Abstract
Resilience following childhood maltreatment has received substantial empirical attention, with the number of studies on this construct growing exponentially in the past decade. While there is ample interest, inconsistencies remain about how to conceptualize and assess resilience. Further, there is a lack of consensus on how developmental stage influences resilience and how protective factors affect its expression. The current systematic review uses a developmental lens to synthesize findings on resilience following child maltreatment. Specifically, this article consolidates the body of empirical literature in a developmentally oriented review, with the intention of inclusively assessing three key areas—the conceptualization of resilience, assessment of resilience, and factors associated with resilience in maltreatment research. A total of 67 peer-reviewed, quantitative empirical articles that examined child maltreatment and resilience were included in this review. Results indicate that some inconsistencies in the literature may be addressed by utilizing a developmental lens and considering the individual’s life stage when selecting a definition of resilience and associated measurement tool. The findings also support developmental variations in factors associated with resilience, with different individual, relational, and community protective factors emerging based on life stage. Implications for practice, policy, and research are incorporated throughout this review.
Child maltreatment is an insidious public health problem affecting millions of youth in the United States. In 2016 alone, there were 676,000 substantiated cases of abuse and neglect (U.S. Department of Health & Human Services [U.S. DHHS], 2018). Broadly, childhood maltreatment encompasses multiple types of offending against children including physical abuse, emotional abuse, sexual abuse, and neglect (U.S. DHHS, 2018). Numerous studies have suggested that childhood maltreatment is directly associated with negative physical and mental health outcomes across the life course (Gomez et al., 2017; Jaffee et al., 2018; Jaffee, Caspi, Moffitt, Polo-Tomas, & Taylor, 2007; Yoon, Cage, Pei, & Barnhart, 2018). Some of the associations between child maltreatment and poor health outcomes may be explained by characteristics of maltreating families and unique challenges these families and individuals face, such as poverty, parental substance use and parenting stress, exposure to community violence and crime, and lack of access to needed services and resources (Jaffee et al., 2007; Maguire-Jack, Cao, & Yoon, 2018; Taylor, Guterman, Lee, & Rathouz, 2009; Yoon et al., 2018). Despite these challenges, some individuals display resilience following childhood maltreatment experiences. It is important to examine resilience through a developmental lens given that people exhibit variable profiles of functioning based on their developmental stage (Cowell, Cicchetti, Rogosch, & Toth, 2015; Geller et al., 2001), and resilience may occur at any point in the life span (Werner, 1989). Thus, the aim of the current study was to review empirical articles to summarize and synthesize the conceptualization, assessment, and features of resilience in child maltreatment research.
Child Maltreatment and Resilience
A substantial proportion of individuals who experience child maltreatment do not exhibit or develop negative outcomes; this phenomenon has often been labeled as resilience (Holmes, Yoon, Voith, Kobulsky, & Steigerwald, 2015). The construct of resilience has received significant attention in the field of trauma research. Notably, studies examining resilience in the context of trauma have suggested that certain (protective) factors bolster resilience, allowing trauma-exposed individuals to recover from adversity (Howell & Miller-Graff, 2014; Luthar & Cicchetti, 2000). At the individual level, self-regulation skills, higher levels of social competence, personal control, problem-solving skills, motivation, self-esteem, and heightened adaptive functioning abilities have been documented as factors associated with positive outcomes or resilience (Child Welfare Information Gateway [CWIG], 2014; Schultz, Tharp-Taylor, Haviland, & Jaycox, 2009). At the relational level, parenting competence, positive peer relationships, religiosity, and caring adults have been noted as potential protective factors (CWIG, 2014; Perkins & Jones, 2004; Schultz et al., 2009). In particular, studies have highlighted caring adults (e.g., mentors, caregivers, teachers) who provide consistent support to be a major factor that contributes to resilience following trauma (Holmes, Yoon, Berg, Cage, & Perzynski, 2018; Zimmerman et al., 2013). At the societal or community level, a positive school climate, close-knit community, safe neighborhoods, and social connections have been identified as factors associated with resilience (CWIG, 2014; Perkins & Jones, 2004). Furthermore, studies have identified social support, positive childhood family environment (e.g., caregiver stability, caregiver well-being), and the child’s preexisting capacities (e.g., lack of preexisting psychopathology) as protective factors that are related to resilience following child maltreatment (Bradley, Davis, Wingo, Mercer, & Ressler, 2013; Goslin, Stover, Berkowitz, & Marans, 2013; Greenfield & Marks, 2010).
Resilience: Conceptualization, Assessment, and Associated Factors
Despite the increasing interest and significant advancements in maltreatment and resilience research over the past several decades, it remains unclear how resilience can be effectively defined and assessed in individuals with histories of child maltreatment. Some researchers conceptualize resilience as a personal trait, specifically as a relatively stable, innate characteristic that is featured by psychological hardiness, ego resilience, and coping efficacy (Block & Block, 1980; Connor & Davidson, 2003). Others have conceptualized resilience as an outcome or process of positive adaptation across multiple domains after experiencing substantial trauma (e.g., Luthar & Cicchetti, 2000). Researchers have also conceptualized resilience using a social–ecological framework and defined resilience as individuals navigating their way to internal and external resources that bolster their well-being (e.g., Ungar, 2004).
