University of Southern California Dissertations and Theses

Mental health outcomes associated with profiles of risk and resilience among military-connected youth

(USC Thesis Other)

Mental health outcomes associated with profiles of risk and resilience among military-connected youth

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content i

Mental Health Outcomes Associated with Profiles of Risk and Resilience among Military-
Connected Youth

Kathrine S. Sullivan, MSW, Ph.D. Candidate

Dissertation

Doctor of Philosophy (SOCIAL WORK)

August 2018 Degree Conferral

University of Southern California

Dissertation Guidance Committee:
Carl Castro, Ph.D. (Chair)
Julie Cederbaum, Ph.D.
Gayla Margolin, Ph.D.

FACULTY OF THE USC GRADUATE SCHOOL

ii
Table of Contents
Tables and Figures ......................................................................................................................... iv
Dedication ....................................................................................................................................... v
Acknowledgements ........................................................................................................................ vi
Chapter 1: Overview of the Three Studies ...................................................................................... 1
Abstract ..................................................................................................................................... 1
Introduction and Significance ................................................................................................... 3
Conceptual Framework ....................................................................................................... 4
Risk and Protective Factors for Military Spouses and Children ......................................... 5
Contributions to the Literature ............................................................................................ 9
Methodology ........................................................................................................................... 10
Overview ........................................................................................................................... 10
Comprehensive Soldier and Family Fitness ..................................................................... 11
DOD Archival Datasets .................................................................................................... 12
Person-Event Data Environment....................................................................................... 12
Big Data ............................................................................................................................ 13
Participants ........................................................................................................................ 14
Analyses ............................................................................................................................ 14
Psychometrics ................................................................................................................... 15
Analytic Plan ..................................................................................................................... 16
References ............................................................................................................................... 17
Chapter 2 (Study 1): Preliminary Psychometrics and Potential Big Data Uses of the U.S. Army
Family Global Assessment Tool ................................................................................................... 29
Abstract ................................................................................................................................... 29
Introduction ............................................................................................................................. 30
Theoretical Foundations of CSF2 and the Family GAT ................................................... 30
Big Data Uses of the Family GAT .................................................................................... 31
Methods................................................................................................................................... 32
Participants and Procedures .............................................................................................. 32
Sample............................................................................................................................... 33
Family GAT Measures ...................................................................................................... 34
Analytic Plan ..................................................................................................................... 37
Results ..................................................................................................................................... 38
Family GAT Structure ...................................................................................................... 38
Reliability .......................................................................................................................... 40
Validity ............................................................................................................................. 40
Discussion ............................................................................................................................... 42
Limitations ........................................................................................................................ 44
Future Directions .............................................................................................................. 45
References ............................................................................................................................... 47
Chapter 3 (Study 2): Mental Health Outcomes Associated with Profiles of Risk and Resilience
among Army Spouses ................................................................................................................... 60
Abstract ................................................................................................................................... 60
Introduction ............................................................................................................................. 61
Theoretical Foundation ..................................................................................................... 62
iii
Empirical Evidence for Risk and Protective Factors ........................................................ 63
The Current Study ............................................................................................................. 66
Methods................................................................................................................................... 67
Family Global Assessment Tool ....................................................................................... 67
Data Linkage ..................................................................................................................... 67
Participants ........................................................................................................................ 68
Measurement ..................................................................................................................... 69
Analyses ............................................................................................................................ 72
Results ..................................................................................................................................... 73
LPA Model........................................................................................................................ 73
Covariates ......................................................................................................................... 74
Distal Outcomes ................................................................................................................ 75
Discussion ............................................................................................................................... 75
Strengths and Limitations ................................................................................................. 80
Conclusions ....................................................................................................................... 81
References ............................................................................................................................... 83
Chapter 4 (Study 3): Mental Health Outcomes Associated with Profiles of Risk and Resilience
among Military-Connected Youth ................................................................................................ 98
Abstract ................................................................................................................................... 98
Introduction ............................................................................................................................. 99
Theoretical Foundation ................................................................................................... 100
Empirical Evidence of Risk and Protective Factors ....................................................... 102
The Current Study ........................................................................................................... 104
Methods................................................................................................................................. 106
Family Global Assessment Tool ..................................................................................... 106
Data Linkage ................................................................................................................... 106
Participants ...................................................................................................................... 107
Measurement ................................................................................................................... 108
Analyses .......................................................................................................................... 111
Results ................................................................................................................................... 112
LPA Model...................................................................................................................... 112
Covariates ....................................................................................................................... 114
Distal Outcomes .............................................................................................................. 114
Discussion ............................................................................................................................. 115
Limitations ...................................................................................................................... 119
Conclusions ..................................................................................................................... 121
References ............................................................................................................................. 123
Chapter 5: Conclusions, Implications, and Future Directions .................................................... 139
Major Findings and Implications for Policy and Practice .................................................... 140
Study 1 ............................................................................................................................ 140
Studies 2 and 3 ................................................................................................................ 141
Future Directions .................................................................................................................. 144
Concluding Thoughts ............................................................................................................ 146
References ............................................................................................................................. 148

iv
Tables and Figures
Figure 1.1. Integrated Conceptual Model ..................................................................................... 28
Table 2.1. Pattern Matrix after Oblique Rotation ......................................................................... 54
Table 2.2. Final Factors and Items ................................................................................................ 56
Table 2.3. Bivariate Correlations between Composite Scores ...................................................... 57
Table 2.4. Simple Theoretically Driven Regression Model ......................................................... 58
Figure 2.1. Moderation Effects ..................................................................................................... 59
Table 3.1. Sample Demographics and Latent Profile Indicator Means and Prevalence ............... 92
Table 3.2. Model Fit Indexes for LPA Model .............................................................................. 94
Table 3.3. Means and Conditional Probabilities for Risk and Protective Factor Indicators ......... 95
Table 3.4. Service Member Covariates Associated with Latent Profiles ..................................... 96
Table 3.5. Latent Profiles Associated with Distal Outcome (Mental Health Diagnosis) ............. 97
Table 4.1. Sample Demographics and Latent Profile Indicator Means and Prevalence ............. 133
Table 4.2. Model Fit Indexes for LPA Model ............................................................................ 135
Table 4.3. Means and Conditional Probabilities for Risk and Protective Factor Indicators ....... 136
Table 4.4. Service Member Covariates Associated with Latent Profiles ................................... 137
Table 4.5. Child Latent Profiles Associated with Distal Outcome (Mental Health Diagnosis) . 138

v
Dedication

To my husband, Kevin – Thank you for literally and figuratively fueling this effort. I would not
be here without your love and support.

To my son, Aidan – Thank you for making all of this worthwhile.

To my Dad – Thank you for setting the bar high and always supporting my dreams in ways both
big and small.

To my Mom – Thank you for being the professional woman, the friend, the mother, and the
human being that I strive to emulate every day. Words cannot describe how much I wish
you were here to celebrate this moment – and so many others – with me.

vi
Acknowledgements

To my cohort (in name and in spirit), Nick Barr, Gordon Capp, Jaih Craddock, Andi Lane
Eastman, Karissa Fenwick, Chung Hyeon Jeong, and Rebecca Lengnick-Hall – Thank
you for skyping me in to class when I was at home with my mom and walking in circles
around my living room with me while I tried to soothe a colicky baby. Life does not stop
when this endeavor begins. I would not have made it over the peaks or out of the valleys
without all of you.

To Dr. Tamika Gilreath, Dr. Gayla Margolin, Dr. Sara Kintzle, Dr. Patricia Lester, Dr. Gary
Bowen, Dr. Stacy Hawkins, and Eric Lindberg – Thank you for your energy, expertise,
and guidance. This project would not be what it is without your input and support.

To Dr. Ron Astor – Thank you for launching me on this journey and guiding me toward a
burgeoning body of work of which I can be proud.

To Dr. Michael Hurlburt – Thank you for your critical eye, your strategic, thoughtful and
rigorous appraisal of my work, and your generosity with your time and experience.

To Dr. Julie Cederbaum – Thank you for being a compass and a sounding board and for
modeling how to do this work with excellence, integrity, balance, and humor.

To Dr. Carl Castro – Thank you for your wisdom, generosity, and candor, for your willingness to
open doors for me, and for your unquestioning trust and belief in my abilities.
1
Chapter 1: Overview of the Three Studies
Abstract
Although many military families cope with stressors effectively (Wadsworth, 2013), a
significant subset of military spouses and children, who number close to 3 million (U.S.
Department of Defense, 2017), experience adverse consequences of wartime military service
including poor health and mental health, increased risk behaviors, suicidality, and substance use
(Chandra, Lara-Cinisomo, et al., 2010; de Burgh, White, Fear, & Iversen, 2011; Flake, Davis,
Johnson, & Middleton, 2009; Gilreath et al., 2016; Gorman, Eide, & Hisle-Gorman, 2010; Lester
et al., 2010; Mansfield et al., 2010; Sullivan et al., 2015). These outcomes are a public health
concern and highlight the need to understand contributing factors to develop policy, prevention,
and intervention efforts to support military families.
This three-paper dissertation was guided by a risk and resilience framework (Black &
Lobo, 2008; Bowen & Martin, 2011; Masten & Narayan, 2012; Palmer, 2008; Rodriguez &
Margolin, 2015), which posits that outcomes are determined by a dynamic balance of (a) the
stressors a family faces and (b) protective factors the family possesses or accesses from
surrounding systems. Although the most recent research examining resilience in military families
has used variable-focused analytic models (e.g., Cederbaum et al., 2014; Gilreath et al., 2013;
Sullivan et al., 2015), this work employs a person- or family-focused approach, defining unique
subgroups of military families based on profiles of risk and protective factors and evaluating
outcomes (Masten, 2001; Nurius & Macy, 2008; Rosato & Baer, 2012). A family-centered
approach has the potential to advance understanding due to its increased focus on the strengths of
military families and the use of analytic models able to examine constellations of risk and
protective factors as they naturally occur (Masten, 2001; Nurius & Macy, 2008; Rosato & Baer,
2
2012; Saltzman et al., 2011). Capitalizing on the U.S. Army’s extensive secondary data holdings,
this dissertation research used Comprehensive Soldier and Family Fitness data along with
additional U.S. Department of Defense archival datasets to understand risk and protective factors
affecting military families and associated mental health outcomes for spouses and children.
Chapter 1 presents the theoretical framework guiding this project, a discussion of the
theoretical and empirical evidence undergirding the risk and protective factors employed in these
analyses, and an overview of the methodology for each of the three studies. Chapter 2 (Study 1)
examines the psychometric properties of the Family Global Assessment Tool, which will inform
protective factor scales employed in subsequent models. Chapter 3 (Study 2) explores patterns of
risk and protective factors affecting military families and their association with mental health
outcomes for military spouses. Because spouse outcomes were hypothesized to have cascading
effects on youth outcomes, Chapter 4 (Study 3) builds on the results of Study 2 by modeling
spouse mental health as a risk factor along with other family-level risk and protective factors
associated with mental health outcomes for military-connected youth. Finally, Chapter 5
explores conclusions that can be drawn from the three studies and outlines broader implications
for future research, policy, and practice.

3
Introduction and Significance
Since the transition to an all-volunteer military in 1973, significant demographic changes
have occurred in the ranks of the U.S. Armed Forces (U.S. Department of Defense, 2017). The
implications of these changes have taken decades to be fully appreciated. Currently, across
service branches, 50.0% of personnel are married and 42.7% have children. Family members
(56.4%) outnumber military personnel (43.6%; U.S. Department of Defense, 2017). Empirical
evidence has demonstrated that the impact of war extends beyond service members to their
families. Many definitions of what constitutes a military family exist; a strict definition includes
the spouses and children of active-duty and reserve-component personnel (Cozza, Haskins, &
Lerner, 2013). Since 2001, these families have experienced the longest sustained overseas
deployment in the history of the all-volunteer force. More than 2 million children have been
affected by the deployment of a parent; many have faced multiple, prolonged deployments.
Extensive media coverage and real-time communication with deployed soldiers have increased
awareness of the dangers faced by loved ones overseas. Families of service members returning
from combat are often confronted with a loved one who has physical injuries or is experiencing
the psychological consequences of combat (Cozza et al., 2013). Although many military families
appear to cope with stressors, evidence has suggested a significant subset of military spouses and
children are experiencing adverse outcomes associated with military family life during wartime
(Chandra, Lara-Cinisomo, et al., 2010; de Burgh et al., 2011; Flake et al., 2009; Gilreath et al.,
2016; Gorman et al., 2010; Lester et al., 2010; Mansfield et al., 2010; Sullivan et al., 2015). This
dissertation addresses two overarching aims: to explore risk and protective factors and associated
outcomes in (a) military spouses and (b) military-connected youth to elucidate possible points of
intervention to better support the healthy development of military-connected youth.
4
Conceptual Framework
This dissertation is framed by an integration of theoretical orientations and constructs,
including a risk and resilience framework, family systems theory, and developmental cascades.
Risk and resilience framework. In contrast to deficit-oriented approaches, family
resilience encompasses the interactional and systemic processes occurring in family systems that
enable families to thrive despite adversity (Walsh, 1996, 2003). The concept of resilience has
been variably defined in the literature as an inherent trait possessed by individuals or family
systems, a process of leveraging protective factors to manage risk, or an outcome of that process.
The risk and resilience perspective guiding this dissertation defines resilience as the sum of
psychological and social processes that permit families to maintain or return to previous levels of
well-being and functioning in response to adversity (Hawkins et al., 2017).
Family systems theory. Family systems theory is a broad perspective that posits: (a) the
family as a unit is greater than the sum of its individual members; (b) family members have an
ongoing, reciprocal impact on one another; and (c) the behavior of individuals is best understood
in the context of the family system (Cox & Paley, 2003; Minuchin, 1974). In military families,
this orientation implies that the experiences of the military parent will have ramifications for
family dynamics, which ultimately affect the functioning of the military spouse and child. Recent
research with military families demonstrated that examining individual functioning and health
outcomes is best undertaken in the context of the family unit (O’Neal, Lucier-Greer, Mancini,
Ferraro, & Ross, 2016).
Developmental cascades. The term developmental cascade refers to “the cumulative
consequences for development of the many interactions and transactions occurring in developing
systems that result in spreading effects across levels, among domains at the same level, and
5
across different systems or generations” (Masten & Cicchetti, 2010, p. 491). With regard to
military families, cascade effects help to explain the “spillover” of stress associated with military
family life from one member of a family system to other members of that system (Masten, 2013).
Although military-related stressors can have a direct impact on all members of the family
system, cascade effects may compound the adverse impact of these stressors.
Integrated framework. Combining the crucial elements of these three perspectives, this
dissertation assumes that: (a) outcomes for military-connected family members (i.e., spouses and
children) will be determined by a dynamic balance of risk and protective factors; (b) these risk
and protective factors will include factors affecting the family system as a whole; and (c)
outcomes for military spouses will have a cascading impact on outcomes for military children,
which compound the direct effects of risk and protective factors. Although many outcomes could
result from this dynamic resilience process, the adverse outcome of focus in this dissertation will
be individual mental health (Bowen & Martin, 2011).

This integrated framework is depicted in
Figure 1.1.
Risk and Protective Factors for Military Spouses and Children
Military spouses are described as the foundation of resilience processes in their family
systems, particularly during wartime, when service members may be absent due to deployment
and work-related obligations (Green, Nurius, & Lester, 2013; Riggs & Riggs, 2011). Outcomes
for spouses may have a cascading effect on children in military families (Masten, 2013). Thus,
an initial focus on factors influencing outcomes for military spouses has direct relevance to
understanding health and positive development among military youth.
Risk factors affecting military spouses. Common stressors associated with military life
include family separations and frequent moves (Burrell, Adams, Durand, & Castro, 2006).
6
Deployment of a service member, in particular, has been found to be a significant risk factor for
military spouses (de Burgh et al., 2011; Mansfield et al., 2010) and has been associated with
adverse outcomes. Reunification, a period that can last 6 months or longer following the end of a
deployment (Pincus, House, Christenson, & Adler, 2001), can be an added stressor, as spouses
reconnect and readjust to sharing roles (McNulty, 2010, 2013). On average, military families
move every 2 to 3 years (Park, 2011). These moves can affect spouses’ ability to maintain
employment (Harrell, Lim, Castaneda, & Golinelli, 2004; U.S. Department of Defense, 2015)
and may be negatively associated with well-being (Burrell et al., 2006; Lincoln, Swift, &
Shorteno-Fraser, 2008; U.S. Department of Defense, 2015). When soldiers exhibit physical or
mental health problems, spouses bear a significant burden and may experience increases in
adverse outcomes (de Burgh et al., 2011). Normative stressors may exacerbate previously
discussed war-related risk factors. In particular, many military families face financial challenges;
these challenges are most evident among enlisted soldiers compared to officers (Lincoln et al.,
2008). Further, families with more children in the home may be at greater risk (Haas &
Pazdernik, 2007; Wiens & Boss, 2006).
Protective factors affecting military spouses. Typical deficit-oriented models have
focused less on protective factors. However, the most commonly cited protective factor for
military spouses is access to and use of social support. Rates of anxiety among spouses were
found to be lower during family separations when they could access both formal support,
through their community or the military, and informal support from family or friends (Bowen &
Martin, 2011; Burrell, Durand, & Fortado, 2003). Recently, perceived military community
support, including feelings of connection to the military community, was associated with military
parent psychosocial functioning (Conforte et al., 2017). Additionally, effective coping and a
7
healthy spousal relationship can be protective factors for military spouses (Huebner, Mancini,
Bowen, & Orthner, 2009; McGuire et al., 2012; Orthner & Rose, 2009; Weins & Boss, 2006).
One recent study highlighted the importance of marital support during deployments and found
daily fluctuations in levels of support based on communication patterns between deployed
spouses and their at-home partners (Wilson et al., 2018). Finally, it has been hypothesized that
families that can draw on a “guiding belief system” or sense of meaning have better outcomes
(Saltzman et al., 2011).
Outcomes for military spouses. Adverse outcomes related to military stressors have
been documented among spouses, including potentially disruptive distress, parenting stress,
mental health diagnoses, substance use, and secondary traumatization (Ahmadi & Green, 2011;
Eaton et al., 2008; Flake et al., 2009; Mansfield et al., 2010; Renshaw et al., 2011; Renshaw,
Rodebaugh, & Rodrigues, 2010). Military spouses have been found to experience adverse
outcomes at rates similar to service members but may seek treatment at higher rates from both
primary and mental health care providers (Eaton et al., 2008).
Cascade effects. The phenomenon of developmental cascades among military families
suggests that if military spouses can successfully perform their daily roles, their children may be
at less risk; in contrast, children of military spouses experiencing increased distress, health, or
mental health problems may be at greater risk of adverse outcomes. Among military families
with young children, the functioning of the at-home parent has been found to be crucial for child
adjustment (Paris, DeVoe, Ross, & Acker, 2010). Young children who’s at-home parent was
stressed experienced greater deterioration in their behavior during parental deployment (Barker
& Berry, 2009). Among older children, military spouses’ mental health significantly affected
behavioral, psychological, and academic outcomes during and following deployment (Chandra,
8
Martin, Hawkins, & Richardson, 2010; Finkel, Kelley, & Ashby, 2003; Lester et al., 2010).
Spousal reports of parenting stress were also associated with higher psychosocial morbidity
among older youth (Flake et al., 2009).
Risk factors affecting military children. In addition to the functioning of the at-home
parent, parental deployment and reunification have been associated with poor outcomes for
military youth (Cederbaum et al., 2014; Chandra, Lara-Cinisomo, et al., 2010; Flake et al., 2009;
Gilreath et al., 2013; Lester et al., 2010). Deployments have been associated with increased
psychosocial morbidity, emotional difficulties, depressive symptoms and suicidal ideation, recent
and lifetime substance use, and decreased academic engagement and achievement (Cederbaum et
al., 2014; Chandra, Lara-Cinisomo, et al., 2010; Engel, Gallagher, & Lyle, 2010; Flake et al.,
2009; Gilreath et al., 2013). Family moves may also be a risk factor for poor academic outcomes
and increased mental distress (Bradshaw, Sudhinaraset, Mmari, Blum, & Hopkins, 2010).
Having a military parent with a lower rank (e.g., enlisted vs. officer) may be a marker of
financial strain in the family system, which has been associated with increased depressive
symptoms among military youth (Lucier-Greer, O’Neal, Arnold, Mancini, & Wickrama, 2014)
and may imply less work control. This has been associated with increased psychological strain
for service members, and by extension, adverse outcomes for youth (Bliese & Castro, 2000).
Finally, exposure to the military parent’s mental health symptoms may adversely affect youth
outcomes (Creech, Hadley, & Borsari, 2014; Foran, Eckford, Sinclair, & Wright, 2017; Holmes,
Rauch, & Cozza, 2013; Jordan et al., 1992; Solomon, Debby-Aharon, Zerach, & Horesh, 2011).
Protective factors affecting military children. At the individual and family levels,
access to social support has been found to be protective for military-connected youth. Living on
a military base with greater access to formal sources of support and more immediate access to
9
informal support, including friendships with other military youth, may lead to better outcomes
(Chandra, Lara-Cinisomo, et al., 2010; Huebner & Mancini, 2010; Mmari, Roche, Sudhinaraset,
& Blum, 2009). The family’s community connectedness and community support has been
associated with greater well-being among military-connected youth (Conforte et al., 2017; Flake
et al., 2009; Lucier-Greer et al., 2014). Finally, the interrelated concepts of family cohesion and
healthy marital and parenting relationships are likely protective for youth (Finkel et al., 2003;
Foran et al., 2017; Paley, Lester, & Mogil, 2013).
Outcomes for military children. Although many military children appear to function
well despite the aforementioned risk factors, many also experience adverse outcomes. Military
connectedness and exposure to military-related stressors, including deployment, have been
associated with physical symptoms including increased heart rate (Barnes, Davis, & Treiber,
2007), behavioral health visits (Gorman et al., 2010; Mansfield, Kaufman, Engel, & Gaynes,
2011), increased rates of mental health disorders (Gorman et al., 2010; Mansfield et al., 2011),
lower academic achievement and poor school performance (Bradshaw et al., 2010; Engel et al.,
2010), substance use (Gilreath et al., 2013; Sullivan et al., 2015), victimization (Gilreath, Astor,
Cederbaum, Atuel, & Benbenishty, 2014; Sullivan et al., 2015), and disrupted relationships with
parents (Holmes et al., 2013; Jordan et al., 1992; Solomon et al., 2011).
Contributions to the Literature
Much research to date has focused on individual risk factors affecting military families.
Rarely have multiple risk factors been considered simultaneously, although some evidence of the
influence of cumulative risk on military youth outcomes exists (Lucier-Greer, Arnold, Mancini,
Ford, & Bryant, 2015). Focusing on individual risk factors may underestimate the impact of
military-related stressors on youth and mask cascading risk from spouses to youth in military
10
families. Further, an emphasis on risk has come at the expense of considering protective factors
that may account for positive outcomes observed among many military youth. The majority of
research has used a variable-focused approach, potentially obscuring heterogeneity in family
structure and experiences (Nurius & Macy, 2008; Rosato & Baer, 2012). Many questions remain
regarding how family members maintain resilient functioning in the face of significant stressors
and what outcomes are associated with different family structures and experiences (Maholmes,
2012).
This dissertation contributes to the scientific literature in three ways. First, it provides a
more comprehensive understanding of how risks occur and co-occur in military families,
including how such risks compound and cascade through family systems. Second, it maintains a
concurrent focus on both risk and protective factors, providing a more complete picture of
military families and highlighting vulnerability and characteristics or skills that may be
protective and can be developed to improve outcomes. Third, its use of person-focused models
captures the heterogeneity in military family systems, generating useful information for
clinicians tasked with supporting families with a variety of characteristics and experiences.
Addressing these gaps in the literature has implications for prevention, intervention, policy
development, and service delivery with this population. Further, lessons learned from these
efforts have broader applicability for work with vulnerable families in different contexts outside
the military.
Methodology
Overview
This dissertation describes a secondary analysis of data available from the U.S.
Department of Defense (DOD) through a memorandum of agreement between the Army
11
Research Facilitation Laboratory, under the Army Analytics Group, and the University of
Southern California. Using linked, deidentified data from the Comprehensive Soldier and Family
Fitness (CSF2) Global Assessment Tool (GAT) and other DOD archival datasets, latent profile
analysis (LPA) was used to identify patterns of risk and protective factors inherent in military
family systems that affect outcomes for military spouses. Military spouse outcomes informed an
LPA model of risk and protective factor profiles for military youth, which were associated with
their outcomes.
Comprehensive Soldier and Family Fitness
Although military leadership has traditionally focused on the physical fitness of service
members, recent empirical evidence highlighting the negative effects of combat exposure on
service members (Hoge et al., 2004) and family separation due to deployment on family
members (Flake et al., 2009) has brought increased attention to the psychological health of both
service members and their families. Traditional approaches to addressing these issues tended to
focus on treating new and existing psychological injury. However, the Army’s CSF2 program
endeavors to prevent adverse outcomes by assessing, promoting, and sustaining the
psychological fitness of service members and their families (Peterson, Park, & Castro, 2011).
The GAT, with specific versions for both service members and families, was developed
by the CSF2 team to measure emotional, social, family, and spiritual fitness, hypothesized to be
elements of overall psychological fitness (Peterson et al., 2011; Seligman & Fowler, 2011). The
Soldier GAT is a mandatory survey completed by all Army soldiers once per year. The Family
GAT is completed voluntarily by Army spouses, who complete the survey by visiting the hosting
website. CSF2, and the Family GAT specifically, are extensively publicized as a critical
component of the Army’s Ready and Resilient Campaign, focused on enhanced performance and
12
mission readiness of the total force. Spouses learn about the program through Army publicity,
their spouse, their spouse’s military unit, or their family readiness group. After completing the
survey, spouses receive feedback about their responses and links to online learning modules to
address identified challenges. Upon completion of both the Soldier and Family GATs,
respondents have the option to give consent for their data to be used for human subjects research.
Data from soldiers and spouses who have provided consent were used in the current analyses.
DOD Archival Datasets
Demographic data were drawn from the service members’ Master Personnel File and
Family Files. These datasets provide information on the age and gender of soldiers and
dependents, the legal relationship between soldiers and spouses, parental status, soldier’s rank
(enlisted or officer), and the family’s current and previous duty stations. The Contingency
Tracking System Deployment files managed by the Defense Manpower Data Center provided
information regarding the timing and length of deployments and reunifications. Further, the
Defense Health Agency maintains records on health care use for all soldiers and covered
dependents enrolled in Tricare Health Insurance. These data include dates of health care visits
and International Statistical Classification of Diseases (ICD-9) codes for diagnoses. Data are
available from both military and civilian inpatient and outpatient providers related to services
covered by insurance.
Person-Event Data Environment
These disparate data sources are linked and analyzed through the Person-Event Data
Environment (PDE). The PDE is a cloud-based, virtual enclave created by the Army Analytics
Group to address challenges associated with data access and linkage across the Army and DOD.
The PDE is a repository of more than 350 manpower, service, personnel, financial, behavioral
13
health, and medical datasets for soldiers and dependents. In the PDE, data are deidentified and
available to authorized users. Because these are real-time active datasets, new data are entered
into the PDE on a regular basis, providing users with access to up-to-date information regarding
soldiers and their families. Access to the PDE allows the linkage of data for soldiers, spouses,
and their children across previously disparate datasets to identify and analyze a wide array of risk
and protective factors simultaneously. This level of access makes possible the use of person-
centered analytic models to explore the heterogeneity in military family structure, experiences,
and outcomes.
Big Data
The term “big data” has been variably defined in the literature (Manyika et al., 2011).
However, big data practically refers to the linkage of previously disparate data sources to achieve
a more accurate, robust, and potentially clinically meaningful representation of an object of study
than could be achieved using one data source alone (Hawkins et al., 2017). These methods
present several challenges, including that data likely were not collected for research purposes and
may not include all potential variables of interest. However, big data provide a breadth and depth
of information that allows for the consideration of complex questions that one data source may
not.
The data linkage capabilities of the PDE, as previously described, and unprecedented
access to both CSF2 data and a wealth of DOD data sources present an unusual opportunity that
allows for nuanced analyses of military family experiences while avoiding many of the
challenges of collecting original data with this population. Previous research focused on military
families has been hampered by access challenges. Both military personnel and children are
considered vulnerable populations (U.S. Department of Defense, 2011), raising potential ethical
14
challenges with primary data collection. Using a big data approach to analyze existing but
previously segregated deidentified DOD datasets overcomes these barriers and provides a more
complete depiction of the lives of service members and their families. Further, a big data
approach avoids misattribution by incorporating information on actual service utilization (e.g.
medical visits) rather than symptom reporting.
Participants
Military spouses are considered the index respondent for each family included in this
project. To be included in analyses, spouses had to: (a) provide consent to have their GAT survey
responses used for research purposes; and (b) be linked to an Army soldier who had also
completed a GAT survey and provided consent for responses to be used in research. Specific
information about the sample included in each of the three studies is presented in Chapters 2
through 4.
Analyses
Previous research examining mechanisms of risk and resilience has taken two distinct
methodological approaches: variable-focused and person-focused models (Masten, 2001). A
variable-focused approach involves the use of statistical tools, including regression or structural
equation modeling, useful for capturing aggregate relationships between variables for an entire
sample and understanding underlying explanatory processes (Nurius & Macy, 2008). By
contrast, person-focused tools, including LPA, are useful for examining relationships at the
person or family level and exploring heterogeneity, because this technique identifies subgroups
in a sample (Rosato & Baer, 2012). Using a family-based approach to identify patterns has
demonstrated utility in understanding the functioning of military families (Oshri et al., 2015).
15
For the purposes of this dissertation, LPA is a useful tool for several reasons. First,
resilience has been conceptualized as a dynamic balance between risk and protective factors.
This process-oriented definition requires a statistical modeling technique that allows for
simultaneous evaluation of both risk and protective factors. Second, military families are not a
homogenous group; rather, subgroups that have different family structures and constellations of
risk and protective factors may have markedly different outcomes. LPA can define unique
subgroups of military families for whom outcomes can then be evaluated.
Psychometrics
Variables drawn from DOD archival datasets are observed variables that do not require
statistical validation. These variables primarily reflect risk factors in proposed models, including
cumulative length of deployments, reunification, or relocation; soldier’s rank; and receipt of a
mental health diagnosis. By contrast, scales drawn from the Family GAT are latent variables that
represent primarily protective factors. However, the provenance of many items on the Family
GAT are unknown. In some cases, individual items have been drawn from previously validated
scales, but often these items have been modified and complete scales are not included, which
presents significant challenges for interpretation. As such, the scales included in models required
validation. The specifics of this effort are described in Chapter 2. Although the approach to
psychometric validation used in this study does not provide a perfect measure of validity, this
approach is in keeping with recent efforts to validate the Soldier GAT (Vie, Scheier, Lester &
Seligman, 2016). Further, the overall strengths inherent in the big data approach described here,
including the opportunity to provide a more rich and robust picture of military families through
the linkage of previously disparate datasets, outweighs the potential psychometric weakness of
the Family GAT scales.
16
Analytic Plan
Following psychometric evaluation, two LPA models were specified sequentially: (a) a
spouse model capturing factors related to spouse outcomes; and (b) a child model capturing
factors related to youth outcomes. To reflect cascade effects, the distal outcome in the spouse
model was treated as a risk factor in the child model. Family GAT completion date served as the
index for all variables involving an element of time (e.g., a reunification during the preceding
year). All items in the Family GAT feature 5-point Likert response options.

