs 
(N=560, M=1.4, SD=1.7) (Dolbier, 2019; In Prep).  
In their seminal work, Felitti and colleagues (1998) found the association between ACEs 
and poorer health outcomes is most notable for individuals with a history of four or more (? 4) 
types of childhood abuse and family dysfunction. Initial results showed that a history of ? 4 
ACEs yielded a 4-12 fold increase in health risks for alcohol and other drug use, depression, 
suicide attempts, and interpersonal violence. Additionally, health risks were at least double for 
smoking, risky sexual behavior, sexually transmitted diseases and poor self-rated health, and 
nearly double for physical inactivity and obesity. In a recent systematic review of health 
outcomes in pediatric samples, ACEs were found to be associated with delays in cognitive 
development, asthma, infection, somatic complaints, and sleep disruption (Oh et al., 2018). Due 
to a variety of poor health outcomes,  increased mortality rates have been observed in individuals 
with elevated ACE scores; specifically, those with six or more ACEs were found to die, on 
average, 20 years earlier than individuals who reported no history of ACEs (Brown et al., 2009). 
Poorer physical and mental health for individuals with a history of childhood adversity is 
seen most clearly through an increased prevalence of chronic health conditions in adulthood. 
Felitti and colleagues (1998) found that the presence of more types of ACEs was associated with 
an increased prevalence of ischemic heart disease, chronic lung disease, cancer, and liver 
disease. Importantly, such findings have been widely replicated. One recent study found that 
higher ACE scores increased risk for diabetes, coronary artery disease, myocardial infarction, 
stroke, and depression (Campbell et al., 2017). When controlling for demographic variables in a 
US sample, there was a dose-response relationship between ACE scores and health outcomes 
including fair/poor general health, frequent mental distress, asthma, diabetes, coronary heart 
9 
 
 
 
disease, stroke, myocardial infarction, and disability (Gilbert et al., 2015). Further, in a nationally 
representative sample of Irish adults, childhood adversity was associated with earlier onset of 
cardiovascular and pulmonary diseases, as well as psychiatric disorders (McCrory, Dooley, 
Layte, & Kenny, 2015). Taken together, this association between early life adversity and worse 
health outcomes demands a greater understanding of pathways to poor health.  
Although population data has been used to describe ACEs prevalence and subsequent 
health risks, the pathways to disease remain largely unknown. One of the most widely adopted 
explanations of poorer health is that of toxic stress. As previously noted, the 
ecobiodevelopmental (EBD) framework is used to examine the various ways that prolonged 
stress activation at an early age can lead to developmental changes. Thus far, these 
developmental changes have been primarily assessed at the neurological level. For instance, 
changes to the amygdala, hippocampus, and prefrontal cortex from toxic stress can reduce 
learning, memory, and executive functioning (Shonkoff et al., 2012). Further, the EBD 
framework offers the assumptions that heightened stress at an early age alters the development of 
adaptive coping responses and provides a basis for altered physiological responses, maladaptive 
coping, and unhealthy lifestyle behaviors (Garner et al., 2012). Similarly, Lovallo (2013) 
proposes a model whereby early adversity alters neurodevelopment of the limbic system (one of 
the brain’s major centers of emotion processing). As such, these alterations may contribute to 
decreased stress reactivity and increased emotion dysregulation, resulting in subsequent 
impulsive and risky behaviors. This pattern of emotion dysregulation has been widely identified 
in the literature as a factor that increases the likelihood of addiction and unhealthy coping 
behaviors (Beauchaine et al., 2018). Thus, both models assume a cascading effect of 
neurodevelopment on subsequent emotion and behavior. Therefore, ideal prevention and 
10 
 
 
 
intervention efforts should aim to identify indicators of stress and physiological dysfunction 
earlier in the process of neurodevelopment before emotion dysregulation and maladaptive coping 
behaviors become manifest. 
Physiological Stress and Dysfunction 
Various biomarkers of stress are used to identify dysfunction in the body’s regulatory 
systems. For the purposes of this study, regulation of the cardiovascular system is of primary 
interest, since a cardiovascular approach to stress reactivity has been only lightly explored in the 
ACEs population despite early evidence of unique psychological factors contributing to ischemic 
heart disease (Dong et al., 2004). Neurodevelopmental changes, as previously noted in the ACEs 
population, interact with the cardiovascular system via the body’s systems of stress reactivity, 
primarily explained via activity in the autonomic nervous system. More specifically, the 
hypothalamic-pituitary-adrenocortical (HPA) axis has been extensively identified as a regulatory 
system that plays a critical role in the body’s response to stress, and alterations to this system 
have been found associated with ACEs (Kalmakis et al., 2015).  
The HPA axis operates as one of the primary stress response systems in the body, and 
chronic activation of this system is associated with various adverse effects. Under normal 
conditions of stress, corticotropin-releasing hormone (CRH) is produced and released by the 
hypothalamus to stimulate adrenocorticotrophic hormone (ACTH) production in the pituitary 
gland. Once ACTH is released into circulation, glucocorticoids (e.g. cortisol) are produced and 
released by the adrenal gland. When this system is chronically activated, this may lead to 
increased basal secretion of cortisol, heightened stress reactivity, and/or adrenal fatigue (Herman 
et al., 2016). Similarly, activation of the sympathetic-adrenal-medullary (SAM) axis during 
sympathetic arousal stimulates the release of catecholamines such as epinephrine and 
11 
 
 
 
norepinephrine. In response to stress, these hormones are released by the adrenal gland to 
increase neural and cardiovascular activity.  
If adequate downregulation via parasympathetic activity does not take place, this can lead 
to chronic dysregulation, which has more recently been referred to as toxic stress (Shonkoff et 
al., 2012). In one systematic review of pediatric health outcomes associated with childhood 
adversity, the authors identified several categories of disease biomarkers associated with ACEs 
as a result of altered HPA and SAM activity (Oh et al., 2018). The authors identify that studies of 
endocrine function, such as cortisol levels, found that participants who experienced childhood 
maltreatment exhibited blunted cortisol responses to stress. In contrast, studies in which 
participants had ACEs related to parental depression indicated slightly increased cortisol 
responses. Taken together, this could indicate that maltreatment is associated with a more 
chronic stress profile, in which cortisol release becomes blunted due to heightened activation and 
exhaustion. Regarding inflammation and immune function, the authors identify various 
associations between ACEs and increased cytotoxicity, such as elevated proinflammatory 
cytokines. Further, the authors discuss biomarkers of autonomic function, and identify a sample 
in which exposure to domestic violence was associated with lower vagal tone (Rigterink et al., 
2010). Given the heterogeneity of adversity experiences among ACE samples, further research is 
needed to better understand the relationships between disease biomarkers and specific, as well as 
collective, ACE items. Additionally, it may be important to identify factors of adversity 
experiences such as frequency and intensity, as this may help distinguish those that are 
associated with acute versus chronic stress responses.  
In recent clinical and research settings, changes to the HPA and SAM systems are most 
commonly measured through cortisol, a widely used biomarker of physiological stress. Due to 
12 
 
 
 
the high accessibility of cortisol in clinical and research settings (measured through sweat, tears, 
urine, blood and hair), many interventions for children with a history of ACEs use cortisol levels 
as a primary outcome (Purewal et al., 2018). Overall, the cortisol literature regarding ACEs and 
stress reactivity has shown that dysregulation is identified by both increases and decreases in 
cortisol levels. Using a college student sample, one study found that higher ACE scores were 
associated with lower concentrations of hair cortisol (Kalmakis et al., 2015). Oh and colleagues 
(2018) found that blunted cortisol levels during childhood were associated with maltreatment, 
and elevated cortisol levels were associated with maternal mental health issues. Further, a meta-
analysis of the association between early-life adversity and cortisol responses to social stress 
found that the effect sizes of cortisol were higher in studies that focused on maltreatment 
compared to other types of adversities (Bunea et al., 2017). Thus, specific types of ACEs may 
have different associations with cortisol and stress reactivity, highlighting the need for additional 
indicators of stress reactivity other than cortisol.  
While there have been multiple reported associations between ACEs and cortisol, this 
information allows for only speculation regarding the relationship between ACEs and 
cardiovascular activity. For instance, one study of lifetime adversity (rather than childhood 
adversity) in adult participants found a similar dose-response rate between experiences of 
adversity with elevations to both cortisol and heart rate (Lovallo, Farag, Sorocco, Cohoon, & 
Vincent, 2012). Although cortisol and heart rate responses to stress are both considered measures 
of ANS activity, studies such as this do not allow for distinctions between heightened cardiac 
reactivity versus vagal withdrawal, or proper cardiac recovery to rest versus prolonged reactivity. 
Further, a recent laboratory study identified that increases in cortisol levels during a stress task 
were preceded by anticipatory decreases in HRV, yet in-task HRV responses were not associated 
13 
 
 
 
with cortisol changes or cortisol recovery (Pulopulos, Vanderhasselt, & De Raedt, 2018). This 
demonstrates how cortisol and HRV are likely related, but provide different information 
regarding the physiological mechanisms of a stress response. While cortisol research in regard to 
ACEs has increased in recent years, significantly fewer studies have looked at outcomes related 
to heart rate variability following a history of childhood adversity; this may address many of the 
limitations of using cortisol measurements as a way to explain the dynamic process of stress 
reactivity over time. To better understand the relationship between ACEs and cardiac reactivity, 
laboratory studies are needed that monitor HRV stress reactivity before, during, and after a stress 
task. 
Heart Rate Variability (HRV) 
Due to the limitations of cortisol research in the ACEs population, heart rate variability 
(HRV) is a more useful measure of cardiovascular stress reactivity, and can provide additional 
information about the influence of psychological stress and ANS dysregulation. As a 
measurement of ANS activity, HRV recording is able to capture the heart-brain interaction and 
provide information about the ANS as an internal regulatory systems (Shaffer & Ginsberg, 
2017). Dysregulation of HRV has been associated with various disease states, including 
hypertension, diabetes, congestive heart failure, renal failure, and subsequent cardiac events after 
a myocardial infarction (Acharya, Joseph, Kannathal, Lim, & Suri, 2006). Using 
electrophysiological methods, HRV is measured as the variation of time intervals between heart 
beats, and is captured during a process of continuous heart rate monitoring. Unlike heart rate, 
HRV is a measurement of the changes in heartbeats over a period of time; thus, there are smaller 
time intervals between rapid heartbeats (low HRV) and greater time intervals between slower 
heartbeats (high HRV). Further, HRV has been described as a non-linear, dynamic measurement 
14 
 
 
 
that is subject to considerable fluctuation (Shaffer & Ginsberg, 2017). Within psychophysiology, 
HRV has been used to examine the function of an individual’s regulatory control system as it 
responds to internal and external stimuli through dynamic processes (Lehrer & Eddie, 2013).  
In their discussion of regulatory systems, Lehrer and Eddie (2013) describe several 
features of health and pathology that can be identified by HRV. Overall, positive functioning is 
characterized by oscillation, and dysfunction can be identified by a lack of oscillatory patterns. 
While flexible responses help a system respond to contextual demands, dysfunctional responses 
may be characterized by simplicity (i.e. dampening), randomness, and heightened reactivity. For 
example, the baroreflex system implements regulatory fluctuations in blood pressure, heart rate, 
and vascular tone. These oscillations, which create variability, help to accomplish modulatory 
control of the baroreflex system in response to demands for blood and oxygen (Lehrer & Eddie, 
2013). Accordingly, adaptive HRV regulation is characterized by adequate reactivity and 
recovery, which signifies proper baroreflex control – this may be contrasted by blunted or 
heightened reactivity and shortened or prolonged recovery. 
The dynamic process of HRV within the baroreflex system is largely influenced by the 
counteracting effects of the sympathetic and parasympathetic nervous systems. As with the 
baroreflex system, these two systems within the ANS act as a negative feedback loop: 
epinephrine and norepinephrine are released by sympathetic fibers to facilitate arousal, while 
acetylcholine is released from the parasympathetic fibers to promote recovery to baseline (Lehrer 
& Eddie, 2013). This system is moderated by the sinoatrial node, which has been referred to as a 
cardiac “pacemaker,” as it is subject to various influences that increase and decrease heart rate. 
Further, the sinoatrial node is innervated by the vagus nerve, therefore an individual’s ability to 
exert control to counteract increases in heart rate is termed vagal control; this action is completed 
15 
 
 
 
via activation of the parasympathetic nervous system. During cardiac reactivity monitoring, the 
high frequency HRV band (HF HRV) is considered a measurement of vagal control (Lehrer & 
Eddie, 2013). Thus, HF HRV is used to measure how well an individual exerts control over 
physiological reactivity in order to respond to changes in external and internal stimuli.  
One explanation of the adaptive function of HRV, and the utility of HRV measurement, 
is given by the polyvagal theory (Porges, 2001). Accordingly, ANS regulation is viewed as the 
result of evolutionary processes, whereby cardiac reactivity functions within a larger system of 
social engagement. The theory posits that behavior is rooted in biological functioning, and 
changes in the ANS mediate social functioning. As such, the use of HRV measurement helps to 
identify stress reactivity in the context of engagement and disengagement with various social 
demands. Optimal functioning is viewed as a proper application of social behavior to 
environmental demands, whereas pathological responses reflect inappropriately low or high 
reactivity. Similar to Lehrer and Eddie’s (2013) explanation of the function of negative feedback 
loops within a control system, Porges (2001) describes how failure to properly respond through 
escape behavior and vagal control results in compensatory chemical processes (e.g. moderation 
of corticosteroids and catecholamines) to decrease one’s stress response. Utilization of and 
dependence on these chemical processes – rather than adequate behavioral responses and vagal 
control – contributes to ANS dysregulation.  
In response to this conceptualization of HRV dysregulation as an indicator of 
maladaptive social responses, HRV has also been regarded as a transdiagnostic biomarker of 
mental health conditions (Beauchaine & Thayer, 2015). Similar to the polyvagal theory (Porges, 
2001), this perspective attempts to identify motivation in the context of evolutionary processes. 
Specifically, the authors discuss the interaction of HF HRV within the behavioral inhibition and 
16 
 
 
 
activation systems (BIS/BAS), whereby autonomic dysregulation and maladaptive behavioral 
patterns interact to contribute to psychopathology. This takes place, in part, because heart rate is 
subject to self-regulatory processes within the autonomic nervous system, which can be 
influenced by various states of psychological distress and well-being. As innate motivational 
systems, the BAS network deals with appetitive motivation characterized by goal-approach and 
active avoidance of punishment; this system is commonly associated with more positive health 
outcomes. In contrast, the BIS network is activated during passive avoidance of punishment, 
stimulates an anxiety response, and is associated with more negative health outcomes (Gray, 
1981).  
Of note, while the BIS/BAS systems provide a rationale for ANS dysregulation, as 
measured by HRV, prior research indicates that self-report measures of BIS/BAS are more 
closely associated with affective states rather than physiological responses (Brenner et al., 2005). 
Therefore, the authors conclude that BIS/BAS self-report measures should not be considered a 
surrogate of HRV measurement, and these two measures are seemingly independent of one 
another. In other words, a participant’s self-evaluation of their response to stress (i.e. inhibition 
versus approach) may not accurately reflect their physiological functioning, but instead, more 
closely reflect their affective state. Nonetheless, Beauchaine and Thayer (2015) propose that 
while HF HRV may be considered a transdiagnostic biomarker of psychopathology, this 
association is likely strongest when HRV recording is completed in contexts that engage with the 
BIS/BAS systems (e.g. social-evaluative stressor task), as these contexts require utilization of 
emotion regulation, cognitive control, and behavioral coping.  
Given this relationship between ACEs, toxic stress, emotion regulation difficulties, ANS 
dysfunction, behavioral inhibition, and psychopathology, each of these factors have been 
17 
 
 
 
proposed as underlying mechanisms that contribute to poorer cardiovascular health outcomes. 
The process by which chronic dysregulation of the autonomic nervous system leads to 
physiological dysfunction has been referred to as allostatic load. Unlike allostasis, or the body’s 
response to acute stressors – where sympathetic activity is down-regulated by parasympathetic 
activity so that the body may return to a state of dynamic equilibrium – chronic stressors may 
result in allostatic load, or a permanent off-setting of homeostasis (McEwen & Stellar, 1993). 
Pathological responses within allostasis (i.e. the process of adjustment to return to homeostasis) 
are a result of repeated and continuous stressors, a lack of adaptation, prolonged stress responses, 
and inadequate stress responses (McEwen, 1998). Consequently, an allostatic load perspective 
suggests that these stress responses facilitate changes to internal structures and contribute to 
pathophysiology; namely, those related to the neuroendocrine, immune, metabolic, 
cardiovascular, and respiratory systems (Juster, McEwen, & Lupien, 2010). Due to the high 
influence of these systems on cardiovascular stress reactivity, HRV could be considered one of 
the possible mechanisms by which psychological stress influences physical health via 
physiological dysregulation, and accordingly, may inform future chronic stress interventions for 
vulnerable populations.
18 
 