In addition to the lack of theoretical agreement and standardized conceptual definition of resilience, there is no methodological consensus on how to best assess resilience following child maltreatment. This is unsurprising given that assessment of resilience is often influenced by the conceptual definition of resilience. Researchers who conceptualize resilience as a personal trait have primarily used personality or coping scales for assessment (Arslan, 2015; Smith, Lenz, & Strohmer, 2017). Researchers who conceptualize resilience as an outcome or process of positive adaptation have typically utilized measures of functioning across multiple areas of competence (e.g., cognitive, emotional, behavioral, and social; Luthar & Cicchetti, 2000).
The lack of uniformity in definitions and measurements of resilience has limited our ability to compare and synthesize previous empirical findings on the factors associated with resilience among individuals who have experienced child maltreatment. Although numerous studies have investigated factors associated with resilience, often referring to them as protective factors, it has been challenging to summarize the findings across studies due to limited comparability that stems from inconsistency in conceptual and operational definitions of resilience (Afifi & MacMillan, 2011). Further, some studies have used the terms resilience and protective factors interchangeably, adding more complexity and confusion. The term protective factors itself has also been in the center of controversy; some researchers have argued that the term promotive factors should be used to describe the variables that are directly associated with positive outcomes (i.e., main effects) and suggested distinguishing promotive factors from protective factors in which the effects of risks are modified and buffered (i.e., interaction effects) to produce positive outcomes (see Zimmerman et al., 2013). Therefore, in the current article, we use the term factors associated with resilience, rather than the term protective factors, to describe the factors that are related to or predictive of resilience.
Developmental Variations
Important gaps remain in the child maltreatment and resilience literature with regard to the role of the developmental stage in which resilience is examined. We do not know if and how definitions, measurements, and features (i.e., factors associated with resilience) of resilience differ depending on developmental stage. Some resilience and developmental scientists have suggested that resilience is a developmental process, highlighting the flexible and mutable nature of this construct (Luthar & Cicchetti, 2000).
The developmental psychopathology perspective (Sroufe & Rutter, 1984) and Erikson’s (1994) theory of psychosocial development emphasize that there exist stage-salient tasks in each developmental stage. The developmental resilience perspective conceives resilience as a developmental process in which competence is acquired through person–environment (e.g., family, community) interactions (Sroufe & Rutter, 1984). Drawing from these perspectives, the conceptualizations and assessments of resilience would look different based on the developmental stage in which resilience is examined. Similarly, the factors that are associated with resilience might differ based on the person’s developmental stage.
Understanding how the conceptualization and assessment of resilience vary as a function of developmental stage (i.e., early childhood, school age, adolescence, adulthood) may inform tailored intervention strategies and policies to promote resilience that might differ by developmental stage. For example, if empirical evidence suggests that resilience during school age is best understood and assessed through children’s social functioning at school, perhaps school-based relationship-building interventions might be a useful approach to promote resilience during this developmental period. The current investigation also has important implications for understanding the prevalence of resilience in different developmental periods and conducting longitudinal studies of resilience trajectories across various developmental stages. Furthermore, understanding the ways in which factors associated with resilience vary across different developmental stages provides valuable information in creating resilience promotion interventions that are developmentally sensitive and appropriate.
Contributions of the Current Study
There have been several previous reviews of child maltreatment and resilience (Afifi & MacMillan, 2011; Klika & Herrenkohl, 2013; Meng, Fleury, Xiang, Li, & D’Arcy, 2018; Walsh, Dawson, & Mattingly, 2010). Afifi and MacMillan (2011) reviewed 27 articles and summarized the findings on protective factors associated with resilient functioning following child maltreatment. Meng, Fleury, Xiang, Li, and D’Arcy (2018) conducted a systematic review of 85 articles to discuss resilience and multi-level (individual, familial, and societal levels) protective factors in people with a history of child maltreatment. Klika and Herrenkohl (2013) reviewed 11 longitudinal studies on resilience in maltreated children to examine the dynamic qualities of resilience across time. Finally, Walsh, Dawson, and Mattingly (2010) reviewed the measurement of resilience following childhood maltreatment in 21 studies. Collectively, these prior reviews found that the measures used in research are notably different across studies and that few studies have employed longitudinal design to investigate resilience over extended periods of development.