17
References
Ahmadi, H., & Green, S. L. (2011). Screening, brief intervention, and referral to treatment for
military spouses experiencing alcohol and substance use disorders: A literature review.
Journal of Clinical Psychology in Medical Settings, 18, 129–136.
https://doi.org/10.1007/s10880-011-9234-7
Barker, L. H., & Berry, K. D. (2009). Developmental issues impacting military families with
young children during single and multiple deployments. Military Medicine, 174, 1033–
1040. https://doi.org/10.7205/MILMED-D-04-1108
Barnes, V. A., Davis, H., & Treiber, F. A. (2007). Perceived stress, heart rate, and blood pressure
among adolescents with family members deployed in Operation Iraqi Freedom. Military
Medicine, 172, 40–43. https://doi.org/10.7205/MILMED.172.1.40
Black, K., & Lobo, M. (2008). A conceptual review of family resilience factors. Journal of
Family Nursing, 14, 33–55. https://doi.org/10.1177/1074840707312237
Bliese, P. D., & Castro, C. A. (2000). Role clarity, work overload and organizational support:
Multilevel evidence of the importance of support. Work & Stress, 14, 65–73.
https://doi.org/10.1080/026783700417230
Bowen, G. L., & Martin, J. A. (2011). The resiliency model of role performance for service
members, veterans, and their families: A focus on social connections and individual
assets. Journal of Human Behavior in the Social Environment, 21, 162–178.
https://doi.org/10.1080/10911359.2011.546198
Bradshaw, C. P., Sudhinaraset, M., Mmari, K., Blum, R. W., & Hopkins, J. (2010). School
transitions among military adolescents: A qualitative study of stress and coping. School
Psychology Quarterly, 39, 84–105.
18
Burrell, L. M., Adams, G. A., Durand, D. B., & Castro, C. A. (2006). The impact of military
lifestyle demands on well-being, Army, and family outcomes. Armed Forces & Society,
33, 43–58. https://doi.org/10.1177/0002764206288804
Burrell, L., Durand, D. B., & Fortado, J. (2003). Military community integration and its effect on
well-being and retention. Armed Forces & Society, 30, 7–24.
https://doi.org/10.1177/0095327X0303000101
Cederbaum, J. A., Gilreath, T. D., Benbenishty, R., Astor, R. A., Pineda, D., DePedro, K. T., …
Atuel, H. (2014). Well-being and suicidal ideation of secondary school students from
military families. Journal of Adolescent Health, 54, 672–677.
https://doi.org/10.1016/j.jadohealth.2013.09.006
Chandra, A., Lara-Cinisomo, S., Jaycox, L. H., Tanielian, T., Burns, R. M., Ruder, T., & Han, B.
(2010). Children on the homefront: The experience of children from military families.
Pediatrics, 125, 16–25. https://doi.org/10.1542/peds.2009-1180
Chandra, A., Martin, L. T., Hawkins, S. A., & Richardson, A. (2010). The impact of parental
deployment on child social and emotional functioning: Perspectives of school staff.
Journal of Adolescent Health, 46, 218–223.
https://doi.org/10.1016/j.jadohealth.2009.10.009
Conforte, A. M., Bakalar, J. L., Shank, L. M., Quinlan, J., Stephens, M. B., Sbrocco, T., &
Tanofsky-Kraff, M. (2017). Assessing military community support: Relations among
perceived military community support, child psychosocial adjustment, and parent
psychosocial adjustment. Military Medicine, 182, e1871–e1878.
https://doi.org/10.7205/MILMED-D-17-00016
Cox, M. J., & Paley, B. (2003). Understanding families as systems. Current Directions in
19
Psychological Science, 12, 193–196. https://doi.org/10.1111/1467-8721.01259
Cozza, S. J., Haskins, R., & Lerner, R. M. (2013). Keeping the promise: Maintaining the health
of military and veteran families and children (Future of Children Policy Brief, Fall 2013).
Retrieved from
http://www.princeton.edu/futureofchildren/publications/docs/23_02_PolicyBrief.pdf
Creech, S. K., Hadley, W., & Borsari, B. (2014). The impact of military deployment and
reintegration on children and parenting: A systematic review. Professional Psychology:
Research and Practice, 45, 452–464. https://doi.org/10.1037/a0035055
de Burgh, H. T., White, C. J., Fear, N. T., & Iversen, A. C. (2011). The impact of deployment to
Iraq or Afghanistan on partners and wives of military personnel. International Review of
Psychiatry, 23, 192–200. https://doi.org/10.3109/09540261.2011.560144
Eaton, K. M., Hoge, C. W., Messer, S. C., Whitt, A. A., Cabrera, O. A., McGurk, D., … Castro,
C. A. (2008). Prevalence of mental health problems, treatment need, and barriers to care
among primary care-seeking spouses of military service members involved in Iraq and
Afghanistan deployments. Military Medicine, 173, 1051–1056.
https://doi.org/10.7205/MILMED.173.11.1051
Engel, R. C., Gallagher, L. B., & Lyle, D. S. (2010). Military deployments and children’s
academic achievement: Evidence from Department of Defense Education Activity
schools. Economics of Education Review, 29, 73–82.
https://doi.org/10.1016/j.econedurev.2008.12.003
Finkel, L. B., Kelley, M. L., & Ashby, J. (2003). Geographic mobility, family, and maternal
variables as related to the psychosocial adjustment of military children. Military
Medicine, 168, 1019–1024. https://doi.org/10.1093/milmed/168.12.1019
20
Flake, E. M., Davis, B. E., Johnson, P. L., & Middleton, L. S. (2009). The psychosocial effects
of deployment on military children. Journal of Developmental & Behavioral Pediatrics,
30, 271–278. https://doi.org/10.1097/DBP.0b013e3181aac6e4
Foran, H. M., Eckford, R. D., Sinclair, R. R., & Wright, K. M. (2017). Child mental health
symptoms following parental deployment: The impact of parental posttraumatic stress
disorder symptoms, marital distress, and general aggression. SAGE Open, 7(3), 1–10.
https://doi.org/10.1177/2158244017720484
Gilreath, T. D., Astor, R. A., Cederbaum, J. A., Atuel, H., & Benbenishty, R. (2014). Prevalence
and correlates of victimization and weapon carrying among military- and nonmilitary-
connected youth in Southern California. Preventive Medicine, 60, 21–26.
https://doi.org/10.1016/j.ypmed.2013.12.002
Gilreath, T. D., Cederbaum, J. A., Astor, R. A., Benbenishty, R., Pineda, D., & Atuel, H. (2013).
Substance use among military-connected youth: The California Healthy Kids Survey.
American Journal of Preventive Medicine, 44, 150–153.
https://doi.org/10.1016/j.amepre.2012.09.059
Gilreath, T. D., Wrabel, S. L., Sullivan, K. S., Capp, G. P., Roziner, I., Benbenishty, R., & Astor,
R. A. (2016). Suicidality among military-connected adolescents in California schools.
European Child and Adolescent Psychiatry, 25, 61–66. https://doi.org/10.1007/s00787-
015-0696-2
Gorman, G. H., Eide, M., & Hisle-Gorman, E. (2010). Wartime military deployment and
increased pediatric mental and behavioral health complaints. Pediatrics, 126, 1058–1066.
https://doi.org/10.1542/peds.2009-2856
Green, S., Nurius, P. S., & Lester, P. (2013). Spouse psychological well-being: A keystone to
21
military family health. Journal of Human Behavior in the Social Environment, 23, 753–
768. https://doi.org/10.1080/10911359.2013.795068
Haas, D. M., & Pazdernik, L. A. (2007). Partner deployment and stress in pregnant women. The
Journal of Reproductive Medicine, 52, 901-906.
Harrell, M. C., Lim, N., Castaneda, L. W., & Golinelli, D. (2004). Working around the military:
Challenges to military spouse employment and education. Santa Monica, CA: RAND
National Defense Research Institute. Retrieved from
https://www.rand.org/content/dam/rand/pubs/monographs/2004/RAND_MG196.pdf
Hawkins, S. A., Sullivan, K. S., Schuyler, A. C., Keeling, M., Kintzle, S., Lester, P. B., &
Castro, C. A. (2017). Thinking “big” about research on military families. Military
Behavioral Health, 5, 335–345. https://doi.org/10.1080/21635781.2017.1343696
Hoge, C. W., Castro, C. A., Messer, S. C., McGurk, D., Cotting, D. I., & Koffman, R. L. (2004).
Combat duty in Iraq and Afghanistan, mental health problems, and barriers to care. New
England Journal of Medicine, 351, 13–22. https://doi.org/10.1056/NEJMoa040603
Holmes, A. K., Rauch, P. K., & Cozza, S. J. (2013). When a parent is injured or killed in combat.
Future of Children, 23, 143–162. https://doi.org/10.1353/foc.2013.0017
Huebner, A. J., & Mancini J. A. (2010). Resilience and vulnerability: The deployment
experiences of youth in military families. Blacksburg, VA: Virginia Polytechnic Institute
and State University.
Huebner, A. J., Mancini, J. A., Bowen, G. L., & Orthner, D. K. (2009). Shadowed by war:
Building community capacity to support military families. Family Relations, 58, 216–
228. https://doi.org/10.1111/j.1741-3729.2008.00548.x
Jordan, B. K., Marmar, C. R., Fairbank, J. A., Schlenger, W. E., Kulka, R. A., Hough, R. L., &
22
Weiss, D. S. (1992). Problems in families of male Vietnam veterans with posttraumatic
stress disorder. Journal of Consulting and Clinical Psychology, 60, 916–926.
https://doi.org/10.1037/0022-006X.60.6.916
Lester, P., Peterson, K., Reeves, J., Knauss, L., Glover, D., Mogil, C., … Beardslee, W. (2010).
The long war and parental combat deployment: Effects on military children and at-home
spouses. Journal of the American Academy of Child & Adolescent Psychiatry, 49, 310–
320. https://doi.org/10.1016/j.jaac.2010.01.003
Lincoln, A., Swift, E., & Shorteno-Fraser, M. (2008). Psychological adjustment and treatment of
children and families with parents deployed in military combat. Journal of Clinical
Psychology, 64, 984–992. https://doi.org/10.1002/jclp.20520
Lucier-Greer, M., Arnold, A. L., Mancini, J. A., Ford, J. L., & Bryant, C. M. (2015). Influences
of cumulative risk and protective factors on the adjustment of adolescents in military
families. Family Relations, 64, 363–377. https://doi.org/10.1111/fare.12123
Lucier-Greer, M., O’Neal, C. W., Arnold, A. L., Mancini, J. A., & Wickrama, K. K. A. S.
(2014). Adolescent mental health and academic functioning: Empirical support for
contrasting models of risk and vulnerability. Military Medicine, 179, 1279–1287.
https://doi.org/10.7205/MILMED-D-14-00090
Maholmes, V. (2012). Adjustment of children and youth in military families: Toward
developmental understandings. Child Development Perspectives, 6, 430–435.
https://doi.org/10.1111/j.1750-8606.2012.00256.x
Mansfield, A. J., Kaufman, J. S., Engel, C. C., & Gaynes, B. N. (2011). Deployment and mental
health diagnoses among children of US Army personnel. Archives of Pediatrics &
Adolescent Medicine, 165, 999–1005. https://doi.org/10.1001/archpediatrics.2011.123
23
Mansfield, A. J., Kaufman, J. S., Marshall, S. W., Gaynes, B. N., Morrissey, J. P., & Engel, C. C.
(2010). Deployment and the use of mental health services among U.S. Army wives. New
England Journal of Medicine, 362, 101–109. https://doi.org/10.1056/NEJMoa0900177
Manyika, J., Chui, M., Brown, B., Bughin, J., Dobbs, R., Roxburgh, C., & Byers, A. H. (2011).
Big data: The next frontier for innovation, competition, and productivity. Retrieved from
https://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/big-data-
the-next-frontier-for-innovation
Masten, A. S. (2001). Ordinary magic: Resilience processes in development. American
Psychologist, 56, 227–238. https://doi.org/10.1037/0003-066X.56.3.227
Masten, A. S. (2013). Afterword: What we can learn from military children and families. Future
of Children, 23, 199–212. https://doi.org/10.1353/foc.2013.0012
Masten, A. S., & Cicchetti, D. (2010). Developmental cascades. Development and
Psychopathology, 22, 491–495. https://doi.org/10.1017/S0954579410000222
Masten, A. S., & Narayan, A. J. (2012). Child development in the context of disaster, war, and
terrorism: Pathways of risk and resilience. Annual Review of Psychology, 63, 227–257.
https://doi.org/10.1146/annurev-psych-120710-100356
McGuire, A., Runge, C., Cosgrove, L., Bredhauer, K., Anderson, R., Waller, M., & Nasveld, P.
(2012). Timor-Leste family study: Summary report. Brisbane, Australia: University of
Queensland, Centre for Military and Veterans’ Health.
McNulty, P. A. F. (2010). Adaptability and resiliency of military families during reunification:
Initial results of a longitudinal study. Federal Practitioner, 27(3), 18–27.
McNulty, P. A. F. (2013). Adaptability and resiliency of military families during reunification:
Results of a longitudinal study. Federal Practitioner, 30(8), 14–22.
24
Minuchin, S. (1974). Families and family therapy. Boston, MA: Harvard University Press.
Mmari, K., Roche, K. M., Sudhinaraset, M., & Blum, R. (2009). When a parent goes off to war:
Exploring the issues faced by adolescents and their families. Youth & Society, 40, 455–
475. https://doi.org/10.1177/0044118X08327873
Nurius, P. S., & Macy, R. J. (2008). Heterogeneity among violence-exposed women. Journal of
Interpersonal Violence, 23, 389–415. https://doi.org/10.1177/0886260507312297
O’Neal, C. W., Lucier-Greer, M., Mancini, J. A., Ferraro, A. J., & Ross, D. B. (2016). Family
relational health, psychological resources, and health behaviors: A dyadic study of
military couples. Military Medicine, 181, 152–160. https://doi.org/10.7205/MILMED-D-
14-00740
Orthner, D. K., & Rose, R. (2009). Work separation demands and spouse psychological well-
being. Family Relations, 58, 392–403. https://doi.org/10.1111/j.1741-3729.2009.00561.x
Oshri, A., Lucier-Greer, M., O’Neal, C. W., Arnold, A. L., Mancini, J. A., & Ford, J. L. (2015).
Adverse childhood experiences, family functioning, and resilience in military families: A
pattern-based approach. Family Relations, 64, 44–63. https://doi.org/10.1111/fare.12108
Paley, B., Lester, P., & Mogil, C. (2013). Family systems and ecological perspectives on the
impact of deployment on military families. Clinical Child and Family Psychology
Review, 16, 245–265. https://doi.org/10.1007/s10567-013-0138-y
Palmer, C. (2008). A theory of risk and resilience factors in military families. Military
Psychology, 20, 205–217. https://doi.org/10.1080/08995600802118858
Paris, R., DeVoe, E. R., Ross, A. M., & Acker, M. L. (2010). When a parent goes to war: Effects
of parental deployment on very young children and implications for intervention.
American Journal of Orthopsychiatry, 80, 610–618. https://doi.org/10.1111/j.1939-
25
0025.2010.01066.x
Park, N. (2011). Military children and families: Strengths and challenges during peace and war.
American Psychologist, 66, 65–72. https://doi.org/10.1037/a0021249
Peterson, C., Park, N., & Castro, C. A. (2011). The Global Assessment Tool. American
Psychologist, 66, 10–18. https://doi.org/10.1037/a0021658
Pincus, S. L., House, R., Christenson, J., & Adler, L. E. (2001). The emotional cycle of
deployment: A military family perspective. U.S. Army Medical Department Journal,
4(4), 9–18. Retrieved from https://www.military.com/spouse/military-
deployment/dealing-with-deployment/emotional-cycle-of-deployment-military-
family.html
Renshaw, K. D., Allen, E. S., Rhoades, G. K., Blais, R. K., Markman, H. J., & Stanley, S. M.
(2011). Distress in spouses of service members with symptoms of combat-related PTSD:
Secondary traumatic stress or general psychological distress? Journal of Family
Psychology, 25, 461–469. https://doi.org/10.1037/a0023994
Renshaw, K. D., Rodebaugh, T. L., & Rodrigues, C. S. (2010). Psychological and marital
distress in spouses of Vietnam veterans: Importance of spouses’ perceptions. Journal of
Anxiety Disorders, 24, 743–750. https://doi.org/10.1016/j.janxdis.2010.05.007
Riggs, S. A., & Riggs, D. S. (2011). Risk and resilience in military families experiencing
deployment: The role of the family attachment network. Journal of Family Psychology,
25, 675–687. https://doi.org/10.1037/a0025286
Rodriguez, A. J., & Margolin, G. (2015). Parental incarceration, transnational migration, and
military deployment: Family process mechanisms of youth adjustment to temporary
parent absence. Clinical Child and Family Psychology Review, 18, 24–49.
26
https://doi.org/10.1007/s10567-014-0176-0
Rosato, N. S., & Baer, J. C. (2012). Latent class analysis: A method for capturing heterogeneity.
Social Work Research, 36, 61–69. https://doi.org/10.1093/swr/svs006
Saltzman, W. R., Lester, P., Beardslee, W. R., Layne, C. M., Woodward, K., & Nash, W. P.
(2011). Mechanisms of risk and resilience in military families: Theoretical and empirical
basis of a family-focused resilience enhancement program. Clinical Child and Family
Psychology Review, 14, 213–230. https://doi.org/10.1007/s10567-011-0096-1
Seligman, M. E. P., & Fowler, R. D. (2011). Comprehensive soldier fitness and the future of
psychology. American Psychologist, 66, 82–86. https://doi.org/10.1037/a0021898
Solomon, Z., Debby-Aharon, S., Zerach, G., & Horesh, D. (2011). Marital adjustment, parental
functioning, and emotional sharing in war veterans. Journal of Family Issues, 32, 127–
147. https://doi.org/10.1177/0192513X10379203
Sullivan, K., Capp, G., Gilreath, T. D., Benbenishty, R., Roziner, I., & Astor, R. A. (2015).
Substance abuse and other adverse outcomes for military-connected youth in California:
Results from a large-scale normative population survey. JAMA Pediatrics, 169, 922–928.
https://doi.org/10.1001/jamapediatrics.2015.1413
U.S. Department of Defense. (2011). Protection of human subjects: Adherence to ethical
standards in DoD-supported research (DOD Instruction 3216.02). Washington, DC:
Author.
U.S. Department of Defense. (2015). Military Family Life Project: Active Duty Spouse Study.
Retrieved from http://download.militaryonesource.mil/12038/MOS/Reports/MFLP-
Longitudinal-Analyses-Report.pdf
U.S. Department of Defense. (2017). 2016 demographics: Profile of the military community.
27
Retrieved from http://download.militaryonesource.mil/12038/MOS/Reports/2016-
Demographics-Report.pdf
Vie, L. L., Scheier, L. M., Lester, P. B., & Seligman, M. E. (2016). Initial validation of the US
Army Global Assessment Tool. Military Psychology, 28(6), 468-487.
Wadsworth, S. M. (2013). Understanding and supporting the resilience of a new generation of
combat-exposed military families and their children. Clinical Child and Family
Psychology Review, 16, 415–420. https://doi.org/10.1007/s10567-013-0155-x
Walsh, F. (1996). The concept of family resilience: Crisis and challenge. Family Process, 35,
261–281. https://doi.org/10.1111/j.1545-5300.1996.00261.x
Walsh, F. (2003). Family resilience: A framework for clinical practice. Family Process, 42, 1–
18. https://doi.org/10.1111/j.1545-5300.2003.00001.x
Weins, T. W., & Boss, P. (2006). Maintaining family resiliency before, during, and after military
separation. In C. A. Castro, A. B. Adler, & T. W. Britt (Eds.), Military life: The
psychology of serving in peace and combat: Vol. 3. The military family (pp. 13–38).
Westport, CT: Praeger Security International.
Wilson, S. R., Marini, C. M., Franks, M. M., Whiteman, S. D., Topp, D., & Wadsworth, S. M.
(2018). Communication and connection during deployment: A daily-diary study from the
perspective of at-home partners. Journal of Family Psychology, 32, 42–48.
https://doi.org/10.1037/fam0000333

28
Figure 1.1. Integrated Conceptual Model

29
Chapter 2 (Study 1): Preliminary Psychometrics and Potential Big Data Uses of the U.S.
Army Family Global Assessment Tool
Abstract
The purpose of this study is to explore the psychometric properties of the U.S. Army’s Family
Global Assessment Tool, which assesses the psychosocial fitness of Army families. With data
from 1,692 Army spouses, we examined the structure, reliability, and validity of the tool using
exploratory and confirmatory factor analysis and two validity studies. Twenty-nine items and
seven factors were retained following exploratory factor analysis. This model provided good fit
and scales demonstrated strong internal consistency. Bivariate correlations and results from a
theoretically driven model provided evidence of validity. Findings support the usefulness of the
instrument for measuring psychosocial fitness.