Proposed Study 
Overview 
The purpose of the proposed study is to examine the association between ACEs and stress 
reactivity from a cardiovascular health perspective, and to better understand the connection 
between childhood adversity and subsequent cardiovascular disease. As previously described, the 
association between ACEs and poorer health, including increased mortality, is well-documented 
(Felitti et al., 1998; Oh et al., 2018; Brown et al., 2009). The negative health effects of ACEs are 
largely due to the prevalence of chronic diseases, and in particular, cardiovascular disease 
(CVD). Further, the ecobiodevelopmental (EBD) framework, which has been adopted by the 
American Academy of Pediatrics, proposes that toxic stress and its associated underlying 
mechanisms are responsible for the progression from elevated ACEs to subsequent poorer health 
(Shonkoff et al., 2012).  
Although research on toxic stress has been increasingly applied to neurological 
development and alteration, very little research to-date has examined the effects of toxic stress 
within the cardiovascular system. A relationship between toxic stress and cardiovascular stress 
reactivity would be expected since other indicators of autonomic dysregulation have already 
been cited in studies of ACEs and cortisol dysregulation (Bunea et al., 2017; Lovallo et al., 
2012). Accordingly, in the proposed study, we seek to identify early signs of cardiovascular 
dysfunction in individuals with elevated ACE scores through the use of HRV monitoring in 
emerging adults, a population in which it is typically considered too early to observe indicators 
of cardiovascular disease.  
The use of ECG laboratory monitoring will be used to derive measurements of HRV. 
Heart rate variability is predominantly associated with dysfunction within domains that involve 
 
 
 
avoidance of perceived social stress, and observed differences in measurement are influenced by 
stress responses within the autonomic nervous system; the habitualization of a stress response 
across various domains is referred to as autonomic dysregulation. While autonomic 
dysregulation takes place at the biophysiological level, it is also influenced by cognitive, 
affective, and behavioral factors. Correspondingly, HRV is influenced by vagal tone – which is 
highly associated with cognitive control – and the motivational systems of behavioral inhibition 
and activation (BIS/BAS) – which help to explain the interaction between social stress, emotion 
regulation, and behavior (Beauchaine & Thayer, 2015). In order to observe this interaction, the 
present study will obtain HRV measurements for participants at rest, during a laboratory stress 
task, and during recovery from the stress task. This will serve as a model through which each 
participant’s stress response can be interpreted as a generalized physiological, affective, and 
behavioral response across various “real-life” contexts. 
In order to generate a stress response that reflects autonomic dysregulation, the present 
study will employ the methodology of the Montreal Imaging Stress Task (MIST) (Dedovic et al., 
2005). The MIST is a standardized computer-based math task that has been used in research 
settings to coincide with HRV monitoring. During task-completion, participants will be asked to 
perform mental computations under time restraint, and response submission will require rapid 
operation of a computer mouse. Additionally, the testing platform provides automated stress-
inducing feedback, in which participants will be frequently alerted of their low performance 
compared to peers. The use of the MIST is warranted for various reasons, but primarily because 
it was designed as a social-evaluative stressor task. This is important, as HRV is conceptualized 
within the polyvagal theory as a measurement of one’s physiological response associated with 
one’s appraisal of social stress and subsequent engagement or disengagement (Porges, 2001). In 
20 
 
 
 
a recent study that compared HRV responses during common laboratory stress tasks (e.g. Trier 
Social Stress Task, Stroop Color-Word Task, speech tasks), the MIST was shown to induce 
higher cardiovascular responses than other tasks, despite similar participant ratings of perceived 
stress across all tasks (Brugnera et al., 2018). Further, the authors of this study noted that stress 
tasks that required participant verbalizations were less able to detect vagal withdrawal, which 
may have been masked due to the influence of respirations. Interpreted alongside of the EBD 
framework, exposure to early toxic stress alters neurodevelopment and leads to maladaptive 
coping, which includes interpreting neutral social stimuli negatively (Iffland, Wiggert, Neuner, 
& Blechert, 2018). Accordingly, elevated exposure to ACEs would suggest an increased 
likelihood of autonomic dysregulation during the MIST, identified by cardiovascular dampening, 
low vagal control, and heightened cardiovascular stress reactivity.  
Since HRV has been conceptualized as a transdiagnostic biomarker of psychopathology 
(Beauchaine & Thayer, 2015), participants will be asked to complete common mental health 
screening measures of depression (PHQ-9), anxiety (GAD-7), post-traumatic stress (PCL-5), and 
attention deficit-hyperactivity (ASRS-5). This may help to identify psychological correlates of 
ACEs and autonomic dysregulation. Additionally, measures of emotion dysregulation will be 
assessed using measures of behavioral inhibition/avoidance (BIS/BAS), emotion difficulties 
(DERS-18), perceived stress (PSS), and resilience (BRS). These are an essential component to 
better understanding the relationship between ACEs and HRV reactivity, as emotion regulation 
difficulties have been shown to be related to neurodevelopmental changes in response to early 
adversity (Lovallo, 2013). 
Further, common types of adaptive and maladaptive behavioral coping styles will be 
assessed using measures of various health and lifestyle behaviors. Participants will be asked to 
21 
 
 
 
complete measures related to exercise (GLTEQ), sleep difficulties (PSQI), alcohol use (AUDIT-
C), tobacco smoking (FTQ), e-cigarette use (EDS), substance misuse (DAST-10), disordered 
eating (ESP), and sexual risk-taking (SRS). These measures may provide important insight about 
the relationship between ACEs, emotion regulation difficulties, and subsequent risk-taking, poor 
health behaviors, and alcohol/substance use risk. To acquire additional information about 
participant’s subjective and objective overall health, the present study will also assess for 
somatic symptoms (PHQ-15), self-rated health, and healthcare utilization. During the lab portion 
of data collection, participants will undergo physical measurements of height, weight, and blood 
pressure. 
Taken together, these measures will allow the present study to identify additional ways 
that psychopathology, cognitive difficulties, emotion dysregulation, and maladaptive behavioral 
coping may be associated with ACEs, autonomic dysregulation, and poorer general health. This 
type of comprehensive assessment will contribute to a greater understanding of the progression 
from childhood adversity to poorer cardiovascular health, which has been well-documented in 
the existing ACEs literature.  
Proposed Aims and Hypotheses 
Aim 1 
To examine the relationship between early adversity and cardiovascular stress reactivity 
in a college student sample, which could provide early indication of cardiovascular disease risk 
among emerging adults with a history of elevated ACEs. 
Hypothesis 1 
22 
 
 
 
Participants who report a history of  ? 4 ACEs on the ACE-Q will demonstrate greater 
autonomic dysregulation during ten minutes of baseline monitoring, measured as significantly 
lower mean scores of HF HRV, compared to participants who report a history of < 4 ACEs. 
Hypothesis 2 
Participants who report a history of  ? 4 ACEs on the ACE-Q will demonstrate greater 
autonomic dysregulation during the MIST stress task, measured as significantly different HF 
HRV, compared to participants who report a history of < 4 ACEs. Two patterns of HRV 
reactivity will be considered maladaptive, as lower cardiovascular reactivity may suggest 
exhaustion and dampening due to chronic stress, and higher cardiovascular reactivity could 
indicate low parasympathetic vagal control, as seen in Figure 1. 
Hypothesis 3 
Compared to participants who report < 4 ACEs on the ACE-Q, participants who report a 
history of  ? 4 ACEs will demonstrate greater autonomic dysregulation during a ten-minute 
recovery period, measured as inadequate return to baseline HRV upon completion of the MIST 
stress task. 
23 
 
 
 
 
Figure 1. Adaptive versus maladaptive HRV responses. The image demonstrates physiological 
responses to stress at baseline, task, and recovery periods. 
Aim 2  
To assess the relationship between ACEs and various forms of psychopathology, as well 
as other transdiagnostic cognitive and affective factors, that are associated with cardiovascular 
stress reactivity. 
Hypothesis 4 
Participants who report a history of  ? 4 ACEs on the ACE-Q will report a higher 
prevalence of psychopathology, measured by clinically-significant self-report scales of 
depression (PHQ-9), anxiety (GAD-7), post-traumatic stress (PCL-5), and attention deficit-
hyperactivity disorder (ASRS-5).  
Hypothesis 5 
Participants who report a history of  ? 4 ACEs on the ACE-Q will have a higher 
prevalence of cognitive and affective difficulties, measured as increased scores of perceived 
24 
 
 
 
stress (PSS), emotion regulation difficulties (DERS-18), disordered eating (ESP), and behavioral 
inhibition/avoidance (BIS/BAS). 
Aim 3 
To evaluate the relationship between ACEs and different lifestyle/health behaviors 
related to maladaptive behavioral coping and poorer physical health. 
Hypothesis 6 
Participants who report a history of  ? 4 ACEs on the ACE-Q will have a higher prevalence of 
maladaptive behavioral coping, measured as decreased exercise (GLTEQ) and increased scores 
of sleep difficulties (PSQI), alcohol use (AUDIT-C), smoking (FTQ), nicotine use (EDS), 
substance misuse (DAST-10), and risky sex behaviors (SRS). 
Hypothesis 7 
Participants who report a history of  ? 4 ACEs on the ACE-Q will have a higher prevalence of 
indicators of poor physical health, such as lower self-rated health (poor, fair, good, very good, 
excellent), and elevated scores of BMI (? 25), blood pressure (systole ? 120; diastole ? 80), 
somatic symptoms (PHQ-15), and healthcare utilization over the past 12 months. 
Proposed Methods 
Participants 
For the proposed study, we seek to enroll approximately 3,000 participants to complete 
online survey measures, and a subsample of 158 participants for an in-person laboratory session 
that includes physical measurements and electrocardiogram (ECG) monitoring. Eligible 
participants will be undergraduate students (ages 18-25) at East Carolina University enrolled in 
25 
 
 
 
the ECU Psychology Department Experimentrak subject pool for their Introductory Psychology 
course. We aim to enroll a sample that is representative of the ECU student population, which is 
approximately 59% female, 66% white/Caucasian, 16% black/African American, 7% 
Hispanic/Latino, 3% Asian, and 8% other/unknown. Additionally, participants will be asked to 
refrain from exercise for 12 hours prior to laboratory testing, and to withhold from ingesting 
products containing tobacco, nicotine, or caffeine for at least one hour before testing. 
Self-Report Measures 
Survey measures will be completed online through Qualtrics, and include a brief 
demographic and medical history questionnaire, a questionnaire of adversity in childhood (ACE-
Q), and measures of psychopathology (PHQ-9, GAD-7, PCL-5, ASRS-5), cognitive and 
affective difficulties (PSS, DERS-18, ESP, BIS/BAS), and resilience (BRS). In addition, 
participants will be asked about lifestyle and health behaviors such as exercise (GLTEQ), sleep 
(PSQI), alcohol use (AUDIT-C), substance misuse (DAST-10), tobacco smoking (FTS), e-
cigarette use (EDS), and sexual risk-taking (SRS). The present study will also assess for 
objective and subjective measures of general physical health by asking participants about 
somatic symptoms (PHQ-15), self-rated health, and healthcare utilization.  
Demographic and Medical History Questionnaire  
Participants will be asked to provide basic demographic information about their age, sex, 
gender, sexual preference, and race/ethnicity. In addition, participants will be screened for a 
history of a diagnosed heart condition or psychological disorder, and will be asked to provide a 
list of active prescription medications; responses will be assessed as potential exclusionary 
criteria by study personnel.  
26 
 
 
 
Adverse Childhood Experiences Questionnaire (ACE-Q) 
 The ACE-Q was initially implemented by Felitti and colleagues (1998), and was 
developed as an epidemiological measure to better understand the relationship between early 
psychosocial stress and subsequent health outcomes. The questionnaire includes ten self-report 
items concerning family dysfunction and abuse/neglect, in which participant total scores are 
quantified as the cumulative number of affirmative responses to instances of adverse events prior 
to age 18; accordingly, each participant’s ACE-Q score will fall between 0 and 10, with higher 
scores indicating more adversity experiences. Extensive research has confirmed a positive dose-
response relationship between ACE scores and negative health outcomes, and a score of four or 
higher has consistently yielded the strongest associations with poorer health (Felitti et al., 1998). 
As such, an ACE-Q score of four has been regarded as a clinical cutoff point.  
While the ACE-Q has been widely disseminated as a screening tool to address a major 
public health concern, the ACEs literature yields a limited understanding of the ACE-Q’s 
psychometric properties. To date, only two studies address the measure’s test-retest reliability, 
and validity statistics have not been well-examined. The earliest study that assessed test-retest 
reliability found that weighted kappa coefficients for ACE score between Time 1 and Time 2 
were acceptable (Cohen’s ? = .64) (Dube, Williamson, Thompson, Felitti, & Anda, 2004). While 
this questionnaire deviated slightly from the ACE-Q, kappa coefficients for individual ACE-Q 
items were emotional abuse (.66), physical abuse (.63), sexual abuse (.69), household substance 
abuse (.75), household mental illness (.51), mother treated violently (.78), parental separation 
(.86), and household member incarceration (.46). Due to the diversity of adversity experiences 
under the ACEs construct, internal consistency between items is not a useful metric to assess the 
measure’s psychometric properties. That said, interrelatedness between items have been 
27 
 
 
 
identified, such that the endorsement of one ACE-Q item increased the odds for endorsing 
additional items; and in one large public health sample (n = 8,629), 87% (median) of the 
respondents who endorsed one ACE-Q item endorsed at least one additional item (Dong et al., 
2004). Furthermore, the adjusted odds of endorsing additional ACE-Q items increased 
significantly as the reported number of ACEs increased, suggesting a coherence within the 
overall ACEs construct.  
Measures of validity and reliability are known obstacles for assessing instances of abuse, 
neglect, maltreatment, and family dysfunction with retrospective self-report questionnaires. That 
said, in one recent sample of college athletes, analysis of test-retest reliability was found to be 
acceptable (r = .71), and a higher stability coefficient was found among items concerning 
household dysfunction (r = .65) compared to items of abuse and neglect (r = .52) (Zanotti et al., 
2018). Nonetheless, issues of validity and test-retest reliability have been found to be a common 
challenge in the childhood abuse and maltreatment literature, and accordingly, several 
consideration must be made while interpreting participant responses. For instance, issues such as 
time lapse since adverse events, subject sensitivity, and memory impairments (potentially 
associated with significant stressful or traumatic events) may be present, and some literature has 
suggested both underestimations and underreporting of instances of abuse among respondents 
(Dube et al., 2004). Additionally, the dichotomous response options on the ACE-Q do not allow 
for information about frequency and intensity of occurrences. The implications of each of these 
limitations must be considered during data collection and statistical analysis.  
Patient Health Questionnaire 9-item (PHQ-9) 
The PHQ-9 (Kroenke, Spitzer, & Williams, 2001) is a brief self-report measure that is 
commonly used to assess depressive symptoms in clinical and research settings. Respondents are 
28 
 
 
 
screened on nine items based on DSM-IV diagnostic criteria for depression over the past two 
weeks, with scores ranging from 0 (Not at all) to 3 (Nearly every day). Scores are then totaled, 
with higher scores reflecting greater levels of depressive symptoms. Total scores of 10 or higher 
were found to have both a sensitivity and specificity of 88% for major depression, and as such, a 
total score of 10 or higher has been accepted as a reliable clinical cutoff point (Kroenke et al., 
2001). 
Generalized Anxiety Disorder 7-item (GAD-7) 
The GAD-7 (Spitzer, Kroenke, Williams, & Löwe, 2006) is a seven-item self-report 
measure that is commonly used to assess anxiety symptoms in clinical and research settings. 
Respondents are asked to rate items from 0 (Not at all sure) to 3 (Nearly every day) to indicate 
the frequency of various anxiety symptoms over the previous two-week period. The 
questionnaire has been found to have high internal consistency (Cronbach ? = .92) and test-retest 
reliability (r = .83). During the validation of the GAD-7, the authors identify a total score of 10 
or higher as a clinical cutoff with good sensitivity (89%) and specificity (82%).   
PTSD Checklist for DSM-5 (PCL-5) 
The PCL-5 is a 20-item PTSD symptom screening tool that asks examiners to report 
issues that have taken place within the past four weeks   (Weathers, Litz, Keane, Palmieri, Marx, 
& Schnurr, 2013). The questionnaire was revised to coincide with the changes in symptom 
criteria between the DSM fourth and fifth editions, and accordingly, each of the 20 items on the 
PCL-5 are derived from DSM-5 symptom criteria for PTSD. Participants are asked to think 
about a stressful experience and rate PTSD-related symptoms over the past month from 0 (Not at 
all) to 5 (Extremely). A participant’s total score ranges from 0-80, and a cutoff score of 31 
29 
 
 
 
indicates the likely presence of sufficient symptoms to meet the DSM-5 criteria for PTSD. The 
psychometric properties of the revised PCL-5 were evaluated following two studies of trauma-
exposed college students, and results demonstrated high internal consistency (Cronbach ? = .94), 
test-retest reliability (r = .82), convergent validity (rs = .74 -.85) and discriminant validity (crs = 
.31-.60) (Blevins, Weathers, Davis, Witte, & Domino, 2015). 
Adult ADHD Self-Report Screening Scale for DSM-5 (ASRS-5) 
The ASRS-5 (Ustun et al., 2017) is a six-item self-report measure of ADHD symptoms 
that asks respondents to rate the frequency of common ADHD symptoms over the past six 
months using a scale of 0 (Never) to 4 (Very Often). The questionnaire was developed as a 
screening tool, and was adapted from a previous version (ASRS-v1.1) to coincide with changes 
to diagnostic criteria upon release of the DSM-5. Use of the updated ASRS-5 yielded an 
identification of ADHD cases from non-cases with 91% sensitivity and 96% specificity in a 
general population sample, and accordingly, a clinical threshold for scores of 14 or higher has 
been recommended. 
Perceived Stress Scale (PSS) 
The PSS (Cohen, Kamarck, & Mermelstein, 1983) is a 10-item self-report measure that 
was developed as a tool to assess how situations and emotions influence levels of perceived 
stress. Respondents are asked to rate items on a scale from 0 (Never) to 4 (Very often) as they 
apply to situations within the past four weeks. After accounting for reverse-scored items (Items 
4, 5, 7 and 8), responses are summed, and total scores are categorized as estimations of low 
stress (0-13), moderate stress (14-26), and high stress (27-40). Initial analysis of internal 
30 
 