While these reviews have provided valuable insight into resilience and/or protective factors associated with resilience following child maltreatment, they were limited in several respects. First, they focused on one aspect of resilience (e.g., measurement of resilience, Walsh et al., 2010; protective factors associated with resilience, Afifi & MacMillan, 2011) rather than inclusively and thoroughly reviewing key areas (i.e., conceptualization, assessment, features) of resilience. Second, prior reviews often used a predetermined definition of resilience (e.g., measures of multidomain functioning: Afifi & MacMillan, 2011; Walsh et al., 2010) and confined the scope of the reviews to those articles that fit the predetermined definition, which may represent a missed opportunity to explore variability in conceptualization and assessment of resilience in the child maltreatment literature. Finally, no previous review has examined developmental variations in the conceptualization, assessment, and features of resilience. We aim to advance our knowledge of resilience by applying a developmental lens in synthesizing prior findings on the conceptualization, assessment, and factors of resilience in child maltreatment research.
Method
For the current systematic review, we conducted comprehensive literature searches using electronic searching databases including Medline, SocINDEX, PsycINFO, CINAHL, and Academic Search Complete to identify studies on child maltreatment and resilience. The research team also examined the reference lists from previous articles to identify any additional eligible studies. Key search terms used for the electronic database searches included child maltreatment, child abuse, neglect, physical abuse, sexual abuse, emotional abuse, protective factors, promotive factors, and resilience/resiliency/resilient. The following search string was used (for more information, see Appendix A): (resilien* or protective factors or promotive factors) AND (child maltreatment or child abuse or neglect or physical abuse or sexual abuse or emotional abuse or maltreat*).
Inclusion criteria were studies that (1) explicitly examined child maltreatment and resilience, (2) were quantitative empirical studies, (3) were published since 2010, (4) were peer-reviewed publications, and (5) were published in English. We limited the scope of our review to articles published since 2010 because two resilience review articles (i.e., Afifi & MacMillan, 2011; Walsh et al., 2010) extensively reviewed studies published before 2010, and a large volume of maltreatment and resilience studies were published in the interim. Searches were limited to peer-reviewed publications to ensure scientific rigor and quality of identified studies.
Through electronic database searches and reference mining, 3,510 unique records were identified after deleting 2,588 duplicates. The lead author and three doctoral-level students completed the screening and eligibility determination processes. First, each reviewer conducted title and abstract screening to determine whether the article matched the inclusion criteria. Studies clearly meeting any of the exclusion criteria were eliminated. During the screening process, 3,201 records were excluded due to irrelevance based on the article title. An additional 169 records were excluded due to irrelevance by abstract, leaving 140 records for full-text review. Of the 140 studies, 74 studies were excluded after full-text screening because they did not meet one or more of the inclusion criteria. Throughout the process, two independent reviewers worked separately, and the results were compared to reduce coder bias and increase reliability. Any disagreement or discrepancy in screening results between coders was reconciled through discussions between the coders. A final set of 67 articles were included in the current review (see Figure 1). Of the 67 articles, 3 articles focused on early childhood (birth to age 5), 7 focused on school age (ages 6–12), 24 focused on adolescence (ages 13–19), and 33 focused on adulthood (age 20 and older). Nine articles had a comparison group, such as a nonmaltreated group. The articles used a wide variety of instruments (e.g., the Childhood Trauma Questionnaire; Maltreatment Classification System; Parent–Child Conflict Tactics Scales) and informants/sources (e.g., self, parents, child welfare workers, official child welfare records) to measure child maltreatment. The majority of the studies focusing on adulthood utilized retrospective self-reported childhood maltreatment history. Table 1 displays the key findings of the reviewed articles.

Flowchart for articles included in the review.
Summary of Critical Study Findings.
Note. CBCL = Child Behavior Checklist; PTSD = posttraumatic stress disorder; fMRI = functional magnetic resonance imaging; CYRM-12 = The Child and Youth Resilience Measure; AAR-C2 = Assessment and Action Record; HAMD = Hamilton Depression Rating Scale; DISC-IV = Diagnostic Interview Schedule for Children Version IV; RSCA = Resiliency Scales for Children and Adolescents; CANS = Child and Adolescent Needs and Strengths; CES-D = Epidemiologic Studies Depression Scale; CD-RISC = Connor–Davidson Resilience Scale; CSA = childhood sexual abuse; RS = Resilience Scale; NSDE = National Study of Daily Experiences; BASC-PRS-II = Behavioral Assessment System for Children, Second Edition; SSRS = Social Skills Rating System; CTQ-SF = Childhood Trauma Questionnaire–Short Form; SRI-25 = Suicide Resilience Inventory; BDI-II = The Beck Depression Inventory–Second Edition; CAQ = College Adjustment Questionnaire; PTGI = Posttraumatic Growth Inventory; ARQ = Adolescent Resilience Questionnaire; PRYM = Pathways to Resilience Youth Measure; CAFAS = Child Adolescent Functional Assessment Scale; RS-25 = Resilience Scale; BRS = Brief Resilience Scale; MDD = Major Depressive Disorder; BDIST = Battelle Developmental Inventory Screening Test; BPCS = Battelle Psychomotor and Cognition Scales; WPPSI-R = Wechsler Preschool and Primary Scale of Intelligence-Revised; MBA = Mini-Battery of Achievement Test, WJ-III = Woodcock–Johnson III Test of Achievement; PLS = Preschool Language Scale; CDI = Children’s Depression Inventory; TSCC = Trauma Symptom Checklist for Children.
a Early childhood = ages 0–5; school age = ages 6–12; adolescence = ages 13–19; adulthood = ages ≥ 20.