30
Introduction
The U.S. Army’s Comprehensive Soldier and Family Fitness (CSF2) program was
created in response to mounting evidence suggesting that U.S. involvement in two protracted
overseas conflicts was taking a toll on the psychological health of our soldiers and their families.
This program was tasked with evaluating the physical, social, emotional, family, and spiritual
health of Army families and implementing universal prevention efforts to bolster soldier and
family resilience (Peterson, Park, & Castro, 2011). Critical to this effort was a valid and reliable
instrument that could efficiently measure these domains as a means to both guide intervention
selection and measure the effectiveness of prevention strategies (Cornum, Matthews, &
Seligman, 2011). Drawing when possible on previously validated measures, the Soldier and
Family Global Assessment Tools (GATs) were created by an expert committee, with input from
the military, academia, and the private sector, to meet this need (Peterson et al., 2011). The
Soldier GAT has recently undergone psychometric validation; the results of this effort supported
its ongoing use as an assessment tool (Vie, Scheier, Lester, & Seligman, 2016). The purpose of
the current study was to present preliminary findings regarding the psychometrics of the Family
GAT.
Theoretical Foundations of CSF2 and the Family GAT
The CSF2 program and the GATs are grounded in the tenets of positive psychology,
which seeks to identify and promote characteristics that enable individuals and communities to
thrive (Seligman & Csikszentmihalyi, 2000). As such, the Soldier and Family GATs assess
positive emotions, personal attributes, and resilient functioning, which contribute to a “full life”
(Peterson, Park, & Seligman, 2005). On the Family GAT, these positive attributes include
character strengths, positive affect, optimism, positive coping, and healthy family and
31
relationship functioning. Additionally, several scales measure aspects of negative explanatory
style, which has its roots in learned helplessness theory (Abramson, Seligman, & Teasdale,
1978). On the Family GAT, these include attributes such as negative affect, catastrophic
thinking, and loneliness.
The Family GAT survey, which is already in use as a key component of the CSF2
program, was designed as a self-development tool and not initially intended for research
purposes (Lester, McBride, & Cornum, 2013). As such, there is little existing information
regarding the validity, reliability, and underlying structure of this instrument, which limits its
usefulness for research or program evaluation. However, because collecting original data from
military families presents logistical and ethical challenges (Castro & Sullivan, 2018), the
richness of the information we have about this population lags far behind the generation of new
knowledge about soldiers (Park, 2011). Thus, efforts to evaluate the psychometric properties of
the Family GAT are critical. Once validated, this survey would offer a unique new avenue
through which to explore the functioning of military family systems, particularly during this
period of increased operational tempo when evidence suggests that some military families may
be struggling (Card et al., 2011; Lester & Flake, 2013).
Big Data Uses of the Family GAT
A particularly useful aspect of the GAT surveys is the capacity to connect these measures
of resilience and psychosocial functioning with objective information regarding soldiers’ service
gathered by the Department of Defense (DOD). These data are stored and accessible in the
Army’s Person-Event Data Environment (PDE), which is a cloud-based, virtual environment that
facilitates data access and linkage across the Army and DOD (Vie et al., 2015; Vie, Griffith,
Scheier, Lester, & Seligman, 2013). The term “big data” has been used to describe the
32
“collection and integration of datasets from multiple disparate sources, covering various unique
topics, to provide a more rich and robust picture of individuals, groups, and systems” (Hawkins
et al., 2017, p. 2). Using the PDE, Family GAT information can be linked to more than 350
manpower, service, personnel, financial, behavioral health, and medical datasets, which can
provide objective information regarding exposure to risk factors including family separations,
reunifications, and relocations.
The focus of the present study was to explore the underlying structure of the Family GAT
and establish preliminary reliability and validity. Ultimately, the goal was to provide evidence of
acceptable psychometric properties such that Family GAT survey data can be reliably included
in big data efforts using the Army’s PDE. Future studies involving these linked datasets have the
potential to provide a rich, nuanced picture of military families and substantially improve the
quality and specificity of research conducted with this population.
Methods
Participants and Procedures
The Family GAT is completed on a voluntary basis by Army spouses, who access the
survey by visiting a hosting website. CSF2, and the Family GAT specifically, are extensively
publicized and a key component of the Army’s Ready and Resilient Campaign. Spouses learn
about the program through Army publicity, their spouse, or their family readiness group. After
the survey, spouses are given feedback about their responses and links to online learning
modules to address identified challenges. Upon completion of the Family GAT, respondents can
give consent for their data to be used for research. Only data from spouses who provided consent
are used in these analyses. Secondary analyses of GAT data were approved by the Army
33
Research, Development and Engineering Center’s Institutional Review Board and the
institutional review board at the [blinded for review].
Sample
To be included in the current sample, participants had to take the Family GAT between
October 2013, when respondents could opt in to research, and December 31, 2016, the most
recent date of available data. Although spouses can take the Family GAT as many times as they
choose, only their first GAT completions were used in the current analyses, which resulted in
2,777 unique Family GAT participants. Additionally, to link survey responses to DOD
administrative and health data, only spouses who could be associated with an Army sponsor
were included in the dataset. This resulted in the loss of 216 participants, reducing the sample to
2,561. Family GAT takers who had a sponsor ID number and could be linked to a spouse with a
sponsor ID were determined to be in a dual military relationship, wherein both spouses were
military service members. These 238 participants were retained in the dataset. Finally, Family
GAT takers who had a sponsor ID number but could not be linked to a spouse with a sponsor ID
were determined to be service members who took the Family GAT in addition to the Soldier
GAT. Because we could not be certain whether these respondents had families or potentially
took the family survey in error, these observations were also eliminated from the dataset,
resulting in a final analytic sample of 1,692 Army spouses.
Sample description. In this sample, 95% of participants were female and were an
average of 36 years old. Slightly more participants were married to enlisted soldiers (55%)
compared to officers (45%). On average, families had 1.76 children in the home, with 34% of
families reporting that their oldest child was school aged and 32% reporting their oldest child
34
was an adolescent. These data did not contain information on the race and ethnicity of spouses,
but 72% of soldiers in these families were White, 11% were Black, and 10% were Hispanic.
Family GAT Measures
The Family GAT survey includes 16 a priori scales, many of which were drawn from or
based on previously validated measures. However, the scales in their current form had not
undergone examination of their structure and psychometric properties. Presented here are the 16
scales, sample items, and information concerning the origin of the items for those based on
previous scales. All scales are coded such that higher scores represent positive functioning.
Character strengths. This scale includes 18 of 240 items from the Values in Action
Inventory of Strengths (Peterson & Seligman, 2004). The scale asks participants to consider how
often they have showed or used the listed qualities in the preceding 4 weeks on a 10-point Likert
scale from never to always. Items include: “creativity – coming up with new ideas” and “self-
control or self-regulation.”
Depression. This scale features five items based on the Patient Health Questionnaire–9
(Kroenke, Spitzer, & Williams, 2001), beginning with the prompt: “In the past four weeks, how
often have you been bothered by any of the following problems?” Response options are on a 5-
point Likert scale from not at all to every day. Items include: “feeling tired or having little
energy” and “little interest or pleasure in doing things.”
Positive and negative affect. This scale features five items drawn from the Positive and
Negative Affect Schedule (Watson, Clark, & Tellegen, 1988). The scale presents “a number of
words that describe different feelings” and asks participants to rate how often they have felt these
emotions during the past 4 weeks on a 5-point Likert scale from never to most of the time. Items
include: “joyful/happy,” “sad,” and “peaceful/calm.”
35
Problem-focused coping. This scale features three items adapted from the Brief COPE
(Carver, 1997; Carver, Scheier, & Weintraub, 1989) and asks participants how well the presented
statements describe them on a 5-point Likert scale from not like me at all to very much like me.
Items include: “For things I cannot change, I accept them and move on” and “When bad things
happen, I try to see the positive sides.”
Catastrophic thinking. This scale uses three items from the Attributional Styles
Questionnaire (Peterson et al., 1982). The prompt asks participants how well the presented
statements describe them on a 5-point Likert scale from not like me at all to very much like me.
Items include: “When bad things happen to me, I expect more bad things to happen” and “I have
no control over things that happen to me.”
Optimism. This scale features three items adapted from the Revised Life Orientation
Test (Scheier & Carver, 1985; Scheier, Carver, & Bridges, 1994). The prompt and response scale
are the same as the catastrophic thinking scale. Items include: “Overall, I expect more good
things to happen to me than bad” and “I am always optimistic about my future.”
Problem management. This scale has four items that ask participants to rate themselves
in terms of “handling the following areas” of their lives on a 5-point Likert scale from poor to
excellent, with an option to select not applicable. Items include: “Handling parenting tasks and
discipline of my children” and “Managing household and chores.”
Loneliness. This scale uses three items adapted from the UCLA Loneliness Scale
(Russell, Peplau, & Ferguson, 1978). The prompt asks participants to be “as honest as possible,”
and response options are on a 5-point Likert scale from never to most of the time. Items include:
“How often do you feel left out?” and “How often do you feel part of a group?”
36
Social support. This 5-item scale asks participants to rate “how well these statements
describe you and your life.” Response options are on a 5-point Likert scale from strongly
disagree to strongly agree. Items include: “There is someone I can turn to for advice on how to
deal with a personal or family problem” and “If I was stranded 10 miles from home, there is
someone I could call who could come and get me.”
Social connections. This scale features five items and asks participants to “think about
your relationship with people in your community and neighborhood (other than family
members).” Participants rate how often they have experienced each item in the past 4 weeks on a
5-point Likert scale from never to most of the time. Items include: “I participated in community
events, activities or meetings” and “I felt close to others in my community.”
Family satisfaction. This scale consists of two items assessing how participants have felt
about their relationship or family during the past 4 weeks. Response options are on a 5-point
Likert scale from not at all satisfied to extremely satisfied, with a not applicable option. The
items are: “How satisfied are you with your marriage/relationship?” and “How satisfied are you
with your family?”
Relationship functioning. This scale consists of eight items that ask participants to
describe their feelings about their partner and their relationship. Response options are on a 5-
point Likert scale from strongly disagree to strongly agree. Items include: “My partner is
emotionally supportive of me” and “My partner and I clearly communicate our expectations for
each other.”
Child functioning. This scale consists of five items that begin with the prompt: “If you
have children, how have they been doing during the past four weeks?” Response options range
37
from poor to excellent, with a not applicable option. For example, items query how children are
doing “socially,” “psychologically” and “at home.”
Support for military. Two items ask participants how strongly they agree or disagree
with the presented statements. Response options are on a 5-point Likert scale, with a not
applicable option. Items include: “I support my partner’s decision to serve in the military” and
“The military meets my family’s needs.”
Family cohesion. Three items based on the McMaster Family Assessment Device
(Epstein, Baldwin, & Bishop, 1983) are used to assess family functioning. The items ask
participants to describe their “family as a whole” and offer response options on a 5-point Likert
scale from strongly disagree to strongly agree. Items include: “My family expresses tenderness”
and “My family confides in each other.”
Meaning. This scale features five items adapted from the Brief Multidimensional
Measure of Religiousness/Spirituality (Fetzer Institute, 2003). The prompt asks participants to
rate how well the statements describe “how you actually live your life.” Response options are on
a 5-point Likert scale from not like me at all to very much like me. Items include: “I have a
purpose in life” and “I believe the things that I do are worthwhile.”
Analytic Plan
To examine the underlying structure of the Family GAT instrument, this study used
exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). To establish
preliminary reliability, the internal consistency of the resulting scales was examined. As this was
a secondary analysis of previously collected data, we were unable to pursue traditional strategies
to establish validity of survey scales. Instead, preliminary validity was evaluated with two
studies using the pattern of bivariate correlations and simple regression to explore the GAT
38
scales’ consistency with theory and previous empirical evidence. All analyses were conducted in
the Army’s PDE using SPSS version 21 and Mplus version 7 (Muthén & Muthén, 2012).
Results
Family GAT Structure
Prior to conducting EFA, five items that referred to the functioning of children were
dropped from analyses. Spouses could select a not applicable option for these items if they did
not have children, which was coded as missing in the final dataset. Missingness on these items
ranged from 444 missing data points for an item assessing handling of parenting tasks and
discipline to 641 for an item assessing how children were functioning at school. Because the goal
of these analyses was to validate the Family GAT survey for use with all Army spouses,
regardless of whether they had children, excluding these items was appropriate and allowed for
the retention of the largest possible sample size.
To examine the underlying structure of the Family GAT, the final sample of 1,692 Army
spouses was randomly split such that half of the observations were used in the EFA phase and
half were used in the CFA phase (DeCoster, 1998). Although adequate sample size to conduct
EFA depends on many factors (Schmitt, 2011), a 10:1 ratio of participants to items is commonly
accepted in the literature (Costello & Osborne, 2005). Thus, a ratio of 807 participants (in one
randomly generated half of the full dataset) to 75 items was deemed adequate. The factorability
of these 75 items was assessed using the Kaiser-Meyer-Olkin measure of sampling adequacy,
which was .957, well above the threshold to proceed, and Bartlett’s test of sphericity, which was
significant (χ
2
[2,775] = 32,942.61, p < .001; Beavers et al., 2013).
EFA was undertaken using the principle axis factor method of factor extraction. This
method was chosen because it is preferable when exploring latent structure (Conway & Huffcutt,
39
2003). Promax rotation was used, because oblique rotations account for expected correlations
between components (Schmitt, 2011). Initially, 15 factors with eigenvalues greater than 1 were
retained. The full pattern matrix following promax rotation is displayed in Table 2.1. To achieve
data reduction, only factor loadings of .63 or higher, considered very good (Tabachnick & Fidell,
2013), were retained in the exploratory stage. Factors with eigenvalues greater than 1 and at least
three indicators with loadings of .63 or higher were retained (Beavers et al., 2013; Costello &
Osborne, 2005). Using these decision rules, eight factors were eliminated because they did not
have at least three items load at .63 or higher. Of the original 75 items, 29 were retained because
they had factor loadings greater than .63 and loaded onto factors with at least two additional
items. Listwise deletion was used to address missing data in SPSS. The 29 items and seven
factors retained following EFA are displayed in Table 2.2.
Results from the exploratory phase were subjected to CFA using the second randomly
generated half of the dataset. A well-fitting model was expected to have a root mean square error
of approximation (RMSEA) less than .05, comparative fit index (CFI) greater than .95, and
standardized root mean square residual (SRMR) less than .08 (Hu & Bentler, 1999). Although
reported, the χ
2
statistic was not used to assess model fit because this measure is highly sensitive
to sample size. For CFA models, the maximum likelihood estimator was used, and missing data
were handled using full information maximum likelihood procedures available in Mplus. CFA
results indicated that the model provided a good fit to the data: χ
2
(831, 361) = 1,067.85, p < .001,
CFI = .955, RMSEA = .049 (90% CI = .045, .052), SRMR = .043. At the CFA stage, factor
loadings, which are displayed in Table 2.2, ranged from .60 to .90.
40
Reliability
Although this study could not establish test–retest reliability because these analyses were
conducted after data had already been collected, internal consistency was established by
calculating Cronbach’s alphas for final scales. All α scores are presented in Table 2.2. These
values ranged from .80 for the positive coping scale to .91 for the meaning scale, suggesting that
the internal consistency for all retained scales was acceptable (Tavakol & Dennick, 2011).
Validity
As this was a secondary analysis of previously collected data, traditional methods of
establishing validity through comparison to previously validated scales were not possible.
Instead, two studies were used to establish the preliminary validity of final scales. Before
proceeding with the first of these studies, composite scores were created for each factor based on
the mean of the items that loaded onto that factor. Means and standard deviations for these
summary variables are displayed in Table 2.2. Although all items on the Family GAT have been
scored such that higher numbers represent positive functioning, for the purposes of establishing
validity and ease of interpretation, this study used a version of the depressive symptoms variable
that was not reverse scored, such that higher scores indicated higher levels of depressive
symptoms.
Study 1. This study followed the approach taken by the Soldier GAT validation team
(Vie et al., 2016). Using this method, preliminary convergent and discriminant validity were
established by examining the pattern of bivariate correlations between scales and comparing this
pattern to expected relationships. Regarding preliminary discriminant validity, all six positive
functioning variables were expected to be significantly inversely related to depressive symptoms
scores. Regarding preliminary convergent validity, all six positive functioning scales were
41
expected to be significantly positively related to each other. Further, the literature has suggested
a relationship between social connections and social support (Heaney & Israel, 2008), between
relationship functioning and family functioning (Katz & Woodin, 2002), and between meaning
making and positive coping skills (Folkman & Moskowitz, 2000). Thus, the magnitude of these
correlations were expected to be higher than other significant relationships.
Bivariate correlations are displayed in Table 2.3. As expected, depressive symptom
scores were significantly negatively correlated with all six positive functioning variables.
Correlations ranged between -.32 (p < .01) for the relationship with family functioning to -.44 (p
< .01) for the relationship with social support. Additionally, all six positive functioning scores
were significantly positively associated (mean r = .44, p < .01). Finally, this study examined
several specific relationships expected to be stronger, based on previous findings. As expected,
the three strongest relationships were between social connections and social support (r = .55, p <
.01), between relationship functioning and family cohesion (r = .52, p < .01), and between
meaning making and positive coping (r = .61, p < .01). The magnitude of these correlations
suggests that these constructs are similar, as expected, but nevertheless conceptually distinct.
Study 2. In Study 2, several scales were selected to test a simple theory driven model.
Family stress theory suggests that stressors, like poor family functioning, are moderated by the
resources (including social support) that families mobilize, which will determine whether
families cope successfully or experience adverse outcomes, including depression (Boss, 2002).
As a measure of the theoretical consistency of the Family GAT scales, this study used multiple
linear regression to test the relationship between family cohesion and spouse depressive
symptoms, moderated by social support.
42
Multivariate regression results are presented in Table 2.4. To avoid issues of
multicollinearity, mean centering was used when creating the interaction term. This model was
significant and provided a good fit to the data: F(3, 1,687) = 162.80, p < .001. Greater family
cohesion (β = -.138, p < .001) and higher levels of social support (β = -.364, p < .001) were
associated with lower levels of depressive symptomatology. Further, the interaction term was
also significant (β = .092, p < .001) indicating that social support moderated the relationship
between family cohesion and depressive symptoms. This interaction is depicted in Figure 2.1,
which suggests that at lower levels of social support, less family cohesion is more strongly
associated with depressive symptoms.
Discussion
Findings from this study provide preliminary evidence to support the reliability, validity,
and multidimensional factor structure of the Family GAT. Following exploratory and
confirmatory factor analysis, 29 items, which loaded highly onto seven factors, were retained.
These factors represent key facets of military family functioning, including social support and
connectedness, coping and meaning making, spousal mental health, and strong and cohesive
marital and family relationships. In addition to the factor structure, preliminary evidence
regarding the reliability and validity of the Family GAT is encouraging. Each of the remaining
scales demonstrated good internal consistency. Further, initial efforts at establishing validity
suggest that the Family GAT scales performed as expected based on theory and previous
empirical evidence.
Although seven cohesive scales emerged from the factor analysis process, many items on
the Family GAT did not exceed the threshold for scientific rigor and are not recommended to be
retained for ongoing research involving military families. Among these were items assessing
43
character, positive and negative affect, and support for the military. Although these items and
scales are limited in their scientific usefulness, they nevertheless represent important constructs
and may continue to contribute to the self-evaluation and development goals of the overarching
CSF2 program.
Previous research suggests that the retained scales from the Family GAT measure critical
aspects of healthy functioning among military families. For example, social support and social
connections have been found to positively influence military family adaptation (Bowen,
Mancini, Martin, Ware, & Nelson, 2003). Further, military parents’ perception of the social
support their families receive has also been associated with positive child psychosocial
functioning (Flake, Davis, Johnson, & Middleton, 2009). Positive coping has been linked with
healthier family functioning and successful management of military separations (Weins & Boss,
2006). Meaning making has been hypothesized as a crucial factor in adaptation to stressors
among military families and children. This element is a critical component of the Families
OverComing Under Stress intervention, a resilience training program for military families
experiencing stressors like deployments (Saltzman et al., 2011).
Extant literature also suggested that aspects of spousal functioning and family
relationships measured by the Family GAT are integral to understanding how military families
are faring in the current operational climate. These associations may not be unidirectional, but
rather may tap into the complex patterns of relationships that reverberate through family systems.
For example, spousal mental health may be adversely affected by exposure to their partners’
combat deployments and the potential consequences of these deployments (de Burgh, White,
Fear, & Iversen, 2011). However, research has also demonstrated the critical role that spousal
mental health plays for the well-being of military-connected youth and family functioning
44
(Green, Nurius, & Lester, 2013). Similarly, deployment may increase the risk of unhealthy
marital relationships (de Burgh et al., 2011), which in turn might affect outcomes for children
and families (Paley, Lester, & Mogil, 2013). Riggs and Riggs (2011) theorized that a healthy
marital relationship contributes to a secure attachment system for the military spouse, which has
a cascade of positive effects on family functioning and child well-being. Further, they suggested
that family cohesion and healthy but permeable family boundaries help families to cope with the
deployment cycle (Riggs & Riggs, 2011). Evidence has suggested that family cohesion may also
be associated with the psychosocial adjustment of military-connected youth (Finkel, Kelley, &
Ashby, 2003).
Limitations
Although the aforementioned research suggests that the scales of the Family GAT survey
may provide information regarding important elements of military family functioning, several
limitations should be noted. Because the analyses presented here were conducted on data
previously collected as a part of the Army’s ongoing CSF2 program, many of these limitations
result from the challenges inherent in secondary data analysis. Some aspects of traditional
psychometric validation were not possible because these data were not available. This study
could not evaluate the test–retest reliability of the Family GAT because data were not collected
at two consistent points to evaluate the uniformity of responses across time. Further, these
analyses did not incorporate previously validated measures to examine the convergent or
discriminant validity of Family GAT scales. Additionally, the Family GAT survey is a self-
report measure and all analyses were conducted using cross-sectional data. Use of multiple
informants and longitudinal data would strengthen validity evidence. Also, the Family GAT is
completed on a voluntary basis and only a small proportion of Army spouses were represented in
45
the dataset, which may introduce bias. However, the demographic profile of GAT completers
was relatively similar to the Army overall, which suggests that this bias should be minimal.
Finally, insufficient sample size prevented examination of measurement invariance across
relevant demographic categories, including gender, race and ethnicity, and service members’
military rank. The Family GAT sample size is expected to increase as Army spouses continue to
take the survey, which may make examining measurement invariance possible in the future.
Future Directions
The overarching goal of these analyses was to provide preliminary evidence of acceptable
psychometric properties such that Family GAT scales could be used alongside DOD archival
datasets to provide a more complete and contextualized picture of military family functioning.
Although these efforts are subject to the aforementioned limitations, the true strength of the
Family GAT data resides in the opportunity this instrument offers to combine psychosocial
indicators with objective information about military-related stressors. Because the Family GAT
was designed to capture elements of positive psychology, most of the scales measure spousal and
family strengths and may offer insight into protective factors that have the potential to counteract
the risks to healthy family functioning inherent in military life (Burrell, Durand, & Fortado,
2006). Meanwhile, the archival data available in the PDE provide a wealth of information about
actual risk factors that families have experienced, including their deployment history and
experiences of relocation. Further, access to soldier and dependent health records provides
concrete health and mental health outcomes for this population, increasing the public health
relevance of these efforts.
Integrating these disparate sources of information using big data methods offers many
opportunities for future research. Ongoing efforts to explore the validity of the Family GAT
46
could be enhanced by examining whether GAT scales can predict health and mental health
outcomes with a reasonable level of accuracy. Further, combining protective factors from the
Family GAT with information about risk factors like deployment history may help us understand
why some military families appear to struggle with the stressors they experience whereas the
majority seem able to cope relatively successfully. Additionally, identifying variables that
mediate or moderate the relationships demonstrated in the literature, particularly between
deployment experience and adverse individual and family outcomes, may shed further light on
meaningful targets for intervention with this population.

47
References
Abramson, L. Y., Seligman, M. E., & Teasdale, J. D. (1978). Learned helplessness in humans:
Critique and reformulation. Journal of Abnormal Psychology, 87, 49–74.
https://doi.org/10.1037/0021-843X.87.1.49
Beavers, A. S., Lounsbury, J. W., Richards, J. K., Huck, S. W., Skolits, G. J., & Esquivel, S. L.
(2013). Practical considerations for using exploratory factor analysis in educational
research. Practical Assessment, Research & Evaluation, 18, 6. Retrieved from
http://www.pareonline.net/getvn.asp?v=18&n=6
Boss, P. (2002). Family stress management: A contextual approach. Thousand Oaks, CA: Sage.
Bowen, G. L., Mancini, J. A., Martin, J. A., Ware, W. B., & Nelson, J. P. (2003). Promoting the
adaptation of military families: An empirical test of a community practice model. Family
Relations, 52, 33–44. https://doi.org/10.1111/j.1741-3729.2003.00033.x
Burrell, L., Durand, D. B., & Fortado, J. (2003). Military community integration and its effect on
well-being and retention. Armed Forces & Society, 30, 7–24.
https://doi.org/10.1177/0095327X0303000101
Card, N. A., Bosch, L., Casper, D. M., Wiggs, C. B., Hawkins, S. A., Schlomer, G. L., &
Borden, L. M. (2011). A meta-analytic review of internalizing, externalizing, and
academic adjustment among children of deployed military service members. Journal of
Family Psychology, 25, 508–520. https://doi.org/10.1037/a0024395
Carver, C. S. (1997). You want to measure coping but your protocol’s too long: Consider the
brief cope. International Journal of Behavioral Medicine, 4, 92–100.
https://doi.org/10.1207/s15327558ijbm0401_6
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A
48
theoretically based approach. Journal of Personality and Social Psychology, 56, 267–
283. https://doi.org/10.1037/0022-3514.56.2.267
Castro, C. A., & Sullivan, K. S. (2018). Military families research: Department of Defense
funding and the elements of a fundable proposal. In L. Hughes-Kirchubel, S. M.
Wadsworth, & D. S. Riggs (Eds.), A battle plan for supporting military families: Lessons
for leaders of tomorrow (pp. 323–332). New York, NY: Springer.
Conway, J. M., & Huffcutt, A. I. (2003). A review and evaluation of exploratory factor analysis
practices in organizational research. Organizational Research Methods, 6, 147–168.
https://doi.org/10.1177/1094428103251541
Cornum, R., Matthews, M. D., & Seligman, M. E. P. (2011). Comprehensive soldier fitness:
Building resilience in a challenging institutional context. American Psychologist, 66, 4–9.
https://doi.org/10.1037/a0021420
Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four
recommendations for getting the most from your analysis. Practice Assessment, Research
& Evaluation, 10, 7. Retrieved from http://www.pareonline.net/pdf/v10n7.pdf
de Burgh, H. T., White, C. J., Fear, N. T., & Iversen, A. C. (2011). The impact of deployment to
Iraq or Afghanistan on partners and wives of military personnel. International Review of
Psychiatry, 23, 192–200. https://doi.org/10.3109/09540261.2011.560144
DeCoster, J. (1998). Overview of factor analysis. Retrieved from http://stat-help.com/factor.pdf
Epstein, N. B., Baldwin, L. M., & Bishop, D. S. (1983). The McMaster Family Assessment
Device. Journal of Marital and Family Therapy, 9, 171–180.
https://doi.org/10.1111/j.1752-0606.1983.tb01497.x
Fetzer Institute. (2003). Multidimensional measurement of religiousness/spirituality for use in
49
health research: A report of the Fetzer Institute/National Institute on Aging Working
Group. Kalamazoo, MI: Author.
Finkel, L. B., Kelley, M. L., & Ashby, J. (2003). Geographic mobility, family, and maternal
variables as related to the psychosocial adjustment of military children. Military
Medicine, 168, 1019–1024. https://doi.org/10.1093/milmed/168.12.1019
Flake, E. M., Davis, B. E., Johnson, P. L., & Middleton, L. S. (2009). The psychosocial effects
of deployment on military children. Journal of Developmental & Behavioral Pediatrics,
30, 271–278. https://doi.org/10.1097/DBP.0b013e3181aac6e4
Folkman, S., & Moskowitz, J. T. (2000). Positive affect and the other side of coping. American
Psychologist, 55, 647–654. https://doi.org/10.1037//0003-066X.55.6.647
Green, S., Nurius, P. S., & Lester, P. (2013). Spouse psychological well-being: A keystone to
military family health. Journal of Human Behavior in the Social Environment, 23, 753–
768. https://doi.org/10.1080/10911359.2013.795068
Hawkins, S. A., Sullivan, K. S., Schuyler, A. C., Keeling, M., Kintzle, S., Lester, P. B., &
Castro, C. A. (2017). Thinking “big” about research on military families. Military
Behavioral Health, 5, 335–345. https://doi.org/10.1080/21635781.2017.1343696
Heaney, C. A., & Israel, B. A. (2008). Social networks and social support. In K. Glanz, B. K.
Rimer, & K. Viswanath (Eds.), Health behavior and health education: Theory, research,
and practice (pp. 189–210). San Francisco, CA: Jossey-Bass.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55.
https://doi.org/10.1080/10705519909540118
Katz, L. F., & Woodin, E. M. (2002). Hostility, hostile detachment, and conflict engagement in
50
marriages: Effects on child and family functioning. Child Development, 73, 636–652.
https://doi.org/10.1111/1467-8624.00428
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, 606–613.
https://doi.org/10.1046/j.1525-1497.2001.016009606.x
Lester, P., & Flake, E. (2013). How wartime military service affects children and families.
Future of Children, 23, 121–141. https://doi.org/10.1353/foc.2013.0015
Lester, P. B., McBride, S., & Cornum, R. L. (2013). Comprehensive Soldier Fitness:
Underscoring the facts, dismantling the fiction. In R. R. Sinclair & T. W. Britt (Eds.),
Building psychological resilience in military personnel: Theory and practice (pp. 267–
309). Washington, DC: APA Press.
Muthén, L. K., & Muthén, B. O. (2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthén
& Muthén.
Paley, B., Lester, P., & Mogil, C. (2013). Family systems and ecological perspectives on the
impact of deployment on military families. Clinical Child and Family Psychology
Review, 16, 245–265. https://doi.org/10.1007/s10567-013-0138-y
Park, N. (2011). Military children and families: Strengths and challenges during peace and war.
American Psychologist, 66, 65–72. https://doi.org/10.1037/a0021249
Peterson, C., Park, N., & Castro, C. A. (2011). The Global Assessment Tool. American
Psychologist, 66, 10–18. https://doi.org/10.1037/a0021658
Peterson, C., Park, N., & Seligman, M. E. P. (2005). Orientations to happiness and life
satisfaction: The full versus the empty life. Journal of Happiness Studies, 6, 25–41.
https://doi.org/10.1007/s10902-004-1278-z
51
Peterson, C., & Seligman, M. E. P. (2004). Character strengths and virtues: A handbook and
classification. New York, NY: Oxford University Press.
Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., & Seligman, M. E.
P. (1982). The Attributional Style Questionnaire. Cognitive Therapy and Research, 6,
287–299. https://doi.org/10.1007/BF01173577
Riggs, S. A., & Riggs, D. S. (2011). Risk and resilience in military families experiencing
deployment: The role of the family attachment network. Journal of Family Psychology,
25, 675–687. https://doi.org/10.1037/a0025286
Russell, D., Peplau, L. A., & Ferguson, M. L. (1978). Developing a measure of loneliness.
Journal of Personality Assessment, 42, 290–294.
https://doi.org/10.1207/s15327752jpa4203_11
Saltzman, W. R., Lester, P., Beardslee, W. R., Layne, C. M., Woodward, K., & Nash, W. P.
(2011). Mechanisms of risk and resilience in military families: Theoretical and empirical
basis of a family-focused resilience enhancement program. Clinical Child and Family
Psychology Review, 14, 213–230. https://doi.org/10.1007/s10567-011-0096-1
Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment and
implications of generalized outcome expectancies. Health Psychology, 4, 219–247.
https://doi.org/10.1037/0278-6133.4.3.219
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism from
neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the Life
Orientation Test. Journal of Personality and Social Psychology, 67, 1063–1078.
https://doi.org/10.1037/0022-3514.67.6.1063
Schmitt, T. A. (2011). Current methodological considerations in exploratory and confirmatory
52
factor analysis. Journal of Psychoeducational Assessment, 29, 304–321.
https://doi.org/10.1177/0734282911406653
Seligman, M. E. P., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction.
American Psychologist, 55, 5–14. https://doi.org/10.1037//0003-066X.55.1.5
Tabachnick, B. G., & Fidell, L. S. (2013). Using multivariate statistics (6th ed.). Boston, MA:
Pearson Education.
Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach’s alpha. International Journal of
Medical Education, 2, 53–55. https://doi.org/10.5116/ijme.4dfb.8dfd
Vie, L. L., Griffith, K. N., Scheier, L. M., Lester, P. B., & Seligman, M. E. P. (2013). The
Person-Event Data Environment: Leveraging big data for studies of psychological
strengths in soldiers. Frontiers in Psychology, 4, 934.
https://doi.org/10.3389/fpsyg.2013.00934
Vie, L. L., Scheier, L. M., Lester, P. B., Ho, T. E., Labarthe, D. R., & Seligman, M. E. P. (2015).
The U.S. Army Person-Event Data Environment: A military–civilian big data enterprise.
Big Data, 3, 67–79. https://doi.org/10.1089/big.2014.0055
Vie, L. L., Scheier, L. M., Lester, P. B., & Seligman, M. E. P. (2016). Initial validation of the
U.S. Army Global Assessment Tool. Military Psychology, 28, 468–487.
https://doi.org/10.1037/mil0000141
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures
of positive and negative affect: The PANAS scales. Journal of Personality and Social
Psychology, 54, 1063–1070. https://doi.org/10.1037/0022-3514.54.6.1063
Weins, T. W., & Boss, P. (2006). Maintaining family resiliency before, during, and after military
separation. In C. A. Castro, A. B. Adler, & T. W. Britt (Eds.), Military life: The
53
psychology of serving in peace and combat: Vol. 3. The military family (pp. 13–38).
Westport, CT: Praeger Security International.