 
 
consistency appeared moderate-high in predominantly college student samples (Cronbach ? = 
.84-.86), and test-retest reliability was also found to be high (r = .85).  
Difficulties in Emotion Regulation Scale 18-item (DERS-18) 
The DERS-18 is an 18-item short form of the longer 36-item DERS questionnaire (Gratz 
& Roemer, 2004). This self-report measure asks respondents to respond to items concerning 
various emotion abilities on a 1 (Almost never) to 5 (Almost always) scale, which address areas 
of emotion regulation such as emotional awareness, emotional acceptance, impulse/behavioral 
control, and use of beneficial emotion regulation strategies. Analysis of the measure’s 
psychometric properties indicated high internal consistency (Cronbach ? = .91) of the combined 
scales, with subscales ranging from .77 to .90 (Victor & Klonsky, 2016). Further, authors 
reported high concurrent validity (r = .98) and significant correlations with difficulty regulating 
both negative and positive emotions.   
Eating Disorder Screen for Primary Care (ESP) 
The ESP is a five-item self-report measure that was developed to screen patients for 
potential eating disorders in a primary care setting (Cotton, Ball, & Robinson, 2003). The 
questionnaire asks respondents to answer Yes/No to common questions related to disordered 
eating, and two or more affirmative responses indicated high sensitivity (100%) and adequate 
specificity (71%) in the primary care setting. Subsequently, the ESP was found to be more 
effective at ruling out disordered eating in a mixed college student and primary care sample 
(Kagan & Melrose, 2003).  
Brief Resilience Scale (BRS) 
31 
 
 
 
The BRS (Smith et al., 2008) is a six-item self-report measure of resilience, or one’s 
ability to recover from stress and adversity. Items are labeled from 1 (Strongly Agree) through 5 
(Strongly Disagree), and responses are averaged, with higher scores indicating greater resilience. 
The authors’ analysis of psychometric properties indicated high internal consistency among four 
groups (Cronbach ? = .80-.91).  
Behavioral Inhibition/Behavioral Activation Scales (BIS/BAS) 
The BIS/BAS scales (Carver & White, 1994) are comprised of 24 self-report items, and 
were developed to assess the influence of emotion on motivational behaviors. Response options 
range from 1 (Very true for me) to 4 (Very false for me), and the questionnaire is subdivided into 
four sections: BAS Drive (four items), which examines respondents’ willingness to pursue goals; 
BAS Fun-Seeking (four items), which provides insight about respondents’ tendency to seek out 
new and exciting stimuli; BAS Reward-Responsiveness (five items), which assesses motivation 
for reward; and BIS (seven items), which is used to evaluates one’s responses to potentially 
punishing events. There are also four filler questions that do not factor onto any of the four 
subscales. Analysis of reliability reflects adequate internal consistency between items (Cronbach 
? = .66-.76) (Carver & White, 1994) and test-retest reliability across scales (r = .68-.72) (Sutton 
& Davidson, 1997). These estimates of reliability were replicated in clinical samples, which also 
yielded appropriate convergent and concurrent validity (Campbell-Sills, Liverant, & Brown, 
2004). Additionally, the same authors found a moderate-strong positive correlation for BIS 
scores with neuroticism (r = .75) and depression (r = .48), and a moderate positive correlation 
between BAS scores and positive affect (r = .43).  
Godin Leisure-Time Exercise Questionnaire (GLTEQ) 
32 
 
 
 
The GLTEQ is a self-report measure that asks respondents about exercise frequency over 
a typical 7-day period (Godin, 1969). Participants are asked to tally the total number of mild, 
moderate, and strenuous bouts of leisure-time exercise (at least 15-minutes in duration), and are 
provided with prototypical activities for each category, respectively. To produce a weighted total 
score, response tallies are multiplied by three (mild), five (moderate) and nine (strenuous), and 
then categorical scores are summed. This yields a Godin Scale Score, and cutoff values are set at 
0-13 (Insufficiently Active/Sedentary), 14-23 (Moderately Active), 24 and above (Active). 
Recent  validation studies have supported the use of the GLTEQ, and conclude that the measure 
has greater sensitivity to detect moderate-strenuous exercise rather than mild exercise intensity 
(Motl et al., 2018).  
Pittsburgh Sleep Quality Index (PSQI) 
The PSQI is a 19-item self-report questionnaire that measures subjective sleep quality 
throughout the past month (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). The PSQI 
assesses several factors related to sleep including subjective sleep quality, sleep latency, sleep 
duration, habitual sleep efficiency, sleep disturbances, use of sleep medication, and daytime 
dysfunction over the last month on a scale from 0 (Not during the past month) to 3 (Three or 
more times a week). The sum of scores are calculated to derive a global PSQI score ranging from 
0-21, with scores greater than 5 reflecting poor sleep quality. While the PSQI has been shown to 
have good internal consistency (Cronbach ? = .80) (Carpenter & Andrykowski, 1998) and test-
retest reliability (r = .78), scores have not been found to reliably predict objective sleep data, 
such as those derived from polysomnography (Backhaus, Junghanns, Broocks, Riemann, & 
Hohagen, 2002). Therefore, the PSQI is most appropriately used as to estimate subjective sleep 
quality.  
33 
 
 
 
Alcohol Use Disorders Identification Test - Consumption (AUDIT-C) 
The AUDIT was developed by the World Health Organization as a self-report measure of 
hazardous or harmful alcohol use, and screens for alcohol intake, dependence, and problems 
related to consumption (WHO, 1982). Since its initial use, the AUDIT has been adapted into 
several shortened versions, and item reductions have been found to maintain satisfactory 
psychometric properties (Meneses-Gaya, Zuardi, Loureiro, & Crippa, 2009). One such 
adaptation is the AUDIT-C, which has been widely validated as a brief screening tool for 
problem drinking, and performed better than the AUDIT for detecting heaving drinking (Bush, 
Kivlahan, McDonell, Fihn, & Bradley, 1998). The AUDIT-C is comprised of three self-report 
items, and responses are marked on a 0-4 scale, yielding potential scores from 0-12. 
Furthermore, for use among college students, the AUDIT-C has been found to have better 
detection of at-risk drinking than the AUDIT; and due to the emergence of gender differences in 
drinking among respondents, the authors recommend that cut-off scores be set at 7 for males and 
5 for females (Demartini & Carey, 2012). These gender-based cut-off scores were further 
supported as an indicator of problematic drinking among college students after evaluation of at-
risk drinking in a university primary care setting  (Campbell, C. E. & Maisto, 2018).  
Drug Abuse Screening Test 10-item (DAST-10) 
The DAST-10 is a 10-item self-report measure that was developed as a condensed 
version of the DAST-28 (Skinner, 1982) to be used as a screening tool to assess both prescription 
and illicit substance misuse. Review of its psychometric properties indicates high internal 
consistency (Cronbach ? = .94) among psychiatric patients (Carey, Carey, & Chandra, 2003) and 
good test-retest reliability (r = .78) when compared to the longer DAST-20 version (r = .71) 
(Cocco & Carey, 1998). Further, the DAST-10 was found to be appropriately sensitive to the 
34 
 
 
 
college undergraduate student population, with 85.6% of respondents endorsing at least one 
DAST-10 item, 49.2% endorsing two or more items, and 25% endorsing three or more items 
(McCabe, Boyd, Cranford, Morales, & Slayden, 2006).  
Fagerstrom Tolerance Questionnaire (FTQ) 
The FTQ (Fagerstrom, 1978) was developed as a tool to measure the severity of nicotine 
dependence among smokers. Accordingly, the FTQ has been found to reliably distinguish 
between light and heavy smokers when biochemically validated by other indicators of nicotine 
use, such as cotinine (Pomerleau, O. F., Fertig, & Shanahan, 1983). The FTQ has also been 
shown to have high test-retest reliability (r = .83) among an American sample (Pomerleau, C. S., 
Carton, Lutzke, Flessland, & Pomerleau, 1994). 
E-Cigarette Dependence Scale 4-item (EDS-4) 
The EDS is a self-report measure of e-cigarette dependence (Morean et al., 2018). Upon 
evaluation of psychometric properties, it performed similarly to the longer 22- and 8-item 
versions by demonstrating high internal consistency (Cronbach ? = .86) and good concurrent 
validity. Items ask participants to respond on a scale from 0 (Never) to 4 (Almost Always), with 
higher scores indicating greater e-cigarette nicotine dependence.  
Sexual Risk Survey (SRS) 
The SRS (Turchik & Garske, 2009) is a 23-item self-report questionnaire that was 
developed to assess sexual risk-taking behaviors among college students. A recent meta-analysis 
of different measurement tools of high-risk sexual behaviors confirmed initial findings that the 
SRS has good internal consistency (Cronbach ? = .61-.93) as well as high test-retest reliability (r 
= .93) (Mirzaei, Ahmadi, Saadat, & Ramezani, 2016).  
35 
 
 
 
Patient Health Questionnaire 15-item (PHQ-15) 
The PHQ-15 is a 15-item self-report measure of somatic symptoms that asks respondents 
to answer on a scale from 0 (Not bothered at all) to 2 (Bothered a lot) regarding their experience 
with physical symptoms over the previous four weeks. The questionnaire was normed with the 
use of a predominantly primary care patient population, and items assess for the most common 
physical complaints reported in outpatient medical settings. Additionally, increased scores on the 
PHQ-15 were associated with decreases in symptom-related difficulty, sick days, healthcare 
utilization, and functional status (Kroenke, Spitzer, & Williams, 2002). As such, somatic 
symptom severity cutoff points are denoted as minimal (0-4), low (5-9), medium (10-14), and 
high (15-30).  
Self-Rated Health  
Self-rated health is a commonly used measure that can be used to estimate participants’ 
subjective understanding of their overall health. Self-rated health measures typically utilize a 
brief single-item format. For the purposes of the present study, participants will be asked “In 
general, would you say your health is,” with the item responses of “(1) poor,” “(2) fair,” “(3) 
good,” “(4) very good,” and “(5) excellent.” This type of informal assessment tool has been 
widely used across healthcare and research applications, and test-retest reliability has been found 
to be moderate (r = .75) when participants from ages 20-80 years were asked on two occasions at 
approximately four weeks apart (Zajacova & Dowd, 2011). 
Healthcare Utilization 
Brief assessments of one’s frequency of engagement with healthcare services can provide 
additional information about participants’ general health status. For the present study, 
36 
 
 
 
participants will be asked to indicate the frequency over the last 12 months with which they have 
received services for both physical and mental health conditions, and responses will be separated 
categorically. Scores will be calculated as the total number of appointments attended, by 
category.  
Laboratory Measurements and Procedures  
Following the completion of online survey measures, participants will be invited into a 
laboratory setting to complete measures of height, weight, and blood pressure, as well as the 
MIST stress task during HRV monitoring. It is likely that the sample will yield more responses 
for the comparison group (< 4 ACEs) than the ACEs group (? 4 ACEs), therefore, an 
oversampling method will be used to capture more participants with clinically significant ACE 
scores. For respondents who report < 4 ACEs, these participants will be randomly selected for 
laboratory testing to act as the comparison group. All participants who report ? 4 ACEs will be 
invited to participate in the laboratory portion. The proportion of participants in each group will 
be monitored and sampling methods increased/decreased, as indicated. 
 Upon arrival for lab testing, participants will be informed by a lab assistant that they are 
being asked to take part in a study about how mathematic abilities are related to other 
psychological factors. Next, participants will also be asked if they have followed requests to 
refrain from exercise within the 12 hours prior to testing and from ingesting tobacco, nicotine, or 
caffeine within 60 minutes prior to testing; responses will be recorded and later analyzed for 
confounding effects. After completing the informed consent process, participants will have their 
height and weight digitally measured on a DETECTO research scale by a lab assistant. Then, 
after a five-minute resting period, participants will have their blood pressure measured with the 
use of a CARESCAPE V100 device, applying the cuff placement and body positioning 
37 
 
 
 
procedural guidelines recommended by the American Heart Association (AHA) (Smith, L., 
2005). Lab assistants will observe standardized procedures throughout testing. 
Next, participants will be asked to complete the Montreal Imaging Stress Task (MIST) 
during HRV monitoring. Upon entry into the monitoring booth, participants will be connected to 
an electrocardiogram recording device and respiratory cuff; HRV will be measured with a Lead 
II ECG using a BIOPAC MP150, and will be processed with the AcqKnowledge software. Prior 
to starting the MIST, participants will be asked to sit quietly and observe a neutral stimulus for 
ten minutes while baseline HRV monitoring is performed. The lab assistant will leave the 
monitoring booth for this portion of psychophysiological monitoring, but upon completion, will 
re-enter and remain seated behind the participant in the booth for the subsequent Training, 
Control, and Experimental phases. Of note, the lab assistant will observe and record the times at 
which the participant begins and ends each task. Following baseline monitoring, a lab assistant 
will provide a brief orientation to the MIST task during the Training phase (2-5 minutes). This 
phase will provide examiners with brief practice, and the MIST program will analyze responses 
and appropriately set the difficulty level and time limit per-item during the Experimental 
(Stressor) phase.  
After the completing the Training phase participants will proceed to a Control and 
Experimental condition, each approximately five minutes in duration, which both require 
participants to complete mental arithmetic computations. The first task is considered a Control 
phase, as it does not apply the same time restrictions and social-evaluative feedback of the 
Experimental phase. After completing the Control phase, participants will be asked by the lab 
assistant to provide a verbal response to a brief measure of task engagement (Level of task 
difficulty: 1-10). After their response is recorded, participants will begin the Experimental phase, 
38 
 
 
 
which is designed to generate the most pronounced stress response, and combines arithmetic 
calculations with an imposed time requirement and a social-evaluative stressor. Throughout this 
phase of the MIST, the computer monitor will occasionally prompt the examiner to improve their 
performance due to less than average correct and timely responses. If a participant questions the 
lab assistant for clarification, they will provide a standardized response such as “Just do your 
best.” Finally, upon completion, the lab assistant will reorient the participant to a neutral 
stimulus and instruct the participant to sit quietly until the lab assistant returns. This will allow 
for ten minutes of HRV recording during the recovery period. 
Once the MIST task has been completed, participants will be debriefed on the purpose of 
the study by a lab assistant. They will also be informed about aspects of deception that were 
utilized throughout data collection, and a lab assistant will explain that the same feedback was 
provided to all participants during the MIST task, regardless of performance. Participants will be 
given the opportunity to ask follow-up questions about their participation, and will be provided 
with a handout that details a list of resources for obtaining healthcare services, if desired.  
Data Cleaning and Preparation 
Following HRV data collection, laboratory assistants will analyze the data using an 
automated software by identifying instances of peak frequency of the high frequency (HF) band 
(0.15–0.4?Hz), known as R peaks. Quality control measures will be taken to correct for incorrect 
identification of R peaks by the AcqKnowledge software. HF power will be calculated in five-
minute increments and then averaged accordingly to create each of the three data collection 
periods (baseline, stressor task, recovery). This method ensures uniformity in the length of each 
recording that contributes to the HF power calculation.  
39 
 
 
 
Once data has been screened and appropriate corrections have been made, HRV scores 
will be transformed using the percent deviation method, as is recommended to reduce the 
potential for Type I and Type II errors (Ellis, Sollers Iii, Edelstein, & Thayer, 2008). Using this 
transformation method, scores at each time period (baseline, stressor task, recovery) will be 
calculated as the deviation from the average HRV score across all time periods. When compared 
to other common transformation methods, a percent deviation transformation has been shown to 
indicate less false positives than reactivity transformations and less false negatives than 
log transformations. All of the collected data, including HRV analysis, blood pressure, physical 
measurements, and self-report survey measures will be compiled in Microsoft Excel and 
prepared for statistical analysis. Associations between covariates and factors of interest will be 
explored using correlation analyses, t-tests or chi square, as indicated. Results of these analyses 
will identify potential covariates. 
Proposed Statistical Analysis 
The primary outcome under examination is the relationship between ACEs and HF HRV, 
a measure of cardiovascular stress reactivity. Accordingly, a 2x3 mixed factorial ANCOVA 
design will be used to explore the interaction of the two levels of the independent variable 
(ACEs_Low [0-3]; ACEs_High [4-10] with measures of cardiovascular reactivity during three 
time periods of a stressor task (Baseline; Stressor Task; Recovery). Prior to ANCOVA analysis, 
covariates will be empirically chosen (e.g., age, sex, gender, BMI, exercise frequency [GLTEQ], 
depression [PHQ-9], anxiety [GAD-7]). Using the GPower.3.1 software, a priori power analysis 
determined that this study will require a total sample size of 158 participants to detect a medium 
effect (f = 0.25) for fixed effects, main effects, and interactions of the two ACE groups and each 
of the three time points (numerator df = 2, ? = 0.05, ? = 0.80).  
40 
 
 
 
Secondary analysis will examine bivariate correlations between ACEs and participant 
scores on measures of psychopathology, as well as other common transdiagnostic measures of 
cognitive and emotion regulation difficulties. Correlation analysis will also examine the 
relationship between ACEs and lifestyle behaviors, in addition to other indicators of poorer 
physical health. Significant correlations will also be analyzed with regression analysis to identify 
potential ways in which ACE-Q scores may predict these various mental and physical health 
factors.
41 
 
Current Study 
Due to logistical barriers imposed by the COVID-19 pandemic, psychophysiological 
laboratory data collection was not completed. Accordingly, study aims, hypotheses, and 
statistical analyses were adjusted to focus on aspects of the research question that could be 
addressed using online survey data only. Therefore, our former focus on heart rate variability 
(HRV) analysis has been removed. The following analyses focus on self-reported symptoms of 
mental health, physical health, and lifestyle health behaviors to better-understand their 
associations with adverse childhood experiences (ACEs) in emerging adults.  
Aims and Hypotheses 
Aim 1 
To assess the relationship between ACEs and various forms of psychopathology, as well 
as other transdiagnostic cognitive, behavioral, and affective factors. 
Hypothesis 1 
Participant scores on the ACE-Q will be positively associated with increased symptoms 
of depression (PHQ-9), anxiety (GAD-7), post-traumatic stress (PCL-5), attention deficit-
hyperactivity disorder (ASRS-5).  
Hypothesis 2  
Participant scores on the ACE-Q will be positively associated with increased cognitive, 
behavioral, and emotion regulation difficulties, measured as increased scores of perceived stress 
(PSS), emotional difficulties (DERS-18), disordered eating (ESP), and behavioral inhibition 
(BIS); we predict that the ACE-Q will be negatively associated with resilience (BRS) and 
behavioral approach (BAS) subscales of Drive, Fun Seeking, and Reward Responsiveness. 
 