Results
Conceptualization of Resilience
Studies typically conceptualized resilience following maltreatment in one of the three ways: (1) as a personality trait, (2) as outcomes related to adaptive functioning, or (3) as socioecological resources. Twenty studies conceptualized resilience as a personality trait. Personality conceptualizations of resilience in the face of childhood maltreatment were further subcharacterized. Several studies considered resilience to be individuals displaying an ability to “bounce back” after adversity and to recover from and cope with trauma (Arslan, 2015; Bradley et al., 2013; Smith et al., 2017). Other studies focused not only on the resistance to the damaging impact of trauma but also on the ability to thrive in the face of adversity (Berkowski & MacDonald, 2014; Beutel et al., 2017). Still others characterized resilience as a personality trait that allowed individuals to emerge from traumatic experiences stronger than they were prior to adversity (Irwin, Beeghly, Rosenblum, & Muzik, 2016).
Slightly less than half (n = 32) of the studies conceptualized resilience as an outcome, primarily defined as adaptive functioning following maltreatment (Bell, Romano, & Flynn, 2013; Oshri, Topple, & Carlson, 2017). Some studies viewed successful resilient outcomes as reaching normative developmental milestones after maltreatment, compared to youth who did not endure similar adversities (Sattler & Font, 2018). Resilience outcomes were often conceptualized as a multifaceted construct and/or included a temporal component, demonstrating how adaptations change and new outcomes are achieved as a process (Dubowitz et al., 2016).
Eleven studies conceptualized resilience in terms of available socioecological resources. This system of resources can span multiple levels including individual, relational, community, and cultural (Collin-Vézina, Coleman, Milne, Sell, & Daigneault, 2011). Within these domains, internal and external resources supportive of successful development during hardship can be found (Davidson-Arad & Navaro-Bitton, 2015; Happer, Brown, & Sharma-Patel, 2017; Seena & Sundaram, 2018). Ultimately, socioecological resource resilience is determined by people’s capacity to find and direct themselves to resources during times of adversity and to take advantage of those resources in a way that is advantageous in the short- and long-term (Sanders, Munford, Thimasarn-Anwar, Liebenberg, & Ungar, 2015; Ungar, Liebenberg, Dudding, Armstrong, & Van de Vijver, 2013).
Assessment of Resilience
Three common approaches were found regarding ways in which the construct of resilience can be assessed (see below for further description of these approaches): (1) lack of psychopathology or negative outcomes, (2) composite scores of multidomain functioning (e.g., social, cognitive, emotional, behavioral), and (3) resilience-specific measures or scales.
Fourteen of the reviewed studies assessed resilience by measuring either a lack of psychopathology or other negative outcomes (e.g., suicide, aggression). Common instruments used in this approach included the Strengths and Difficulties Questionnaire, which measures child socioemotional and behavior problems (Bell et al., 2013; Bell, Romano, & Flynn, 2015). Many studies also employed the Child Behavior Checklist to capture behavior problems including aggression (Holmes et al., 2015), externalizing symptoms (Yoon, 2018), and overall symptomology (Williams & Nelson-Gardell, 2012). Measures of internalizing symptomology were also common and included the Beck Depression Inventory (Lumley & McArthur, 2016) and the Symptom Checklist-90 (Lind et al., 2018).
The second approach to assess resilience was creating a multidomain composite resilience score, which was evident in 11 of the reviewed articles. Specific instruments and measurements differed across studies based on the functional outcomes of interest, the developmental stage being examined, and available measures. The domains of functioning examined (e.g., social, cognitive, emotional, behavioral, occupational) and corresponding measures varied substantially, ranging from two to eight domains and measures. For instance, some studies focused on three domains such as academic, social, and emotional realms (Maples, Park, Nolen, & Rosén, 2014) or emotional, behavioral, and educational indicators of resilience (Walsh et al., 2010). Another study looked at four domains including social functioning, lack of depression, satisfaction with life, and positive affect (Kaye-Tzadok & Davidson-Arad, 2016). One study examined seven domains/outcomes of functioning in adulthood: high school completion, college attendance, incarceration, average income, substance use, depressive symptoms, and expectations for the future (Topitzes, Mersky, Dezen, & Reynolds, 2013). Researchers using the approach of “multidomain composite scores” used multiple measures and created a summative index or total score to use as an indicator of resilience.