54
Table 2.1. Pattern Matrix after Oblique Rotation
Item 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Comm
Joyful .48 .63
Sad .52 .59
Peaceful/calm .48 .57
Hopeful .43 .65
Scared/fearful .65 .47
Feel left out .47 .54
Feel close to people .77 .73
Feel part of a group .77 .76
Participate community events .82 .67
Help out others in neighborhood .88 .73
Close to others in community .84 .82
Good relationship with neighbors .67 .55
Make a difference in community .75 .67
Little interest in doing things .52 .68
Feeling tired or little energy .86 .68
Poor appetite or overeating .75 .60
Feeling bad about yourself .42 .65
Trouble concentrating .76 .66
How satisfied with family .45
How satisfied with relationship .87 .79
Family expresses tenderness .71 .56
Family confides in each other .88 .85
Family shares opinions .66 .63
Contact with friends and family .47 .51
Someone for advice .71 .57
Someone to have lunch with .77 .73
Someone to call if stranded .83 .60
Someone to call for help if sick .82 .65
Fam adjusts to military demands .48 .51
Wish I wasn’t in this relationship .73 .53
Relationship has serious problems .89 .78
Partner emotionally supportive .80 .67
Emotionally distant from partner .86 .72
Communicate expectations .71 .65
Partner doesn’t understand me .84 .66
Trusting relationship with partner .60 .46
Get on each other’s nerves .73 .48
Managing stress effectively .40 .54 .74
Managing household and chores .48 .44
Managing unexpected things .60 .64
Creativity – new ideas .30 .16
Curiosity or interest .55 .31
Critical thinking, good judgement .53 .34
Love of learning .41 .21
Perspective or wisdom .41 .30
Bravery or courage .34 .20
Honesty .10
Zest or enthusiasm .16
Teamwork .23
Fairness -.37 .25
Leadership .40 .18
Forgiveness or mercy .47 .23
Modesty or humility .54 .29
Prudence or caution .51 .27
Self-control or self-regulation .36 .19
Gratitude and thankfulness -.38 .20
Kindness or generosity -.45 .21
Playfulness or humor -.35 .15
Things I can’t change, I accept .60 .42
See the positive side of bad things .78 .61
Expect more bad things happen .59 .55
Effective ways to deal with stress .53 .61
I have no control over things .23
Make things worse under stress .54 .49
In uncertain times expect the best .75 .63
Expect more good than bad things .74 .66
55
Optimistic about my future .60 .68
My life has meaning .78 .75
My life is connected to humanity .46 .44
Partners military job has meaning .50 .41
The things I do are worthwhile .80 .76
There is a purpose for my life .99 .83
How much others pay attention .61 .53
Support partners military service .72 .46
Military meets my family’s needs .60 .42
Note. Factor loadings below .30 are not shown.
a
Comm = communalities.

56
Table 2.2. Final Factors and Items
Factor and Items CFA Loading M SD α
Relationship functioning 4.07 0.86 .93
How satisfied are you with your marriage/relationship? .87
I wish I had not gotten into this relationship. .73
Our relationship has serious problems. .86
My partner is emotionally supportive of me. .77
I feel emotionally distant from my partner. .86
My partner and I clearly communicate our expectations. .77
My partner does not understand me. .83
My partner and I get on each other’s nerves .60
Positive coping 3.72 0.92 .80
When bad things happen, I try to see the positive sides .65
In uncertain times, I usually expect the best. .80
Overall, I expect more good things to happen to me than bad. .81
Depressive symptoms 3.74 0.98 .83
Feeling tired or having little energy .79
Poor appetite or overeating .79
Trouble concentrating on things .75
Social support 3.92 0.93 .86
If I was sick, I could find someone to help me with my daily chores. .73
There is someone I can turn to for advice on personal/family probs. .84
If I wanted to have lunch, I could find someone to join me. .78
If I was stranded 10 miles from home, there is someone I could call. .78
Social connections 3.19 1.10 .91
I participated in community events, activities or meetings .78
I felt like I could make a difference in the community. .82
I helped out others in my neighborhood. .90
I felt close to others in my community. .77
I had a good relationship with people in my neighborhood. .76
Meaning 4.18 0.88 .91
My life has meaning. .89
I believe the things that I do are all worthwhile. .83
I have a purpose in life. .91
Family cohesion 4.16 0.74 .86
My family expresses tenderness .82
My family confides in each other. .87
When my family makes decisions, we all share our opinions. .81

57
Table 2.3. Bivariate Correlations between Composite Scores

1 2 3 4 5 6
1. Depressive symptoms

2. Social connections -.382

3. Social support -.442 .552

4. Relationship functioning -.387 .289 .418

5. Family cohesion -.320 .360 .425 .520

6. Meaning making -.457 .472 .485 .411 .460

7. Positive coping -.422 .432 .475 .343 .407 .605
Note. All correlations significant at p < .01.

58
Table 2.4. Simple Theoretically Driven Regression Model
Model 1 Model 2
b SE β b SE β
Family cohesion -0.201 0.030 -.161 -0.173 0.030 -.138
Social support -0.370 0.024 -.374 -0.360 0.024 -.364
Family cohesion × social support 0.104 0.025 .092
R
2
.217 .225
F for change in R
2
233.781* 16.542*
*p < .001.

59
Figure 2.1. Moderation Effects

1
1.5
2
2.5
3
3.5
4
4.5
5
Low family functioning High family functioning
Depressive Symptoms
Moderation Effects
Low social
support
60
Chapter 3 (Study 2): Mental Health Outcomes Associated with Profiles of Risk and
Resilience among Army Spouses
Abstract
The current study examined patterns of risk and protective factors among military families and
the association of these patterns with mental health diagnoses among U.S. Army spouses.
Spouses (N = 3,036) completed a survey of family psychosocial fitness, the Comprehensive
Soldier and Family Fitness Global Assessment Tool, which informed protective factors including
coping, family cohesion, and social support. Survey results were linked with U.S. Department of
Defense archival data, which informed military-specific risks—including relocation,
deployments, and reunification—and health care utilization and diagnoses. The three-step
method of latent profile analysis was used to explore patterns and associations with covariates
and distal outcomes. Analyses identified six profiles, suggesting significant heterogeneity in
military families with respect to their access to resources and exposure to risk. The largest profile
of families (43.74% of the sample) had limited risk exposure and considerable strengths. Overall,
15.5% of Army spouses received a mental health diagnosis in the year following survey
completion. Variability in risk and protection across profiles was associated with statistically
significant differences in the prevalence of mental health diagnoses among spouses (χ
2
=
126.600, df = 5, p < .001). The highest rates of diagnoses (45.4%) were observed in the profile
with the lowest levels of protective factors. Findings point to the importance of evaluating both
cumulative risk and protective factors and suggest that most military families are faring well.
Increasing access to resources may be a fruitful avenue for prevention among families that are
struggling.

61
Introduction
The adverse impact of war on military service members has been well documented (e.g.,
Hoge, Auchterlonie, & Milliken, 2006; Hoge et al., 2004). As U.S. involvement in overseas
conflicts continues, the negative effects on military families are receiving increased attention.
Recent estimates indicated that 50% of approximately 2.1 million U.S. service members are
married (U.S. Department of Defense, 2017), suggesting a significant number of military spouses
may be experiencing the consequences of extended exposure to war-related stressors. Although
many military families appear to cope successfully with stressors, evidence has suggested that
spouses may be at increased risk of a wide array of poor outcomes (Ahmadi & Green, 2011;
Eaton et al., 2008; Flake, Davis, Johnson, & Middleton, 2009; Mansfield et al., 2010; Renshaw
et al., 2011; Renshaw, Rodebaugh, & Rodrigues, 2010). Further, military spouses may report
more stress compared to service members (Allen, Rhoades, Stanley, & Markman, 2011; Eaton et
al., 2008). These outcomes are particularly concerning considering the documented negative
impact that poor spousal functioning has on military-connected children (Barker & Berry, 2009;
Chandra et al., 2010; Flake et al., 2009; Lester et al., 2010; Paris, DeVoe, Ross, & Acker, 2010).
Although attention to this topic is increasing, most studies have focused solely on the
impact of wartime deployments on spouses and families (de Burgh, White, Fear, & Iversen,
2011), despite evidence that military spouses experience multiple, concurrent layers of stress
(Green, Nurius, & Lester, 2013). Further, there have been increasing calls to include the impact
of many aspects of military family life that may be protective, including access to a broad
spectrum of high-quality formal and informal supports (Easterbrooks, Ginsburg, & Lerner, 2013;
Cozza, Haskins, & Lerner, 2013; Hosek & Wadsworth, 2013). The present study addressed these
62
gaps in the literature using latent profile analysis (LPA) to model the impact of both cumulative
stressors and concurrent protective factors on mental health outcomes for U.S. Army spouses.
Theoretical Foundation
Family systems theory. Family systems theory suggests that family members have an
ongoing, reciprocal impact on one another and posits that the behavior of individuals is best
understood in the context of the family system (Cox & Paley, 2003; Minuchin, 1974). Recent
research with veteran families demonstrated that veteran functioning affects other individuals in
the family through changes in family functioning (Sullivan et al., in press). Further, evidence has
suggested examining individual functioning and health outcomes in military families is best
undertaken in the context of the family system (O’Neal, Lucier-Greer, Mancini, Ferraro, & Ross,
2016). Ultimately, this perspective suggests that risk and protective factors that affect one
member of the family system will likely have consequences for the family as a whole and
eventually for other members of the system.
Risk and protective factors. Risk factors are conceptualized as experiences that increase
the likelihood of a negative outcome, whereas protective factors decrease this likelihood (Fergus
& Zimmerman, 2005). Cumulative or multiple risk exposure is most frequently discussed in the
context of youth development, with less focus on the consequences of multiple exposures to risk
among adults. However, the concept of a “pileup” of stressors is useful to understand the
cumulative impact of multiple-risk exposure on the functioning of military families and military
spouses specifically (Lavee, McCubbin, & Patterson, 1985; McCubbin & Patterson, 1983). This
pileup of demands has been associated with poor family adaptation among U.S. Army soldiers
and their spouses (Lavee et al., 1985). Further, similar to findings in the child development
literature, evidence has suggested that exposure to multiple, concurrent risks has consequences
63
beyond those of any risk factor on its own (Evans & Kim, 2010). The adverse impact of
cumulative stress may be buffered by access to protective factors, including social support
(Green et al., 2013; Lavee et al., 1985). Protective factors operate at multiple social ecological
levels, including intrapersonal, familial, and community systems (Paley, Lester, & Mogil, 2013).
Resilience. Although studies have consistently found an elevated risk of adverse
outcomes among military spouses and children, the majority of these individuals and families are
nevertheless reporting relatively healthy functioning (e.g., Lester et al., 2016; Sullivan et al.,
2015). These findings point to the need to consider resilience processes in military family
systems and their impact on outcomes (Easterbrooks et al., 2013). The present study relied on a
family resilience perspective, which encompasses a dynamic balance of risk and protective
factors operating at individual and family levels, allowing a family to maintain or return to a
previous level of functioning in the face of adversity (Hawkins et al., 2017; Walsh, 1996, 2003).
This theoretical orientation suggests the need to consider the simultaneous impact of both risk
and protective factors on outcomes for military spouses.
Empirical Evidence for Risk and Protective Factors
Risk factors affecting military spouses. The most commonly discussed stressors
associated with military family life include separations, either for training or combat
deployments, and family relocations (Burrell, Adams, Durand, & Castro, 2006). Empirical
evidence has suggested that among these stressors, combat deployments are a particularly
significant risk factor associated with adverse outcomes for military spouses (de Burgh et al.,
2011; Mansfield et al., 2010). Service member deployments have been linked with general
distress (Flake et al., 2009; Lester et al., 2010); increased parenting stress (Flake et al., 2009);
increased depressive and trauma symptoms (Lester et al., 2010); increased feelings of sadness,
64
anxiety, anger, and somatic problems (Paley et al., 2013); and increased mental health diagnoses
among military spouses (Mansfield et al., 2010). Cumulative deployment experiences in
particular appear to be a potent predictor of outcomes for military spouses (Lara-Cinisomo et al.,
2012; Rodriguez & Margolin, 2015). Reintegration or reunification is a period that can last 6
months or longer beyond the end of a deployment (Pincus, House, Christenson, & Adler, 2001).
Although this phase has sometimes been associated with fewer mental health symptoms among
spouses compared to during a deployment (Lester et al., 2010), it can also be accompanied by
additional stressors, as families navigate changing roles and responsibilities (McNulty, 2010;
McNulty, 2013). Finally, military families experience a relocation, or permanent change of
station (PCS), every 2 to 3 years on average (Park, 2011). Spouses may struggle to maintain
employment as a result of these frequent moves (Harrell, Lim, Castaneda, & Golinelli, 2004) and
may experience associated decreases in overall well-being (Burrell et al., 2006; Lincoln, Swift,
& Shorteno-Fraser, 2008).
Protective factors affecting military spouses. Protective factors that reduce the risk of
negative outcomes among military spouses have been less frequently studied. Previous work
with military families suggested that protective factors may exist at several social ecological
levels, including intrapersonal factors, family factors, and community or social network factors
(Bowles et al., 2015; Paley et al., 2013). At the intrapersonal level, the use of positive, active
coping strategies has been associated with healthy spouse and family outcomes, particularly
during military separations (Weins & Boss, 2006). Further, it has been hypothesized that the
capacity to articulate a sense of purpose or meaning may predict better outcomes both for
spouses and children and for overall family functioning (Saltzman et al., 2011). Similarly, sense
of coherence, defined as the meaningfulness of daily life, has been associated with greater
65
contentment among Army spouses, particularly in the context of deployment (Everson, Darling,
& Herzog, 2013). At the familial level, healthy and cohesive family functioning and intimate
relationship quality have been associated with better mental health and lower psychological
distress among military spouses (McGuire et al., 2012). Lower levels of negative communication
and overall marital satisfaction have been associated with lower levels of stress for Army
spouses (Allen et al., 2011). Finally, at the community or network level, formal and informal
social connections are hypothesized to affect the functioning of both service members and their
spouses (Bowen & Martin, 2011; Weins & Boss, 2006). Lower levels of isolation and increased
social support, particularly during deployments, have been associated with decreased stress
among military spouses (Andres, Moelker, & Soeters, 2012; Van Winkle & Lipari, 2015).
Military community support, including connection to the military community, has also been
associated with military parent psychosocial functioning and ultimately with child outcomes
(Conforte et al., 2017).
Outcomes for military spouses. Recent empirical evidence has documented elevated
rates of adverse outcomes among military spouses. These outcomes include general distress,
disrupted parenting, substance use, secondary traumatization, and diagnosed mental health
conditions (Ahmadi & Green, 2011; Eaton et al., 2008; Flake et al., 2009; Mansfield et al., 2010;
Renshaw et al., 2011; Renshaw et al., 2010). Although many adverse outcomes have been
observed, the present study explored rates of mental health diagnoses in this population. Military
spouses may experience similar rates of mental health challenges as service members but are
more likely to seek treatment either from primary care providers or mental health specialists
(Eaton et al., 2008).
66
The Current Study
When exploring resilience processes operating in family systems, two approaches have
been suggested: variable-focused methods and person- or family-focused methods (Masten,
2001). Previous investigations with military families have tended to take a variable-focused
approach, using regression or structural equation modeling to explore the impact of risk on
health or mental health outcomes (Lucier-Greer, Arnold, Mancini, Ford, & Bryant, 2015; Lucier-
Greer, O’Neal, Arnold, Mancini, & Wickrama, 2014; Wadsworth et al., 2016). The current study
employed person-centered methodology to explore patterns of family-level risk and protective
factors and their association with military spouse outcomes. LPA, a form of mixture modeling
employing continuous variables, has been used to uncover these patterns as they naturally occur
in Army families (Nurius & Macy, 2008; Rosato & Baer, 2012).
LPA is a useful methodology to model the impact of cumulative stressors and protective
factors on mental health outcomes for U.S. Army spouses for two reasons. First, the present
study conceptualized resilience as a dynamic balance between risk and protective factors,
requiring a modeling technique that can evaluate these elements simultaneously. Second, by
defining unique subgroups in this sample, LPA can uncover less common configurations of risk
and protective factors among military families that may be masked using variable-focused
modeling (Masten, 2001). Outcomes for these different subgroups can then be explored. Two
specific hypotheses guided these analyses:
1. Heterogeneity will exist in levels of protective factors and exposure to risk across
military families.
2. Rates of spouse mental health diagnoses will be higher in families that experience
higher levels of risk and lower levels of protective factors.
67
Methods
Family Global Assessment Tool
This work is a secondary analysis of Comprehensive Soldier and Family Fitness (CSF2)
data obtained through a partnership with the U.S. Army’s Research Facilitation Laboratory,
under the Army Analytics Group. The CSF2 program was designed by the U.S. Army to foster
five domains of physical and psychosocial fitness among Army soldiers and their families. As a
part of this program, the Family Global Assessment Tool (GAT) was created to evaluate the
physical, social, emotional, family, and spiritual health of Army families (Peterson, Park, &
Castro, 2011). The GAT survey is completed via an online portal by Army spouses on a
voluntary basis and can be completed as often as the spouse chooses. More information about the
GAT has previously been published (Hawkins et al., 2017). This instrument recently underwent
psychometric evaluation, resulting in seven scales with preliminary evidence of good reliability
and validity (Sullivan et al., in process).
Data Linkage
Family GAT data were merged with several other U.S. Department of Defense (DOD)
archival data sources, including data on personnel, military transitions, operations, and medical
outcomes. These data were accessed and merged in the Person-Event Data Environment, a
cloud-based, virtual enclave in which deidentified data from disparate sources within the DOD
are housed, linked, and analyzed. Once given approval to access data for a cohort of Family
GAT completers, these cohort data were merged with Family GAT survey data, using a unique
identification number and identifying only the first GAT completions for those in the approved
cohort.
68
Next, administrative datasets were identified and merged. DOD administrative records
are recorded in “snapshots,” which refers to the process of recording a new line of data on a
quarterly, biannual, or annual basis (depending on the dataset) only if something has changed in
a service member’s record. Active Duty and Reserve Master Personnel, Family Sponsor, and
Family Dependent files were merged using the unique identifier. The closest snapshot date
before each spouse’s Family GAT completion date was retained from each record. Finally,
deployment data were merged using a process in which deployments were sorted in
chronological order and all deployments for each family dating back to the beginning of 2009
(the earliest date of available deployment data) were retained in the final dataset, regardless of
snapshot date.
Finally, four medical datasets, representing inpatient and outpatient medical records from
both military and civilian facilities, were merged following the same process. Members of the
approved cohort were identified using their unique identification number. Only visits that
occurred in the year following GAT completion were retained. Duplicate records were discarded
such that only one record of each unique diagnosis for any given spouse was retained in the
dataset.
Participants
Spouses of U.S. Army soldiers are the index respondent for each family in the sample. To
be included in analyses, spouses had to: (a) complete a Family GAT survey; (b) provide consent
to have their GAT survey responses used for research purposes; and (c) be linked to an Army
soldier using a unique electronic identifier. Following data merging, the final sample was 3,036
U.S. Army families that completed the survey online between 2014 and the end of 2016.
69
Secondary analyses were approved by the Army Research, Development and Engineering Center
Institutional Review Board and the institutional review board at the [blinded for review].
Measurement
Data that inform risk factors, protective factors, covariates, and mental health outcomes
were drawn from different sources, described here.
Risk factors. Based on theory and empirical evidence previously reviewed, three
military-specific risk factors were included in models: (a) experiencing a family move
(permanent change of station [PCS]) in the last year; (b) cumulative days of deployment since
the beginning of 2009; and (c) experiencing reunification following a deployment in the last
year. Datasets managed by the Defense Manpower Data Center (DMDC) were employed to
operationalize these variables.
Recent PCS move. Using the DMDC’s Master Personnel File, a dichotomous item
reflecting a PCS move in the year preceding GAT completion was created by comparing each of
the family’s arrival dates at permanent duty stations to the date the spouse completed the Family
GAT. If an arrival date was within 365 days before the spouse took the GAT, the family was
deemed to have moved in the year preceding GAT completion (yes vs. no).
Cumulative days of deployment. Using DMDC’s Contingency Tracking System
Deployment data, a continuous variable reflecting cumulative days of deployment between 2009
(the earliest date of available deployment data) and the date of GAT completion was created. All
deployment days that occurred up to but not including the date of GAT completion were
summed to create this item.
Recent reunification. Using DMDC’s Contingency Tracking System Deployment data, a
dichotomous item reflecting a recent reunification following deployment was created by
70
comparing deployment end dates with GAT completion dates. If any deployment end date was
within 365 days before the spouse took the GAT, the family was deemed to have experienced a
recent reunification (yes vs. no).
Protective factors. Based on theory and empirical evidence previously discussed, six
protective factor scales, representing three social ecological levels, were drawn from the Family
GAT and included in these analyses. Responses to individual items were recoded such that
higher scores indicated positive functioning and averaged to create composite scores.
Intrapersonal factors. Three items assessed positive coping. Each item had five Likert-
type response options ranging from not like me at all to very much like me. Items included:
“When bad things happen, I try to see the positive sides” and “In uncertain times, I usually
expect the best.” Internal consistency in this sample was good (α = .80). Three items assessed
meaning making. Each item had five Likert-type response options ranging from not like me at all
to very much like me. Items included: “I have a purpose in life” and “I believe the things that I do
are worthwhile.” Internal consistency in this sample was excellent (α = .90).
Familial factors. Eight items assessed relationship functioning. All items had five Likert-
type responses ranging from strongly disagree to strongly agree. Items included: “I feel
emotionally distant from my partner” and “My partner and I clearly communicate our
expectations for each other.” Internal consistency in this sample was excellent (α = .93). Three
items assessed family cohesion. Each item had five Likert-type response options ranging from
strongly disagree to strongly agree. Items included: “My family expresses tenderness” and “My
family confides in each other.” Internal consistency in this sample was good (α = .86).
Community factors. Four items assessed social support. Each item had five Likert-type
response options ranging from strongly disagree to strongly agree. Items included: “There is
71
someone I could turn to for advice on how to deal with a personal or family problem” and “If I
was sick, I could find someone to help with my daily chores.” Internal consistency in this sample
was good (α = .86). Five items assessed social connections. Each item had five Likert-type
response options ranging from never to most of the time. Items included: “I have a good
relationship with people in my neighborhood” and “I participated in community events,
activities, or meetings.” Internal consistency in this sample was excellent (α = .91).
Covariates. Using data from the Master Personnel File, three variables were created as
covariates: (a) service member race and ethnicity (White, non-White); (b) service member
educational attainment (high school or below, some college or above); and (c) service member
military rank (enlisted, officer).
Mental health outcomes. Whether a military spouse received a mental health diagnosis
was the outcome of interest in these analyses. Spouses were deemed to have received a mental
health diagnosis if they were assigned a mental health-related code from the International
Classification of Diseases, 9th Revision (ICD-9) or International Classification of Diseases, 10th
Revision (ICD-10) during an inpatient or outpatient medical visit within 1 year following GAT
survey completion. To identify mental health-related codes, this study followed the procedures
used by Mansfield et al. (2010), who developed a comprehensive list of ICD-9 codes for 17
categories of mental health disorders. Because data from this cohort spanned the transition to
ICD-10, all codes on this list were converted to ICD-10 codes and both ICD-9 and ICD-10 codes
were used to create the outcome variable. Ultimately, the presence of any of these codes was
represented with one dichotomous item reflecting receipt of any mental health diagnosis or not,
from any one of the four medical datasets included in analyses, in the year following GAT
completion.
72
Analyses
To combine an LPA model with covariates or a distal outcome (presence of a mental
health diagnosis, here), a three-step method is preferable, which proceeds as follows: (a) an LPA
model is specified using only indicator variables; (b) a nominal most likely class variable is
created and the classification uncertainty rate is calculated; and (c) considering the uncertainty of
classification, the most likely class variable is associated with covariates or the outcome variable
(Asparouhov & Muthén, 2014). Analyses began with a latent profile model including risk and
protective factors that have been empirically associated with spouse outcomes. All continuous
and categorical variables were entered into the LPA model simultaneously, using summary
scores for protective factors to preserve power. To determine the appropriate number of classes,
an initial 1-profile model was assessed and compared to models with an increasing number of
profiles. Model fit was determined by considering four criteria: (a) low Bayesian information
criterion (BIC); (b) significant Lo-Mendell-Rubin likelihood ratio test (LMR-LRT); (c)
significant bootstrap likelihood ratio test (BLRT); and (d) conceptual and theoretical
considerations (Nylund, Asparouhov, & Muthen, 2007). When specific model fit statistics
indicated different solutions, substantive considerations guided final model selection. Steps 2 and
3 of the three-step method were conducted using the AUXILIARY option in Mplus (Asparouhov
& Muthén, 2014; Muthén & Muthén, 2012). The R3STEP command was used for covariates and
the DCAT command was used for the distal outcome. SPSS version 21 was used for data
cleaning and descriptive statistics; Mplus version 7 was used for LPA models (Muthen &
Muthen, 2012). Full information maximum likelihood, available in Mplus, was employed to
handle missing data in LPA models. Listwise deletion was used in models including covariates,
which resulted in elimination of 326 cases or 10.7% of the sample.
73
Results
Sample demographics are presented in Table 3.1. Most service members were male,
White, and enlisted soldiers. Most spouses were female. The majority of families had two or
fewer children. Families were relatively evenly split between those whose oldest child was age
11 or younger and those whose oldest child was 12 or older. Approximately 15% of families had
experienced a recent relocation and 7% experienced a recent reunification following a
deployment. On average, families reported 139 cumulative days of deployment. Approximately
16% of military spouses received a mental health diagnosis in the year following GAT survey
completion.
LPA Model
Model fit indexes for the LPA model are included in Table 3.2; means and conditional
probabilities are presented in Table 3.3. Although BIC continued to decrease in the 7-profile
solution, a significant LMR-LRT in this solution indicated a preference for the 6-profile model.
This model was more parsimonious and clear distinctions between classes were evident, so the 6-
profile solution was chosen as best fitting.
Profile 1 (moderate protection and low risk) accounted for 18.19% of the sample.
Relatively few spouses in this group experienced a recent move or reunification, and mean
cumulative deployment days were low compared to other profiles. Further, spouses endorsed
moderate to high levels of all six protective factors. Profile 2 (lowest protection and low risk)
accounted for 2.67% of the sample. Very few spouses in this group experienced a recent move or
reunification, and cumulative deployment days were low on average. However, spouses in this
profile endorsed the lowest levels of all six protective factor indicators across profiles. Profile 3
(high protection and moderate risk) accounted for 20.03% of the sample overall. Spouses in this
74
group had a higher probability of experiencing a recent move or reunification compared to
Profile 1 and had 322 cumulative deployment days on average. Further, spouses in Profile 3
reported high levels of all six protective factors. Profile 4 (moderate protection and moderate
risk) accounted for 8.14% of the sample. Spouses in this profile reported slightly lower levels of
protective factors compared to Profile 1. Further, spouses in this profile were slightly more likely
to experience risk factors and had slightly more cumulative deployment days on average
compared to Profile 3. Profile 5 (high protection and highest risk) accounted for 6.52% of the
sample. Spouses in this category reported relatively high levels of protective factors but were
also the most likely be exposed to risk factors and experienced the most cumulative deployment
days across groups. Finally, Profile 6 (high protection and low risk) was the largest group,
accounting for 43.74% of the sample overall. Spouses in this group were less likely to be
exposed to risk factors and reported high levels of protective factors.
Covariates
Associations between covariates and latent profiles are displayed in Table 3.4. Families
of service members who completed at least some college were more likely to be in the High
Protection and Moderate Risk profile than the High Protection and Low Risk profile compared to
families of service members with less education (b = .62, p < .01). Compared to families of non-
White service members, families of White service members were more likely to be in any profile
other than the High Protection and Low Risk profile except the Moderate Protection and Low
Risk profile. Finally, compared to the families of officers, families of enlisted service members
were more likely to be in the Lowest Protection and Low Risk (b = .1.27, p < .05), High
Protection and Moderate Risk (b = -.39, p < .05), or High Protection and Highest Risk (b = -.47,
p < .05) categories versus High Protection and Low Risk.
75
Distal Outcomes
Distal outcome results are presented in Table 3.5. Omnibus chi-square results indicated
significant differences in the prevalence of mental health diagnoses across latent profiles (χ
2
=
126.600, df = 5, p < .001). Individual chi-square tests evaluated statistically significant
differences in the prevalence of mental health diagnoses between two profiles at a time. These
results, displayed in Table 3.4, suggest that many differences in mental health prevalence
between individual profiles were also significant. Overall, the highest prevalence of mental
health diagnoses was observed in the Lowest Protection and Low Risk profile (45.4%) and the
Moderate Protection and Moderate Risk profile (30.5%), whereas the lowest prevalence rates
were observed in the Moderate Protection and Low Risk (8.1%) and High Protection and Low
Risk (14.3%) profiles.
Discussion
This study aimed to describe naturally occurring patterns of risk and protective factors
affecting U.S. Army spouses and the association of these patterns with the prevalence of mental
health diagnoses. Overall, 15.5% of Army spouses in this sample received a mental health
diagnosis during a 1-year period following GAT survey completion. Mansfield et al. (2010)
found an overall rate of 34.7% during their 4-year study from 2003 to 2006. Given the
differences in time span, the results presented here represent relatively similar although perhaps
slightly elevated findings. Building on previous work, these findings highlight the critical role
that both risk and protective factors have in understanding the potential for adverse mental health
outcomes among military-connected spouses.
Confirming the first hypothesis, these results suggest that significant variation exists
among military families in their access to protective factors and exposure to risk factors. LPA
76
identified six distinct groups with markedly different patterns of risk and protective factors. In
the LPA results, the largest group, accounting for 44% of the sample, was composed of families
that reported comparatively little exposure to risk factors and high levels of protective factors.
This finding echoes recent studies that highlighted a small subset of families that exhibit adverse
outcomes and a much larger group of families that appear to be coping with stressors
successfully (Easterbrooks et al., 2013; Sullivan et al., 2015). Further, although there was
variability in levels of protective factors across profiles, even the small group of spouses who
reported the lowest levels of protective factors (Profile 2; 2.67%) nevertheless appeared to have
some intrapersonal, familial, or community resources on which to draw.
Generally, protective factors appeared to vary together, such that families that reported
low levels of one protective factor tended to report low levels across all six protective factors
evaluated. Recent research using a latent class analysis approach among military families had
similar findings (Trail, Meadows, Miles, & Karney, 2017). Although protective factors covaried,
current findings suggest that high levels of protective factors are not necessarily associated with
low levels of risk, which may distinguish these results from findings in the civilian community.
The emergence of Profile 2 (lowest protection and low risk) and Profile 5 (high protection and
highest risk) categories suggest that protective factors and risk factors do not necessarily covary.
Trail and colleagues (2017) theorized that exogenous stressors that are associated specifically
with military service may not be as inherently linked to endogenous resources and
vulnerabilities, which are a product of the family system. This decoupling between
vulnerabilities and resources may be a unique feature of military families and may explain the
lack of covariation between risk and protective factors. Mental health providers may find family
resources could be useful for prevention or intervention efforts, even among the most risk-
77
exposed families. Finally, although protective factors generally covaried, Army spouses
consistently reported lower levels of social connectedness compared to other possible resources
evaluated. This finding is consistent with previous work, which suggested that overall integration
into military communities may be low (Burrell, Durand, & Fortado, 2003).
Regarding risk factors, approximately two thirds of this sample were grouped into three
profiles with relatively low levels of risk. Notably, however, there was some low-level exposure
to risk, even in these three profiles. Likely, this low but consistent risk reflects the systemic
nature of stressors that accompany military service. The risk factors evaluated in this study are,
to a certain degree, an expected part of military life (Burrell et al., 2006). Nevertheless, in the
context of U.S. involvement in multiple overseas conflicts and the accompanying high
operational tempo, many military families are experiencing exposure to risk that goes beyond
this expected level. These experiences are reflected in the three profiles in which spouses were
more likely to endorse a recent move or recent reunification and reported many more cumulative
days of deployment. For example, in Profile 5 (high protection and highest risk), one third of
spouses had experienced a reunification following deployment in the preceding year, and this
group experienced upward of 600 cumulative days of deployment on average. Although this
profile was small, at 6.5% of the overall sample, these families exemplify the high operational
tempo that has characterized recent wartime military family life.
Findings regarding covariates suggest broadly that the families of White and enlisted
service members were less likely to be classified in the High Protection and Low Risk profile
and more likely to be classified in profiles with lower levels of protective factors or greater
exposure to risk. There were limited findings regarding service member education, which may be
because this variable was conflated with rank. Preliminary bivariate correlations and chi-square
78
tests (not shown) suggested that weak or nonsignificant relationships existed between covariates
and protective factors, but that both enlisted rank and nonminority status were associated with
increased days of deployment. It is likely that more deployments and the related experience of
more recent reunification following deployment are driving these relationships.
The association of latent profiles with mental health diagnoses among spouses further
clarifies the relative impact of risk and protective factors on adverse outcomes. The variability in
risk and protective factors across profiles was associated with statistically significant differences
in rates of mental health diagnoses among Army spouses. Confirming the second hypothesis, the
highest rate of mental health diagnoses was observed in Profile 2 (lowest protection and low
risk), which was characterized by the lowest levels of protective factors but also relatively low
levels of risk exposure. Although this was a very small group, at only 2.7% of the sample, more
than 45% of spouses in this group received a mental health diagnosis in the year following GAT
survey completion. This finding seems to suggest that limited access to protective factors may be
driving the relationship with adverse outcomes more than elevated exposure to risk factors.
The lowest rates of mental health diagnoses were observed in Profile 6 (high protection
and low risk), which was characterized by the highest levels of protective factors and relatively
low risk, and Profile 1 (moderate protection and low risk), which was also characterized by
relatively low risk but only moderate levels of protective factors. Although the lower rates of
mental health diagnoses among the high protection and low risk group is not surprising, the
finding regarding the moderate protection and low risk category is more nuanced. It is possible
that this finding suggests an optimal nexus between low exposure to risk and sufficient access to
resources. Outcomes in this profile are in contrast with the outcomes for spouses in Profile 4
(moderate protection and moderate risk), which is characterized by similar levels of protective
79
factors as in Profile 1, but more risk exposure. Although only 8.1% of spouses received a mental
health diagnosis in Profile 1, 30.5% received a diagnosis in Profile 4. These findings may
suggest that when exposure to risk increases, moderate levels of protective factors no longer
provide a sufficient balancing effect, leading to increases in adverse outcomes. These results call
to mind the definition of resilience as a dynamic balance of risk and protective factors, which
allows a family to maintain or return to a previous level of functioning in the face of adversity.
The balance between risk and protection observed in Profile 1 may be emblematic of these
hypothesized resilience processes.
Finally, the results presented here point to the usefulness of the LPA approach alongside
more traditional variable-focused methods. Using LPA uncovered significant heterogeneity
among military families in this sample, which would likely be masked by other statistical
approaches. In particular, this method highlighted the large group of military families with
relatively little exposure to risk and considerable access to protective factors. Although variable-
focused methods are useful for drawing a direct link between exposure to particular risks and
elevated rates of adverse outcomes, these statistical findings are likely driven by a smaller group
of families with much higher risk exposure and possibly fewer protective factors. These variable-
focused results are critical for targeting resources toward a population in need, but they may have
the unintended consequence of overpathologizing military families more generally (Cozza et al.,
2013; Rosato & Baer, 2012). A comprehensive approach to understanding a particular problem
must employ both person- and variable-focused strategies. For example, using a person-focused
approach, these results highlight that families experiencing similar levels of multiple, concurrent
risks are likely to have similar rates of adverse outcomes. It might also be useful to know which
risks among those examined here are particularly useful predictors of negative outcomes, a
80
related question that would be better answered using variable-focused methods. Future research
using variable-focused methods should consider the relative impact of cumulative deployments,
recent reunifications, and family moves and the potentially moderating effect of access to
protective factors on spouse mental health outcomes.
Strengths and Limitations
This study has a number of strengths. The use of big data methods, which link multiple
datasets from different sources, allowed for a more complex and nuanced picture of military
families that includes both cumulative risk and strengths. Using multiple data sources, including
self-report and health records, avoided common method bias, which may plague studies in which
military spouses are the only source of information about family functioning. Using diagnosed
mental health conditions avoided misattribution that can arise from using symptom reporting as
an outcome. Finally, the use of person-centered methods highlighted the heterogeneity in
military families, which is often overlooked in studies that use a variable-focused approach.
Despite these strengths, this study has limitations that need to be considered. First, U.S.
Army spouses take the GAT survey on a voluntary basis. Although this is a national sample and
the demographics are similar to the population of the Army overall, the voluntary nature of the
survey may limit generalizability of these results. Additionally, although temporality was
considered when linking archival data, these data are still essentially cross-sectional, so care
must be taken when considering causality. Finally, a particular challenge of big data methods is
that data are not originally collected for research. As a result, potential variables of interest may
not be available for inclusion in models. Certain military-specific risks that could affect spouse
mental health, including service member mental health and spousal employment, could not be
included here. Further, recent research with military families suggested that in addition to
81
military-specific risk factors, normative stressors that affect all families may be critical for
predicting negative outcomes (Lucier-Greer et al., 2014). Although this study examined the
effects of minority status, lower educational attainment, and service member rank on profile
membership, data on risk factors were drawn from DOD archival records. As a result, additional
normative, nonmilitary-related stressors could not be included in models. Future research should
consider taking a similar approach to examining cumulative normative and military-specific
stressors. Further, because DOD archival data are collected with the service member as the focal
point, these datasets include less detailed information about spouses. Although service member
race and ethnicity and education were available, corresponding information was not available for
spouses.
Conclusions
Ultimately, the results of this study highlight that military families are heterogenous in
their exposure to risk and access to resources. These findings cast doubt on characterizations of
military families as either monolithically pathological or inherently resilient. Doctors, mental
health professionals, and other front-line clinicians serving military families must be aware of the
elevated risk of poor mental health outcomes among military spouses. However, in addition to
the need to assess for mental health symptomatology, clinicians must also be careful not to
overpathologize, particularly among families with lower exposure to risk and sufficient
resources. Assessing spouses’ access to protective factors, in particular, may be useful to
understand the potential for adverse outcomes and inform intervention.
The presence of risk across groups in this study suggests that a certain amount of risk is
systemic in military families. Further, these results highlight the critical importance of access to
protective factors in determining association with mental health outcomes. Even when risk
82
cannot be avoided altogether, which is likely the case in most military families, increasing access
to protective factors may be a fruitful avenue toward prevention and intervention with this
population. Military programs that improve spouses’ internal resources, marital and family
functioning, and social connectivity may effectively counteract systemic risk. Preventive
interventions, like Families OverComing Under Stress (FOCUS), that target many of these
protective factors may be useful in reducing adverse outcomes and improving overall family
functioning (Saltzman et al., 2011).