 
 
Aim 2  
To evaluate the relationship between ACEs and lifestyle behaviors associated with poorer 
physical health, as well as other indicators of poorer health.  
Hypothesis 3 
Participant scores on the ACE-Q will be positively associated with lifestyle behaviors 
associated with poorer physical health, measured as decreased self-reported exercise (GLTEQ) 
and increased scores of sleep difficulties (PSQI), alcohol use (AUDIT-C), e-cigarette use (EDS), 
substance misuse (DAST-10), and risky sex behaviors (SRS). 
Hypothesis 4 
Participant scores on the ACE-Q will be positively associated with indicators of poor 
physical health, such as lower self-rated health (poor, fair, good, very good, excellent), and 
increased self-report of somatic symptoms (PHQ-15) and healthcare utilization over the past 12 
months. 
Aim 3 
Through exploratory analysis based on the results of Aims 1 and 2, to better understand 
the association between the ACE-Q subscales – categorized as Childhood Maltreatment (items 1-
5) and Household Dysfunction (items 6-10) – and measures of mental and physical health that 
were significantly associated with the ACE-Q.  
Hypothesis 5 
Participant scores on the Child Maltreatment subscale, compared to scores on the 
Household Dysfunction subscale, will be more strongly associated with scores on mental health 
43 
 
 
 
measures of depression (PHQ-9), anxiety (GAD-7), post-traumatic stress (PCL-5), attention 
deficit-hyperactivity disorder (ASRS-5), emotion dysregulation (DERS-18), perceived stress 
(PSS), and disordered eating (ESP). 
Hypothesis 6 
Participant scores on the Child Maltreatment subscale, compared to scores on the 
Household Dysfunction subscale, will be more strongly associated with scores on physical health 
measures of sleep difficulties (PSQI), illicit substance use (DAST-10), e-cigarette use (EDS), 
and somatic symptoms (PHQ-15).
44 
 
Methods 
Participants and Demographics 
For the present study, 236 undergraduate students between the ages of 18 and 23 years 
completed an online survey. The survey was made accessible via the ECU Psychology 
Department Experimentrak subject pool, and upon completion, students received participation 
credit for an Introductory Psychology course. Participant data were collected over the span of the 
Spring 2020 semester. Upon analysis of demographic information, the sample was found to be 
51.8% female sex-at-birth, 48.2% male sex-at-birth, and 0.0% intersex. Additionally, 52.6% of 
respondents identified as female gender, 47.4% identified as male gender, and 0.0% identified as 
transgender, gender-nonconforming, or preferred not to answer. The median age of the total 
sample was 19.0 years. Racial/ethnic characteristics were 66.2% White/Caucasian, 22.1% 
Black/African American, 4.4% multiracial, 3.9% Hispanic or Latino, 3.1% Asian, 0.9% 
American Indian or Alaska Native, and 0.4% Other. These findings are consistent with recent 
student racial/ethnic demographic data provided by the university, which suggests a sample that 
is representative of the overall student population. Responses on sexual orientation showed that 
participants were 91.7% heterosexual, 4.4% bisexual, 1.8% questioning, and 0.9% gay/lesbian; 
1.3% of participants preferred not to answer. On a proxy measure of SES, 8.6% of participants 
reported that their families receive some form of government financial assistance, including 
various food and housing resources.  
Data Cleaning 
Following the data collection period, the final dataset was exported to SPSS for further 
data cleaning and analysis. Prior to analysis, participant data was thoroughly checked for 
validity. Accordingly, participants whose response times fell within the lowest quartile (less than 
 
 
 
17.2 minutes) and those with less than 90% survey completion were removed from the sample, 
reducing the sample from 236 to 170 participants. Of the remaining sample, data was checked 
for duplicates and invalid responses, such as outliers and nonsensical text. When necessary, 
invalid responses to individual survey items were deleted. Additional data cleaning was 
performed so that participant text responses were usable for analysis. When appropriate, text 
responses were converted to numeric responses (e.g., everyday = 7 days) and means were used to 
replace number ranges that were entered as text (e.g., 3-5 = 4 days).  
Descriptive Statistics 
 After correcting for unreliable, invalid, and missing data, descriptive statistics were 
analyzed for each measure of the online survey. To highlight certain measures of greater 
importance, some measures that did not reach statistical or clinical significance have been 
removed from this section.  
Adverse Childhood Experience Questionnaire (ACE-Q) 
 Descriptive statistics of ACE-Q scores in the final sample (n=170) yielded a mean score 
of 1.75 out of 10 (sd=2.08). The frequency of ACE-Q scores were: 0 (63; 37.1%), 1 (35; 20.6%), 
2 (27; 15.9%), 3 (17; 10.0%), 4 (7; 4.1%), 5 (9; 5.3%), 6 (5; 2.9%), 7 (3; 1.8%), 8 (3; 1.8%), 9 
(0; 0.0%), 10 (1; 0.6%). Using the suggested clinical cutoff score of ? 4, 28 participants (16.5%) 
reached this threshold, and 142 (83.5%) reported an ACE score from 0-3. Additionally, within 
the final sample, more participants endorsed affirmative responses on items within the 
Household Dysfunction subcategory (ACE-Q items 6-10: 96; 56.5%) than items within the 
Childhood Maltreatment subcategory (ACE-Q items 1-5: 64; 37.6%). 
46 
 
 
 
       
Figure 2. Frequency of Adverse Childhood Experiences.  
Patient Health Questionnaire 9-item (PHQ-9) 
 Upon analysis of depression symptoms endorsed on the PHQ-9, the final sample revealed 
a mean score of 6.41 out of 27 (sd=5.63). Based on the suggested clinical cutoffs, 80 participants 
(47.1%) did not screen positive for depression, while 47 (27.6%) screened positive for mild 
depression, 26 (15.3%) screened positive for moderate depression, 9 (5.3%) screened positive for 
moderately severe depression, and 8 (4.7%) screened positive for severe depression. 
Generalized Anxiety Disorder 7-item (GAD-7) 
 Analysis of participant self-reported anxiety symptoms on the GAD-7 indicated a mean 
score of 5.25 out of 21 (sd=4.92). Using the suggested clinical cutoff scores, 92 participants 
(54.1%) did not exhibit clinically significant anxiety symptoms, 41 (24.1%) exhibited mild 
anxiety symptoms, 29 (17.1%) exhibited moderate anxiety, and 8 (4.7%) exhibited severe 
anxiety. 
47 
 
 
 
PTSD Checklist for DSM-5 (PCL-5) 
 Self-reported post-traumatic stress symptoms on the PCL-5 revealed a mean score of 
13.26 out of 80 (sd=14.44). Using the clinical cutoff of 33, which has been used to represent a 
provisional diagnosis of PTSD, 147 participants (86.5%) fell below the cutoff score and 23 
(13.5%) met criteria for PTSD.  
Adult ADHD Self-Report Screening Scale for DSM-5 (ASRS-5) 
 Participant responses to items on a measure of ADHD symptoms indicated a mean score 
of 6.99 out of 24 (sd=4.26). Using the clinical cutoff score (?14), 158 participants (92.9%) fell 
below, and 12 (7.1%) exceeded, this threshold.  
Perceived Stress Scale (PSS) 
 Responses on items of the PSS revealed a mean score of 17.85 out of 40 (sd=6.38). Of 
these responses, 28 participants (16.5%) recorded scores within the low stress range (0-13), 136 
participants (80.0%) recorded scores within the moderate stress range (14-26), and 6 participants 
(3.5%) recorded scores within the high stress range (27-40).  
Difficulties with Emotion Regulation Scale 18-item (DERS-18) 
 Responses on this measure of emotion dysregulation showed a mean score of 38.62 out 
of 90 (sd=13.63). While there is no suggested clinical cutoff score, 134 participants (78.8%) 
recorded scores below, and 36 (21.2%) recorded scores above, 50; scores above 50 could be 
interpreted to indicate concerning levels of emotion regulation difficulties.  
Eating Disorder Screen for Primary Care (ESP) 
48 
 
 
 
 Responses on a measure of disordered eating symptoms indicated a mean score of 1.38 
out of 5 (sd=0.10). Based on the suggested cutoff score (?2) used to indicate concern for 
disordered eating in the primary care setting, the distribution of scores showed that 97 
participants (57.0%) endorsed no concerns for disordered eating history, and 73 participants 
(43.0%) endorsed some concerns for disordered eating history. 
Brief Resilience Scale (BRS) 
 Due to the inclusion of participants with ? 90% of survey completion, responses on the 
BRS were taken from 68 participants in the final sample (n=170). Responses to items of 
resilience showed a mean score of 20.66 out of 30 (sd=5.68), with higher scores representing 
greater participant resilience.  
Behavioral Inhibition and Behavioral Activation Scales (BIS/BAS) 
 Participant responses to items of behavioral inhibition yielded a mean score of 19.79 out 
of 28 (sd=4.03). Responses to items of behavioral activation are broken down by subscale; 
responses to the BAS-Drive subscale demonstrated a mean score of 11.43 out of 16 (sd=2.20), 
responses to the BAS-Reward Responsiveness subscale revealed a mean score of 17.49 out of 20 
(sd=1.87), and responses to the BAS-Fun Seeking subscale revealed a mean score of 12.44 out of 
16 (sd=2.20). 
Godin Leisure-Time Exercise Questionnaire (GLTEQ) 
 Responses on a measure of exercise over the past week were broken down into categories 
of mild, moderate, and strenuous exercise. After scores were weighted and totaled, results 
indicated a mean score of 49.69 (sd=28.39). Twelve participants (7.1%) fell within the Sedentary 
49 
 
 
 
range (0-13), 14 participants (8.2%) fell within the Moderately Active range (14-23), and 144 
participants (84.7%) fell within the Active range (24 and above).  
Pittsburgh Sleep Quality Index (PSQI) 
 Due to the inclusion of participants with ? 90% of survey completion, responses on the 
PSQI were taken from 139 participants in the final sample (n=170). Responses on a measure of 
sleep difficulties demonstrated a mean score of 6.98 out of 21 (sd=3.78). Using the suggested 
cutoff score of ? 5 as an indicator of poor sleep quality, 94 participants (67.6%) reached this 
threshold.   
Alcohol Use Disorders Identification Test – Consumption (AUDIT-C) 
 Due to the inclusion of participants with ? 90% of survey completion, responses on the 
AUDIT-C were taken from 138 participants in the final sample (n=170). Responses on a measure 
of alcohol use indicated a mean score of 4.01 out of 12 (sd=2.29). Sixty-eight participants 
(49.3%) scored below the suggested cutoff for hazardous drinking (?4), and 70 participants 
(50.7%) met the suggested criteria for hazardous drinking.  
Drug Abuse Screening Test 10-item (DAST-10) 
 Participant responses on this measure of illicit substance use demonstrated a mean score 
of 1.74 out of 10 (sd=0.12). Responses were distributed such that 13 participants (7.6%) scored 
within the “no problem” range, 123 participants (72.4%) scored in the “low problem” range, 28 
participants (16.5%) scored in the “moderate problem” range, 5 participants (3.0%) scored in the 
“substantial problem” range, and 1 participant (0.6%) scored in the “severe problem” range.   
Fagerstrom Tolerance Questionnaire (FTQ) 
50 
 
 
 
 Only eight participants (4.7%) in the total sample (n=170) reported that they smoke 
cigarettes on most days. Due to a low number of responses this measure was not included in the 
results.  
E-cigarette Dependence Scale 4-item (EDS-4) 
 Forty-nine participants (28.8%) indicated that they use an e-cigarette on most days. 
Reponses to items on this measure of e-cigarette dependence yielded a mean score of 2.10 out of 
16 (sd=0.28). One hundred eleven participants (65.3%) indicated that they “never” experience 
symptoms of e-cigarette dependence; on average, 21 participants (12.4%) endorsed that they 
“rarely” experience symptoms of e-cigarette dependence, 17 participants (10.1%) endorsed that 
they “sometimes” experience symptoms of e-cigarette dependence, and 4 participants (2.4%) 
endorsed that the “often” experience symptoms of e-cigarette dependence.  
Sexual Risk Survey (SRS) 
 Due to the inclusion of participants with ? 90% of survey completion, responses on the 
SRS were taken from 134 participants in the final sample (n=170). Participant responses were 
separated into four ordinal categories to reduce variability, in which higher scores indicated 
higher sexual risk-taking behaviors. Fifty-seven participants (42.5%) scored in the lowest risk 
category, 36 participants (26.9%) scored in the second lowest risk category, 26 participants 
(19.4%) scored in the second highest risk category, and 15 participants (11.2 %) scored in the 
highest risk category. 
Patient Health Questionnaire 15-item (PHQ-15) 
 Due to the inclusion of participants with ? 90% of survey completion, responses on the 
PHQ-15 were taken from 140 participants in the final sample (n=170). Self-reported somatic 
51 
 
 
 
symptoms on this measure yielded a mean score of 6.58 out of 30 (sd=4.60). The breakdown of 
participant responses were such that 49 participants (35.0%) fell in the category of minimal 
somatic symptom severity (Total Score=0-4), 58 participants (41.4%) fell in the category of low 
symptom severity (Total Score=5-9), 25 participants (17.9%) fell in the category of medium 
symptom severity (Total Score=10-14), and 8 participants (5.7%) fell in the category of high 
symptom severity (Total Score=15-30).  
Self-Rated Health 
 On a subjective measure of health, 28 participants (16.5%) rated their health as 
“excellent”, 15 participants (8.8%) rated their health as “very good”, 54 participants (31.8%) 
rated their health as “good”, 72 participants (42.4%) rated their health as “fair”, and 1 participant 
(0.6%) rated their health as “poor”. 
Healthcare Utilization 
 On a measure of previous 12-month healthcare utilization, calculated as the total number 
of physical and mental health visits, 71 participants (42.0%) reported 0 visits, 60 participants 
(35.5%) reported 1-3 visits, 22 participants (13.0%) reported 4-9 visits, and 16 participants 
(9.5%) reported 10 or more visits.
52 
 
Statistical Analysis and Results 
Aim 1 
The primary aim of this study was to examine the relationship between ACEs and various 
forms of psychopathology in emerging adults, as well as other transdiagnostic cognitive, 
behavioral, and affective factors. For Hypothesis 1, Pearson correlations and hierarchical linear 
regression analyses were used to examine the relationships between ACE scores and self-
reported symptoms of depression, anxiety, PTSD, and ADHD. Concerning Hypothesis 2, the 
same correlation and regression analyses were used to assess the relationship between ACE 
scores and measures of emotion dysregulation, perceived stress, resilience, and behavioral 
inhibition/approach.  
Hypothesis 1 – Correlation Analysis 
Consistent with Hypothesis 1, correlation analysis demonstrated moderate-strong positive 
correlations between scores on the ACE-Q and symptoms of depression (PHQ-9; r=.42), anxiety 
(GAD-7;  r=.41), PTSD (PCL-5; r=.48), and ADHD (ASRS-5; r=.40); all were significant at the 
p<.01 level. These associations can be seen in Table 1. 
  ACE-Q PHQ-9 GAD-7 PCL-5 ASRS-5 
ACE-Q  —              
PHQ-9  0.421 *** —           
GAD-7  0.412 *** 0.724 *** —        
PCL-5  0.479 *** 0.723 *** 0.712 *** —     
ASRS-5  0.397 *** 0.663 *** 0.652 *** 0.647 *** —  
Table 1. Hypothesis 1 Correlation Matrix. Note: * p < .05, ** p < .01, *** p < .001 
Hypothesis 1 – Regression Analysis 
 
 
 