Finally, 34 of the reviewed studies used the approach of assessing resilience via resilience-specific measures and scales. Whereas the previous two approaches utilized measures that were not specifically developed to assess resilience (e.g., psychopathology measures and developmental measures), the studies using this approach utilized measures and scales that were created to purposely assess the construct of resilience. Most commonly, one of the three versions (2, 10, and 25 items) of the Connor–Davidson Resilience Scale was utilized (e.g., Berkowski & MacDonald, 2014; Carli et al., 2011; Dale et al., 2015; Ding et al., 2017; Howell & Miller-Graff, 2014; Irwin et al., 2016; Ressel, Lyons, & Romano, 2018; Sexton, Hamilton, McGinnis, Rosenblum, & Muzik, 2015; Wingo et al., 2010). Five studies used general resilience scales and measures, including the Brief Resilience Coping Scale (Beutel et al., 2017), the Resilience Quotient Test (Choi et al., 2013), and the Resilience Scale-25 (Schulz et al., 2014; Tlapek et al., 2017). Two measures specific to resilience in adults were used the Adult Resilience Measure (Arslan, 2015) and the Brief Resilience Scale (Smith et al., 2017). More frequently observed were measures specific to children, youth, and adolescents, including The Child and Youth Resilience Measure (CYRM; Collin-Vézina et al., 2011; Sanders et al., 2015; Ungar et al., 2013), the Resilience Scales for Children and Adolescents (Day & Kearney, 2016; Deblinger, Runyon, & Steer, 2014), the Resilience and Youth Development Module (RYDM; Davidson-Arad & Navaro-Bitton, 2015), Child and Adolescent Needs and Strengths (CANS; Go, Chu, Barlas, & Chng, 2017), and the Adolescent Resilience Questionnaire (Pérez-González, Guilera, Pereda, & Jarne, 2017).
Developmental Variations in Conceptualization and Assessment of Resilience
Research focusing on early childhood (birth to age 5) was limited with only three studies but consistently conceptualized resilience as the outcome of displaying adaptive or successful functioning. In terms of the assessment of resilience, one study focused on a single developmental domain—language and academic functioning—and used measures of language development and academic achievement (Holmes et al., 2018). The remaining two studies assessed resilience using multidomain composite scores derived from multiple measures across various domains of functioning (Dubowitz et al., 2016: behavioral, social, and developmental; Sattler & Font, 2018: cognitive, social).
The research on school-age (ages 6−12) resilience following child maltreatment differed little from early childhood studies. Similar to early childhood resilience research, all studies focusing on school-aged children consistently conceptualized resilience as outcomes or processes related to adaptive functioning (Bell et al., 2013, 2015; Yoon, 2018). Additionally, studies focusing on school-aged children primarily assessed resilience either as a lack of psychopathology/negative outcomes (Bell et al., 2013, 2015; Holmes et al., 2015) or competence across multiple domains of functioning (Cicchetti & Rogosch, 2012; Walsh et al., 2010).
Whereas studies focusing on earlier developmental stages conceptualized resilience exclusively as outcomes of adaptive functioning, adolescent resilience studies (ages 13−19) conceptualized resilience in a more diverse manner. In addition to studies defining resilience as outcomes of adaptive functioning (43%; n = 10), there were studies that conceptualized resilience as a personal trait (22%; n = 5) or dynamic systems of socioecological resources that promote well-being (35%; n = 8). Similar to research looking at younger ages, the adaptive functioning outcome conceptualization of resilience included single and multidomain inspection of dynamic, positive adaptation despite experiencing maltreatment (Asgeirsdottir, Gudjonsson, Sigurdsson, & Sigfusdottir, 2010; Dang, 2014; Williams & Nelson-Gardell, 2012). For studies conceptualizing resilience as a personal trait, individual characteristics, such as self-efficacy, mastery, relatedness, and emotional reactivity, were personal traits tied to resilience (Day & Kearney, 2016; Deblinger et al., 2014). The final category of resilience conceptualization found in the adolescent maltreatment literature was social–ecological, which involved utilization and influence of multidomain, socioecological resources, and factors beyond the individual- and family-level systems (Banyard, Hamby, & Grych, 2017; Collin-Vézina et al., 2011; Davidson-Arad & Navaro-Bitton, 2015; Seena & Sundaram, 2018). In addition to the expansion in conceptualization, the assessment of resilience in adolescence was extended, with some studies utilizing resilience-specific scales such as the CYRM (Sanders et al., 2015), RYDM (Davidson-Arad & Navaro-Bitton, 2015), and CANS (Go et al., 2017).
In studies focusing on adulthood, resilience was conceptualized more often as a personal trait (50%) than an adaptive functioning outcome (40%) or socioecological resources (10%). The application of adaptive functioning to explain resilience as a dynamic, multidimensional process is fundamentally the same as described for prior developmental groups except for the addition of more diverse, age-specific outcomes such as college attendance, incarceration, average income, and life satisfaction (Kaye-Tzadok & Davidson-Arad, 2016; Topitzes et al., 2013). The descriptions of personality trait conceptualization were more direct in language about an adult’s innate ability to bounce back, recover, or thrive (Arslan, 2015; Berkowski & MacDonald, 2014; Beutel et al., 2017; Schulz et al., 2014; Smith et al., 2017). Consideration of how present adaptive coping skills provide for coping with future adversity was a component only applied to adult resilience (Irwin et al., 2016). The socioecological resources conceptualization continued to view resilience as an interplay between an individual- and multi-level systems to access and utilize resources that promote well-being (Banyard et al., 2017).