83
References
Ahmadi, H., & Green, S. L. (2011). Screening, brief intervention, and referral to treatment for
military spouses experiencing alcohol and substance use disorders: A literature review.
Journal of Clinical Psychology in Medical Settings, 18, 129–136.
https://doi.org/10.1007/s10880-011-9234-7
Allen, E. S., Rhoades, G. K., Stanley, S. M., & Markman, H. J. (2011). On the home front: Stress
for recently deployed Army couples. Family Process, 50, 235–247.
https://doi.org/10.1111/j.1545-5300.2011.01357.x
Andres, M., Moelker, R., & Soeters, J. (2012). A longitudinal study of partners of deployed
personnel from the Netherlands’ armed forces. Military Psychology, 24, 270–288.
https://doi.org/10.1080/08995605.2012.678237
Asparouhov, T., & Muthén, B. (2014). Auxiliary variables in mixture modeling: Three-step
approaches using Mplus. Structural Equation Modeling, 21, 329–341.
https://doi.org/10.1080/10705511.2014.915181
Barker, L. H., & Berry, K. D. (2009). Developmental issues impacting military families with
young children during single and multiple deployments. Military Medicine, 174, 1033–
1040. https://doi.org/10.7205/MILMED-D-04-1108
Bowen, G. L., & Martin, J. A. (2011). The resiliency model of role performance for service
members, veterans, and their families: A focus on social connections and individual
assets. Journal of Human Behavior in the Social Environment, 21, 162–178.
https://doi.org/10.1080/10911359.2011.546198
Bowles, S. V., Pollock, L. D., Moore, M., Wadsworth, S. M., Cato, C., Dekle, J. W., … Bates,
M. J. (2015). Total force fitness: The military family fitness model. Military Medicine,
84
180, 246–258. https://doi.org/10.7205/MILMED-D-13-00416
Burrell, L. M., Adams, G. A., Durand, D. B., & Castro, C. A. (2006). The impact of military
lifestyle demands on well-being, Army, and family outcomes. Armed Forces & Society,
33, 43–58. https://doi.org/10.1177/0002764206288804
Burrell, L., Durand, D. B., & Fortado, J. (2003). Military community integration and its effect on
well-being and retention. Armed Forces & Society, 30, 7–24.
https://doi.org/10.1177/0095327X0303000101
Chandra, A., Lara-Cinisomo, S., Jaycox, L. H., Tanielian, T., Burns, R. M., Ruder, T., & Han, B.
(2010). Children on the homefront: The experience of children from military families.
Pediatrics, 125, 16–25. https://doi.org/10.1542/peds.2009-1180
Conforte, A. M., Bakalar, J. L., Shank, L. M., Quinlan, J., Stephens, M. B., Sbrocco, T., &
Tanofsky-Kraff, M. (2017). Assessing military community support: Relations among
perceived military community support, child psychosocial adjustment, and parent
psychosocial adjustment. Military Medicine, 182, e1871–e1878.
https://doi.org/10.7205/MILMED-D-17-00016
Cox, M. J., & Paley, B. (2003). Understanding families as systems. Current Directions in
Psychological Science, 12, 193–196. https://doi.org/10.1111/1467-8721.01259
Cozza, S. J., Haskins, R., & Lerner, R. M. (2013). Keeping the promise: Maintaining the health
of military and veteran families and children (Future of Children Policy Brief, Fall 2013).
Retrieved from
http://www.princeton.edu/futureofchildren/publications/docs/23_02_PolicyBrief.pdf
de Burgh, H. T., White, C. J., Fear, N. T., & Iversen, A. C. (2011). The impact of deployment to
Iraq or Afghanistan on partners and wives of military personnel. International Review of
85
Psychiatry, 23, 192–200. https://doi.org/10.3109/09540261.2011.560144
Easterbrooks, M. A., Ginsburg, K., & Lerner, R. M. (2013). Resilience among military
youth. The Future of Children, 23(2), 99-120.
Eaton, K. M., Hoge, C. W., Messer, S. C., Whitt, A. A., Cabrera, O. A., McGurk, D., … Castro,
C. A. (2008). Prevalence of mental health problems, treatment need, and barriers to care
among primary care-seeking spouses of military service members involved in Iraq and
Afghanistan deployments. Military Medicine, 173, 1051–1056.
https://doi.org/10.7205/MILMED.173.11.1051
Evans, G. W., & Kim, P. (2010). Multiple risk exposure as a potential explanatory mechanism
for the socioeconomic status–health gradient. Annals of the New York Academy of
Sciences, 1186, 174–189. https://doi.org/10.1111/j.1749-6632.2009.05336.x
Everson, R. B., Darling, C. A., & Herzog, J. R. (2013). Parenting stress among US Army spouses
during combat-related deployments: The role of sense of coherence. Child & Family
Social Work, 18, 168–178. https://doi.org/10.1111/j.1365-2206.2011.00818.x
Fergus, S., & Zimmerman, M. A. (2005). Adolescent resilience: A framework for understanding
healthy development in the face of risk. Annual Review of Public Health, 26, 399–419.
https://doi.org/10.1146/annurev.publhealth.26.021304.144357
Flake, E. M., Davis, B. E., Johnson, P. L., & Middleton, L. S. (2009). The psychosocial effects
of deployment on military children. Journal of Developmental & Behavioral Pediatrics,
30, 271–278. https://doi.org/10.1097/DBP.0b013e3181aac6e4
Green, S., Nurius, P. S., & Lester, P. (2013). Spouse psychological well-being: A keystone to
military family health. Journal of Human Behavior in the Social Environment, 23, 753–
768. https://doi.org/10.1080/10911359.2013.795068
86
Harrell, M. C., Lim, N., Castaneda, L. W., & Golinelli, D. (2004). Working around the military:
Challenges to military spouse employment and education. Santa Monica, CA: RAND
National Defense Research Institute. Retrieved from
https://www.rand.org/content/dam/rand/pubs/monographs/2004/RAND_MG196.pdf
Hawkins, S. A., Sullivan, K. S., Schuyler, A. C., Keeling, M., Kintzle, S., Lester, P. B., &
Castro, C. A. (2017). Thinking “big” about research on military families. Military
Behavioral Health, 5, 335–345. https://doi.org/10.1080/21635781.2017.1343696
Hoge, C. W., Auchterlonie, J. L., & Milliken, C. S. (2006). Mental health problems, use of
mental health services, and attrition from military service after returning from
deployment to Iraq or Afghanistan. Journal of the American Medical Association, 295,
1023–1032. https://doi.org/10.1001/jama.295.9.1023
Hoge, C. W., Castro, C. A., Messer, S. C., McGurk, D., Cotting, D. I., & Koffman, R. L. (2004).
Combat duty in Iraq and Afghanistan, mental health problems, and barriers to care. New
England Journal of Medicine, 351, 13–22. https://doi.org/10.1056/NEJMoa040603
Hosek, J., & Wadsworth, S. (2013). Economic conditions of military families. Future of
Children, 23, 41–59. https://doi.org/10.1353/foc.2013.0009
Lara-Cinisomo, S., Chandra, A., Burns, R. M., Jaycox, L. H., Tanielian, T., Ruder, T., & Han, B.
(2012). A mixed-method approach to understanding the experiences of non-deployed
military caregivers. Maternal and Child Health Journal, 16, 374–384.
https://doi.org/10.1007/s10995-011-0772-2
Lavee, Y., McCubbin, H. I., & Patterson, J. M. (1985). The double ABCX model of family stress
and adaptation: An empirical test by analysis of structural equations with latent variables.
Source Journal of Marriage and Family, 47, 811–825. https://doi.org/10.2307/352326
87
Lester, P., Liang, L., Milburn, N., Mogil, C., Woodward, K., Nash, W., … Beardslee, W. (2016).
Evaluation of a family-centered preventive intervention for military families: Parent and
child longitudinal outcomes. Journal of the American Academy of Child and Adolescent
Psychiatry, 55, 14–24. https://doi.org/10.1016/j.jaac.2015.10.009
Lester, P., Peterson, K., Reeves, J., Knauss, L., Glover, D., Mogil, C., … Beardslee, W. (2010).
The long war and parental combat deployment: Effects on military children and at-home
spouses. Journal of the American Academy of Child & Adolescent Psychiatry, 49, 310–
320. https://doi.org/10.1016/j.jaac.2010.01.003
Lincoln, A., Swift, E., & Shorteno-Fraser, M. (2008). Psychological adjustment and treatment of
children and families with parents deployed in military combat. Journal of Clinical
Psychology, 64, 984–992. https://doi.org/10.1002/jclp.20520
Lucier-Greer, M., Arnold, A. L., Mancini, J. A., Ford, J. L., & Bryant, C. M. (2015). Influences
of cumulative risk and protective factors on the adjustment of adolescents in military
families. Family Relations, 64, 363–377. https://doi.org/10.1111/fare.12123
Lucier-Greer, M., O’Neal, C. W., Arnold, A. L., Mancini, J. A., & Wickrama, K. K. A. S.
(2014). Adolescent mental health and academic functioning: Empirical support for
contrasting models of risk and vulnerability. Military Medicine, 179, 1279–1287.
https://doi.org/10.7205/MILMED-D-14-00090
Mansfield, A. J., Kaufman, J. S., Marshall, S. W., Gaynes, B. N., Morrissey, J. P., & Engel, C. C.
(2010). Deployment and the use of mental health services among U.S. Army wives. New
England Journal of Medicine, 362, 101–109. https://doi.org/10.1056/NEJMoa0900177
Masten, A. S. (2001). Ordinary magic: Resilience processes in development. American
Psychologist, 56, 227–238. https://doi.org/10.1037/0003-066X.56.3.227
88
McCubbin, H. I., & Patterson, J. M. (1983). The family stress process. Marriage & Family
Review, 6, 7–37. https://doi.org/10.1300/J002v06n01_02
McGuire, A., Runge, C., Cosgrove, L., Bredhauer, K., Anderson, R., Waller, M., & Nasveld, P.
(2012). Timor-Leste family study: Summary report. Brisbane, Australia: University of
Queensland, Centre for Military and Veterans’ Health.
McNulty, P. A. F. (2010). Adaptability and resiliency of military families during reunification:
Initial results of a longitudinal study. Federal Practitioner, 27(3), 18–27.
McNulty, P. A. F. (2013). Adaptability and resiliency of military families during reunification:
Results of a longitudinal study. Federal Practitioner, 30(8), 14–22.
Minuchin, S. (1974). Families and family therapy. Boston, MA: Harvard University Press.
Muthén, L. K., & Muthén, B. O. (2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthén
& Muthén.
Nurius, P. S., & Macy, R. J. (2008). Heterogeneity among violence-exposed women. Journal of
Interpersonal Violence, 23, 389–415. https://doi.org/10.1177/0886260507312297
Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2007). Deciding on the number of classes in
latent class analysis and growth mixture modeling: A Monte Carlo simulation study.
Structural Equation Modeling, 14, 535–569. https://doi.org/10.1080/10705510701575396
O’Neal, C. W., Lucier-Greer, M., Mancini, J. A., Ferraro, A. J., & Ross, D. B. (2016). Family
relational health, psychological resources, and health behaviors: A dyadic study of
military couples. Military Medicine, 181, 152–160. https://doi.org/10.7205/MILMED-D-
14-00740
Paley, B., Lester, P., & Mogil, C. (2013). Family systems and ecological perspectives on the
impact of deployment on military families. Clinical Child and Family Psychology
89
Review, 16, 245–265. https://doi.org/10.1007/s10567-013-0138-y
Paris, R., DeVoe, E. R., Ross, A. M., & Acker, M. L. (2010). When a parent goes to war: Effects
of parental deployment on very young children and implications for intervention.
American Journal of Orthopsychiatry, 80, 610–618. https://doi.org/10.1111/j.1939-
0025.2010.01066.x
Park, N. (2011). Military children and families: Strengths and challenges during peace and war.
American Psychologist, 66, 65–72. https://doi.org/10.1037/a0021249
Peterson, C., Park, N., & Castro, C. A. (2011). The Global Assessment Tool. American
Psychologist, 66, 10–18. https://doi.org/10.1037/a0021658
Pincus, S. L., House, R., Christenson, J., & Adler, L. E. (2001). The emotional cycle of
deployment: A military family perspective. U.S. Army Medical Department Journal,
4(4), 9–18. Retrieved from https://www.military.com/spouse/military-
deployment/dealing-with-deployment/emotional-cycle-of-deployment-military-
family.html
Renshaw, K. D., Allen, E. S., Rhoades, G. K., Blais, R. K., Markman, H. J., & Stanley, S. M.
(2011). Distress in spouses of service members with symptoms of combat-related PTSD:
Secondary traumatic stress or general psychological distress? Journal of Family
Psychology, 25, 461–469. https://doi.org/10.1037/a0023994
Renshaw, K. D., Rodebaugh, T. L., & Rodrigues, C. S. (2010). Psychological and marital
distress in spouses of Vietnam veterans: Importance of spouses’ perceptions. Journal of
Anxiety Disorders, 24, 743–750. https://doi.org/10.1016/j.janxdis.2010.05.007
Rodriguez, A. J., & Margolin, G. (2015). Military service absences and family members’ mental
health: A timeline followback assessment. Journal of Family Psychology, 29, 642–648.
90
https://doi.org/10.1037/fam0000102
Rosato, N. S., & Baer, J. C. (2012). Latent class analysis: A method for capturing heterogeneity.
Social Work Research, 36, 61–69. https://doi.org/10.1093/swr/svs006
Saltzman, W. R., Lester, P., Beardslee, W. R., Layne, C. M., Woodward, K., & Nash, W. P.
(2011). Mechanisms of risk and resilience in military families: Theoretical and empirical
basis of a family-focused resilience enhancement program. Clinical Child and Family
Psychology Review, 14, 213–230. https://doi.org/10.1007/s10567-011-0096-1
Sullivan, K., Capp, G., Gilreath, T. D., Benbenishty, R., Roziner, I., & Astor, R. A. (2015).
Substance abuse and other adverse outcomes for military-connected youth in California:
Results from a large-scale normative population survey. JAMA Pediatrics, 169, 922–928.
https://doi.org/10.1001/jamapediatrics.2015.1413
Sullivan, K. S., Kintzle, K., Barr, N., Gilreath, T. D., & Castro, C. A. (in press). Veterans’
social/emotional and physical functioning informs perceptions of family and child
functioning. Journal of Military, Veteran, and Family Functioning.
Sullivan, K. S., Hawkins, S. A., Gilreath, T. D., Castro, C. A. (in process). Preliminary
psychometrics and potential big data uses of the U.S. Army Family Global Assessment
Tool.
Trail, T. E., Meadows, S. O., Miles, J. N., & Karney, B. R. (2017). Patterns of vulnerabilities and
resources in U.S. Military families. Journal of Family Issues, 38, 2128–2149.
https://doi.org/10.1177/0192513X15592660
U.S. Department of Defense. (2017). 2016 demographics: Profile of the military community.
Retrieved from http://download.militaryonesource.mil/12038/MOS/Reports/2016-
Demographics-Report.pdf
91
Van Winkle, E. P., & Lipari, R. N. (2015). The impact of multiple deployments and social
support on stress levels of women married to active duty servicemen. Armed Forces &
Society, 41, 395–412. https://doi.org/10.1177/0095327X13500651
Wadsworth, S. M., Cardin, J.-F., Christ, S., Willerton, E., Flittner O’Grady, A., Topp, D., …
Mustillo, S. (2016). Accumulation of risk and promotive factors among young children in
US military families. American Journal of Community Psychology, 57, 190–202.
https://doi.org/10.1002/ajcp.12025
Walsh, F. (1996). The concept of family resilience: Crisis and challenge. Family Process, 35,
261–281. https://doi.org/10.1111/j.1545-5300.1996.00261.x
Walsh, F. (2003). Family resilience: A framework for clinical practice. Family Process, 42, 1–
18. https://doi.org/10.1111/j.1545-5300.2003.00001.x
Weins, T. W., & Boss, P. (2006). Maintaining family resiliency before, during, and after military
separation. In C. A. Castro, A. B. Adler, & T. W. Britt (Eds.), Military life: The
psychology of serving in peace and combat: Vol. 3. The military family (pp. 13–38).
Westport, CT: Praeger Security International.