Next, hierarchical linear regression analyses were completed to further examine the 
relationship between the ACE-Q and screening measures for depression, anxiety, PTSD, and 
ADHD. In the first step of each regression model, we controlled for demographic covariates – 
including gender, race/ethnicity, and SES variables – to see if they significantly accounted for 
variance on scores for each screening measure. In the second step of each model, the ACE-Q was 
added as a predictor variable to assess for additional variance in the scores for each screening 
measure, respectively. The second step was also assessed for significant predictor variables in 
the total model. The results of these hierarchical linear regression analyses are reported below.  
Results for the PHQ-9 demonstrated that demographic covariates in the first step 
accounted for 7.7% of variance in depression scores, which was significantly different than zero 
(F(3, 166) = 4.63, p<.01). When the ACE-Q was added in the second step, this accounted for an 
additional 15.3% of variance (F?(1,165) = 32.87, p<.01). In the total model, only ACE-Q scores 
(B = 1.10, p<.01) and race/ethnicity (B = 2.23, p<.01) were significant predictors of scores on the 
PHQ-9, such that being White (vs non-White) and higher ACE-Q scores predicted higher self-
reported depressive symptoms over the past two weeks. See Appendix A for the effects of each 
predictor variable. 
For the GAD-7, results showed that demographic covariates in the first step accounted for 
7.6% of variance in anxiety scores, which was significantly different than zero (F(3, 166) = 4.58, 
p<.01). When the ACE-Q was added in the second step, this accounted for an additional 15.4% 
of variance (F?(1,165) = 33.10, p<.01). In the overall model, ACE-Q scores (B = 0.97, p<.01) 
and race/ethnicity (B = 2.23, p<.01) significantly predicted participant scores on the GAD-7, 
such that higher ACE-Q scores and being White (vs non-White) predicted higher self-reported 
54 
 
 
 
anxiety symptoms over the past two weeks. See Appendix A for the effects of each predictor 
variable.  
Regression analysis for the PCL-5 indicated that demographic covariates accounted for 
2.8% of variance in PTSD scores, but this first step of the model was not statistically significant 
(F(1,166) = 1.61, p = .19). When the ACE-Q was added in the second step, this accounted for an 
additional 22.8% of variance, which was significantly different than zero (F?(1,165) = 50.69, 
p<.01). Only ACE-Q scores were a significant predictor of PCL-5 scores in the overall model (B 
= 3.44, p<.01), such that higher ACE-Q scores predicted higher self-reported distress from PTSD 
symptoms over the past month. See Appendix A for the effects of each predictor variable. 
Results for the ASRS-5 demonstrated that the demographic covariates accounted for 
5.3% of variance in ADHD scores, and this was significantly different from zero F(1,166) = 
3.07, p=.03). When ACE-Q scores were added in the second step, this accounted for an 
additional 14.0% of total variance, and this step was also statistically significant (F(4, 165) = 
28.51, p<.01). In the total model, ACE-Q scores (B = 0.79, p<.01) and race/ethnicity (B = 1.43, 
p=.03) significantly predicted scores on the ASRS-5, such that higher ACE-Q scores and being 
White (vs non-White) predicted higher self-reported ADHD symptoms within the last 6 months. 
See Appendix A for the effects of each predictor variable.  
Hypothesis 2 – Correlation Analysis 
Correlation analysis for Hypothesis 2 yielded a moderate-strong positive correlation 
between scores on the ACE-Q and emotion dysregulation (DERS-18; r=.42). Additionally, 
results indicated a moderate positive association between the ACE-Q and disordered eating 
(ESP; r=.26) and perceived stress (PSS; r=.24). We observed a moderate negative association 
55 
 
 
 
between the ACE-Q and scores of behavioral approach-reward responsiveness (BAS-Reward 
Responsiveness; r=-.23). All were significant at the p<.01 level. Correlations between the ACE-
Q and scores of behavioral inhibition, behavioral approach-drive, and behavioral approach-fun 
seeking did not reach significance. Further details concerning these associations can be seen in 
Figure 4. 
BAS_ BAS_ BAS_ 
  ACE-Q DERS18 PSS ESP BRS BIS 
Reward Drive Fun 
ACE-Q  —                          
DERS18  0.420 *** —                       
PSS  0.241 ** 0.406 *** —                    
ESP  0.292 *** 0.458 *** 0.338 *** —                 
BRS  -0.232  -0.752 *** -0.450 *** -0.453 *** —              
BIS  0.052  0.385 *** 0.272 *** 0.340 *** -0.613 *** —           
BAS_Reward  -0.234 ** -0.187 * 0.017  -0.050  0.228  0.209 ** —        
BAS_Drive  -0.062  -0.190 * -0.011  -0.104  0.268 * -0.070  0.449 *** —     
BAS_Fun  0.024  0.056  0.036  -0.013  0.097  -0.131  0.364 *** 0.512 *** —  
Table 2. Hypothesis 2 Correlation Matrix. Note: * p < .05, ** p < .01, *** p < .001 
 
 Hypothesis 2 – Regression Analysis 
Next, hierarchical linear regression analyses were completed to examine the relationship 
between the ACE-Q and emotion dysregulation, perceived stress, resilience, and behavioral 
inhibition/approach. In the first step of each regression model, we controlled for demographic 
covariates (gender, race/ethnicity, and SES) to see if they significantly accounted for variance on 
scores for each measure. In the second step of each model, the ACE-Q was added as a predictor 
variable to assess for additional variance in the scores for each measure, respectively. The second 
step was also assessed for significant predictor variables in the total model. The results of these 
hierarchical linear regression analyses are reported below.  
56 
 
 
 
Results for the DERS-18 demonstrated that demographic covariates in the first step 
accounted for 4.6% of variance in depression scores, but this did not reach statistical significance 
(F(3,166) = 2.64, p=.51). When the ACE-Q was added in the second step, this accounted for an 
additional 15.5% of variance, which was significantly different than zero (F?(1,165) = 31.99, 
p<.01). In the total model, only the ACE-Q (B = 2.68, p<.01) was a significant predictor of 
scores on the DERS-18, such that higher ACE-Q scores predicted higher self-reported emotion 
dysregulation. See Appendix A for the effects of each predictor variable. 
Regression analysis for the ESP showed that demographic covariates in the first step 
accounted for 12.3% of variance in scores of self-reported disordered eating, and this was 
significantly greater than zero (F(3, 166) = 7.76, p<.01). When the ACE-Q was added in the 
second step, this accounted for an additional 6.9% of variance, which was also statistically 
significant (F?(1,165) = 14.05, p<.01). In the total model, ACE-Q scores (B = 0.17, p<.01) and 
gender (B = 0.69, p<.01) were significant predictors of ESP scores, such that higher ACE-Q 
scores and being female (vs male) predicted more self-reported symptoms of disordered eating 
history. See Appendix B for the effects of each predictor variable. 
Regression analysis for the PSS indicated that demographic covariates in the first step 
accounted for 6.5% of variance in perceived stress scores, which was significantly different than 
zero (F(3,166) = 3.85, p=.01). When the ACE-Q was added in the second step, this accounted for 
an additional 5.2% of variance, and this was also significant (F?(1,165) = 9.70, p<.01). In the 
total model, ACE-Q scores (B = 0.73, p<.01) and race/ethnicity (B = 2.48, p<.01) were 
significant predictors of scores on the PSS, such that being White (vs non-White) and higher 
ACE-Q scores predicted higher self-reported perceived stress during the past month. See 
Appendix A for the effects of each predictor variable. 
57 
 
 
 
For the BRS, demographic covariates in the first step accounted for 21.3% of variance in 
resilience scores, which was significantly different than zero (F(3,64) = 5.78, p<.01). When the 
ACE-Q was added in the second step, this accounted for an additional 3.6% of variance, though 
this step was not statistically significant (F?(1,63) = 3.03, p=.09). In the total model, only 
race/ethnicity significantly predicted BRS scores (B = 4.70, p<.01), such that being non-White 
(vs White) predicted higher scores of resilience. See Appendix A for the effects of each predictor 
variable. 
Results for the BIS showed that demographic covariates in the first step accounted for 
14.6% of variance in scores of behavioral inhibition, which was significantly different than zero 
(F(3,166) = 9.44, p<.01). When the ACE-Q was added in the second step, this accounted for an 
additional 0.1% of variance, and this step did not reach statistical significance (F?(1,165) = 0.14, 
p=.71). In the total model, gender was the only significant predictor of BIS scores (B = 2.85, 
p<.01), such that being female (vs male) predicted higher behavioral inhibition scores. See 
Appendix A for the effects of each predictor variable. 
For the BAS-Reward Responsiveness subscale, demographic covariates in the first step 
accounted for 2.1% of variance in scores of behavioral approach, but this fell short of statistical 
significance (F(3,166) = 1.79, p=.17). When the ACE-Q was added in the second step, this 
accounted for an additional 7.4% of variance, which was significantly different than zero 
(F?(1,165) = 13.54, p<.01). In the total model, both ACE-Q scores (B = 0.25, p<.01) and gender 
(B = 0.69, p=.02) significantly predicted BAS-Reward Responsiveness scores, such that lower 
ACE-Q scores and being female (vs male) predicted higher behavioral approach scores. See 
Appendix A for the effects of each predictor variable. 
58 
 
 
 
Regression analysis for the BAS-Drive subscale revealed that demographic covariates in 
the first step accounted for 1.8% of variance in scores of behavioral approach, which was not 
significantly different than zero (F(3, 166) = 0.99, p=.40). When the ACE-Q was added in the 
second step, this did not account for any additional variance, falling short of statistical 
significance (F?(1,165) = 0.04, p=.85). In the total model, none of the predictor variables 
significantly predicted BAS-Drive scores. See Appendix A for the effects of each predictor 
variable. 
For the BAS-Fun Seeking subscale, demographic covariates in the first step accounted 
for 2.6% of variance in scores of behavioral approach, which was not significantly different than 
zero (F(3, 166) = 1.47, p=.22). When the ACE-Q was added in the second step, this accounted 
for an additional 0.2% of variance, but this step also fell short of significance (F?(1,165) = 0.41, 
p=.52). In the total model, none of the predictor variables significantly predicted BAS-Fun 
Seeking scores. See Appendix A for the effects of each predictor variable. 
Aim 2 
The secondary aim of this study was to examine the relationship between ACEs and 
physical health in emerging adults. For Hypothesis 3, Pearson correlations and hierarchical linear 
regression analyses were used to examine the relationships between ACE scores and lifestyle 
health behaviors such as sleep, exercise, alcohol use, substance use, e-cigarette use, disordered 
eating, and risky sexual behaviors. Regarding Hypothesis 4, the same correlation and regression 
analyses were used to assess the relationship between ACE scores and other indicators of poorer 
physical health.  
Hypothesis 3 – Correlation Analysis 
59 
 
 
 
Correlation analysis for Hypothesis 3 yielded mixed results. We observed a strong 
positive correlation between scores on the ACE-Q and self-reported sleep difficulties on the 
PSQI (r=.50, p<.01). Additionally, results indicated mild-moderate correlations between the 
ACE-Q and scores of illicit substance use (DAST-10; r=.20) and e-cigarette use (EDS; r=.22); 
both were significant at the p<.05 level. Correlations between the ACE-Q and exercise, alcohol 
use, and risky sex behaviors did not reach significance. Further details regarding these 
associations can be seen in Figure 5. 
  ACE-Q GLTEQ PSQI AUDIT-C DAST-10 EDS SRS 
ACE-Q  —                    
GLTEQ  -0.124  —                 
PSQI  0.485 *** -0.027  —              
AUDIT-C  -0.093  0.027  0.024  —           
DAST-10  0.231 ** -0.041  0.117  0.304 *** —        
EDS  0.211 ** 0.037  0.081  0.369 *** 0.517 *** —     
SRS  0.151  -0.058  0.008  0.310 *** 0.253 ** 0.405 *** —  
Table 3. Hypothesis 3 Correlation Matrix. Note: * p < .05, ** p < .01, *** p < .001 
 Hypothesis 3 – Regression Analysis 
Next, hierarchical linear regression analyses were completed to further examine the 
relationship between the ACE-Q and lifestyle behaviors. In the first step of each regression 
model, we controlled for demographic covariates (gender, race/ethnicity, and SES) to see if they 
significantly accounted for variance on scores for each measure. In the second step of each 
model, the ACE-Q was added as a predictor variable to assess for additional variance in the 
scores for each lifestyle measure, respectively. The second step was also assessed for significant 
predictor variables in the total model. The results of these hierarchical linear regression analyses 
are reported below.  
60 
 
 
 
Results for the PSQI demonstrated that demographic covariates in the first step accounted 
for 3.9% of variance in scores of sleep difficulties, but this was not statistically significant (F(3, 
135) = 1.81, p=.15). When the ACE-Q was added in the second step, this accounted for an 
additional 22.7% of variance, which was significantly different than zero (F?(1,134) = 41.48, 
p<.01). In the total model, ACE-Q scores were the only significant predictor of PSQI scores (B = 
.89, p<.01), such that higher ACE-Q scores predicted increased sleep difficulties during the past 
month. See Appendix B for the effects of each predictor variable. 
Regression analysis for the GLTEQ revealed that demographic covariates in the first step 
accounted for 8.1% of variance in scores of self-reported exercise, which was significantly 
greater than zero (F(3, 166) = 4.90, p<.01). When the ACE-Q was added in the second step, this 
was not statistically significant, and only accounted for an additional 0.3% of variance 
(F?(1,165) = 0.58, p=.45). In the total model, race/ethnicity (B = 10.25, p=.03) and SES (B = 
17.72, p=.03) were the only significant predictors of GLTEQ scores, such that being non-White 
(vs White) and receiving government financial assistance (vs no assistance) predicted lower self-
reported exercise over a seven-day period. See Appendix B for the effects of each predictor 
variable. 
For the AUDIT-C, regression analysis revealed that demographic covariates in the first 
step accounted for 16.3% of variance in scores of self-reported alcohol use, which was 
significantly greater than zero (F(3, 134) = 8.72, p<.01). When the ACE-Q was added in the 
second step, this did not account for any additional variance (F?(1,133) = 0.00, p=.95). In the 
total model, gender (B = 1.54, p<.01) and race/ethnicity (B = 0.87, p=.03) were the only 
significant predictors of alcohol use, such that being male (vs female) and White (vs non-White) 
61 
 
 
 
predicted higher self-reported alcohol use within the past year. See Appendix B for the effects of 
each predictor variable. 
Regression analysis for the DAST-10 showed that demographic covariates in the first 
step accounted for 2.3% of variance in scores of self-reported illicit substance use, which was not 
significantly greater than zero (F(3, 166) = 1.32, p=.27). When the ACE-Q was added in the 
second step, this accounted for an additional 7.1% of variance, which was statistically significant 
(F?(1,165) = 13.03, p<.01). In the total model, ACE-Q scores (B = 0.20, p<.01) and SES (B = 
0.94, p=.03) were significant predictors of DAST-10 scores, such that higher ACE-Q scores and 
not receiving government financial assistance (vs receiving assistance) predicted higher self-
reported illicit substance use within the past year. See Appendix B for the effects of each 
predictor variable. 
For the EDS-4, regression analysis showed that demographic covariates in the first step 
accounted for 2.5% of variance in scores of self-reported e-cigarette dependence, which was not 
significantly greater than zero (F(3, 166) = 1.42, p=.24). When the ACE-Q was added in the 
second step, this accounted for an additional 6.9% of variance, which was statistically significant 
(F?(1,165) = 12.56, p<.01). In the total model, ACE-Q scores (B = 0.47, p<.01) and gender (B = 
1.23, p=.03) were significant predictors of EDS-4 scores, such that higher ACE-Q scores and 
being male (vs female) predicted higher self-reported e-cigarette dependence. See Appendix B 
for the effects of each predictor variable. 
For the SRS, regression analysis showed that demographic covariates in the first step 
accounted for 1.5% of variance in scores of risky sex behaviors, which was not significantly 
greater than zero (F(3, 130) = 0.67, p=.57). When the ACE-Q was added in the second step, this 
accounted for an additional 3.8% of variance, which was statistically significant (F?(1,129) = 
62 
 
 
 
5.23, p=.02). In the total model, only ACE-Q scores were a significant predictor of SRS scores 
(B = 0.12, p=.02), such that higher ACE-Q scores predicted higher self-reported history of risky 
sex behaviors. See Appendix B for the effects of each predictor variable. 
Hypothesis 4 – Correlation Analysis 
Correlation analysis for Hypothesis 4 indicated a strong positive association between 
ACE-Q scores and self-reported somatic symptoms on the PHQ-15 (r=.50, p<.01). Correlations 
between the ACE-Q, self-rated health, and healthcare utilization over the past 12 months were 
not statistically significant. Further details regarding these associations can be seen in Figure 6. 
  ACE-Q Self-Rated Health Healthcare Utilization PHQ-15 
ACE-Q  —           
Self-Rated Health  -0.097  —        
Healthcare Utilization  0.039  -0.134  —     
PHQ-15  0.492 *** -0.146  0.138  —  
Table 4. Hypothesis 4 Correlation Matrix. Note: * p < .05, ** p < .01, *** p < .001 
  
Hypothesis 4 – Regression Analysis 
Next, hierarchical linear regression analyses were completed to examine the relationship 
between the ACE-Q and indicators of physical health, including somatic symptoms, self-rated 
health, and healthcare utilization. In the first step of each regression model, we controlled for 
demographic covariates – including gender, race/ethnicity, and SES variables – to see if they 
significantly accounted for variance on scores for each measure. In the second step of each 
model, the ACE-Q was added as a predictor variable to assess for additional variance in the 
scores for each physical health measure, respectively. The second step was also assessed for 
63 
 
 
 
significant predictor variables in the total model. The results of these hierarchical linear 
regression analyses are reported below.  
For the PHQ-15, regression analysis showed that demographic covariates in the first step 
accounted for 14.7% of variance in scores of self-reported somatic symptoms, which was 
significantly greater than zero (F(3, 136) = 7.84, p<.01). When the ACE-Q was added in the 
second step, this accounted for an additional 17.7% of variance, which was also statistically 
significant (F?(1,135) = 35.35, p<.01). In the total model, ACE-Q scores (B = 1.00, p<.01) and 
gender (B = 2.61, p<.01) were significant predictors of PHQ-15 scores, such that higher ACE-Q 
scores and being female (vs male) predicted higher self-reported somatic symptoms over the past 
month. See Appendix B for the effects of each predictor variable. 
Regression analysis of a self-rated health measure showed that demographic covariates in 
the first step accounted for 5.2% of variance in scores of subjective health, which was 
significantly greater than zero (F(3, 166) = 3.03, p=.03). When the ACE-Q was added in the 
second step, this accounted for an additional 0.7% of variance, but this step was not statistically 
significant (F?(1,165) = 1.18, p=.28). In the total model, only race/ethnicity was a significant 
predictor of self-rated health (B = 0.38, p=.04), such that being White (vs non-White) predicted 
lower subjective ratings of personal health. See Appendix B for the effects of each predictor 
variable. 
Regression analysis of the healthcare utilization questionnaire demonstrated that 
demographic covariates in the first step accounted for 7.2% of variance in 12-month healthcare 
utilization, which was significantly greater than zero (F(3, 165) = 4.28, p<.01). When the ACE-
Q was added in the second step, this did not account for any additional variance, and this step 
was not statistically significant (F?(1,164) = 0.03, p=.87). In the total model, only gender was a 
64 
 
 
 
significant predictor of healthcare utilization (B = 2.74, p<.01), such that being female (vs male) 
predicted increased healthcare utilization over the previous 12 months. See Appendix B for the 
effects of each predictor variable. 
Aim 3 
The final aim of the current study was to better understand the respective influence of 
childhood maltreatment and household dysfunction items of the ACE-Q on mental and physical 
health. In Hypothesis 5, Pearson correlations and hierarchical linear regression analyses were 
used to explore the associations between the Childhood Maltreatment subscale (ACE-Q items 1-
5), Household Dysfunction subscale (ACE-Q items 6-10), and mental health measures that were 
clinically and statistically significant in Aim 1. The same correlation and regression analyses 
were also used in Hypothesis 6 to examine the relationships between the ACE-Q subscales and 
physical health measures that were clinically and statistically significant in Aim 2.  
Hypothesis 5 – Correlation Analysis 
For the mental health measures defined in Hypothesis 5, correlation analysis revealed that 
scores on the Childhood Maltreatment (CM) and Household Dysfunction (HD) subscales were 
both significantly associated with scores on the PHQ-9 (CM: r=.35; HD: r=.40), GAD-7 (CM: 
r=.40; HD: r=.34), PCL-5 (CM: r=.44; HD: r=.41), ASRS-5 (CM: r=.32; HD: r=.38), DERS-18 
(CM: r=.38; HD: r=.37), and ESP (CM: r=.31, p<.01; HD: r=.21, p<.01). All of these 
associations were significant at the p<.01 level, and can be seen in Figure 7. 
 