Developmental Variations in Factors Associated With Resilience
Caregiver relational characteristics of warmth, emotional support, and cognitive stimulation were the primary factors associated with resilience among maltreated children in early childhood (Holmes et al., 2018; Sattler & Font, 2018). One study also found that lower levels of caregiver depressive symptoms were associated with resilience in children during early childhood (Dubowitz et al., 2016).
Caregiver well-being was associated with resilience in school-aged maltreated children (Holmes et al., 2015; Yoon, 2018). Parental engagement, characterized as positive and effective, was also found to be a key factor associated with resilience in the school-age period (Bell et al., 2013, 2015). Attribution of resilience to individual characteristics, including child prosocial behavior and internalized well-being, also received recognition as factors associated with resilience in school-age youth (Holmes et al., 2015; Yoon, 2018).
The factors associated with resilience for adolescents who have experienced child maltreatment included parental/caregiver support (Asgeirsdottir et al., 2010; Williams & Nelson-Gardell, 2012) and quality of the youth–caregiver relationship (Guibord, Bell, Romano, & Rouillard, 2011). Similarly, having paternal acceptance was associated with resilience among maltreated adolescents (Davidson-Arad & Navaro-Bitton, 2015). School-related variables were highlighted as critical factors associated with resilience in adolescence. Positive attitudes toward school (Asgeirsdottir et al., 2010), higher school engagement (Williams & Nelson-Gardell, 2012), and participation in sports and other extracurricular activities (Asgeirsdottir et al., 2010; Guibord et al., 2011) were all identified as factors associated with resilience among adolescents with a history of childhood maltreatment. Self-esteem was another salient factor in maltreated adolescents (Asgeirsdottir et al., 2010; Davidson-Arad & Navaro-Bitton, 2015; Maples et al., 2014).
In studies focusing on adulthood, a positive sense of community including belonging, relating, and mattering was associated with resilience (Bradley et al., 2013; Greenfield & Marks, 2010). Belonging is also important in other areas such as religion, family, and friends with the support provided through those relationships acting as an invaluable factor associated with resilience (Banyard et al., 2017; Goldstein, Faulkner, & Wekerle, 2013). A positive childhood family environment was also associated with resilience in adults with a history of childhood maltreatment (Bradley et al., 2013). In addition, interpersonal strength (Kapoor et al., 2018), positive reframing (Mohr & Rosén, 2017), and having an attitude of acceptance and a positive outlook (Lumley & McArthur, 2016; Mohr & Rosén, 2017) were related to resilience in adulthood following childhood maltreatment.
Discussion
Decades of research have produced a vast body of literature on positive and negative functioning that individuals exhibit following exposure to child maltreatment (Cowell et al., 2015; Keiley, Howe, Dodge, Bates, & Pettit, 2001). In the past 10 years alone, research on resilience has transformed from a relatively niche topic to a flourishing discipline. The fast pace at which resilience research has advanced may have contributed to some of the growing challenges and confusions currently experienced in the field. As evident in this review, the definitional inconsistencies and discrepancies in how the construct is assessed reflect the lack of a common understanding of resilience in the field.
This review focused on resilience-related research conducted since 2010 and indicated that in the span of just 9 years, nearly 70 quantitative studies have been published on resilience following childhood maltreatment. The sheer number of studies completed in such a short span of time highlights a shift in the field from examining problems following maltreatment to assessing strengths in the midst of adversity. While the overall amount of research conducted on resilience is staggering, there were developmental groups that continue to be underrepresented. Specifically, only a handful of studies have targeted resilience in early childhood (i.e., age ≤ 5) even though these children represent the largest age-group experiencing maltreatment (U.S. DHHS, 2018). Further, the work that has been done for this developmental stage was often narrow in scope. Such findings suggest the immediate need for deeper investigation of resilience in early childhood. Further, very little research has included a comparison group or tracked participants over time to gather data on variations in resilience as individuals pass through developmental stages or encounter other life challenges. Thus, while the overall amount of resilience research suggests that this is a topic of great interest, gaps remain that require more focused attention.
With regard to the conceptualization of resilience following childhood maltreatment, this review found that studies do not use common definitional terminology and there is currently no consensus on how to best define resilience. Unlike research on psychopathology that can turn to standardized and agreed upon symptoms to define a disorder, the study of resilience does not have a “diagnostic manual” that clearly shows who meets criteria for resilience. There was no singular emerging conceptualization of resilience in the studies reviewed. Instead, resilience was defined by some researchers as a personal trait, by others as socioecological resource acquisition, and by more than half of researchers as an adaptive outcome. These are vastly different ways of defining resilience, but when studies were divided by developmental stage, some conceptual consistency emerged.