92
Table 3.1. Sample Demographics and Latent Profile Indicator Means and Prevalence
n (%) M (SD) Min Max
Service member age 33.16 (9.65) 17 66
Service member sex
Male 2,513 (83.1)
Female 295 (9.8)
Service member education
Completed high school or below 1,282 (42.2)
Completed some college or above 1,459 (48.2)
Service member rank
Enlisted 1,959 (64.8)
Officer 849 (28.1)
Service member race and ethnicity
White 1,769 (58.5)
Black 512 (16.9)
Asian 95 (3.4)
American Indian or Alaskan Native 15 (0.5)
Hawaiian or Pacific Islander 23 (0.8)
Hispanic 351 (11.6)
Family demographics
Spouse age 34.45 (8.95) 17 69
Spouse sex
Male 126 (4.2)
Female 1,951 (64.5)
Number of children in the home
0 541 (17.9)
93
1 496 (16.4)
2 601 (19.9)
3 331 (10.9)
4 or more 204 (6.7)
Age of oldest child
0 –5 390 (12.8)
6 –11 425 (14.0)
12 –19 638 (21.0)
20 or older 179 (5.9)
Recent PCS move
Yes 458 (15.1)
No 2,567 (84.9)
Recent reunification
Yes 208 (6.8)
No 2,828 (93.0)
Cumulative days of deployment 0 1,639
Social connectedness 138.58 (202.09) 1 5
Social support 3.26 (1.09) 1 5
Relationship functioning 3.98 (0.89) 1 5
Family cohesion 4.01 (0.86) 1 5
Meaning making 4.09 (0.76) 1 5
Positive coping 4.24 (0.86) 1 5
Spouse mental health diagnosis 3.77 (0.92)
Yes 470 (15.5)
No 2,570 (84.5)

94
Table 3.2. Model Fit Indexes for LPA Model
Profiles BIC LMR-LRT BLRT Entropy
1 85,087.798
2 81,559.652 .0000 .0000 .720
3 80,578.702 .0177 .0000 .707
4 80,024.519 .0063 .0000 .731
5 79,451.129 .0412 .0000 .786
6 78,292.660 .0200 .0000 .839
7 77,841.906 .8057 .0000 .855
Note. Best fitting model indicated in bold. BIC = Bayesian information criterion; LMR-LRT = Lo-Mendell-Rubin
likelihood ratio test; BLRT = bootstrap likelihood ratio test.

95
Table 3.3. Means and Conditional Probabilities for Risk and Protective Factor Indicators
Moderate
Protection,
Low Risk
Lowest
Protection,
Low Risk
High
Protection,
Moderate Risk
Moderate
Protection,
Moderate Risk
High
Protection,
Highest Risk
High
Protection,
Low Risk
Class prevalence 18.91% 2.67% 20.03% 8.14% 6.52% 43.74%
Protective factors
Social connections 2.67 1.76 3.59 2.28 3.48 3.83
Social support 3.47 2.54 4.31 3.07 4.13 4.47
Relationship functioning 3.61 3.03 4.24 3.34 4.29 4.35
Family cohesion 3.69 2.88 4.36 3.57 4.28 4.40
Positive coping 3.36 2.12 4.07 2.83 3.85 4.28
Meaning making 3.91 2.01 4.55 3.23 4.38 4.74
Risk factors
Recent relocation
No 0.88 0.90 0.80 0.72 0.73 0.89
Yes 0.12 0.10 0.20 0.28 0.27 0.11
Recent reunification
No 0.99 0.99 0.88 0.81 0.67 0.99
Yes 0.01 0.01 0.12 0.19 0.33 0.01
Total days of deployment 9.89 10.28 321.58 341.26 634.02 7.60
Note. Figures represent means for protective factors and days of deployment, and conditional probabilities for recent
relocation and recent reunification.

96
Table 3.4. Service Member Covariates Associated with Latent Profiles
Moderate Protection,
Low Risk
Lowest Protection,
Low Risk
High Protection,
Moderate Risk
Moderate Protection,
Moderate Risk
High Protection,
Highest Risk
Covariates b SE OR b SE OR b SE OR b SE OR b SE OR
Some college (vs. high school) -0.28 0.19 0.75 0.22 0.32 1.25 0.62** 0.16 1.86 0.54 0.19 1.72 0.39 0.23 1.48
White (vs. non-White) 0.13 0.15 1.14 0.61* 0.30 1.85 0.33* 0.13 1.39 0.82** 0.19 2.28 0.84** 0.21 2.32
Enlisted (vs. officer) 0.12 0.23 1.13 1.27* 0.51 3.56 -0.39* 0.16 0.68 0.29 0.22 1.33 -0.47* 0.23 0.63
Note. High protection and low risk (Profile 6) was the reference category. Due to listwise deletion, n = 2,710 in these analyses.
*p < .05. **p < .001.

97
Table 3.5. Latent Profiles Associated with Distal Outcome (Mental Health Diagnosis)
Outcome 1 2 3 4 5
% (SE) χ
2
χ
2
χ
2
χ
2
χ
2

1. Moderate Protection,
Low Risk
8.1 (0.008)
2. Lowest Protection,
Low Risk
45.4 (0.070) 28.480***
3. High Protection,
Moderate Risk
17.2 (0.024) 13.488*** 16.719***
4. Moderate Protection,
Moderate Risk
30.5 (0.028) 60.564*** 3.781* 11.890**
5. High Protection,
Highest Risk
26.0 (0.034) 26.708*** 5.936* 4.249* 1.078
6. High Protection, Low
Risk
14.3 (0.016) 11.087** 19.115*** 1.018 25.109*** 9.636**
Note. Degrees of freedom = 1 for all individual chi-square results.
*p < .05. **p < .01. ***p < .001.

98
Chapter 4 (Study 3): Mental Health Outcomes Associated with Profiles of Risk and
Resilience among Military-Connected Youth
Abstract
The purpose of this study was to describe patterns of risk and protective factors affecting U.S.
Army families and the association of these patterns with mental health diagnoses among
military-connected children. Protective factors, including marital and family functioning, social
connections, and social support, were drawn from a survey of Army spouses, the Comprehensive
Soldier and Family Fitness Global Assessment Tool. Risk factors, including the mental health of
the at-home parent, family moves, and deployment and reunification experiences, were drawn
from U.S. Department of Defense archival data. Rates of mental health diagnoses among
military-connected youth were pulled from U.S. Department of Defense health records. Using
latent profile analysis, this study identified five distinct profiles with significant variability in
access to resources and exposure to risk. The largest group of families (40% of the sample)
reported considerable strengths and limited exposure to risk. Overall, 17.5% of military-
connected youth received a mental health diagnosis in the year following survey completion. The
three-step method was used to associate latent profiles with covariates and distal outcomes.
Statistically significant differences in the prevalence of mental health diagnoses among military-
connected youth were observed across profiles (χ
2
= 19.893, df = 4, p < .001), with the lowest
rates (10.6%) found in the group with the highest levels of protective factors and lowest exposure
to risk. Findings suggest that most military families are faring well. Further, results highlight the
importance of a thorough assessment that evaluates both the stressors military families face and
the strengths and resources they possess.

99
Introduction
In today’s all volunteer U.S. Armed Forces, 40.5% of the 2.1 million military personnel
are parents. In these military families, almost 70% of children are 11 years old or younger,
suggesting that for most military-connected children, our nation has been at war for longer than
they have been alive (U.S. Department of Defense, 2017). As U.S. engagement in international
conflicts has continued through a second decade, concern about how military families and
military-connected children are managing wartime stressors has increased. A growing body of
empirical evidence has suggested that wartime military service, and the deployment cycle in
particular, are associated with increased risk of adverse health and mental health outcomes for
military-connected youth, including increased mental health diagnoses, substance use, poor
academic achievement, victimization, weapon-carrying, and problematic behavior (Cederbaum
et al., 2014; Engel, Gallagher, & Lyle, 2010; Flake, Davis, Johnson, & Middleton, 2009;
Gorman, Eide, & Hisle-Gorman, 2010; Lester et al., 2010; Sullivan et al., 2015). These empirical
findings represent a public health concern because poor developmental outcomes can have
compounding negative effects into adulthood.
More recently, findings have suggested that rather than exposure to any one stressor, it
may be cumulative risk exposure that determines outcomes for military-connected youth (Lucier-
Greer, Arnold, Mancini, Ford, & Bryant, 2015; Wadsworth et al., 2016). Further, experts have
noted that frequently studies fail to account for the many strengths of military families, including
the wide array of both informal and formal supports available to them (Cozza, Haskins, &
Lerner, 2013; Hosek & Wadsworth, 2013). Although much knowledge has been generated about
the impact of individual stressors on military-connected youth outcomes, less attention has
focused on the unique constellation of risk and protective factors experienced by this population.
100
Using a latent profile analysis (LPA) approach, the present study addressed this gap in the
literature by exploring patterns of both cumulative risk and concurrent protective factors in U.S.
Army families and the association of these patterns with mental health outcomes among military-
connected youth.
Theoretical Foundation
Family systems theory. Family systems theory is a broad perspective that describes
reciprocal patterns of influence among family members and highlights the importance of
understanding the family context for individual behavior (Cox & Paley, 2003; Minuchin, 1974).
Recent research with veteran families demonstrated the mediating role of family functioning in
the relationship between veteran functioning and mental health outcomes for other system
members (Sullivan et al., in press). Further, evidence has highlighted the importance of
considering the reciprocal nature of family relationships and the need to consider individual
behavior in a family system context (O’Neal, Lucier-Greer, Mancini, Ferraro, & Ross, 2016).
Ultimately, the family systems perspective suggests that risk and protective factors likely affect
families at both the individual and family levels.
Risk and protective factors. Risk factors have been defined as conditions that increase
the likelihood of an adverse outcome, whereas promotive or protective factors decrease this
likelihood (Fergus & Zimmerman, 2005). It has been consistently demonstrated in the civilian
literature that the cumulative effects of risk factors exceed the impact of individual stressors
(Zimmerman et al., 2013). These findings have been replicated among military-connected youth
(Lucier-Greer et al., 2015; Lucier-Greer, O’Neal, Arnold, Mancini, & Wickrama, 2014;
Wadsworth et al., 2016). Among younger children (aged 10 and younger), 11% of those exposed
to one risk factor were considered at risk of adverse developmental or mental health outcomes,
101
compared to 40% of those exposed to four or more risk factors (Wadsworth et al., 2016). Among
older adolescents, higher levels of cumulative risk, both military-specific and more normative
risk factors, were associated with increased depressive symptoms and lower levels of academic
performance (Lucier-Greer et al., 2015).
Resilience. Despite evidence suggesting an increased risk of adverse outcomes, many
military families appear to fare well despite exposure to stressors, suggesting that resilience
processes may be affecting outcomes (Cozza et al., 2013; Easterbrooks, Ginsburg, & Lerner,
2013). In contrast to deficit-oriented approaches, the present study relied on a family resilience
perspective, which encompasses the interactional and systemic processes occurring in family
systems that enable families to thrive despite adversity (Walsh, 2003). The concept of resilience
has been variably defined in the literature. The perspective guiding this study conceptualizes
resilience as a dynamic balance of risk and protective factors that permit families to maintain or
return to previous levels of well-being and functioning in response to adversity (Hawkins et al.,
2017). This theoretical perspective points to the importance of examining the simultaneous
impact of both risk and protective factors. Rather than finding they operate additively, empirical
research with military-connected youth has suggested that protective factors buffer the adverse
impact of cumulative risk (Wadsworth et al., 2016).
Developmental cascades. Developmental cascades have been defined as “the cumulative
consequences for development of the many interactions and transactions occurring in developing
systems that result in spreading effects across levels, among domains at the same level, and
across different systems or generations” (Masten & Cicchetti, 2010, p. 491). Cascade effects
capture the spillover of stress associated with military service from service members to other
members of the family system, potentially compounding direct effects of these stressors (Masten,
102
2013). Military spouses, in particular, have been conceptualized as the lynchpin of healthy
family functioning (Green, Nurius, & Lester, 2013), and their well-being provides a secure
foundation for a healthy attachment system across the deployment cycle (Riggs & Riggs, 2011).
Empirically, poor spousal adjustment, particularly during deployments, is a potent cascading risk
factor that has been associated with more adverse outcomes for children in these family systems
(Chandra, Lara-Cinisomo, et al., 2010; Flake et al., 2009; Lester et al., 2010).
Empirical Evidence of Risk and Protective Factors
Risk factors affecting military children. In addition to the cascading effects of the
mental health of the at-home parent, the deployment cycle has been demonstrated to have
negative consequences for a significant subset of military youth, including increased depressive
symptoms, suicidal ideation, increased substance use, and lowered academic engagement and
achievement (Cederbaum et al., 2014; Chandra, Martin, Hawkins, & Richardson, 2010; Engel et
al., 2010; Flake et al., 2009; Gilreath et al., 2013; Lester et al., 2010). Cumulative experiences of
deployment, which often reflect the high operational tempo of current overseas conflicts, appear
to be a particularly potent predictor of negative outcomes for youth (Lester et al., 2010;
Mansfield, Kaufman, Engel, & Gaines, 2011). Less research has focused on the impact of family
reunification following deployment, but there appears to be an association between emotional
challenges in military-connected youth and difficulties experienced during reintegration
following a deployment (Chandra, Lara-Cinisomo, et al., 2010). Qualitative evidence has
suggested that this is a stressful period, often characterized by worry for military-connected
youth (Esposito-Smythers et al., 2011; Huebner, Mancini, Wilcox, Grass, & Grass, 2007).
Further, particularly in families in which the soldier’s mental health has been impacted,
reintegration has had negative effects on parenting and overall family adjustment, which may
103
affect outcomes among youth (Allen, Rhoades, Stanley, & Markman, 2010; Gewirtz, Polusny,
DeGarmo, Khaylis, & Erbes, 2010; Sayers, Farrow, Ross, & Oslin, 2009). Finally, family moves
and associated changes in schools may be a risk factor for poor academic outcomes and
increased mental health difficulties (Bradshaw, Sudhinaraset, Mmari, Blum, & Hopkins, 2010;
Richardson, Mallette, O’Neal, & Mancini, 2016).
Protective factors affecting military children. Although they have been less frequently
studied in this population, evidence exists for protective factors that reduce the likelihood of
adverse outcomes among military-connected youth. Specifically, access to social support has
been found to be protective. Living on a military base where access may be more immediate to
both formal support programs and informal support from military peers appears to be associated
with fewer negative outcomes (Chandra, Lara-Cinisomo, et al., 2010). In qualitative work,
military-connected youth highlighted the importance of social support from other military
children as particularly important in coping with deployment stressors (Huebner & Mancini,
2010; Mmari, Roche, Sudhinaraset, & Blum, 2009). Additionally, youth from military families
that are more connected to their community and perceive more support from these connections
may be less likely to experience adverse outcomes (Conforte et al., 2017; Flake et al., 2009;
Lucier-Greer et al., 2014; Richardson et al., 2016). Finally, several interrelated family-level
processes, including family cohesion and healthy marital and parenting relationships, are likely
protective for youth, although it is challenging to tease apart the complexities of these
relationships in a family system (Finkel, Kelley, & Ashby, 2003; Foran, Eckford, Sinclair, &
Wright, 2017; Paley, Lester, & Mogil, 2013).
Outcomes for military children. Although there is evidence of resilience processes
among many military-connected youth, research also has suggested that a significant subset of
104
this population experience a wide array of adverse outcomes. Being from a military family
during wartime and exposure to specific military stressors, including deployment, have been
associated with changes in physical health including increased heart rate (Barnes, Davis, &
Treiber, 2007) and greater likelihood of physical injury (Hisle-Gorman et al., 2015); more
frequent outpatient behavioral health visits and higher prevalence of diagnosed mental health
disorders (Gorman et al., 2010; Mansfield et al., 2011); increased depressive symptoms and
suicidality (Cederbaum et al., 2014; Gilreath et al., 2016); decreased engagement in school and
poor academic performance (Bradshaw et al., 2010; Engel et al., 2010);

increased recent and
lifetime use of substances (Gilreath et al., 2013; Sullivan et al., 2015); experiences of violence
perpetration and victimization (Gilreath, Astor, Cederbaum, Atuel, & Benbenishty, 2014;
Sullivan et al., 2015); and disrupted attachment relationships (Holmes, Rauch, & Cozza, 2013;
Jordan et al., 1992; Solomon, Debby-Aharon, Zerach, & Horesh, 2011). Although many adverse
outcomes have been observed in this population, the present study focused on the prevalence of
mental health diagnoses among military-connected youth.
The Current Study
Masten (2001) discussed two distinct approaches to resilience-focused studies aimed at
understanding variation in adverse outcomes among youth: variable-focused methods and
person-focused methods. Most previous investigations of cumulative risk have followed a
variable-focused approach in which variables are dichotomized at meaningful cutoff points and
each risk or protective factor is assigned a score of 1, which represents the presence of this risk,
or 0, which represents little to no risk. Scores across multiple risk or protective factor variables
are then summed to provide a cumulative measure (Lucier-Greer et al., 2015, 2014; Wadsworth
et al., 2016), which is associated with outcomes using regression or structural equation modeling
105
approaches. The present study used a person-centered approach, LPA, to operationalize the
impact of multiple concurrent stressors and protective factors on youth outcomes. LPA is a form
of mixture modeling, a useful methodology to uncover meaningful heterogeneity in a population
(Nurius & Macy, 2008; Rosato & Baer, 2012), and has previously been used to explore military
family functioning and youth outcomes (Okafor, Lucier-Greer, & Mancini, 2016; Oshri et al.,
2015).
For this study, LPA is a useful tool for several reasons. First, resilience has been
conceptualized as a dynamic balance between risk and protective factors. This process-oriented
definition requires a statistical modeling technique that allows for simultaneous evaluation of
both risk and protective factors. Second, this approach allows for the exploration of naturally
occurring patterns of risk and protective factors and can uncover uncommon configurations that
may be masked using more traditional variable-focused methods (Masten, 2001). LPA can define
unique subgroups of military families for whom outcomes can then be evaluated. Three specific
hypotheses guided these analyses:
1. Significant heterogeneity would exist in levels of protective factors and exposure to
risk factors among military families.
2. A higher prevalence of mental health disorders would exist among military-connected
youth from families that reported lower levels of protective factors and higher
exposure to risk.
3. Elevated levels of mental health diagnoses among at-home parents, which represent
cascade effects in this model, would emerge as a particularly potent risk factor that
defines latent profiles and is associated with higher levels of mental health diagnoses
among military-connected youth.
106
Methods
Family Global Assessment Tool
Comprehensive Soldier and Family Fitness is a U.S. Army program designed to enhance
the physical, social, emotional, family, and spiritual health of Army soldiers and families. To
evaluate these domains of fitness, the Family Global Assessment Tool (GAT) was developed
based on expert opinion and previously validated measures (Peterson, Park, & Castro, 2011).
Army spouses access the survey via an online portal and receive immediate feedback about their
family’s well-being. Participation is voluntary, and spouses can complete the instrument as many
times as they choose. More information about the GAT survey procedures has been published
elsewhere (Hawkins et al., 2017). Recent psychometric validation of the Family GAT identified
seven scales measuring family strengths and functioning with preliminary evidence of good
reliability and validity (Sullivan et al., in process).
Data Linkage
Family GAT data were linked to additional U.S. Department of Defense (DOD) data
sources, which provided information on personnel, military transitions, operations, and medical
outcomes, using the DOD’s Person-Event Data Environment. This cloud-based, virtual
environment stores the DOD’s vast administrative data holdings and offers researchers a means
to link and analyze data from disparate sources.
Initially, data from an approved cohort of spouses, who completed the Family GAT
survey between 2014 and 2016, were identified and merged with Family GAT survey data using
a unique identification number and retaining only first GAT completions. Data linkage
proceeded with merging several DOD administrative datasets. These records are organized by
“snapshots,” in which a new line of data is created for each individual on a quarterly, biannual,
107
or annual basis (depending on the dataset) only if something has changed in their record. Using a
unique identification number to link members of the cohort, Active Duty and Reserve Master
Personnel, Family Sponsor, and Family Dependent files were linked to the cohort dataset.
Because multiple rows of data could exist for each family, information from the closest snapshot
date before each spouse’s Family GAT completion date was retained. Finally, deployment data
for each family were identified using the unique identifier. Information on all deployments
dating back to the beginning of 2009 (the earliest date of available data) was retained in the final
dataset, regardless of snapshot date.
Finally, medical data, which is recorded in four datasets for military and civilian inpatient
and outpatient facilities, were merged. This process, which was the same for each dataset,
proceeded in three steps. Initially, only medical records for members of the approved cohort and
their dependents were identified and retained using their unique identification number. Next,
visits for military spouses that occurred in the year preceding GAT completion were retained. In
this group, duplicate records were discarded such that only one record of each unique diagnosis
for any given spouse was retained in the dataset. Finally, visits for military dependents that
occurred in the year following GAT completion were retained. In this group, duplicated records
were discarded such that only one record of each unique diagnosis for any given dependent was
retained.
Participants
The primary respondent for each U.S. Army family in the sample was a civilian spouse.
Spouses were included in the sample if: (a) they completed a Family GAT survey between 2014
and 2016; (b) they provided consent to have their GAT survey responses used for research
purposes; (c) they could be linked to an Army soldier using a unique electronic identifier; and (d)
108
archival records indicated that the family had at least one dependent child. The final sample
consisted of 1,630 U.S. Army families. Secondary analyses were approved by the Army
Research, Development and Engineering Center Institutional Review Board and the institutional
review board at the [blinded for review].
Measurement
Data that informed risk factors, protective factors, covariates, and mental health outcomes
were drawn from different sources, described here.
Risk factors. Based on previous research, four military-specific risk factors were
included in models: (a) experiencing a family move (permanent change of station [PCS]) in the
last year; (b) cumulative days of deployment; (c) experiencing a reunification following
deployment in the last year; and (d) whether the at-home parent received a mental health
diagnosis in the preceding year. Datasets managed by the Defense Manpower Data Center were
employed to operationalize these variables.
Recent PCS move. Using Master Personnel File data, each of the family’s permanent
duty station arrival dates was compared to the date of GAT survey completion to create a
dichotomous variable (yes/no). Arrival dates that fell within 365 days preceding GAT
completion were deemed to reflect a recent PCS move.
Cumulative deployment days. Using Contingency Tracking System Deployment data, the
difference in days between each deployment’s begin and end dates was calculated and then all
deployment days were summed to create a cumulative, continuous measure of deployment. All
deployment days between 2009 (the earliest date of available data) and the date of GAT survey
completion were included in these calculations.
109
Recent reunification. Using Contingency Tracking System Deployment data,
deployment end dates were compared with GAT completion dates to create a dichotomous
variable (yes/no). Deployment end dates that fell within 365 days preceding GAT completion
were deemed to reflect a recent reunification.
At-home parent mental health diagnosis. Using Defense Health Agency data, a
dichotomous variable was created to reflect whether the at-home parent (military spouse)
received a mental health diagnosis in the year preceding GAT completion. A medical record
containing a mental-health related International Classification of Diseases, 9th Revision (ICD-9)
or International Classification of Diseases, 10th Revision (ICD-10) diagnosis code was deemed
to reflect the presence of a mental health diagnosis. To identify mental health codes, this study
employed procedures used by Mansfield and colleagues (2011, 2010), who developed a
comprehensive list of ICD-9 codes for 17 categories of mental health disorders. This list was
expanded to include ICD-10 codes, because data from the cohort spanned the transition from
ICD-9 to ICD-10. Ultimately, the presence of any of these codes in the medical record of a
military spouse in the year preceding GAT completion was represented with one dichotomous
item (yes/no).
Protective factors. Based on theory and empirical evidence previously discussed, four
protective factor scales drawn from the Family GAT were included in these analyses. Military
spouses reported on protective factors affecting the family system. Responses to individual items
were recoded such that higher scores indicated positive functioning. Composite scores were
created by averaging scores on individual items.
Relationship functioning. The functioning of the marital relationship was assessed with
eight Likert-type items, all of which had five responses options, from strongly disagree to
110
strongly agree. Items included: “I feel emotionally distant from my partner” and “My partner
and I clearly communicate our expectations for each other.” These items had excellent internal
consistency in this sample (α = .93).
Social support. Social support for the family was assessed with four Likert-type items, all
of which had five response options, from strongly disagree to strongly agree. Items included:
“There is someone I could turn to for advice on how to deal with a personal or family problem”
and “If I was sick, I could find someone to help with my daily chores.” These items had good
internal consistency in this sample (α = .86).
Social connections. The family’s social connectedness was assessed with five Likert-
type items, all of which had five response options, from never to most of the time. Items
included: “I have a good relationship with people in my neighborhood” and “I participated in
community events, activities, or meetings.” These items had excellent internal consistency in this
sample (α = .91).
Family cohesion. Family cohesion was assessed with three Likert-type items, all of
which had five response options, from strongly disagree to strongly agree. Items included: “My
family expresses tenderness” and “My family confides in each other.” These items had good
internal consistency in this sample (α = .86).
Covariates. Using data from the Defense Manpower Data Center’s Master Personnel
File, three variables served as covariates: (a) service member race and ethnicity (White, non-
White); (b) service member educational attainment (high school or below, some college or
above); and (c) service member military rank (enlisted, officer).
Mental health outcomes. Whether a military-connected youth was given a mental health
diagnosis was the outcome of interest in these analyses. To create this outcome variable, a
111
similar procedure to that used to capture the presence of mental health diagnoses for the at-home
parent was used. Youth were deemed to have received a mental health diagnosis if they were
assigned an ICD-9 or ICD-10 mental health-related code during an inpatient or outpatient
medical visit within 1 year following GAT completion. The presence of any of these codes was
represented with one dichotomous item reflecting receipt of a mental health diagnosis or not.
Analyses
The three-step method of associating an LPA model with covariates and distal outcomes
(presence of a mental health diagnosis among military-connected youth in these analyses) was
used. In Step 1 of this process, an LPA model is specified using risk and protective factor
indicator variables. In Step 2, a nominal most likely class variable is created, along with a
measure of the error in classification. In Step 3, the most likely class variable is associated with
the covariates or distal outcome while accounting for classification error (Asparouhov &
Muthén, 2014). In these analyses, Step 1 involved the inclusion of all continuous and categorical
risk and protective factor variables in the LPA model simultaneously, using summary scores for
protective factors to preserve power. An initial 1-profile solution was compared to successive
models with an increasing number of profiles using statistical indicators to determine model fit.
Indicators included: (a) low Bayesian information criterion (BIC); (b) significant Lo-Mendell-
Rubin likelihood ratio test (LMR-LRT); (c) significant bootstrap likelihood ratio test (BLRT);
and (d) conceptual and theoretical considerations (Nylund, Asparouhov, & Muthén, 2007).
Substantive conceptual and theoretical considerations guided model selection when statistical
indicators suggested different solutions. The AUXILIARY option in Mplus was used to
complete Steps 2 and 3 of the three-step process (Asparouhov & Muthén, 2014; Muthén &
Muthén, 2012). For covariates, the R3STEP command was utilized; for distal outcomes, the
112
DCAT command was used. SPSS version 21 was used for data cleaning and descriptive
statistics; Mplus version 7 was used for LPA models (Muthén & Muthén, 2012). Full
information maximum likelihood, available in Mplus, was employed to handle missing data in
LPA models. Models including covariates used listwise deletion, which resulted in the
elimination of 65 cases (4.0% of the sample).
Results
Sample demographics are presented in Table 4.1. The majority of service members in this
sample were White, male, and enlisted; most reported completing at least some college
education. Most of the spouses in the sample were female. Approximately one-third of families
reported having one child, one-third reported having two children, and one-third had three or
more children. In the year preceding GAT completion, the at-home parent had a mental health
diagnosis in approximately 20% of families; 19% of families experienced a move; and 9% of
families experienced a reunification following a deployment. On average, families reported 194
cumulative days of deployment. Approximately 18% of children received a mental health
diagnosis in the year following GAT survey completion.
LPA Model
Fit indexes for the LPA model are displayed in Table 4.2; conditional probabilities and
means for LPA indicators by profiles are displayed in Table 4.3. BIC continued to decrease
beyond the 5-profile solution, and BLRT remained significant. LMR-LRT suggested a 5-profile
solution, but this indicator may not be trustworthy in this model because significance did not
remain consistent beyond the 2-profile solution. Nevertheless, the 5-profile solution was chosen
because this model was parsimonious, clinically meaningful, and offered good separation
between classes, as indicated by an entropy value of .855.
113
Profile 1 (low protection and low risk) accounted for 11.84% of the sample. Although
there was limited variability among the dichotomous risk indicators across profiles, at-home
parents in this profile had higher probability of a mental health diagnosis. Families had a lower
probability of experiencing a recent move or reunification and mean cumulative deployment
days were low compared to other profiles. Further, spouses endorsed comparatively lower levels
of all four protective factors. Profile 2 (low protection and moderate risk) accounted for 10.25%
of the sample. Spouses in this profile endorsed very similar levels of protective factors compared
to Profile 1. However, in Profile 2, at-home parents had the highest probability of having a
mental health diagnosis and families were more likely to have experienced a recent move or
reunification. Further cumulative days of deployment were elevated, on average, compared to
Profile 1. Profile 3 (high protection and moderate risk) accounted for 29.26% of the sample
overall. On average, families in this profile experienced similar cumulative days of deployment
and were similarly likely to have experienced a recent reunification compared to Profile 2.
However, at-home parents in this profile had a lower probability of a mental health diagnosis and
families had a slightly lower probability of a recent move. Spouses in Profile 3 also reported
higher levels of protective factors compared to Profile 2. Profile 4 (high protection and low risk)
was the largest group, accounting for 40.0% of the sample. Spouses in this profile reported the
highest levels of protective factors. Similarly, at-home parents in this group had the lowest
probability of a mental health diagnosis, families had the lowest probability of a recent move or
reunification, and families had similar cumulative deployment days, on average, to Profile 1.
Finally, Profile 5 (high protection and high risk) accounted for 8.65% of the sample. Spouses in
this category reported high levels of protective factors, similar to Profiles 3 and 4, but had a
higher probability of a mental health diagnosis and a recent move. Families in Profile 5 also
114
experienced many more cumulative deployment days on average compared to other profiles and
had a higher probability of a recent reunification than any other profile.
Covariates
Associations between covariates and latent profiles are displayed in Table 4.4. Overall,
there were relatively few significant relationships between covariates and latent profiles.
Families of service members who completed at least some college were more likely to be in the
High Protection and Moderate Risk profile than the High Protection and Low Risk profile,
compared to families of service members with less education. Compared to families of non-
White service members, families of White service members were more likely to be in the Low
Protection and Moderate Risk profile than the High Protection and Low Risk profile. Finally,
compared to the families of officers, families of enlisted service members were more likely to be
in the High Protection and High Risk profile compared to the High Protection and Low Risk
profile.
Distal Outcomes
Distal outcome results are presented in Table 4.5. Omnibus chi-square results indicate
significant differences in the prevalence of mental health diagnoses among youth across latent
profiles (χ
2
= 19.893, df = 4, p < .001). Individual chi-square tests evaluated statistically
significant differences in the prevalence of mental health diagnoses between two profiles. These
results, displayed in Table 4.5, indicate significant differences in the prevalence of youth mental
health diagnoses between Profile 4 (high protection and low risk) and all other profiles. Overall,
rates of mental health diagnoses among military-connected youth were relatively similar across
four of five profiles, ranging from 19.7% in the Profile 1 (low protection and low risk) to 28.0%
115
in Profile 3 (high protection and moderate risk). In Profile 4, mental health diagnoses were
significantly less prevalent (10.6%).
Discussion
The purpose of the present study was to describe patterns of risk and protective factors
affecting Army families and the association of these patterns with mental health diagnoses
among military-connected children. Approximately 18% of military-connected youth in this
sample received a mental health diagnosis at an inpatient or outpatient medical visit in the year
following survey completion. Mansfield et al. (2011) found an overall rate of 16.7% during their
4-year study from 2003 to 2006. The rate of mental health diagnoses observed here indicates an
increase in prevalence, particularly when considering the difference in time span between the
two studies (1 year in the current study versus 4 years in the Mansfield study). Although various
explanations are possible, this increase in observed prevalence of mental health conditions
among military-connected youth may be reflective of longer cumulative exposure to war-related
stressors, because data for this study were collected 10 years later. Building on prior findings,
this work highlights the relevance of family-level risk and protective factors to understanding the
potential for adverse mental health outcomes among military-connected youth.
Findings also reflect considerable heterogeneity among military families, confirming the
first hypothesis. LPA identified five distinct groups with significant variation in their access to
protective factors and exposure to risk factors. In the LPA results, the largest profile, accounting
for 40% of the sample, was composed of families that experienced comparatively little exposure
to risk factors and high levels of protective factors. Further, although there was variation in the
levels of protective factors across profiles, even in the lowest protection categories (Profiles 1
and 2), families reported moderate levels of intrafamilial and community-level resources. These
116
findings highlight the strengths in military families, even among those exposed to higher levels
of risk. These protective factors represent potentially useful resources on which to draw for the
purposes of prevention and intervention with this population.
Regarding risk factors, slightly more than half of the families in this sample were
grouped into two categories (Profiles 1 and 4) with relatively low levels of risk. Even in these
profiles, however, there was some low-level exposure to risk, with the prevalence of mental
health diagnoses among at-home parents ranging between 12% and 19% and the probability of
experiencing a recent move between 12% and 18%. Military service involves exposure to a
certain level of risk, which is increasingly understood to apply more broadly to the families of
service members (Burrell et al., 2006). The families in these two profiles likely represent this low
level of systemic risk consistent with military service. In contrast, just under half of the families
in the sample were grouped into three profiles (Profiles 2, 3, and 5) with higher levels of risk
exposure, including higher prevalence of mental health disorders among at-home parents, greater
probability of experiencing a recent relocation or reunification, and considerably more
cumulative days of deployment. For example, in Profile 5 (high protection and high risk), a third
of families recently reunified following a deployment and also reported more than 600
cumulative days of deployment on average. The experiences of families in these profiles likely
typify the high operational tempo that has characterized current overseas conflicts.
Although limited significant associations existed between covariates and latent profiles,
the relationship between profiles and the prevalence of mental health diagnoses among military-
connected youth highlights the importance of examining risk and protective factors
simultaneously. Overall, the heterogeneity across profiles was associated with a statistically
significant difference in the prevalence of mental health diagnoses among military-connected
117
youth. Specifically, military-connected youth from families in the largest profile (Profile 4; high
protection and low risk) were significantly less likely than those in all other profiles to receive a
mental health diagnosis. Only 10.6% of youth from families in this category received a
diagnosis, compared to rates ranging from 19.7% to 28.0% across the other four, smaller
profiles. Because families in this group had the highest levels of protective factors and the lowest
exposure to risk factors, these results confirm the second hypothesis. From a policy perspective,
this finding suggests that many military families are likely faring well and have good internal
and external resources to drawn on during periods of increased stress. The need for military-
sponsored support programs is likely limited in this group. Thus, resources being directed to
universal prevention programs, which would be important if a broader group of families
appeared to be at risk, might be better used for targeted prevention and intervention services
directed specifically at those families with higher risk exposure and fewer resources.
Importantly, for military-connected youth in this sample, the combination of both lower
risk exposure and higher levels of protective factors constituted a critical difference. Profiles
with low levels of risk alone (e.g., Profile 1) or high levels of protection alone (e.g., Profiles 3
and 5) exhibited significantly higher rates of mental health diagnoses among youth. This finding
is consistent with the definition of resilience that guided these analyses. Although many varied
definitions of resilience exist, this study defined resilience as a dynamic balance of risk and
protective factors that allow a family to maintain or return to a previous level of functioning
(Hawkins et al., 2017; Walsh, 1996, 2003). These findings suggest a critical balance between
low to moderate risk exposure and considerable access to protective factors that is significantly
associated with lower rates of adverse outcomes. Further, these results highlight the importance
118
of evaluating both stressors and resources when assessing youth from military families, because
both may be crucial to determining possible outcomes.
The third hypothesis was not confirmed. Based on previous literature that has
demonstrated a strong connection between the functioning of the at-home parent and the well-
being of military-connected youth (Chandra, Lara-Cinisomo, et al., 2010; Flake et al., 2009;
Green et al., 2013; Lester et al., 2010) and the theoretical concept of developmental cascades
(Masten & Cicchetti, 2010), diagnosed mental health conditions among at-home parents were
expected be a particularly potent risk factor in defining latent profiles and associated with more
mental health conditions among youth. Rather than this risk factor standing alone to define a
particular profile or profiles, families in this sample had a low to moderate probability of
experiencing this risk across profiles. Profile 2 (low protection and moderate risk) had the
highest rates of mental health diagnoses among at-home parents at 35%. Had the hypothesis been
correct, higher rates of mental health diagnoses would have existed among military-connected
youth in this category, but this was not borne out by the data. Rather, rates among military-
connected youth from families in this profile were lower and not statistically different from rates
in profiles in which the prevalence of diagnoses among at-home parents was much lower. Likely
the diffusion of this risk factor across profiles explains the lack of statistically significant
associations with mental health outcomes. Given this diffusion, variable-focused methods that
can zero in on the relationship between these two variables are likely a better approach to further
explore this relationship. Other possible explanations may explicate why these low to moderate
rates were observed across groups. First, this finding may reflect the impact of systemic risk that
accompanies military service. Additionally, growing evidence has suggested that military
spouses are more likely to have accumulated more adverse childhood experiences prior to
119
marriage (Oshri et al., 2015), which has been associated with elevated rates of mental health
conditions in adulthood (Chapman et al., 2004; Logan-Greene, Green, Nurius, & Longhi, 2014).
If these spouses are evenly spread across groups, this may account for the pattern observed here.
Finally, these results highlight the usefulness of considering person-focused methods in
addition to more traditional variable-focused approaches. Although variable-focused statistical
techniques are helpful for understanding causality and the relationships among variables, these
methods may mask important differences between subgroups of military families, which can
have the unintended consequence of overpathologizing military-connected youth (Rosato &
Baer, 2012). Using regression or structural equation modeling, empirical research has
consistently demonstrated an elevated risk of poor health and mental health outcomes among
military-connected youth (Gorman et al., 2010; Mansfield et al., 2011). Although these findings
are accurate and point to a critical need for services for this population, the large group of
military families that emerged in this study that have good supports and relatively little risk
exposure are often overlooked. Using both person- and variable-focused methods can provide a
more comprehensive approach to understanding the functioning of military families. For
example, the current study explored the collective impact of multiple stressors simultaneously on
rates of adverse outcomes. A complementary variable-focused study may be able to tease out
whether particular stressors, like deployment or the mental health of the at-home parent, are
more or less important for predicting outcomes.
Limitations
Despite the strengths of this study, there are limitations to consider. This sample was
composed entirely of Army families, which may complicate comparisons to other branches of
the military. This was also a voluntary sample of families that completed the Family GAT
120
survey. Although these data were drawn from across the country and the demographics are
relatively similar to the demographics of the military more broadly (U.S. Department of Defense,
2017), care should still be taken in generalizing these results. Additionally, these analyses were
cross-sectional, although temporality was considered when linking archival data. As a result,
associations between latent profile membership and mental health diagnoses cannot be
considered causal. Further, because DOD archival data were collected with the service member
as the focal point, there was limited information available about children in these families. As
such, this study could not explore the impact of important demographic characteristics of
military-connected youth, including age and gender.
Finally, because DOD archival data were used to inform risk in these analyses, many
potential variables of interest were not available for inclusion in models, including military-
specific stressors like the mental health of the service member. Further, recent evidence has
highlighted the importance of normative stressors, which may affect outcomes for all families,
beyond the impact of military-specific stressors alone (Lucier-Greer et al., 2014). These analyses
did not find many significant effects of minority status, low educational attainment of the
military parent, or military rank on latent profile membership. Nevertheless, other normative
stressors, not included here, could have had a significant impact on results. For example, as
previously mentioned, preliminary evidence has suggested that both military service members
and military spouses are more likely to have experienced childhood adversity (Blosnich, Dichter,
Cerulli, Batten, & Bossarte, 2014; Oshri et al., 2015). Adversity in the early life of a parent and
the health and behavioral health outcomes associated with these experiences have the potential to
affect outcomes for youth in these families (Sun et al., 2017). Future research should consider
121
patterns of both normative and military-specific stressors alongside protective factors in this
population.
Conclusions
Overall, the results presented here highlight that military families are not homogenous in
their access to resources and exposure to risk. Clearly, there is a need to be concerned about
military-connected youth, as evidenced by the approximately 18% of youth in this sample who
received a mental health diagnosis in the year following survey completion. However, these
findings also highlight the large group of military families that have relatively minimal
comparative exposure to risk and report high levels of protective factors. The significantly lower
observed rates of mental health conditions in this group underscore the critical importance of
conducting a thorough assessment with military families that evaluates both the stressors these
families face and the strengths and resources they possess. Clinicians serving this population
should be aware of the elevated risk of adverse outcomes among military-connected youth but
must also be careful not to overpathologize.
The nexus of high protective factors and low exposure to risk associated with
significantly lower rates of diagnosed mental health conditions among military youth in these
analyses points to the critical importance of building resources in and among military families.
Although reducing exposure to military-specific risk may be challenging because some level of
risk exposure is likely unavoidable given the nature of military service, increasing military
families’ access to protective resources, like those identified in this study, is likely much more
achievable. The military hierarchy at multiple levels has the opportunity and obligation to
support prevention and intervention efforts that promote healthy marital and family functioning
122
and build supportive relationships in military communities. These efforts have the potential to
significantly improve outcomes for military-connected youth and families.