 
 
65 
 
 
 
ACE-Q ACE-Q 
  ACE-Q PHQ-9 GAD-7 PCL-5 ASRS-5 DERS-18 PSS ESP 
(CM) (HD) 
ACE-Q  —                             
ACE-Q  0.871 *** —                          
(CM) 
ACE-Q  0.900 *** 0.569 *** —                       
(HD) 
PHQ-9  0.421 *** 0.346 *** 0.398 *** —                    
GAD-7  0.412 *** 0.400 *** 0.335 *** 0.724 *** —                 
PCL-5  0.479 *** 0.437 *** 0.414 *** 0.723 *** 0.712 *** —              
ASRS-5  0.397 *** 0.323 *** 0.377 *** 0.663 *** 0.652 *** 0.647 *** —           
DERS-18  0.420 *** 0.375 *** 0.370 *** 0.704 *** 0.681 *** 0.674 *** 0.657 *** —        
PSS  0.241 ** 0.226 ** 0.202 ** 0.454 *** 0.534 *** 0.506 *** 0.476 *** 0.406 *** —     
ESP  0.292 *** 0.314 *** 0.210 ** 0.502 *** 0.497 *** 0.452 *** 0.370 *** 0.458 *** 0.338 *** —  
Table 5. Hypothesis 5 Correlation Matrix. Note: * p < .05, ** p < .01, *** p < .001 
 
Hypothesis 5 – Regression Analysis 
Next, hierarchical linear regression analyses were used to further examine these 
relationships. In the first step of each regression model, we controlled for demographic 
covariates (gender, race/ethnicity, and SES variables) to see if they significantly accounted for 
variance on scores for each mental health measure. In the second step of each model, scores on 
the Childhood Maltreatment and Household Dysfunction subscales were added to evaluate 
whether both subscales were significant predictors of each mental health measure, or if the 
significant relationship between the ACE-Q and each mental health measure was more strongly 
influenced by one subscale. For this analysis, results concerning demographic covariates were 
not reported, as these were previously discussed in the results for Aims 1 and 2. Further, our 
objective was to explore the influence of each ACE-Q subscale, and reporting on demographic 
covariates is peripheral to this aim.   
Regression analyses confirmed that step 2 (CM and HD subscales) significantly 
accounted for a proportion of variance in scores on the PHQ-9, GAD-7, PCL-5, ASRS-5, DERS-
18, and ESP, as demonstrated in Aim 1. In the overall model, both ACE-Q subscales 
66 
 
 
 
significantly predicted scores on the PHQ-9 (CM: ?=0.18, p=.04; HD: ?=0.28, p<.01), PCL-5 
(CM: ?=0.29, p<.01; HD: ?=0.27, p<.01) and DERS-18 (CM: ?=0.26, p<.01; HD: ?=0.20, 
p=.02); only the Childhood Maltreatment subscale significantly predicted scores on the GAD-7 
(CM: ?=0.31, p<.01; HD: ?=0.15, p=.07) and ESP (CM: ?=0.23, p<.01; HD: ?=0.08, p=.39); 
only the Household Dysfunction subscale significantly predicted scores on the ASRS-5 (CM: 
?=0.17, p=.05; HD: ?=0.27, p<.01). See Appendix C for the effects of each predictor variable.  
Hypothesis 6 – Correlation Analysis 
For the physical health measures defined in Hypothesis 6, correlation analysis revealed 
that scores on the Childhood Maltreatment (CM) and Household Dysfunction (HD) subscales 
were both significantly associated with scores on the PSQI (CM: r=.51, p<.01; HD: r=.36, 
p<.01), and PHQ-15 (CM: r=.46, p<.01; HD: r=.41, p<.01); only the Household Dysfunction 
subscale was significantly associated with scores on the DAST-10 (CM: r=.13, p=.19; HD: 
r=.23, p<.01) and EDS (CM: r=.14, p=.06; HD: r=.23, p<.01). These associations can be seen in 
Figure 8. 
  ACE-Q ACE-Q (CM) ACE-Q (HD) PSQI DAST10 EDS PHQ-15 
ACE-Q  —                    
ACE-Q (CM)  0.871 *** —                 
ACE-Q (HD)  0.900 *** 0.569 *** —              
PSQI  0.485 *** 0.509 *** 0.361 *** —           
DAST10  0.231 ** 0.131  0.270 *** 0.117  —        
EDS  0.211 ** 0.139  0.230 ** 0.081  0.517 *** —     
PHQ-15  0.492 *** 0.459 *** 0.407 *** 0.646 *** 0.330 *** 0.168 * —  
Table 6. Hypothesis 6 Correlation Matrix. Note: * p < .05, ** p < .01, *** p < .001 
 Hypothesis 6 – Regression Analysis 
67 
 
 
 
Next, hierarchical linear regression analyses were used to further examine these 
relationships. In the first step of each regression model, we controlled for demographic 
covariates – including gender, race/ethnicity, and SES variables – to see if they significantly 
accounted for variance on scores for each physical health measure. In the second step of each 
model, scores on the Childhood Maltreatment and Household Dysfunction subscales were added 
to examine whether both subscales, or only one subscale, significantly predicted scores on 
physical health measures. Once again, results concerning demographic covariates were not 
reported. 
Regression analyses confirmed that step 2 (CM and HD subscales) significantly 
accounted for a proportion of variance in scores on the PSQI, DAST-10, EDS, and PHQ-15, as 
demonstrated in Aim 2. In the overall model, both ACE-Q subscales significantly predicted 
scores on the PHQ-15 (CM: ?=0.31, p<.01; HD: ?=0.20, p=.03); only the Childhood 
Maltreatment subscale significantly predicted scores on the PSQI (CM: ?=0.45, p<.01; HD: 
?=0.12, p=.21); only the Household Dysfunction subscale significantly predicted scores on the 
DAST-10 (CM: ?=0.04, p=.65; HD: ?=0.35, p<.01) and EDS (CM: ?=0.03, p=.72; HD: ?=0.27, 
p<.01). See Appendix C for the effects of each predictor variable.
68 
 
Discussion 
Adverse childhood experiences (ACEs) are an important public health concern, and the 
past two decades of research have continued to highlight the need for early prevention and 
intervention strategies that address the heightened chronic disease risks associated with ACEs in 
adulthood. Presently, there is a need for further research that utilizes an etiological framework to 
help identify the mechanisms responsible for poorer health outcomes. Accordingly, the current 
study sought to better understand physical and mental health characteristics associated with 
ACEs in the emerging adult population – by targeting this age group, results may help elucidate 
clinically significant health issues within this stratum of the population. Previous research has 
largely focused on the aggregated adult population and the pediatric population, with 
comparatively much less being known about the health issues in emerging adults with a history 
of ACEs. 
Mental Health Factors  
Our results provide important information about the mental health of emerging adults 
with a history of ACEs, and we believe it is essential to compare these results with previous 
research on pediatric, emerging adult, and aggregated adult samples. The pediatric literature has 
most consistently demonstrated a relationship between ACEs and depression, anxiety, and 
ADHD (Horn et al., 2019; Bright et al., 2016). In addition to depression, anxiety, and ADHD, the 
aggregated adult literature has also demonstrated associations between ACEs and PTSD 
(Kalmakis & Chandler, 2015; Felitti et al., 1998). Recent findings within a young adult sample 
suggest a higher prevalence of worsening stress and mental health over the course of the 
semester for college students with high (versus low) ACE scores, though these results were only 
limited to measurement with the PHQ-9 (Karatekin, 2018).  
 
 
 
In comparison, results within our emerging adult sample demonstrated associations 
between ACEs and depression, anxiety, ADHD, PTSD, disordered eating, emotion 
dysregulation, and perceived stress. Thus, our findings suggest a wider array of mental health 
concerns than what has been previously reported. This is likely due, in part, to our decision to 
analyze our mental health measures continuously rather than using a suggested clinical cutoff 
score to categorize participants as screening “positive” or “negative” for each respective mental 
health concern. That said, our results suggest numerous mental health concerns for emerging 
adults with multiple ACEs, even if those concerns are screened as subclinical. Taken together, 
we contend that this helps to identify aspects of pathology and potential mechanisms related to 
increased mental health symptomology.  
Demographic Differences in Mental Health Factors 
Interestingly, our analysis of demographic covariates revealed important information 
about potential cultural differences in the presentation of mental health symptoms among 
emerging adults. For instance, when looking at race/ethnicity, being White predicted higher self-
reported symptoms of depression, anxiety, ADHD, and perceived stress. Additionally, 
membership of a non-White racial/ethnic group predicted higher resilience scores. This trend 
could indicate that membership of a minority racial/ethnic group is related to greater resilience, a 
protective factor against psychopathology and mental distress. However, this may also be 
partially explained by well-documented mental health stigma among certain minority 
racial/ethnic minority groups (Miranda et al., 2015), potentially contributing to the under-
reporting of mental health symptoms. Of note, demographic covariates such as gender and 
socioeconomic status did not appear to significantly predict psychopathology or transdiagnostic 
mental health factors. 
70 
 
 
 
Physical Health Factors 
Concerning physical health factors, it is also important to distinguish the differences in 
ACE-related symptoms among pediatric, emerging adult, and aggregated adult populations. 
Research with pediatric samples has identified various health complaints associated with ACEs, 
including asthma, infection, somatic complaints, and sleep disruption (Oh et al., 2018). 
Additionally, others have documented an increased prevalence of weight gain/loss, enuresis, 
constipation, and poorer diabetes control (Purewal et al., 2016). Evidence for physical health 
symptoms is much more robust among aggregated adult samples, suggesting that ACEs are 
associated with cardiovascular issues such as ischemic heart disease, coronary artery disease, 
diabetes, myocardial infarction, and stroke (Campbell et al., 2017), in addition to chronic lung 
disease, cancer, and liver disease (Felitti et al., 1998). Concerning younger adult samples, recent 
findings have identified higher rates of chronic disease onset for individuals between ages 18 and 
34 (Sonu et al., 2019). Within our emerging adult sample, our findings demonstrated that ACEs 
had a strong positive relationship with somatic complaints, although many of these other 
physical health concerns were not explored.  
To our knowledge, these results provide new evidence that ACEs are associated with 
somatic complaints in a healthy emerging adult population. Although somatic complaints were 
previously found in a systematic review of pediatric studies (Oh et al., 2018), we contend that 
these symptoms are important to note in an emerging adult sample, since they have also not yet 
been identified in ACEs studies with aggregated adult samples. Further, there are key differences 
in the types of somatic complaints made by pediatric participants and emerging adults. For 
instance, research with children and early adolescents are more likely to focus on stomachaches, 
vomiting, headaches, dizziness, and nightmares, and are largely dependent on parental report 
71 
 
 
 
(Flaherty et al., 2009). Based on our results, the somatic symptoms most frequently endorsed 
included racing heart/pounding chest, dizziness, trouble sleeping, low energy/fatigue, headaches, 
back pain, and shortness of breath. Importantly, these somatic symptoms are also associated with 
psychological distress (i.e., symptoms of depression, anxiety, panic, PTSD), suggesting an area 
in which physical complaints could benefit from mental health interventions. Future research 
could further illuminate differences in somatic symptoms between pediatric, emerging adult, and 
adult populations.  
Lifestyle Health Behaviors 
Within our emerging adult sample, we identified a strong association between ACEs and 
sleep difficulties, and more modest associations between ACEs and measures of illicit substance 
use and e-cigarette dependence. Our results surrounding ACEs and health behaviors in emerging 
adults build on a very limitted literature, and draw several distinctions from previous research 
findings. In the past, health behaviors like cigarette smoking, over-eating, and alcohol/drug use 
have been theorized as mechanisms by which childhood adversity contributes to downstream 
chronic health conditions. From this perspective, individuals with more childhood adversity 
disproportionately engage in unhealthy coping behaviors to compensate for chronic stress 
(Felitti, 2009). Of note, these lifestyle health behaviors have been mostly studied with older 
adults, and we continue to highlight the need for disaggregated adult samples that are stratified 
by narrower age-groups to better understand the disproportionate development of chronic disease 
in later adulthood.  
Interestingly, we did not observe significant relationships between ACEs and indicators 
of poorer physical health highlighted elsewhere in the adult literature, such as cigarette smoking, 
low physical activity, heavy alcohol use, risky sex behaviors, and lower self-rated health. In 
72 
 
 
 
contrast, our findings suggest that sleep difficulties and somatic complaints may play a larger 
role than previously understood, and particularly for emerging adults. Concerning this trend, we 
suggest that emerging adulthood may act as a transitional period, and that engagement in 
unhealthy lifestyle behaviors (e.g., over-eating, low physical activity, cigarette smoking, heavy 
alcohol use, risky sex behaviors) may become more prominent over time, if unaddressed. 
Additionally, evidence suggests that unhealthy coping behaviors like sexual risk-taking, heavy 
alcohol use, and illicit substance use are more normative among college students than the general 
adult population (Krieger et al., 2018), which could mask the significance of problematic 
engagement in these behaviors for emerging adults with a higher ACE history.  
Compared to the results of other emerging adult samples, our results also demonstrate  
differences concerning ACEs and lifestyle behaviors. In a recent study on college students 
(Windle et al., 2018), the authors found that ACE scores predicted higher cigarette use, alcohol 
use, marijuana use, and BMI, and lower self-reported hours of sleep and fruit and vegetable 
consumption. In contrast, we did not observe a significant relationship between ACEs and 
cigarette use or alcohol use; though, this may be due to survey differences and our smaller 
sample size. Additionally, we did not assess for BMI and fruit/vegetable consumption, which 
limits the ability to compare these findings. While the two studies were designed similarly (e.g., 
online survey method among college students), key methodological differences could have also 
contributed to this difference in findings. For instance, the questionnaires used in the current 
study were much more robust, and data was collected using common measures of lifestyle health 
behaviors (e.g., PSQI, AUDIT-C, DAST-10) rather than single-item questionnaires.  
73 
 
 
 
Key Differences by ACE-Q Subscale 
The current study also provides further insight into the respective influences of the ACE-
Q subscales (childhood maltreatment and household dysfunction) on measures of mental and 
physical health; this approach was adapted from a recent study of mental health issues based on 
latent ACE classes (child maltreatment, household dysfunction, community violence, low 
adversity) by Lee et al. (2020). When comparing scores on the child maltreatment and household 
dysfunction subscales, the authors reported that only the child maltreatment subscale was 
associated with a greater odds ratio of depression, anxiety, and PTSD. In comparison, the results 
of the present study showed that both subscales significantly predicted scores of depression, 
PTSD, and emotion dysregulation; however, only the childhood maltreatment subscale predicted 
scores of anxiety and disordered eating, and only the household dysfunction subscale predicted 
scores of ADHD. This finding could indicate etiological differences between specific mental 
health issues in the ACEs population, but more research is needed.  
The aforementioned authors (Lee et al., 2020) did not assess the relationships between 
ACE-Q subscales and measures of physical health in their sample of young adults – therefore, 
this became the focus of our third study aim. In the present study, both the childhood 
maltreatment and household dysfunction subscales predicted self-reported somatic symptoms. 
Only the childhood maltreatment subscale predicted sleep difficulties, and this relationship 
seemed to be particularly strong. On measures of illicit substance use and e-cigarette 
dependence, analysis of the ACE-Q subscales revealed that household dysfunction was more 
predictive of substance use and dependence. This finding provides further explanation about 
substance use risk in the ACEs population, and suggests that environmental factors (e.g., 
household substance use) could pose a greater risk for subsequent substance use compared to 
74 
 