Studies conducted on early childhood consistently conceptualized resilience as outcomes or processes related to adaptive functioning (Dubowitz et al., 2016; Holmes et al., 2018; Sattler & Font, 2018) rather than considering resilience as a static state. This view holds that resilience is a dynamic process affecting how young children reach successful outcomes despite adversity (Cicchetti & Rogosch, 2012). It may be that the enduring and abstract nature of personal traits is inconsistent with how scientists have viewed young children, who are typically seen as relatively changeable during this developmental stage (Shiner & Caspi, 2003).
Similarly, studies focusing on school-aged participants consistently conceptualized resilience as outcomes or processes related to positive and adaptive functioning (Bell et al., 2013, 2015; Cicchetti & Rogosch, 2012; Holmes et al., 2015; Yoon, 2018). Researchers may have primarily conceptualized resilience as successful adaptation and functioning for school-aged children because meeting age-expected milestones across multiple domains of functioning (e.g., social, cognitive, behavioral, social) is a particularly salient task during this developmental stage (Luthar & Cicchetti, 2000).
Notably, over one third of the studies focusing on adolescence conceptualized resilience as socioecological resources that reinforce well-being, whereas none of the studies focusing on early childhood or school-age periods used this conceptualization. It may be that younger children do not have the autonomy or knowledge to take advantage of such resources yet, thus making it unlikely for them to access an array of resources in their environment (Masten & Reed, 2002). With the onset of adolescence, resilience conceptualizations widened. Individuals at this developmental stage begin to form a more solidified identity and manifest greater agency, which fits well with resilience definitions that are more inclusive of personality traits and socioecological resources.
Transitioning into adulthood, resilience was most frequently conceptualized as a personality trait, which highlights a developmental shift toward greater autonomy and a clearer individual identity in adulthood (Erikson, 1994). The perspective that resilience is an internal state and individual response to hardship was most prevalent during adulthood. In sum, our findings suggest that the conceptualization of resilience evolves in a manner that mirrors life span development theory (Erikson, 1994). During earlier life stages, resilience is viewed as external to oneself in which age-specific milestones are attained and, as youth mature into adolescence, resilience becomes less about how one adapts to his/her environment and more about how individual traits emerge. As a person transitions into adulthood, it is expected that they will gain autonomy and prosocial skills, both of which are important for resilience defined as a personality trait and access to resources.
Although developmental variations in the conceptualization and assessment of resilience emerged, there may be other explanations for these results. For instance, it makes sense that resilience at the developmental stage of early childhood would not be assessed using measures of life satisfaction, as infants and children do not have language or capacity to comprehend and report life satisfaction. However, viewing the results through a lens of developmental stage is helpful in terms of understanding how researchers make definitional distinctions and choose to measure resilience. A developmental framework not only informs our understanding of current literature around resilience following maltreatment but also shapes next steps for future study and responsive, age-appropriate interventions to build resilience capacity.
Given the array of resilience conceptualizations evident in this review, it was not surprising to find that the assessments of resilience also showed great variability. While definitions provide understanding of what is meant by the term resilience, operationalization and assessment strategies refer to the ways in which resilience can be measured or evaluated. In this way, it is apparent that conceptual definitions of resilience will directly influence the strategies for measuring resilience as it is defined. For instance, in conceptualizations of resilience focused on positive outcomes despite adversity, assessment strategies designed to capture positive outcomes will be employed. What can be gleaned from this review is that for every manner of conceptualizing resilience, there is a multitude of means for assessing the construct.
No consensus has been reached on how best to measure resilience, thus contributing to the variability in assessment approaches in the existing literature. Indeed, domains measured ranged from language and academic functioning to job performance and life satisfaction. Further, the number of measures used to assess resilience has blossomed in recent years, with dozens of assessment instruments available (e.g., socioecological resources at the individual, relational, community, and cultural levels and general well-being or lack of psychopathology).
From a life span perspective, an individual’s developmental stage actively influences researchers’ decisions regarding how resilience is assessed. Often (especially when the participants are children) studies use measures of psychopathology to capture resilience, such that youth who exhibit normative scores or a lack of clinically significant psychopathology are deemed resilient. Resilience-specific scales are largely developed for, and utilized with, adolescent and adult populations. While the availability of psychometrically sound resilience measures has grown in recent years, these instruments have primarily been created for adolescents and adults, with little attention given to measures that capture resilience in early childhood and the school-age years. The measures for childhood that do exist have not been refined with regard to reliability and validity to the same degree as measures for adults and adolescents, indicating a need for more focused research on scales capturing childhood resilience.
The final phase of this review identified different factors associated with resilience from studies that have examined resilience at various developmental stages. Characteristics of the child’s caregiver, such as warmth, engagement, and well-being, were highlighted as central factors associated with resilience during childhood (Holmes et al., 2015). These findings are congruent with the conceptualization of resilience in early childhood and the school-age years as an adaptive response to the environment. It may suggest that the factors associated with resilience for these two developmental groups are primarily conditions of the child’s external setting rather than internal traits, although it is likely that resilience is shaped by an ongoing interaction between the child and his/her environment at all ages. The factors associated with resilience for adolescents reflect their growing exposure to additional systems (e.g., school) and also the heightened introspective qualities (e.g., self-esteem) developmentally appropriate for this age. For example, school setting and academic aptitude, as well as high self-esteem, become key factors associated with resilience during this period. Finally, factors associated with resilience in adulthood align with stage-salient aspects of this period, during which making meaning of one’s life is dominant (Kroger, 2006). Religion, community connectedness, and social support are central factors associated with resilience for adults (Simpson, 2010). In sum, the series of studies reviewed suggest that some factors (e.g., family support) are protective elements across all developmental stages, but other factors are prominent during unique developmental junctures.
Conclusions
This review utilizes a developmental lens to consolidate recent empirical studies on resilience in youth exposed to child maltreatment. By integrating research on how resilience is defined and assessed, this review suggests that some inconsistencies in the literature may be addressed by utilizing a developmental framework when assessing resilience. Despite ongoing debates of how best to conceptualize and assess resilience, this review has highlighted key themes and central factors that offer directions for advancing the field.
Implications for Practice, Policy, and Research
A variety of implications arise from the findings synthesized in this review. Clearly, additional attention is needed in understanding both the conceptualization and assessment of resilience in young children who have experienced maltreatment. High rates of maltreatment occur during the formative childhood years, so this is a time in which initial evidence of resilience may emerge. Yet available definitions of resilience are limited to adaptive outcomes, and there are not any (resilience-specific) measures tailored for this developmental period. An instrument or approach that takes into consideration unique facets of early childhood, including key factors that may be protective for this stage (e.g., caregiver warmth and emotional support; Holmes et al., 2018; Sattler & Font, 2018), would make a meaningful contribution to practitioners and researchers alike.
The studies reviewed in this article had several limitations, which inform directions for future research. First, only a handful of studies included a comparison or control group of nonmaltreated individuals. Further, many of the articles used cross-sectional data, with no repeated measures of resilience outcomes or assessment of premaltreatment status, potentially introducing confounding effects. Taken together, the lack of control group and absence of longitudinal data in many of the reviewed studies significantly limited our ability to draw any causal inferences or conclusions from this review. Future resilience research may benefit from the use of more rigorous study design such as utilization of a comparison group and longitudinal design.
Another methodological limitation of reviewed studies is that those studies focusing on childhood oftentimes relied exclusively on parent report of youth resilience, raising concerns of unreliable results. Future research should consider using multiple informants, including parents, teachers, and youth, to assess youth resilience. Additionally, the reviewed studies have often failed to include trauma-specific measures when assessing resilience following child maltreatment. Given that maltreated children are at higher risk for developing trauma symptoms, it would be important to develop trauma-specific/trauma-informed resilience measures for this population. Next, while available research has focused on resilience as an outcome and factors that may be associated with more or less resilience, very little research exists on the mechanisms underlying pathways to resilient functioning. Specifically, longitudinal research on moderating and mediating variables that promote resilience would offer novel directions for intervention. Finally, the studies reviewed largely failed to address issues of diversity such as race/ethnicity, socioeconomic status, gender identity, and sexual orientation. Future research should explore how resilience following maltreatment might be associated with aspects of individual differences and how factors associated with resilience may vary across diverse groups.
With regard to practice implications, our findings add support to the fields’ efforts to expand the use of trauma-informed interventions to promote children’s resilience and adaptive functioning following childhood maltreatment and other traumatic experiences. Multiple trauma-informed treatments and interventions such as the Child and Family Traumatic Stress Intervention (Berkowitz, Stover, & Marans, 2011) and Trauma-Focused Cognitive Behavioral Therapy (Cohen & Mannarino, 2008) have been developed and successfully implemented to both reduce trauma symptoms and enhance resilience in maltreated children.
The results of this review further suggest that it is critical to incorporate a developmental perspective into the creation and implementation of any resilience-promotion policy and practice efforts. This review identified developmentally salient factors associated with resilience; some factors have been consistently associated with resilience at certain developmental stages. For example, studies have consistently reported school-related factors (e.g., higher school engagement, positive attitudes toward school, and participation in sports and other extracurricular activities) as well as high self-esteem to be central to resilience in adolescents. Given these findings, resilience promotion interventions for adolescents may benefit from incorporating these elements. Similar strategies may be employed at each developmental stage to maximize treatment benefit by helping clients acquire developmentally salient protective skills that are associated with resilience.
The findings from this review also carry implications for how several existing policies may be adapted and administered. The Child Abuse Prevention and Treatment Act Reauthorization Act of 2010 sanctions the use of federal funds for the development and delivery of maltreatment intervention programs for affected youth (Child Welfare Information Gateway, 2015). Results of this review detailing the strategies for understanding, assessing, and promoting resilience within a developmental context could provide guidance around allocation of these funds to foster the development and implementation of developmentally sensitive intervention services to promote resilience in maltreated children.
Footnotes
Appendix A
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