123
References
Allen, E. S., Rhoades, G. K., Stanley, S. M., & Markman, H. J. (2010). Hitting home:
Relationships between recent deployment, posttraumatic stress symptoms, and marital
functioning for Army couples. Journal of Family Psychology, 24, 280–288.
https://doi.org/10.1037/a0019405
Asparouhov, T., & Muthén, B. (2014). Auxiliary variables in mixture modeling: Three-step
approaches using Mplus. Structural Equation Modeling, 21, 329–341.
https://doi.org/10.1080/10705511.2014.915181
Barnes, V. A., Davis, H., & Treiber, F. A. (2007). Perceived stress, heart rate, and blood pressure
among adolescents with family members deployed in Operation Iraqi Freedom. Military
Medicine, 172, 40–43. https://doi.org/10.7205/MILMED.172.1.40
Blosnich, J. R., Dichter, M. E., Cerulli, C., Batten, S. V., & Bossarte, R. M. (2014). Disparities in
adverse childhood experiences among individuals with a history of military service.
JAMA Psychiatry, 71, 1041–1048. https://doi.org/10.1001/jamapsychiatry.2014.724
Bradshaw, C. P., Sudhinaraset, M., Mmari, K., Blum, R. W., & Hopkins, J. (2010). School
transitions among military adolescents: A qualitative study of stress and coping. School
Psychology Quarterly, 39, 84–105. Retrieved from
http://search.proquest.com/openview/34443878a484897ab37abae00de6d097
Burrell, L. M., Adams, G. A., Durand, D. B., & Castro, C. A. (2006). The impact of military
lifestyle demands on well-being, Army, and family outcomes. Armed Forces & Society,
33, 43–58. https://doi.org/10.1177/0002764206288804
Cederbaum, J. A., Gilreath, T. D., Benbenishty, R., Astor, R. A., Pineda, D., DePedro, K. T., …
Atuel, H. (2014). Well-being and suicidal ideation of secondary school students from
124
military families. Journal of Adolescent Health, 54, 672–677.
https://doi.org/10.1016/j.jadohealth.2013.09.006
Chandra, A., Lara-Cinisomo, S., Jaycox, L. H., Tanielian, T., Burns, R. M., Ruder, T., & Han, B.
(2010). Children on the homefront: The experience of children from military families.
Pediatrics, 125, 16–25. https://doi.org/10.1542/peds.2009-1180
Chandra, A., Martin, L. T., Hawkins, S. A., & Richardson, A. (2010). The impact of parental
deployment on child social and emotional functioning: Perspectives of school staff.
Journal of Adolescent Health, 46, 218–223.
https://doi.org/10.1016/j.jadohealth.2009.10.009
Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., & Anda, R. F.
(2004). Adverse childhood experiences and the risk of depressive disorders in adulthood.
Journal of Affective Disorder, 82, 217–225. https://doi.org/10.1016/j.jad.2003.12.013
Conforte, A. M., Bakalar, J. L., Shank, L. M., Quinlan, J., Stephens, M. B., Sbrocco, T., &
Tanofsky-Kraff, M. (2017). Assessing military community support: Relations among
perceived military community support, child psychosocial adjustment, and parent
psychosocial adjustment. Military Medicine, 182, e1871–e1878.
https://doi.org/10.7205/MILMED-D-17-00016
Cox, M. J., & Paley, B. (2003). Understanding families as systems. Current Directions in
Psychological Science, 12, 193–196. https://doi.org/10.1111/1467-8721.01259
Cozza, S. J., Haskins, R., & Lerner, R. M. (2013). Keeping the promise: Maintaining the health
of military and veteran families and children (Future of Children Policy Brief, Fall 2013).
Retrieved from
http://www.princeton.edu/futureofchildren/publications/docs/23_02_PolicyBrief.pdf
125
Easterbrooks, M. A., Ginsburg, K., & Lerner, R. M. (2013). Resilience among military
youth. The Future of Children, 23(2), 99-120.
Engel, R. C., Gallagher, L. B., & Lyle, D. S. (2010). Military deployments and children’s
academic achievement: Evidence from Department of Defense Education Activity
schools. Economics of Education Review, 29, 73–82.
https://doi.org/10.1016/j.econedurev.2008.12.003
Esposito-Smythers, C., Wolff, J., Lemmon, K. M., Bodzy, M., Swenson, R. R., & Spirito, A.
(2011). Military youth and the deployment cycle: Emotional health consequences and
recommendations for intervention. Journal of Family Psychology, 25, 497–507.
https://doi.org/10.1037/a0024534
Fergus, S., & Zimmerman, M. A. (2005). Adolescent resilience: A framework for understanding
healthy development in the face of risk. Annual Review of Public Health, 26, 399–419.
https://doi.org/10.1146/annurev.publhealth.26.021304.144357
Finkel, L. B., Kelley, M. L., & Ashby, J. (2003). Geographic mobility, family, and maternal
variables as related to the psychosocial adjustment of military children. Military
Medicine, 168, 1019–1024. https://doi.org/10.1093/milmed/168.12.1019
Flake, E. M., Davis, B. E., Johnson, P. L., & Middleton, L. S. (2009). The psychosocial effects
of deployment on military children. Journal of Developmental & Behavioral Pediatrics,
30, 271–278. https://doi.org/10.1097/DBP.0b013e3181aac6e4
Foran, H. M., Eckford, R. D., Sinclair, R. R., & Wright, K. M. (2017). Child mental health
symptoms following parental deployment: The impact of parental posttraumatic stress
disorder symptoms, marital distress, and general aggression. SAGE Open, 7(3), 1–10.
https://doi.org/10.1177/2158244017720484
126
Gewirtz, A. H., Polusny, M. A., DeGarmo, D. S., Khaylis, A., & Erbes, C. R. (2010).
Posttraumatic stress symptoms among National Guard soldiers deployed to Iraq:
Associations with parenting behaviors and couple adjustment. Journal of Consulting and
Clinical Psychology, 78, 599–610. https://doi.org/10.1037/a0020571
Gilreath, T. D., Astor, R. A., Cederbaum, J. A., Atuel, H., & Benbenishty, R. (2014). Prevalence
and correlates of victimization and weapon carrying among military- and nonmilitary-
connected youth in Southern California. Preventive Medicine, 60, 21–26.
https://doi.org/10.1016/j.ypmed.2013.12.002
Gilreath, T. D., Cederbaum, J. A., Astor, R. A., Benbenishty, R., Pineda, D., & Atuel, H. (2013).
Substance use among military-connected youth: The California Healthy Kids Survey.
American Journal of Preventive Medicine, 44, 150–153.
https://doi.org/10.1016/j.amepre.2012.09.059
Gilreath, T. D., Wrabel, S. L., Sullivan, K. S., Capp, G. P., Roziner, I., Benbenishty, R., & Astor,
R. A. (2016). Suicidality among military-connected adolescents in California schools.
European Child and Adolescent Psychiatry, 25, 61–66. https://doi.org/10.1007/s00787-
015-0696-2
Gorman, G. H., Eide, M., & Hisle-Gorman, E. (2010). Wartime military deployment and
increased pediatric mental and behavioral health complaints. Pediatrics, 126, 1058–1066.
https://doi.org/10.1542/peds.2009-2856
Green, S., Nurius, P. S., & Lester, P. (2013). Spouse psychological well-being: A keystone to
military family health. Journal of Human Behavior in the Social Environment, 23, 753–
768. https://doi.org/10.1080/10911359.2013.795068
Hawkins, S. A., Sullivan, K. S., Schuyler, A. C., Keeling, M., Kintzle, S., Lester, P. B., &
127
Castro, C. A. (2017). Thinking “big” about research on military families. Military
Behavioral Health, 5, 335–345. https://doi.org/10.1080/21635781.2017.1343696
Hisle-Gorman, E., Harrington, D., Nylund, C. M., Tercyak, K. P., Anthony, B. J., & Gorman, G.
H. (2015). Impact of parents’ wartime military deployment and injury on young
children’s safety and mental health. Journal of the American Academy of Child &
Adolescent Psychiatry, 54(4), 294–301. https://doi.org/10.1016/j.jaac.2014.12.017
Holmes, A. K., Rauch, P. K., & Cozza, S. J. (2013). When a parent is injured or killed in combat.
Future of Children, 23, 143–162. https://doi.org/10.1353/foc.2013.0017
Hosek, J., & Wadsworth, S. (2013). Economic conditions of military families. Future of
Children, 23, 41–59. https://doi.org/10.1353/foc.2013.0009
Huebner, A. J., & Mancini J. A. (2010). Resilience and vulnerability: The deployment
experiences of youth in military families. Blacksburg, VA: Virginia Polytechnic Institute
and State University.
Huebner, A. J., Mancini, J. A., Wilcox, R. M., Grass, S. R., & Grass, G. A. (2007). Parental
deployment and youth in military families: Exploring uncertainty and ambiguous loss.
Family Relations, 56, 112–122. https://doi.org/10.1111/j.1741-3729.2007.00445.x
Jordan, B. K., Marmar, C. R., Fairbank, J. A., Schlenger, W. E., Kulka, R. A., Hough, R. L., &
Weiss, D. S. (1992). Problems in families of male Vietnam veterans with posttraumatic
stress disorder. Journal of Consulting and Clinical Psychology, 60, 916–926.
https://doi.org/10.1037/0022-006X.60.6.916
Lester, P., Peterson, K., Reeves, J., Knauss, L., Glover, D., Mogil, C., … Beardslee, W. (2010).
The long war and parental combat deployment: Effects on military children and at-home
spouses. Journal of the American Academy of Child & Adolescent Psychiatry, 49, 310–
128
320. https://doi.org/10.1016/j.jaac.2010.01.003
Logan-Greene, P., Green, S., Nurius, P. S., & Longhi, D. (2014). Distinct contributions of
adverse childhood experiences and resilience resources: A cohort analysis of adult
physical and mental health. Social Work in Health Care, 53, 776–797.
https://doi.org/10.1080/00981389.2014.944251
Lucier-Greer, M., Arnold, A. L., Mancini, J. A., Ford, J. L., & Bryant, C. M. (2015). Influences
of cumulative risk and protective factors on the adjustment of adolescents in military
families. Family Relations, 64, 363–377. https://doi.org/10.1111/fare.12123
Lucier-Greer, M., O’Neal, C. W., Arnold, A. L., Mancini, J. A., & Wickrama, K. K. A. S.
(2014). Adolescent mental health and academic functioning: Empirical support for
contrasting models of risk and vulnerability. Military Medicine, 179, 1279–1287.
https://doi.org/10.7205/MILMED-D-14-00090
Mansfield, A. J., Kaufman, J. S., Engel, C. C., & Gaynes, B. N. (2011). Deployment and mental
health diagnoses among children of US Army personnel. Archives of Pediatrics &
Adolescent Medicine, 165, 999–1005. https://doi.org/10.1001/archpediatrics.2011.123
Mansfield, A. J., Kaufman, J. S., Marshall, S. W., Gaynes, B. N., Morrissey, J. P., & Engel, C. C.
(2010). Deployment and the use of mental health services among U.S. Army wives. New
England Journal of Medicine, 362, 101–109. https://doi.org/10.1056/NEJMoa0900177
Masten, A. S. (2001). Ordinary magic: Resilience processes in development. American
Psychologist, 56, 227–238. https://doi.org/10.1037/0003-066X.56.3.227
Masten, A. S. (2013). Afterword: What we can learn from military children and families. Future
of Children, 23, 199–212. https://doi.org/10.1353/foc.2013.0012
Masten, A. S., & Cicchetti, D. (2010). Developmental cascades. Development and
129
Psychopathology, 22, 491–495. https://doi.org/10.1017/S0954579410000222
Minuchin, S. (1974). Families and family therapy. Boston, MA: Harvard University Press.
Mmari, K., Roche, K. M., Sudhinaraset, M., & Blum, R. (2009). When a parent goes off to war:
Exploring the issues faced by adolescents and their families. Youth & Society, 40, 455–
475. https://doi.org/10.1177/0044118X08327873
Muthén, L. K., & Muthén, B. O. (2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthén
& Muthén.
Nurius, P. S., & Macy, R. J. (2008). Heterogeneity among violence-exposed women. Journal of
Interpersonal Violence, 23, 389–415. https://doi.org/10.1177/0886260507312297
Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2007). Deciding on the number of classes in
latent class analysis and growth mixture modeling: A Monte Carlo simulation study.
Structural Equation Modeling, 14, 535–569. https://doi.org/10.1080/10705510701575396
O’Neal, C. W., Lucier-Greer, M., Mancini, J. A., Ferraro, A. J., & Ross, D. B. (2016). Family
relational health, psychological resources, and health behaviors: A dyadic study of
military couples. Military Medicine, 181, 152–160. https://doi.org/10.7205/MILMED-D-
14-00740
Okafor, E., Lucier-Greer, M., & Mancini, J. A. (2016). Social stressors, coping behaviors, and
depressive symptoms: A latent profile analysis of adolescents in military families.
Journal of Adolescence, 51, 133–143. https://doi.org/10.1016/j.adolescence.2016.05.010
Oshri, A., Lucier-Greer, M., O’Neal, C. W., Arnold, A. L., Mancini, J. A., & Ford, J. L. (2015).
Adverse childhood experiences, family functioning, and resilience in military families: A
pattern-based approach. Family Relations, 64, 44–63. https://doi.org/10.1111/fare.12108
Paley, B., Lester, P., & Mogil, C. (2013). Family systems and ecological perspectives on the
130
impact of deployment on military families. Clinical Child and Family Psychology
Review, 16, 245–265. https://doi.org/10.1007/s10567-013-0138-y
Peterson, C., Park, N., & Castro, C. A. (2011). The Global Assessment Tool. American
Psychologist, 66, 10–18. https://doi.org/10.1037/a0021658
Richardson, E. W., Mallette, J. K., O’Neal, C. W., & Mancini, J. A. (2016). Do youth
development programs matter? An examination of transitions and well-being among
military youth. Journal of Child and Family Studies, 25, 1765–1776.
https://doi.org/10.1007/s10826-016-0361-5
Riggs, S. A., & Riggs, D. S. (2011). Risk and resilience in military families experiencing
deployment: The role of the family attachment network. Journal of Family Psychology,
25, 675–687. https://doi.org/10.1037/a0025286
Rosato, N. S., & Baer, J. C. (2012). Latent class analysis: A method for capturing heterogeneity.
Social Work Research, 36, 61–69. https://doi.org/10.1093/swr/svs006
Sayers, S. L., Farrow, V. A., Ross, J., & Oslin, D. W. (2009). Family problems among recently
returned military veterans referred for a mental health evaluation. Journal of Clinical
Psychiatry, 70, 163–170. https://doi.org/10.4088/JCP.07m03863
Solomon, Z., Debby-Aharon, S., Zerach, G., & Horesh, D. (2011). Marital adjustment, parental
functioning, and emotional sharing in war veterans. Journal of Family Issues, 32, 127–
147. https://doi.org/10.1177/0192513X10379203
Sullivan, K., Capp, G., Gilreath, T. D., Benbenishty, R., Roziner, I., & Astor, R. A. (2015).
Substance abuse and other adverse outcomes for military-connected youth in California:
Results from a large-scale normative population survey. JAMA Pediatrics, 169, 922–928.
https://doi.org/10.1001/jamapediatrics.2015.1413
131
Sullivan, K. S., Kintzle, K., Barr, N., Gilreath, T. D., & Castro, C. A. (in press). Veterans’
social/emotional and physical functioning informs perceptions of family and child
functioning. Journal of Military, Veteran, and Family Functioning.
Sullivan, K. S., Hawkins, S. A., Gilreath, T. D., Castro, C. A. (in process). Preliminary
psychometrics and potential big data uses of the U.S. Army Family Global Assessment
Tool.
Sun, J., Patel, F., Rose-Jacobs, R., Frank, D. A., Black, M. M., & Chilton, M. (2017). Mothers’
adverse childhood experiences and their young children’s development. American
Journal of Preventive Medicine, 53, 882–891.
https://doi.org/10.1016/j.amepre.2017.07.015
U.S. Department of Defense. (2017). 2016 demographics: Profile of the military community.
Retrieved from http://download.militaryonesource.mil/12038/MOS/Reports/2016-
Demographics-Report.pdf
Wadsworth, S. M., Cardin, J.-F., Christ, S., Willerton, E., Flittner O’Grady, A., Topp, D., …
Mustillo, S. (2016). Accumulation of risk and promotive factors among young children in
US military families. American Journal of Community Psychology, 57, 190–202.
https://doi.org/10.1002/ajcp.12025
Walsh, F. (1996). The concept of family resilience: Crisis and challenge. Family Process, 35,
261–281. https://doi.org/10.1111/j.1545-5300.1996.00261.x
Walsh, F. (2003). Family resilience: A framework for clinical practice. Family Process, 42, 1–
18. https://doi.org/10.1111/j.1545-5300.2003.00001.x
Zimmerman, M. A., Stoddard, S. A., Eisman, A. B., Caldwell, C. H., Aiyer, S. M., & Miller, A.
(2013). Adolescent resilience: Promotive factors that inform prevention. Child
132
Development Perspectives, 7, 215–220. https://doi.org/10.1111/cdep.12042

133
Table 4.1. Sample Demographics and Latent Profile Indicator Means and Prevalence
n (%) M (SD) Min Max
Service member age 37.07 (7.95) 17 66
Service member sex
Male 1,497 (91.8)
Female 133 (8.2)
Service member education
Completed high school or below 588 (36.1)
Completed some college or above 1,003 (61.5)
Service member rank
Enlisted 1,021 (62.6)
Officer 609 (37.4)
Service member race and ethnicity
White 1,059 (65.0)
Black 281 (17.2)
Asian 46 (2.8)
American Indian or Alaskan Native 10 (0.6)
Hawaiian or Pacific Islander 16 (1.0)
Hispanic 190 (11.7)
Family demographics
Spouse age 35.90 (7.98) 18 66
Spouse sex
Male 79 (4.8)
Female 1,453 (89.1)
Number of children in the home
1 501 (30.7)
134
2 595 (36.5)
3 333 (20.4)
4 or more 201 (12.3)
Age of oldest child
0 –5 390 (12.8)
6 –11 425 (14.0)
12 –19 638 (21.0)
20 or older 179 (5.9)
At-home parent mental health diagnosis
Yes 318 (19.5)
No 1,312 (80.5)
Recent relocation
Yes 307 (18.8)
No 1,323 (81.2)
Recent reunification
Yes 144 (8.8)
No 1,486 (91.2)
Cumulative days of deployment 194.48 (217.96) 0 1,639
Social connectedness 3.31 (1.10) 1 5
Social support 3.96 (0.91) 1 5
Relationship functioning 4.04 (0.86) 1 5
Family cohesion 4.15 (0.72) 1 5
Child mental health diagnosis
Yes 285 (17.5)
No 1,345 (82.5)

135
Table 4.2. Model Fit Indexes for LPA Model
Profiles BIC LMR-LRT BLRT Entropy
1 41,357.001
2 40,153.695 .0000 .0000 .730
3 39,907.511 .0637 .0000 .653
4 39,700.215 .0032 .0000 .734
5 39,382.374 .0447 .0000 .855
6 39,309.542 .1394 .0000 .829
Note. Best fitting model indicated in bold. BIC = Bayesian information criterion; LMR-LRT = Lo-Mendell-Rubin
likelihood ratio test; BLRT = bootstrap likelihood ratio test.