 
 
childhood abuse and neglect. Taken together, analysis of these subscales suggests that 
participants with a history of childhood maltreatment may demonstrate different physical health 
concerns related to trauma and chronic stress (e.g., sleep difficulties) compared to participants 
with a history of household dysfunction (e.g., e-cigarette and illicit substance use). However, 
further research is needed to better understand these relationships. 
Conclusion 
Altogether, our results suggest that even among healthy emerging adults, ACEs are 
associated with both mental and physical health difficulties. These findings provide important 
information about the pathway from early adversity to chronic disease onset. Previously, the 
ACEs literature thoroughly addressed the increased risk of chronic diseases in older adulthood 
for individuals with a history of higher childhood adversity; in particular, those with ? 4 ACEs 
have been consistently found to have an increased risk of chronic conditions like depression, 
cardiovascular disease, and cancer, among others (Felitti et al., 1998). Of note, our results 
support recent findings that these same chronic health difficulties are disproportionately present 
among younger adults (18-34 years) with a higher ACEs history (Sonu et al., 2019), and not just 
among individuals in later adulthood. That is, our findings identify numerous mental and 
physical chronic health concerns associated with ACEs even among healthy emerging adults (18-
25 years old). To our knowledge, this is the most comprehensive study to concurrently assess for 
both mental and physical health concerns related to ACEs in the emerging adult population.  
After consideration of the current ACEs literature, we believe that the most important 
contribution of the present study is the growing understanding that ACEs are associated with 
sleep difficulties and somatic complaints in emerging adults. These findings are particularly 
relevant for healthcare providers that oversee the care of this age group (e.g., primary care and 
75 
 
 
 
mental health clinicians), since sleep and somatic symptoms are likely to be overlooked in 
healthy emerging adults. Adverse childhood experiences have already been identified as a 
widespread public health concern, contributing to increased risk for chronic physical and mental 
health concerns throughout the lifespan. From a population health perspective, there are vast 
benefits to identifying a patient population with a higher prevalence of these symptoms, such as 
the cost-effectiveness of early treatments that could potentially reduce chronic disease onset and 
healthcare utilization in later adulthood. Integrated primary care clinics could be an ideal setting 
to address sleep difficulties and pain symptoms in emerging adults, as behavioral health 
interventions are already being implemented to target these issues. Through an increased 
understanding of the health risks related to childhood adversity, providers can more effectively 
screen for potential health concerns and implement prevention and early intervention strategies.  
Future directions within this line of research could expand on our etiological focus, and 
we support further attempts to connect ACE-related health risks with specific biological 
mechanisms, physical symptoms, and lifestyle health behaviors. Felitti (2009) famously 
distinguished that the two most likely pathways from ACEs to chronic illness are compensatory 
coping through poor health behaviors and inflammatory biological responses due to chronic 
stress. Our results suggest both of these to be accurate, with e-cigarette and illicit substance use 
pertaining to the former, and somatic complaints relating to the latter. Further, we believe that 
sleep difficulties could be more accurately understood as a health issue related to psychological 
distress due to the common presentation of sleep difficulties as a PTSD symptom. Regarding 
sleep difficulties, researchers may want to further examine the relationship between ACEs and 
individual PSQI components (e.g., subjective sleep quality, latency, duration, efficiency, 
disturbances, sleep medication use, daytime dysfunction), as well as other sleep-related factors 
76 
 
 
 
(e.g., sleep hygiene, unhelpful beliefs about sleep). Given the strong associations between ACEs 
and sleep and pain symptomology, future research may want to evaluate these health concerns 
from a chronic stress perspective.  
Limitations 
Of note, the present study was met with several limitations. We encountered substantial 
logistical obstacles due to restrictions related to the COVID-19 pandemic, which prevented us 
from collecting in-person laboratory participant data to explore our original aim – the 
relationship between ACEs and heart rate variability (HRV). Due to autonomic dysregulation 
caused by chronic/toxic stress, we identified HRV as a possible correlate of childhood adversity, 
and sought to understand HRV dysregulation as a potential mechanism of downstream chronic 
disease onset. In the absence of laboratory data, our study aims were adjusted accordingly, and 
we directed our attention to other physical and mental health factors assessed via online survey 
that could influence chronic disease risk in later adulthood. Although unintended, this deviation 
allowed us to look deeper at the relationships between ACEs and other physical and mental 
health factors. In addition to HRV monitoring, we were also unable to measure BMI and blood 
pressure, and recommend that these measurements be collected in future research on ACEs and 
physical health in emerging adults. 
While our results concerning various physical and mental health concerns align with the 
existing ACEs literature, we also recognize the potential influence of factors related to the 
COVID-19 pandemic. For instance, the COVID-19 pandemic has been widely recognized as a 
substantial stressor, with the potential for increases in psychological distress (e.g., depression, 
anxiety, perceived stress) and maladaptive coping (e.g., increased sleep difficulties and substance 
misuse; decreased physical activity). Accordingly, due to the limitations of self-report survey 
77 
 
 
 
data, participant responses may have been unduly impacted by current events. To address this, 
future analyses with this dataset could include a COVID-19 moderator variable to explore 
differences in participant responses from before and after the start of the pandemic. However, 
due to the high effect sizes of many of our statistically significant findings, we contend that our 
conclusions remain valid.  
We also note some of the limitations of using certain survey measures and screening tools 
in an online questionnaire format. For instance, we were surprised that our results did not 
endorse gender as a significant predictor of PTSD symptoms within our sample, as this 
relationship has been widely acknowledged in the literature (Tolin & Foa, 2006). Since the PCL-
5 was completed as part of an online questionnaire, we were unable to provide participants with 
detailed instructions, though they were provided with a prompt to answer questions with regard 
to a specific stressor. This degree of ambiguity may have contributed to responses that more 
generally referenced a significant stressor rather than a previous DSM-5 traumatic event. 
Additionally, due to the layout of the Sexual Risk Survey within our online Qualtrics survey 
(e.g., use of string, rather than numeric, variables), this seemingly contributed to a higher rate of 
invalid responses. Thus, we interpret our results with caution, such that other methods may have 
identified an association between ACEs and risky sex behaviors in emerging adults.  
Finally, some of our results may have been limited by our modest sample size (n=236). 
Due to the low prevalence of self-reported daily tobacco smoking in our sample, we did not have 
sufficient power to assess for associations between ACEs and cigarette smoking. However, a 
recent study of ACEs and smoking behaviors reported increasing rates of cigarette smoking 
among youths with higher ACE scores, despite decreasing rates of cigarette smoking in 
participants with no ACEs (Parks et al., 2018). Therefore, in addition to risky sex behaviors, we 
78 
 
 
 
contend that this lifestyle health behavior should be further explored in the emerging adult ACEs 
population.
79 
 
References 
Backhaus, J., Junghanns, K., Broocks, A., Riemann, D., & Hohagen, F. (2002). Test–retest 
reliability and validity of the pittsburgh sleep quality index in primary insomnia. Journal of 
Psychosomatic Research, 53(3), 737-740. 
Beauchaine, T. P., Crowell, S. E., Garland, E. L., Bell, S., Atchley, R., & Froeliger, B. (2018). 
Emotion dysregulation in addiction Oxford University Press. 
Beauchaine, T. P., & Thayer, J. F. (2015). Heart rate variability as a transdiagnostic biomarker of 
psychopathology. International Journal of Psychophysiology, 98(2), 338-350. 
Benjamin, E. J., Muntner, P., Alonso, A., Bittencourt, M. S., Callaway, C. W., Carson, A. P., . . . 
On behalf of the American Heart Association Council on Epidemiology and Prevention 
Statistics Committee and Stroke Statistics Subcommittee. (2019). Heart disease and stroke 
Statistics—2019 update: A report from the american heart association. Circulation, 139(10), 
e66. 
Bethell, C. D., Carle, A., Hudziak, J., Gombojav, N., Powers, K., Wade, R., & Braveman, P. 
(2017). Methods to assess adverse childhood experiences of children and families: Toward 
approaches to promote child well-being in policy and practice. Academic Pediatrics, 17(7), 
S5-S69. 
Blevins, C. A., Weathers, F. W., Davis, M. T., Witte, T. K., & Domino, J. L. (2015). The 
posttraumatic stress disorder checklist for DSM?5 (PCL?5): Development and initial 
psychometric evaluation. Journal of Traumatic Stress, 28(6), 489-498. 
 
 
 
Brenner, S. L., Beauchaine, T. P., & Sylvers, P. D. (2005). A comparison of psychophysiological 
and self?report measures of BAS and BIS activation. Psychophysiology, 42(1), 108-115. 
Bright, M. A., Knapp, C., Hinojosa, M. S., Alford, S., & Bonner, B. (2016). The comorbidity of 
physical, mental, and developmental conditions associated with childhood adversity: a 
population based study. Maternal and child health journal, 20(4), 843-853. 
Brown, D. W., Anda, R. F., Tiemeier, H., Felitti, V., Edwards, V. J., Croft, J. B., & Giles, W. H. 
(2009). Adverse childhood experiences and the risk of premature mortality. American 
Journal of Preventive Medicine, 37(5), 389-396. 
Brugnera, A., Zarbo, C., Tarvainen, M. P., Marchettini, P., Adorni, R., & Compare, A. (2018). 
Heart rate variability during acute psychosocial stress: A randomized cross-over trial of 
verbal and non-verbal laboratory stressors. International Journal of Psychophysiology, 127, 
17-25. 
Bunea, I. M., Szentágotai-T?tar, A., & Miu, A. C. (2017). Early-life adversity and cortisol 
response to social stress: A meta-analysis. Translational Psychiatry, 7(12), 1274-8. 
Bush, K., Kivlahan, D. R., McDonell, M. B., Fihn, S. D., & Bradley, K. A. (1998). The AUDIT 
alcohol consumption questions (AUDIT-C): An effective brief screening test for problem 
drinking. Archives of Internal Medicine, 158(16), 1789-1795. 
Buysse, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The 
Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. 
Psychiatry Research, 28(2), 193-213. 
81 
 
 
 
Campbell, C. E., & Maisto, S. A. (2018). Validity of the AUDIT-C screen for at-risk drinking 
among students utilizing university primary care. Journal of American College Health : J of 
ACH, 66(8), 774-782. 
Campbell, J., Walker, R., & Egede, L. (2016). Associations between adverse childhood 
experiences, high-risk behaviors, and morbidity in adulthood. American Journal of 
Preventive Medicine, 50(3), 344-352. 
Campbell-Sills, L., Liverant, G. I., & Brown, T. A. (2004). Psychometric evaluation of the 
behavioral inhibition/behavioral activation scales in a large sample of outpatients with 
anxiety and mood disorders. Psychological Assessment, 16(3), 244-254. 
Carey, K. B., Carey, M. P., & Chandra, P. S. (2003). Psychometric evaluation of the alcohol use 
disorders identification test and short drug abuse screening test with psychiatric patients in 
india. The Journal of Clinical Psychiatry, 64(7), 767-774. 
Carpenter, J. S., & Andrykowski, M. A. (1998). Psychometric evaluation of the pittsburgh sleep 
quality index. Journal of Psychosomatic Research, 45(1), 5-13. 
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective 
responses to impending reward and punishment: The BIS/BAS scales. Journal of 
Personality and Social Psychology, 67(2), 319-333. 
Chaddha, A., Robinson, E. A., Kline-Rogers, E., Alexandris-Souphis, T., & Rubenfire, M. 
(2016). Mental health and cardiovascular disease. American Journal of Medicine, The, 
129(11), 1145-1148. 
82 
 
 
 
Cocco, K. M., & Carey, K. B. (1998). Psychometric properties of the drug abuse screening test in 
psychiatric outpatients. Psychological Assessment, 10(4), 408-414. 
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. 
Journal of Health and Social Behavior, 24(4), 385-396. 
Cotton, M., Ball, C., & Robinson, P. (2003). Four simple questions can help screen for eating 
disorders. Journal of General Internal Medicine, 18(1), 53-56. 
Dedovic, K., Renwick, R., Mahani, N. K., Engert, V., Lupien, S. J., & Pruessner, J. C. (2005). 
The montreal imaging stress task: Using functional imaging to investigate the effects of 
perceiving and processing psychosocial stress in the human brain. Journal of Psychiatry & 
Neuroscience : JPN, 30(5), 319-325.  
Demartini, K. S., & Carey, K. B. (2012). Optimizing the use of the AUDIT for alcohol screening 
in college students. Psychological Assessment, 24(4), 954-963. 
Dong, M., 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. 
Dong, M., Giles, W. H., Felitti, V. J., Dube, S. R., Williams, J. E., Chapman, D. P., & Anda, R. 
F. (2004). Insights into causal pathways for ischemic heart disease: Adverse childhood 
experiences study. Circulation, 110(13), 1761-1766. 
83 
 
 
 
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. 
Ellis, R. J., Sollers Iii, J. J., Edelstein, E. A., & Thayer, J. F. (2008). Data transforms for spectral 
analyses of heart rate variability. Biomedical Sciences Instrumentation, 44, 392-397.  
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. 
Flaherty, E. G., MD, Thompson, R., PhD, Litrownik, A. J., PhD, Zolotor, Adam J., MD, MPH, 
Dubowitz, Howard, MD, MS, Runyan, Desmond K., MD, DrPH, . . . Everson, M. D., PhD. 
(2009). Adverse childhood exposures and reported Child Health at age 12. Academic 
Pediatrics, 9(3), 150-156. 
Franke, H. A. (2014). Toxic stress: Effects, prevention and treatment. Children (Basel, 
Switzerland), 1(3), 390-402. 
Garner, A. S., Shonkoff, J. P., Committee on Early Childhood, Adoption,and Dependent Care, 
Committee on Psychosocial Aspects of Child and Family Health, Section on Developmental 
and Behavioral Pediatrics, & Committee on Psychosocial Aspects of Child and Family 
Health, Committee on Early Childhood, Adoption, and Dependent Care,and Section on 
Developmental and Behavioral Pediatrics. (2012). Early childhood adversity, toxic stress, 
and the role of the pediatrician: Translating developmental science into lifelong health. 
Pediatrics, 129(1), e224-e231. 
84 
 
 
 
Gilbert, L. K., Breiding, M. J., Merrick, M. T., Thompson, W. W., Ford, D. C., Dhingra, S. S., & 
Parks, S. E. (2015). Childhood adversity and adult chronic disease: An update from ten 
states and the district of columbia, 2010. American Journal of Preventive Medicine, 48(3), 
345.  
Godin, G. (1969). The godin-shephard leisure-time physical activity questionnaire. The Health & 
Fitness Journal of Canada, 4(1). 
Gratz, K. L., & Roemer, L. (2004). Multidimensional assessment of emotion regulation and 
dysregulation: Development, factor structure, and initial validation of the difficulties in 
emotion regulation scale. Journal of Psychopathology and Behavioral Assessment, 26(1), 
41-54. 
Herman, J. P., McKlveen, J. M., Ghosal, S., Kopp, B., Wulsin, A., Makinson, R., . . . Myers, B. 
(2016). Regulation of the hypothalamic-pituitary-adrenocortical stress response. 
Comprehensive Physiology, 6(2), 603-621. 
Horn, S. R., Leve, L. D., Levitt, P., & Fisher, P. A. (2019). Childhood adversity, mental health, 
and oxidative stress: A pilot study. PloS One, 14(4), e0215085. 
Hughes, K., Bellis, M. A., Hardcastle, K. A., Sethi, D., Butchart, A., Mikton, C., . . . Dunne, M. 
P. (2017). The effect of multiple adverse childhood experiences on health: A systematic 
review and meta-analysis. The Lancet Public Health, 2(8), e366. 
85 
 
 
 
Iffland, B., Wiggert, N., Neuner, F., & Blechert, J. (2018). Neutral is negative enough—Peer 
victimization influences cardiac, facial-muscular and experiential reactions to both negative 
evaluative and neutral social stimuli. Biological Psychology, 139, 152-162. 
Jin, M. J., Kim, J. S., Kim, S., Hyun, M. H., & Lee, S. (2018). An integrated model of emotional 
problems, beta power of electroencephalography, and low frequency of heart rate variability 
after childhood trauma in a non-clinical sample: A path analysis study. Frontiers in 
Psychiatry, 8. 
Juster, R., McEwen, B. S., & Lupien, S. J. (2010). Allostatic load biomarkers of chronic stress 
and impact on health and cognition. Neuroscience and Biobehavioral Reviews, 35(1), 2-16. 
Kagan, S., & Melrose, C. (2003). The SCOFF questionnaire was less sensitive but more specific 
than the ESP for detecting eating disorders. Evidence Based Nursing, 6(4), 118. 
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. 
Kalmakis, K. A., Meyer, J. S., Chiodo, L., & Leung, K. (2015). Adverse childhood experiences 
and chronic hypothalamic-pituitary-adrenal activity. Stress (Amsterdam, Netherlands), 
18(4), 446-450. 
Karatekin, C. (2018). Adverse childhood experiences (ACEs), stress and mental health in college 
students. Stress and Health, 34(1), 36-45. 
86 
 
 
 
Krieger, H., Young, C. M., Anthenien, A. M., & Neighbors, C. (2018). The epidemiology of 
binge drinking among college-age individuals in the united states. Alcohol Research, 39(1), 
23-30. 
Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2002). The PHQ-15: Validity of a new 
measure for evaluating the severity of somatic symptoms. Psychosomatic Medicine, 64(2), 
258-266. 
Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9: Validity of a brief 
depression severity measure. Journal of General Internal Medicine, 16(9), 606-613. 
Lee, H., Kim, Y., & Terry, J. (2020). Adverse childhood experiences (ACEs) on mental 
disorders in young adulthood: Latent classes and community violence exposure. Preventive 
Medicine, 134, 106039. 
Lehrer, P., & Eddie, D. (2013). Dynamic processes in regulation and some implications for 
biofeedback and biobehavioral interventions. Applied Psychophysiology and Biofeedback, 
38(2), 143-155. 
Lovallo, W. R. (2013). Early life adversity reduces stress reactivity and enhances impulsive 
behavior: Implications for health behaviors. International Journal of Psychophysiology, 
90(1), 8-16. 
Lovallo, W. R., Farag, N. H., Sorocco, K. H., Cohoon, A. J., & Vincent, A. S. (2012). Lifetime 
adversity leads to blunted stress axis reactivity: Studies from the oklahoma family health 
patterns project. Biological Psychiatry, 71(4), 344-349. 
87 
 
 
 
McCabe, S. E., Boyd, C. J., Cranford, J. A., Morales, M., & Slayden, J. (2006). A modified 
version of the drug abuse screening test among undergraduate students. Journal of 
Substance Abuse Treatment, 31(3), 297-303. 
McCrory, C., Dooley, C., Layte, R., & Kenny, R. A. (2015). The lasting legacy of childhood 
adversity for disease risk in later life. Health Psychology : Official Journal of the Division of 
Health Psychology, American Psychological Association, 34(7), 687-696. 
McEwen, B. S. (1998). Protective and damaging effects of stress mediators. The New England 
Journal of Medicine, 338(3), 171-179. 
McEwen, B. S., & Stellar, E. (1993). Stress and the individual: Mechanisms leading to disease. 
Archives of Internal Medicine, 153(18), 2093-2101. 
Meneses-Gaya, C. d., Zuardi, A. W., Loureiro, S. R., & Crippa, J. A. S. (2009). Alcohol use 
disorders identification test (AUDIT): An updated systematic review of psychometric 
properties. Psychology & Neuroscience, 2(1), 83-97. 
Miranda, R., Soffer, A., Polanco-Roman, L., Wheeler, A., & Moore, A. (2015). Mental health 
treatment barriers among Racial/Ethnic minority versus white young adults 6 months after 
intake at a college counseling center. Journal of American College Health, 63(5), 291-298. 
Mirzaei, M., Ahmadi, K., Saadat, S., & Ramezani, M. A. (2016). Instruments of high risk sexual 
behavior assessment: A systematic review. Materia Socio-Medica, 28(1), 46-50. 
88 
 
 
 
Morean, M. E., Krishnan-Sarin, S., Sussman, S., Foulds, J., Fishbein, H., Grana, R., & O’Malley, 
S. S. (2018). Psychometric evaluation of the E-cigarette dependence scale. Nicotine & 
Tobacco Research, 21(11), 1556-1564. 
Motl, R. W., Sandroff, B. M., Wingo, B. C., McCroskey, J., Pilutti, L. A., Cutter, G. R., . . . 
McAuley, E. (2018). Phase-III, randomized controlled trial of the behavioral intervention 
for increasing physical activity in multiple sclerosis: Project BIPAMS. Contemporary 
Clinical Trials, 71, 154-161. 
Oh, D. L., Jerman, P., Silvério Marques, S., Koita, K., Purewal Boparai, S. K., Burke Harris, N., 
& Bucci, M. (2018). Systematic review of pediatric health outcomes associated with 
childhood adversity. BMC Pediatrics, 18(1), 83. 
Parks, M. J., Davis, L., Kingsbury, J. H., & Shlafer, R. J. (2020). Adverse childhood experiences 
and youth cigarette use in 2013 and 2016: emerging disparities in the context of declining 
smoking rates. Nicotine & Tobacco Research, 22(1), 124-129. 
Pomerleau, C. S., Carton, S. M., Lutzke, M. L., Flessland, K. A., & Pomerleau, O. F. (1994). 
Reliability of the fagerstrom tolerance questionnaire and the fagerstrom test for nicotine 
dependence. 
Pomerleau, O. F., Fertig, J. B., & Shanahan, S. O. (1983). Nicotine dependence in cigarette 
smoking: An empirically-based, multivariate model. Pharmacology, Biochemistry and 
Behavior, 19(2), 291-299. 
89 
 
 
 
Porges, S. W. (2001). The polyvagal theory: Phylogenetic substrates of a social nervous system. 
International Journal of Psychophysiology, 42(2), 123-146. 
Pulopulos, M. M., Vanderhasselt, M., & De Raedt, R. (2018). Association between changes in 
heart rate variability during the anticipation of a stressful situation and the stress-induced 
cortisol response. Psychoneuroendocrinology, 94, 63-71. 
Purewal, S. K., Au, V., Koita, K., Oh, D. L., Briner, S., Burke Harris, N., & Bucci, M. (2018). 
Ameliorating the biological impacts of childhood adversity: A review of intervention 
programs. Child Abuse & Neglect, 81, 82-105. 
Purewal, S. K., Bucci, M., Gutiérrez Wang, L., Koita, K., Silvério Marques, S., Oh, D., & Burke 
Harris, N. (2016). Screening for adverse childhood experiences (ACEs) in an integrated 
pediatric care model. Zero to three, 37(1), 10-17. 
Rajendra Acharya, U., Paul Joseph, K., Kannathal, N., Lim, C. M., & Suri, J. S. (2006). Heart 
rate variability: A review. Medical & Biological Engineering & Computing, 44(12), 1031-
1051. 
Rigterink, T., Katz, L. F., & Hessler, D. M. (2010). Domestic violence and longitudinal 
associations with children's physiological regulation abilities. Journal of Interpersonal 
Violence, 25(9), 1669-1683. 
Shaffer, F., & Ginsberg, J. P. (2017). An overview of heart rate variability metrics and norms. 
Frontiers in Public Health, 5, 258. 
90 
 
 
 
Shonkoff, J. P., Garner, A. S., Committee on Early Childhood, Adoption,and Dependent Care, 
Committee on Psychosocial Aspects of Child and Family Health, Section on Developmental 
and Behavioral Pediatrics, & THE COMMITTEE ON PSYCHOSOCIAL ASPECTS OF 
CHILD AND FAMILY HEALTH, COMMITTEE ON EARLY CHILDHOOD, 
ADOPTION, AND DEPENDENT CARE,AND SECTION ON DEVELOPMENTAL AND 
BEHAVIORAL PEDIATRICS. (2012). The lifelong effects of early childhood adversity 
and toxic stress. Pediatrics, 129(1), e246.  
Skinner, H. A. (1982). The drug abuse screening test. Addictive Behaviors, 7(4), 363-371. 
Smith, B. W., Dalen, J., Wiggins, K., Tooley, E., Christopher, P., & Bernard, J. (2008). The brief 
resilience scale: Assessing the ability to bounce back. International Journal of Behavioral 
Medicine, 15(3), 194-200.  
Smith, L. (2005). New AHA recommendations for blood pressure measurement. American 
Family Physician, 72(7), 1391-1396.  
Sonu, S., Post, S., & Feinglass, J. (2019). Adverse childhood experiences and the onset of 
chronic disease in young adulthood. Preventive Medicine, 123, 163-170. 
Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006). A brief measure for 
assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166(10), 
1092-1097. 
91 
 
 
 
Steffen, P. R., Fidalgo, L., Schmuck, D., Tsui, Y., & Brown, T. (2014). Psychotherapy 
participants show increased physiological responsiveness to a lab stressor relative to 
matched controls. Frontiers in Psychology, 5, 795. 
Su, S., Jimenez, M. P., Roberts, C. T. F., & Loucks, E. B. (2015). The role of adverse childhood 
experiences in cardiovascular disease risk: A review with emphasis on plausible 
mechanisms. Current Cardiology Reports, 17(10), 1-10. 
Sutton, S. K., & Davidson, R. J. (1997). Prefrontal brain asymmetry: A biological substrate of 
the behavioral approach and inhibition systems. Psychological Science, 8(3), 204-210. 
Thayer, J. F., Ahs, F., Fredrikson, M., Sollers, J. J., & Wager, T. D. (2012). A meta-analysis of 
heart rate variability and neuroimaging studies: Implications for heart rate variability as a 
marker of stress and health. Neuroscience and Biobehavioral Reviews, 36(2), 747.  
Thayer, J. F., Hansen, A. L., Saus-Rose, E., & Johnsen, B. H. (2009). Heart rate variability, 
prefrontal neural function, and cognitive performance: The neurovisceral integration 
perspective on self-regulation, adaptation, and health. Annals of Behavioral Medicine, 37(2), 
141-153. 
Thayer, J. F., & Lane, R. D. (2007). The role of vagal function in the risk for cardiovascular 
disease and mortality. Biological Psychology, 74(2), 224-242. 
Tolin, D. F., & Foa, E. B. (2006). Sex differences in trauma and posttraumatic stress disorder: a 
quantitative review of 25 years of research. Psychological Bulletin, 132 (6), 959- 
992 
92 
 
 
 
Turchik, J. A., & Garske, J. P. (2009). Measurement of sexual risk taking among college 
students. Archives of Sexual Behavior, 38(6), 936-948. 
Ustun, B., Adler, L. A., Rudin, C., Faraone, S. V., Spencer, T. J., Berglund, P., . . . Kessler, R. C. 
(2017). The world health organization adult attention-deficit/hyperactivity disorder self-
report screening scale for DSM-5. JAMA Psychiatry, 74(5), 520-526. 
Victor, S. E., & Klonsky, E. D. (2016). Validation of a brief version of the difficulties in emotion 
regulation scale (DERS-18) in five samples. Journal of Psychopathology and Behavioral 
Assessment, 38(4), 582-589. 
Zajacova, A., & Dowd, J. B. (2011). Reliability of self-rated health in US adults. American 
Journal of Epidemiology, 174(8), 977-983. 
Zanotti, D. C., Kaier, E., Vanasse, R., Davis, J. L., Strunk, K. C., & Cromer, L. D. (2018). An 
examination of the test-retest reliability of the ACE-SQ in a sample of college athletes. 
Psychological Trauma : Theory, Research, Practice and Policy, 10(5), 559-562. 
 
93 
 
Appendix A: Aim 1 Hierarchical Linear Regression Results (SPSS) 
PHQ-9 
 
 
 
 
 
 
 
 
GAD-7 
 
 
 
 
 
 
95 
 
 
 
PCL-5 
 
 
 
 
 
96 
 
 
 
ASRS-5 
 
 
 
 
 
 
97 
 
 
 
DERS-18 
 
 
 
 
 
98 
 
 
 
PSS 
 
 
 
 
99 
 
 
 
ESP 
 
 
 
 
 
100 
 
 
 
BRS 
 
 
 
 
101 
 
 
 
BIS 
 
 
 
 
 
102 
 
 
 
BAS–Reward Responsiveness 
 
 
 
 
 
103 
 
 
 
BAS–Drive 
 
 
 
 
104 
 
 
 
BAS–Fun Seeking 
 
 
 
105 
 
Appendix B: Hierarchical Linear Regression Results (SPSS) 
PSQI 
 
 
 
 
 
 
GLTEQ 
 
 
 
 
 
107 
 
 
 
AUDIT-C 
 
 
 
 
 
108 
 
 
 
DAST-10 
 
 
 
 
 
109 
 
 
 
EDS-4 
 
 
 
 
 
110 
 
 
 
SRS 
 
 
 
 
 
111 
 
 
 
PHQ-15 
 
 
 
 
 
112 
 
 
 
Self-Rated Health 
 
 
 
 
 
113 
 
 
 
Healthcare Utilization 
 
 
 
114 
 
Appendix C: Hierarchical Linear Regression Results (SPSS) 
PHQ-9 
 
 
 
 
 
 
 
GAD-7 
 
 
 
 
116 
 
 
 
PCL-5 
 
 
 
 
117 
 
 
 
ASRS-5 
 
 
 
 
118 
 
 
 
DERS-18 
 
 
 
 
119 
 
 
 
ESP 
 
 
 
 
120 
 
 
 
PSQI 
 
 
 
121 
 
 
 
DAST-10 
 
 
 
122 
 
 
 
EDS-4 
 
 
 
123 
 
 
 
PHQ-15 
 
 
 
124 
 
Appendix D: IRB Approval Letter 
 
EAST  CAROLINA  UNIVERSITY 
University & Medical Center Institutional Review Board   
4N-64 Brody Medical Sciences Building· Mail Stop 682 
600 Moye Boulevard · Greenville, NC 27834 
Office 252-744-2914  · Fax 252-744-
 
2284  · rede.ecu.edu/umcirb/  
 
Notification of Initial Approval: Expedited 
 
From: Biomedical IRB 
To: Alexander Capiaghi 
CC:  Matthew Whited 
Date: 2/17/2020  
Re: UMCIRB 19-003332 
Adverse Childhood Experiences and Heart Rate Variability 
 
I am pleased to inform you that your Expedited Application was approved. Approval of the study and 
any consent form(s) occurred on  2/14/2020. The research study is eligible for review under 
expedited category # 4,7. The Chairperson (or designee) deemed this study no more than minimal 
risk. 
 
Changes to this approved research may not be initiated without UMCIRB review except when 
necessary to eliminate an apparent immediate hazard to the participant.  All unanticipated problems 
involving risks to participants and others must be promptly reported to the UMCIRB.  The investigator 
must submit a Final Report application to the UMCIRB prior to the Expected End Date provided in the 
IRB application.  If the study is not completed by this date, an Amendment will need to be submitted 
to extend the Expected End Date.  The Investigator must adhere to all reporting requirements for this 
study. 
 
Approved consent documents with the IRB approval date stamped on the document should be used 
to consent participants (consent documents with the IRB approval date stamp are found under the 
Documents tab in the study workspace). 
 
The approval includes the following items: 
 
Name Description 
Capiaghi - Thesis Proposal Study Protocol or Grant Application 
Debriefing Statement Debriefing Statement 
Laboratory Consent Form Consent Forms 
List of Self-Report Measures Surveys and Questionnaires 
Online Consent Form Consent Forms 
Qualtrics Survey Surveys and Questionnaires 
Recruitment Script Recruitment Documents/Scripts 
 
 
 
 
 
  
For research studies where a waiver of HIPAA Authorization has been approved, each of the waiver 
criteria in 45 CFR 164.512(i)(2)(ii) has been met. Additionally, the elements of PHI to be collected as 
described in items 1 and 2 of the Application for Waiver of Authorization have been determined to be 
the minimal necessary for the specified research. 
 
The Chairperson (or designee) does not have a potential for conflict of interest on this study. 
 
 
 
  
IRB00000705 East Carolina U IRB #1 (Biomedical) IORG0000418 
IRB00003781 East Carolina U IRB #2 (Behavioral/SS) IORG0000418 
 
 
 
 
126 
 
Appendix E: Informed Consent for Online Portion of Study 
Dear Participant, 
  
            I am a student at East Carolina University in the Department of Psychology. I am asking you 
to take part in my research study entitled, “Overcoming Adverse Childhood Experiences.”  
  
The purpose of this research is to better understand the relationship between early life experiences 
and health. By doing this research, I hope to explore ways that health interventions may be improved 
for emerging adults. Your participation is completely voluntary.  
  
You are being invited to take part in this research because you are a student between 18 and 25 
years of age in an ECU introductory psychology course. The amount of time it will take you to 
complete this survey is approximately 45 to 60 minutes.  
  
If you agree to take part in this survey, you will be asked questions that relate to childhood 
experiences and various components of mental and physical health. Once you complete the online 
survey portion of this study, you may also opt to sign up and take part in the in-person laboratory 
portion. 
  
This research is overseen by the University and Medical Center Institutional Review Board(UMCIRB) 
at ECU. Therefore, some of the UMCIRB members or the UMCIRB staff may need to review your 
research data. Your identity will be evident to those individuals who see this information. However, I 
will take precautions to ensure that anyone not authorized to see your identity will not be given that 
information 
  
Identifiers might be removed from the identifiable private information and, after such removal, the 
information could be used for future research studies or distributed to another investigator for future 
research studies without additional informed consent from you or your Legally Authorized 
Representative (LAR). However, there still may be a chance that someone could figure out the 
information is about you. 
  
Please contact Alex Capiaghi at 845-216-8751 or capiaghia18@students.ecu.edu for any research 
related questions. If you have questions about your rights when taking part in this research, call the 
University and Medical Center Institutional Review Board (UMCIRB) at252-744-2914 (Monday-
Friday, 8:00AM-5:00PM). If you would like to report a complaint or concern about this research 
study, call the Director of Human Research Protections at 252-744-2914. 
  
You do not have to take part in this research, and you can stop at any time. If you decide you are 
willing to take part in this study, check the AGREE box below and the research questions will 
appear. 
  
Thank you for taking the time to participate in my research. 
  
Sincerely, Alex Capiaghi, Principal Investigator 
  
 
By clicking "I agree," I certify that: 
 
1) I have read all of the above information 
2) I am between 18 and 25 years of age 
3) I am currently enrolled in an Introductory Psychology course 
4) I know that I can stop taking part in this study at any time