136
Table 4.3. Means and Conditional Probabilities for Risk and Protective Factor Indicators
Low Protection
and Low Risk
Low Protection
and Moderate
Risk
High
Protection and
Moderate Risk
High
Protection and
Low Risk
High
Protection and
High Risk
Class prevalence 11.84% 10.25% 29.26% 40.00% 8.65%
Protective factors
Social connections 2.42 2.33 3.59 3.70 3.55
Social support 2.92 2.94 4.31 4.36 4.20
Relationship
functioning
3.19 3.07 4.31 4.38 4.38
Family cohesion 3.47 3.48 4.39 4.37 4.37
Risk factors
At-home parent mental
health diagnosis

No 0.81 0.65 0.78 0.88 0.78
Yes 0.19 0.35 0.22 0.12 0.22
Recent relocation
No 0.82 0.72 0.79 0.88 0.71
Yes 0.18 0.28 0.21 0.12 0.29
Recent reunification
No 0.99 0.84 0.88 0.98 0.67
Yes 0.01 0.16 0.12 0.02 0.33
Days of deployment 14.45 361.93 322.045 14.91 642.916
Note. Figures represent means for protective factors and days of deployment, and conditional probabilities for at-
home parent mental health diagnosis, recent relocation, and recent reunification.

137
Table 4.4. Service Member Covariates Associated with Latent Profiles
Low Protection and
Low Risk
Low Protection and
Moderate Risk
High Protection and
Moderate Risk
High Protection and High
Risk
b SE OR b SE OR b SE OR b SE OR
Some college (vs. high school) 0.14 0.27 1.14 -0.01 0.23 0.99 0.56** 0.19 1.75 0.04 0.28 1.04
White (vs. non-White) 0.15 0.23 1.16 0.34* 0.21 1.40 0.23 0.16 1.26 0.86 0.25 2.36
Enlisted (vs. officer) 0.45 0.30 1.57 0.73 0.28 2.08 -0.19 0.18 0.83 -0.40** 0.27 0.67
Note. High Protection and Low Risk profile (Profile 4) was the reference category. Due to listwise deletion, n = 1,565 in these analyses.
*p < .05. **p < .01.

138
Table 4.5. Child Latent Profiles Associated with Distal Outcome (Mental Health Diagnosis)
Outcome 1 2 3 4
% (SE) χ
2
χ
2
χ
2
χ
2

1. Low protection and low risk 19.7 (0.037)
2. Low protection and moderate risk 21.9 (0.050) 0.121
3. High protection and moderate risk 28.0 (0.040) 1.875 0.719
4. High protection and low risk 10.6 (0.021) 4.171* 4.374* 13.976**
5. High protection and high risk 23.9 (0.040) 0.560 0.096 0.511 6.458*
Note. Degrees of freedom = 1 for all individual chi-square results.
*p < .05. **p < .001.

139
Chapter 5: Conclusions, Implications, and Future Directions
The goal of this dissertation was to explore risk and protective factors and associated
outcomes among military spouses and military-connected youth to elucidate possible points of
intervention to better support the healthy development of military-connected youth. Previous
work has consistently demonstrated a connection between military-related stressors, particularly
during wartime, and an elevated risk of adverse outcomes among the spouses and children of
military service members (Eaton et al., 2008; Gilreath et al., 2016; Mansfield, Kaufman, Engel,
& Gaynes, 2011; Mansfield et al., 2010; Renshaw et al., 2011; Sullivan et al., 2015). However,
these prior efforts have resulted in an incomplete picture of military family functioning. In
particular, earlier empirical examinations have tended to focus on the impact of one stressor at a
time, most frequently focusing on the negative outcomes of overseas combat deployments.
Further, these studies have often overlooked the potential effects of protective factors, which
may lead to overpathologizing of these families. This dissertation built on previous work to more
completely describe the functioning of these families. Study 1 (Chapter 2) focused on evaluating
the psychometric properties of a U.S. Army survey of family psychosocial fitness, which yielded
several strong measures of family-level protective factors. Study 2 (Chapter 3) explored the
impact of multiple concurrent stressors and the simultaneous effect of protective factors,
developed in Study 1, on the mental health of Army spouses. Study 3 (Chapter 4) explored the
impact of cumulative stress, including poor mental health of the at-home parent—which was the
outcome of Study 2—and the simultaneous impact of family-level protective factors on the
mental health of military-connected youth.
140
Major Findings and Implications for Policy and Practice
Study 1
The purpose of Study 1 was to examine the psychometric properties of the U.S. Army’s
Comprehensive Soldier and Family Fitness (CSF2) Family Global Assessment Tool (GAT).
Though the soldier version of this assessment tool was previously validated (Vie, Scheier, Lester,
& Seligman, 2016), the psychometrics of the family survey were unknown prior to this effort.
Using data from 1,692 Army spouses, exploratory and confirmatory factor analysis and two
validity studies established the underlying structure, reliability, and preliminary validity of the
Family GAT. Of the original 75 survey items, 29 were retained. These items loaded onto seven
factors, representing positive coping, meaning making, depression, relationship functioning,
family cohesion, social connections, and social support. Currently, the full Family GAT,
including all 75 original items, is still in use as a self-development and program evaluation tool.
These findings raise major policy concerns about the usefulness of the full survey. Items and
scales that were eliminated because they did not meet rigorous psychometric standards for
research purposes may continue to have face validity to the military hierarchy. Although these
items may retain utility for CSF2 program use, this utility must be balanced against ethical
concerns that arise from offering recommendations to individual military families based on
results of survey items that have not been validated.
From a research perspective, the scales that exceeded the stringent threshold for scientific
rigor in this study represent critical protective factors that are likely to be important to healthy
family functioning. This was a central prerequisite for Studies 2 and 3, because data on
protective factors were a critical counterbalance to U.S. Department of Defense (DOD) archival
data that informed risk factors to build more comprehensive models of military family
141
functioning. Further, these protective factor scales offer important nuance that can contribute to
future big data efforts, given the otherwise limited information about psychosocial factors in the
DOD’s data holdings. Additionally, Studies 2 and 3 provided further evidence of the validity of
protective factor scales established in Study 1; families that endorsed fewer protective factors
were more likely to have poor mental health outcomes.
Studies 2 and 3
The purpose of Studies 2 and 3 was to examine patterns of risk and protective factors
affecting military families and the association of these patterns with outcomes for military
spouses (Study 2) and military-connected youth (Study 3). In both studies, latent profile analysis
uncovered significant heterogeneity in the distribution of risk and protective factors across
families. This heterogeneity emphasizes the importance of supplementing variable-focused
approaches with person-centered methods, like latent profile analysis. Variable-focused methods
alone can fall victim to two pitfalls, which may inadvertently mislead policy makers. First,
relationships between variables that may hold true for the sample overall may be inaccurate for
certain subsamples of a population. Second, smaller subgroups of a population may hold undue
influence over variable-focused findings if their scores on variables of interest are sufficiently
extreme as to have a larger influence on summary statistics (Nurius & Macy, 2008; Rosato &
Baer, 2012). It is possible that either or both pitfalls have characterized previous research
involving military families, leaving policy makers with the impression that many or most
military families are struggling, which the present findings suggest is not the case.
Although profiles in both studies indicated some families were likely to be exposed to
high risk and have relatively few protective factors, the largest groups in both analyses were
composed of families that reported significant strengths and resources and relatively little
142
exposure to risk factors. Across studies, these groups exhibited the lowest rates of mental health
diagnoses among both military spouses and military-connected youth. From a policy perspective,
this finding suggests that many military families are likely faring well and have good internal
and external resources to drawn on during periods of increased stress. The need for military-
sponsored support programs is likely limited in this group. Thus, resources being directed to
universal prevention programs, which would be important if a broader group of families
appeared to be at risk, might be better used for targeted prevention and intervention services
directed specifically at those families with higher risk exposure and fewer resources.
Additionally, in both studies, profiles emerged in which a decoupling between strengths
and vulnerabilities was observed. Previous work has suggested that limited resources and greater
vulnerabilities may cluster together such that families exposed to more risk may also have access
to fewer protective factors (Trail, Meadows, Miles, & Karney, 2017). Although a small group of
these families was observed in each study and outcomes in this group were consistently poorer,
most families appeared to have some access to protective factors, even those in higher-risk
groups. This is an important finding that points to the need to assess for both risk and protective
factors in clinical settings serving this population. Understanding the stressors that families face
may not be sufficient to understand their likelihood of experiencing adverse outcomes. Further,
awareness of the strengths families possess may provide a solid foundation on which to build
intervention efforts. For example, a family stressed by multiple prolonged deployments and a
spouse suffering from a diagnosed mental health condition may nevertheless have good social
support. A mental health provider might find ways for the family to engage its support network
to reduce the stress burden on the at-home parent.
143
Finally, a guiding framework of this dissertation was a process-oriented definition of
resilience that encompassed a dynamic balance between risk and protective factors. The
literature on family functioning is rife with debate about a singular definition of resilience. Some
scholars strongly argue for operationalizing this concept as a trait, a process, or an outcome of
that process (Bowen & Martin, 2011). In contrast to static, trait-based definitions, these results
lend support to dynamic characterizations of resilience that incorporate an interplay between
exposure to stressors and access to internal or external protective factors. In Studies 2 and 3,
profiles emerged in which a nexus between low to moderate exposure to risk and considerable or
sufficient access to resources was associated with a significantly decreased prevalence of poor
mental health outcomes. Neither low levels of protective factors nor high risk exposure alone
was consistently sufficient to determine outcomes among spouses or children.
Further, these findings lend support for conceptualizing resilience at the family level. In
describing family resilience, Froma Walsh (2003) wrote that this concept “extends beyond
seeing individual family members as potential resources for individual resilience to focusing on
risk and resilience in the family as a functional unit” (p. 3). Particularly in Study 3, in which
protective factors were assessed from the perspective of one parent in the family and risk factors
largely centered on the other parent, findings demonstrated that these strengths and
vulnerabilities nevertheless operated at the family level and ultimately had significant effects on
the functioning of children in these systems. These findings point to the usefulness of
considering the family unit as the focal point for prevention and intervention efforts, rather than
providing individual services to multiple members of the same family unit. Effecting meaningful
change at the family level has the potential to influence outcomes across family members.
Traditionally, the military has focused minimal resources on families relative to the extensive
144
resources devoted to the well-being and mission readiness of service members. Findings like
those presented here suggest that increased attention to the well-being of military families may
have relevance for both service members and their dependents.
Future Directions
Both findings from these studies and the limitations to the approach taken in this
dissertation suggest directions for future research. First, person-focused methods offer
methodological strengths that have already been detailed. However, these strengths are most
useful when combined with results from variable-focused studies. In particular, this dissertation
set out to evaluate the impact of multiple, concurrent stressors on outcomes, in contrast to
previous research that has largely isolated outcomes associated with one factor. Findings
highlighted the clustering of risks and the impact of exposure to multiple risks on both spouses
and children. Although this information is useful on its own, this methodology is unable to
untangle the relative effects of one risk factor versus another. To add more nuance and depth to
these findings, future research with this dataset could rely on a structural equation modeling
approach. This would provide standardized parameter estimates that could identify the relative
strength of associations among experiences of family moves, deployments, and reunifications.
This approach may be particularly useful for further exploring the concept of cascade
effects. Study 3 found limited support for the impact of cascade effects on outcomes for military-
connected youth, in contrast to much previous empirical work (Chandra et al., 2010; Flake,
Davis, Johnson, & Middleton, 2009; Lester et al., 2010). In all likelihood, the lack of significant
findings was because rates of mental health diagnoses among at-home parents, which
represented cascades effects in these models, were diffuse across all five profiles and likely
indistinguishable from the impact of other stressors. Using a regression or structural equation
145
modeling approach will complement the latent profile analysis results presented here by focusing
on the relationship between the variables representing parent and child mental health that
characterize the sample overall. Unlike latent profile analysis, this approach does not explicitly
consider the heterogeneity in this population that may detract from these potential effects.
Further a variable-focused approach can consider moderation, which would allow the
exploration of buffering relationships between risk and protective factors. Although evidence
exists of the impact of both stressors and protective factors, person-centered methods are not
suited to teasing out whether these effects are cumulative, similar to some findings in the civilian
literature (Gabalda, Thompson, & Kaslow, 2010), or whether some buffering effects exist.
Buffering, which has most frequently been discussed with regard to the impact of social support,
refers to the idea that internal or external resources protect against the pathogenic influence of
stressors, rather than having a direct effect on well-being (Cohen & Wills, 1985). Some evidence
has suggested the buffering effect of protective factors exists among military families
(Wadsworth et al., 2016), but latent profile analysis cannot explicitly test for these effects. Using
a regression or structural equation modeling approach would complement the findings presented
here, further exploring the mechanisms by which protective factors influence outcomes among
military families.
Finally, the big data methods employed in these analyses limited the selection of risk and
protective factor variables to those available in either the CSF2 GAT survey or DOD archival
data holdings. Although these integrative methods allow for the inclusion of a breadth of
information about military families from multiple sources, some important areas of family
functioning are more difficult to capture. In particular, recent research has emphasized the
critical importance of normative stressors that all families may face in determining outcomes for
146
military-connected spouses and children (Lucier-Greer et al., 2014). Although many potential
stressors could be of interest in future research, experiencing adversity during childhood,
including exposure to maltreatment and household violence, is associated with significant
consequences for development as well as adverse health and mental health outcomes across the
lifespan (Anda et al., 2003; Felitti et al., 1998; Nurius, Green, Logan-Greene, & Borja, 2015).
Adversity in the early life of either parent and the behavioral health outcomes associated with
these experiences can influence family-level processes, including relationship functioning,
family communication, and parenting, which may increase risk of exposure to adversity in the
next generation (Barrett, 2009; Bifulco et al., 2002; O’Neal, Richardson, Mancini, & Grimsley,
2016; Oshri et al., 2015; Spieker, Oxford, Fleming, & Lohr, 2018). These outcomes are of
particular concern in military families, in light of evidence that both service members and
spouses are more likely than civilians to have experienced childhood adversity (Blosnich,
Dichter, Cerulli, Batten, & Bossarte, 2014; Oshri et al., 2015). Thus, in addition to other
potential normative stressors, future work with this population should consider examining the
effects of childhood adversity in the early life of service members and spouses on the functioning
of military family systems, because these experiences may fundamentally alter how families
manage the military-specific stressors to which they are exposed.
Concluding Thoughts
After 17 uninterrupted years of conflict, the capabilities and resources of our nation’s
service members and their loved ones have been taxed in ways seen and unseen. These families
have sacrificed greatly, and the majority continue to thrive despite these challenges. This project
continues an ongoing dialogue about what our military families need and how their country,
which benefits from their sacrifices, can support them. Future projects will continue to refine the
147
strategies presented here, but a fundamental message arises from these findings. Military families
live every day with risk and need adequate supports to manage the stressors they face. When
families are not able to muster sufficient resources to cope with risks on their own, we have an
obligation to provide these resources to avoid poor outcomes.

148
References
Anda, R. F., Dube, S. R., Felitti, V. J., Dong, M., Chapman, D. P., & Giles, W. H. (2003).
Childhood abuse, neglect, and household dysfunction and the risk of illicit drug use: The
Adverse Childhood Experiences Study. Pediatrics, 111, 564–572.
https://doi.org/10.1542/peds.111.3.564
Barrett, B. (2009). The impact of childhood sexual abuse and other forms of childhood adversity
on adulthood parenting. Journal of Child Sexual Abuse, 18, 489–512.
https://doi.org/10.1080/10538710903182628
Bifulco, A., Moran, P. M., Ball, C., Jacobs, C., Baines, R., Bunn, A., & Cavagin, J. (2002).
Childhood adversity, parental vulnerability and disorder: examining inter-generational
transmission of risk. Journal of Child Psychology and Psychiatry, 43, 1075–1086.
https://doi.org/10.1111/1469-7610.00234
Blosnich, J. R., Dichter, M. E., Cerulli, C., Batten, S. V., & Bossarte, R. M. (2014). Disparities in
adverse childhood experiences among individuals with a history of military service.
JAMA Psychiatry, 71, 1041–1048. https://doi.org/10.1001/jamapsychiatry.2014.724
Bowen, G. L., & Martin, J. A. (2011). The resiliency model of role performance for service
members, veterans, and their families: A focus on social connections and individual
assets. Journal of Human Behavior in the Social Environment, 21, 162–178.
https://doi.org/10.1080/10911359.2011.546198
Chandra, A., Lara-Cinisomo, S., Jaycox, L. H., Tanielian, T., Burns, R. M., Ruder, T., & Han, B.
(2010). Children on the homefront: The experience of children from military families.
Pediatrics, 125, 16–25. https://doi.org/10.1542/peds.2009-1180
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis.
149
Psychological Bulletin, 98, 310–357. https://doi.org/10.1037/0033-2909.98.2.310
Eaton, K. M., Hoge, C. W., Messer, S. C., Whitt, A. A., Cabrera, O. A., McGurk, D., … Castro,
C. A. (2008). Prevalence of mental health problems, treatment need, and barriers to care
among primary care-seeking spouses of military service members involved in Iraq and
Afghanistan deployments. Military Medicine, 173, 1051–1056.
https://doi.org/10.7205/MILMED.173.11.1051
Felitti, V. J., Anda, R. F., Nordenburg, D., Williamson, D. F., Spitz, A. M., Edwards, V., …
Marks, J. S. (1998). Relationship of childhood abuse and household dysfunction to many
of the leading causes of death in adults: The Adverse Childhood Experiences (ACE)
Study. American Journal of Preventive Medicine, 14, 245–258.
https://doi.org/10.1016/S0749-3797(98)00017-8
Flake, E. M., Davis, B. E., Johnson, P. L., & Middleton, L. S. (2009). The psychosocial effects
of deployment on military children. Journal of Developmental & Behavioral Pediatrics,
30, 271–278. https://doi.org/10.1097/DBP.0b013e3181aac6e4
Gabalda, M. K., Thompson, M. P., & Kaslow, N. J. (2010). Risk and protective factors for
psychological adjustment among low-income, African American children. Journal of
Family Issues, 31, 423–444. https://doi.org/10.1177/0192513X09348488
Gilreath, T. D., Wrabel, S. L., Sullivan, K. S., Capp, G. P., Roziner, I., Benbenishty, R., & Astor,
R. A. (2016). Suicidality among military-connected adolescents in California schools.
European Child and Adolescent Psychiatry, 25, 61–66. https://doi.org/10.1007/s00787-
015-0696-2
Lester, P., Peterson, K., Reeves, J., Knauss, L., Glover, D., Mogil, C., … Beardslee, W. (2010).
The long war and parental combat deployment: Effects on military children and at-home
150
spouses. Journal of the American Academy of Child & Adolescent Psychiatry, 49, 310–
320. https://doi.org/10.1016/j.jaac.2010.01.003
Lucier-Greer, M., O’Neal, C. W., Arnold, A. L., Mancini, J. A., & Wickrama, K. K. A. S.
(2014). Adolescent mental health and academic functioning: Empirical support for
contrasting models of risk and vulnerability. Military Medicine, 179, 1279–1287.
https://doi.org/10.7205/MILMED-D-14-00090
Mansfield, A. J., Kaufman, J. S., Engel, C. C., & Gaynes, B. N. (2011). Deployment and mental
health diagnoses among children of US Army personnel. Archives of Pediatrics &
Adolescent Medicine, 165, 999–1005. https://doi.org/10.1001/archpediatrics.2011.123
Mansfield, A. J., Kaufman, J. S., Marshall, S. W., Gaynes, B. N., Morrissey, J. P., & Engel, C. C.
(2010). Deployment and the use of mental health services among U.S. Army wives. New
England Journal of Medicine, 362, 101–109. https://doi.org/10.1056/NEJMoa0900177
Nurius, P. S., Green, S., Logan-Greene, P., & Borja, S. (2015). Life course pathways of adverse
childhood experiences toward adult psychological well-being: A stress process analysis.
Child Abuse & Neglect, 45, 143–153. https://doi.org/10.1016/j.chiabu.2015.03.008
Nurius, P. S., & Macy, R. J. (2008). Heterogeneity among violence-exposed women. Journal of
Interpersonal Violence, 23, 389–415. https://doi.org/10.1177/0886260507312297
O’Neal, C. W., Richardson, E. W., Mancini, J. A., & Grimsley, R. N. (2016). Parents’ early life
stressful experiences, their present well-being, and that of their children. American
Journal of Orthopsychiatry, 86, 425–435.
Oshri, A., Lucier-Greer, M., O’Neal, C. W., Arnold, A. L., Mancini, J. A., & Ford, J. L. (2015).
Adverse childhood experiences, family functioning, and resilience in military families: A
pattern-based approach. Family Relations, 64, 44–63. https://doi.org/10.1111/fare.12108
151
Renshaw, K. D., Allen, E. S., Rhoades, G. K., Blais, R. K., Markman, H. J., & Stanley, S. M.
(2011). Distress in spouses of service members with symptoms of combat-related PTSD:
Secondary traumatic stress or general psychological distress? Journal of Family
Psychology, 25, 461–469. https://doi.org/10.1037/a0023994
Rosato, N. S., & Baer, J. C. (2012). Latent class analysis: A method for capturing heterogeneity.
Social Work Research, 36, 61–69. https://doi.org/10.1093/swr/svs006
Spieker, S. J., Oxford, M. L., Fleming, C. B., & Lohr, M. J. (2018). Parental childhood adversity,
depressive symptoms, and parenting quality: Effects on toddler self-regulation in child
welfare services-involved families. Infant Mental Health Journal, 39, 5–16.
https://doi.org/10.1002/imhj.21685
Sullivan, K., Capp, G., Gilreath, T. D., Benbenishty, R., Roziner, I., & Astor, R. A. (2015).
Substance abuse and other adverse outcomes for military-connected youth in California:
Results from a large-scale normative population survey. JAMA Pediatrics, 169, 922–928.
https://doi.org/10.1001/jamapediatrics.2015.1413
Trail, T. E., Meadows, S. O., Miles, J. N., & Karney, B. R. (2017). Patterns of vulnerabilities and
resources in U.S. Military families. Journal of Family Issues, 38, 2128–2149.
https://doi.org/10.1177/0192513X15592660
Vie, L. L., Scheier, L. M., Lester, P. B., & Seligman, M. E. P. (2016). Initial validation of the
U.S. Army Global Assessment Tool. Military Psychology, 28, 468–487.
https://doi.org/10.1037/mil0000141
Wadsworth, S. M., Cardin, J.-F., Christ, S., Willerton, E., Flittner O’Grady, A., Topp, D., …
Mustillo, S. (2016). Accumulation of risk and promotive factors among young children in
US military families. American Journal of Community Psychology, 57, 190–202.
152
https://doi.org/10.1002/ajcp.12025
Walsh, F. (2003). Family resilience: A framework for clinical practice. Family Process, 42, 1–
18. https://doi.org/10.1111/j.1545-5300.2003.00001.x

Abstract (if available)

Abstract Although many military families cope with stressors effectively (Wadsworth, 2013), a significant subset of military spouses and children, who number close to 3 million (U.S. Department of Defense, 2017), experience adverse consequences of wartime military service including poor health and mental health, increased risk behaviors, suicidality, and substance use (Chandra, Lara-Cinisomo, et al., 2010

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An examination of sexual harassment, gender discrimination, stalking, and sexual assault among female and male veterans and associations with PTSD and depression

PDF

Parental military service and adolescent mental and behavioral health: the role of adolescent-civilian parent interactions

PDF

Exploring the social determinants of health in a population with similar access to healthcare: experiences from United States active-duty army wives

PDF

U.S. Latinx youth development and substance use risk: adversity and strengths

PDF

Not just talk: observed communication in adolescent friendship and its implications for health risk behavior

PDF

Disclosure of lesbian, gay, bisexual, and transgender (LGBT) identity among U.S. service members after repeal of LGBT military service bans

PDF

Mindfulness and resilience: an investigation of the role of mindfulness in post-9/11 military veterans' mental health-related outcomes

PDF

Impacts of caregiving on wellbeing among older adults and their spousal caregivers in the United States

PDF

Cultural resources and health among Asian Americans: results from the National Latino and Asian American Study

PDF

Empowering and building resilience with youth in congregate care

PDF

Family relationships and their influence on health outcomes over time among older adults in rural China

PDF

The role of school climate in the mental health and victimization of students in military-connected schools

PDF

Maltreated adolescents and their families: a longitudinal examination of family functioning, parenting attitudes, & youth mental health

PDF

Pain and multi-morbidity among veterans: theory-guided, data-driven, and narrative approaches

PDF

Mental health advocacy and navigation partnerships: the case for a community collaborative approach

PDF

An ecological mixed-methods analysis of homeless students’ school experience at the state, district, and school levels

PDF

Childhood cancer survivorship: parental factors associated with survivor's follow-up care behavior and mental health

PDF

Offline social functioning and online communication: how social competence translates to an online context

PDF

Connecting students to wellness: student parents empowering parents)

PDF

Adolescent conduct problems and substance use: an examination of the risk pathway across the transition to high school

Asset Metadata

Creator Sullivan, Kathrine Symonds (author)

Core Title Mental health outcomes associated with profiles of risk and resilience among military-connected youth

School School of Social Work

Degree Doctor of Philosophy

Degree Program Social Work

Publication Date 07/03/2018

Defense Date 06/08/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag latent class analysis,Mental Health,military families,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Castro, Carl (committee chair), Cederbaum, Julie (committee member), Margolin, Gayla (committee member)

Creator Email kate.sullivan@usc.edu,katefigge@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-509366

Unique identifier UC11268510

Identifier etd-SullivanKa-6370.pdf (filename),usctheses-c40-509366 (legacy record id)

Legacy Identifier etd-SullivanKa-6370.pdf

Dmrecord 509366

Document Type Dissertation

Format application/pdf (imt)

Rights Sullivan, Kathrine Symonds

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA