Exploring the Link Between Adverse Childhood Experiences and Adolescent Dating Violence Perpetration and Victimization: A Review of the Literature

Abstract

Background and Purpose: Adverse childhood experiences (ACEs) have been linked with multiple adverse outcomes, including adolescent dating violence (ADV). ADV is a serious public health issue, with the U.S. rates ranging from 6.4% to 73% for victimization across all types and from 11% to 77% for perpetration. Given the high prevalence of ADV and the fact that it has been frequently linked with ACEs earlier in life, it is crucial to thoroughly understand this relation and apply this information to prevention and intervention efforts. However, to date, there are no known reviews synthesizing the link between ACEs and ADV, nor exploring which ACEs have been included across other studies. Methods: To address these gaps, this study included three comprehensive searches to identify the following: (1) the measurement tools used to measure ACEs; (2) the type of ACEs significantly linked with ADV; and (3) the risk and protective factors considered in this relation. Results: A total of 34 studies were included. There were 24 different assessment tools used to measure the ACEs of adolescents, with the number of ACEs included ranging from 1 to 93. All but one study linked at least one ACE to ADV. The types of ACEs linked with ADV consisted of victimization (e.g., child abuse and bullying), witnessing violence (i.e., home and school), and household dysfunction (e.g., caregiver substance abuse). Several risk and protective factors were considered. Conclusions/Implications: This study found the measurement of ACEs in the literature to be inconsistent, making it difficult to synthesize and compare the findings across studies. Despite the measurement inconsistencies, however, most studies linked ACEs with ADV, emphasizing the importance of considering this relation in prevention/intervention programming.

1. Introduction

1.1. Adolescent Dating Violence

Threatening, stalking, and aggression (physical, sexual, and/or psychological) between adolescent romantic partners, known as adolescent dating violence (ADV), is a serious public health concern [1]. In the United States alone, the data have revealed lifetime victimization reporting rates ranging from 6.4% to 73%, with a great deal of variability in reporting based on gender and the type of ADV included in each study [2,3]. This same kind of variability is observed with ADV perpetration, with the studies reporting rates ranging from 11% to 77% [4,5]. Given the high prevalence of ADV, there is a large body of literature examining the links between ADV perpetration and victimization and various negative outcomes, including, but not limited to, bullying [6], suicidal ideation [7], substance abuse [8], depression [9], anxiety [10], and intimate partner violence (IPV) in adulthood [11].

1.2. Adverse Childhood Experiences

According to the Center for Disease Control, adverse childhood experiences (ACEs) are events occurring in childhood (0–17 years) that are potentially traumatic [12]. In many earlier studies, ACEs were traditionally operationalized as events, such as child abuse, neglect, and witnessing interparental violence [13,14]. These “traditional” ACEs have been linked with multiple adverse outcomes in the literature, including sexual risk behaviors, drug abuse, IPV in adulthood, and suicide [13,15]. More recently, studies have begun to expand their definition of ACEs to include things like community violence exposure and discrimination [16]. Both of these additional ACEs have been found to be related to many deleterious outcomes as well, including depression [17], posttraumatic stress disorder (PTSD; [18]), and juvenile delinquency [19].

1.3. Theoretical Orientation

ACE Framework. According to the ACE framework, the more ACEs that a child experiences, especially in the absence of adult support, the more likely a child is to experience toxic stress [20]. Toxic stress, defined as the intense, frequent activation of an individual’s stress response system, can result in the disruption of brain development, as well as damage to a person’s mental, behavioral, and physical health [20]. This damage can inhibit an individual’s ability to manage stress in the future, and potentially result in unhealthy lifestyle choices and behaviors used as coping mechanisms for that stress, as well as potential re-victimization similar to initial exposure [21]. Research into the ACE framework has found multiple links between ACEs and numerous high-risk behaviors, often through a pathway involving toxic stress [20,21]. Given that IPV in adulthood has frequently been found to be an outcome of ACEs [15], it stands to reason that ACEs in early childhood could be a contributing factor to the occurrence of ADV as well.

Social Learning Theory. In terms of ACEs such as violence exposure being considered as possible antecedents to ADV, one of the most prominent theories in the literature is social learning theory [22]. The social learning theory suggests that the people around a child (e.g., family, teachers, and community members) are the primary source of socialization and learning in a child’s life, and therefore, the behavior exhibited by those individuals provide a behavioral model that the child then learns as an acceptable way to act [23]. Furthermore, the theory notes that the consequences of those behaviors can alter a child’s desire to model the behavior, so if the child observes more positive consequences of the behavior, then the child will develop positive expectations for their behavior if they act similarly, and the same goes for negative consequences [23]. Therefore, if a child observes positive consequences, such as increased power and control when someone perpetrates IPV, it could potentially influence that child to also want to perpetrate violence when they enter a romantic relationship.

1.4. Existing Gaps

With the ever-growing popularity of the ACE framework in research with adolescents, it is important to understand the scope of the prospective measurement of ACEs with an adolescent population as opposed to retrospective measurement with adults 18 and over, especially if the research findings are to be applied to various prevention and intervention programs with youth. To date, there are no known reviews that examine the measurement tools used to measure ACEs and how they do so. Furthermore, though ACEs have been considered as potential predictors of many outcomes [13,15], to the authors’ knowledge, the literature evaluating the different types of ACE as predictors of ADV has yet to be synthesized in the form of a review. Given that there is already a robust body of literature evaluating the link between ACEs and ADV, and the methods used between the studies evaluating this link vary greatly, it is crucial that this information is clearly synthesized before moving forward with the next steps in the field.

1.5. Current Study

Given the current gaps in the literature, the present study seeks to answer the following research questions:

• RQ1: What are the existing ACE measurement tools that have been used with adolescents under the age of 18?
• RQ2: How do ACEs impact the likelihood of adolescent dating violence?
• RQ3: What mediators, moderators, and protective factors have been found to influence the relationship between ACEs and ADV?

2. Methods

2.1. Search Strategy

Three comprehensive literature searches were conducted in order to identify relevant articles for each of the research questions. The searches were conducted in eight electronic databases (ERIC, Humanities International Complete, Medline, PsychINFO, Psychology and Behavioral Sciences Collection, Social Work Abstracts, SocINDEX, and Social Science Citation Index) for articles published until January 2024.

Search terms. All the search terms can be found in

Table 1. Search terms for each concept.

Search Terms
ADV	ACEs in Aggregate	ACEs by Category
		Victimization	Witnessing Violence	Loss/ Separation	Family/Household Dysfunction
adolescent dating violence OR teen dating violence OR adolescent dating abuse	ACE OR adverse childhood experience OR childhood adversity	victim* OR experience* AND abus* OR violen* OR conflict OR discrimin* OR neglect	expos* OR witness* AND abus* OR violen* OR conflict	los* OR separat*	house* OR home* OR famil* OR parent* AND disrupt* OR dysfunction* OR problem* OR issue

. After librarian consultation, the key search terms used for RQ1 included “(ACE OR adverse childhood experience OR childhood adversity) AND (adolescen* OR youth OR teen)”. For ACEs in aggregate (i.e., summary score), the key search terms used for RQ2 and RQ3 included “(ACE OR adverse childhood experience OR childhood adversity) AND (adolescent dating violence OR teen dating violence OR adolescent dating abuse)”. In order to capture the studies that evaluated the link between individual ACEs and ADV, four categories of individual ACE types were created based on the ACEs included in the measurement tools found for RQ1. This approach was used given the sheer number of ACEs, limiting the authors’ ability to individually search each possible ACE. The four categories and the search terms used for RQ2 and RQ3 for each category were as follows:

(1) Victimization (e.g., child abuse, neglect, discrimination, or bullying);

-

“(victim* or experience*) AND (abus* OR violen* OR conflict OR discrimin* OR neglect) AND (adolescent dating violence OR teen dating violence OR adolescent dating abuse)”.

(2) Witnessing Violence (e.g., school, home, or community violence);

-

“(expos* or witness*) AND (abus* OR violen* OR conflict) AND (child* OR youth OR teen OR adolescen*) AND (adolescent dating violence OR teen dating violence OR adolescent dating abuse)”.

(3) Loss/Separation (e.g., foster care, deportation, or the incarceration of family member);

-

“(los* or separat*) AND (adolescent dating violence OR teen dating violence OR adolescent dating abuse)”.

(4) Family/Household Dysfunction (e.g., substance abuse or mental illness in home, parental unemployment, or the severe illness or disability of parent or child);

-

“(house* OR home* OR famil* OR parent*) AND (disrupt* OR dysfunction* OR problem* OR issue) AND (adolescent dating violence OR teen dating violence OR adolescent dating abuse)”.

2.2. Inclusion and Exclusion Criteria

The inclusion criteria for RQ1 were as follows: (1) the measurement tool measures adverse childhood experiences; (2) the measurement tool are used on adolescents between the ages 10 and 18; and (3) the measurement tools are self-reported or caregiver-reported. The inclusion criteria for RQ2 and RQ3 were: (1) ADV is an outcome; (2) ACEs (in aggregate or individual) are included as independent variable; and (3) the majority of the sample is within the target age range (age 10–18). Articles were excluded regardless of the research question if the full-text articles were not available in English, if they utilized adult retrospective reports of ADV, and/or if they were published in non-peer-reviewed sources (i.e., grey literature).

2.3. Screening and Data Extraction

All citations were transferred into the Mendeley bibliographic software for screening by two screeners. After duplicates and the grey literature were excluded, titles for the remaining citations were reviewed and coded for the inclusion and exclusion criteria. Once all the excluded citations from the title review were removed, the abstracts were reviewed for the remaining citations to determine eligibility, and when necessary, full-text articles were reviewed. Following this second round of eligibility screening, full-text articles were reviewed for all the remaining citations, and a final eligibility decision was made. The data from each of the retained articles were input into an excel spreadsheet for the research question they pertained to.

3. Results

Overall, the three literature searches identified a total of 211,455 citations for screening. After the de-duplication of citations and the removal of the grey literature, 11,444 citations remained. Following the screening and exclusion procedures, 34 studies were included in this review (24 for RQ1 and 10 for RQ2/RQ3).

3.1. ACE Measurement Tools

As noted, the literature search yielded a total of 24 studies, each including a different assessment tool used to measure ACEs with adolescents. ACE measurement was incredibly inconsistent across the studies, with many studies using a study-specific tool to measure them. Using the ACE categories noted above for organization, 21 of the measurement tools assessed victimization, 19 assessed witnessing violence, 16 assessed loss/separation, and 20 assessed family/household dysfunction. The number of ACE-related questions ranged from one [24] to ninety-three [25]. The majority of the measurement tools (n = 13) were self-report surveys, four were self-report interviews [26,27,28,29], five could be used as a self-report survey or interview [16,30,31,32,33], and two could be regarded as a caregiver-report survey or interview [14,34]. The response types varied across measurement tools, with sixteen requiring “yes” or “no” responses, one with “yes” or “no” responses followed by the frequency for the “yes” responses [35], four with the frequency rated on a Likert scale [29,32,36,37], two with a count of the ACEs experienced [38,39], and one of them circling the “worst” ACE experienced [24]. Based on this information, it is clear that the measurement of ACEs varies across studies, with no consistent decision on the number of items, the administration method, which ACEs to include, or the response type.

3.2. ACEs and ADV

ADV Measurement. Of the ten studies included that examined the link between ACEs and ADV, five of the studies measured ADV using the Conflict in Adolescent Dating Relationships Inventory (CADRI; [40]), with three measuring both perpetration and victimization [41,42,43] and two measuring just perpetration [44,45]. All of the studies that utilized the CADRI measured five types of ADV (sexual, physical, relational, verbal, and threatening). Four of the studies utilized varying versions of the Conflict Tactics Scale (CTS; [14]) to measure ADV [46,47,48,49]. Black et al. [46] and Tschann et al. [49] both measured physical and psychological perpetration and victimization, Lavoie et al. [47] measured physical and psychological perpetration, and Stocker and Richmond [48] measured psychological perpetration and victimization. The remaining study utilized a combination of the Safe Date Psychological Abuse Victimization Survey [50], the National Intimate Partner and Sexual Violence Survey [51], and the Sexual Experiences Survey [52] to measure emotional, physical, and sexual ADV perpetration and victimization [53]. All but two of the studies were longitudinal [46,53].

Study Findings. Only one of the included studies looked at ACEs in aggregate, and that study found the ACEs overall to be a significant predictor of ADV, but did not identify the type of ACE [45]. Of the nine studies that examined the link between the different types of ACEs and ADV, all but one (n = 8) found at least one ACE to be significantly related to ADV perpetration and/or victimization [43]. Six of the studies found ACEs in the victimization category (i.e., child abuse, neglect, discrimination, or bullying) to be significantly related to some form of ADV [41,44,46,47,48,53], six linked ACEs in the witnessing violence category (i.e., school, home, or community violence) with at least one form of ADV [41,42,44,46,48,49], and one study found ACEs in the family/household dysfunction category (i.e., substance abuse or mental illness in home, parental unemployment, or he severe illness or disability of a parent or a child) to be related to some form of ADV [42]. No studies examined the link between ACEs in the loss/separation category (i.e., foster care, deportation, or the incarceration of family member) and any type of ADV.

Mediators, Moderators, and Protective Factors. Of the ten studies, three included mediators in their analysis [42,47,49], with two of the studies reporting significant results. Livingston et al. [42] found that maternal warmth, self-regulation, externalizing behaviors, and problems with siblings significantly mediated in the relation between witnessing IPV and ADV, as well as between parental mental health issues and ADV. Tschann et al. [49] found that emotional distress and the appraisal of IPV mediated the link between witnessing non-violent IPV and ADV.

Two studies included moderators in their analysis, with Tschann et al. [49] finding gender and ethnicity to significantly moderate in the relation between witnessing IPV and ADV and Stover et al. [43] finding secure attachment and avoidant attachment styles to be significant moderators in the link between child maltreatment and ADV. Finally, Davis et al. [45] was the only study to examine the protective factors and found that social support, parental monitoring, school belonging, and academic achievement were all significant protective factors in the relationship between ACEs and ADV. However, though not assessed as protective factors, some of the mediators and moderators included in the studies already noted (i.e., maternal warmth, self-regulation, and secure attachment) could also be considered protective factors [42,43].

4. Discussion

Overall, the findings from this study make two things clear. First, the measurement of both the ACEs and ADV is inconsistent across the studies, with twenty-four ACE measurement tools for use with adolescents and five ADV measurement tools used between the 10 included studies. This inconsistency makes the findings difficult to synthesize and compare across the studies, as they could be measuring the variables in completely different ways. For example, one study could be measuring ADV as physical perpetration only, while another study may be measuring ADV as sexual perpetration and victimization combined. To make the data in the field more comparable, it is advised that a standard is determined for the measurement of ADV, so that future research can be more easily synthesized and digested. In this field specifically, if attempting to elucidate which types of ADV are predicted by the many different ACEs, perhaps for the purpose of prevention or intervention efforts, it is recommended to utilize an assessment tool that considers all the forms of ADV (threatening, stalking, physical, psychological, and sexual). If participant burden is a concern due to the number of questions in many of the existing assessment tools, using a short form that reduces the quantity of questions, but still assesses for each type is recommended. However, this can reduce the reliability and validity of the measurement of those constructs, unless using a validated short form. In terms of ACE measurement, it is difficult to suggest a standard tool, as new ACEs are frequently being added as the definition of an ACE shifts. Though this can complicate synthesis across studies, trauma is a subjective concept; therefore, the inclusion of several potential ACEs allows for everyone to report on their own experience of adversity, rather than fitting within a box of what trauma “typically” looks like. However, it is recommended that the tools used to measure ACEs utilize a similar response type across studies so the results can be compared, especially when the same ACEs are used (e.g., all frequency or all “yes” or “no” responses). For instance, given that many of the existing ACE questionnaires implement a count of the ACEs experienced, one recommendation could be that even if another type of score is calculated for a particular study (e.g., frequency or severity), a count of the ACEs experienced could also be included so that data across studies can be synthesized.

Second, despite the inconsistency in measurement, the majority of the included studies found a link between at least one type of ACE and the occurrence of ADV. However, with most of the studies evaluating the types of ACEs typically considered “traditional” (child abuse, neglect, and witnessing IPV), more research is warranted, examining additional forms of ACEs, such as those included in the loss/separation and family/household dysfunction categories (e.g., the death of a parent, deportation, parental substance abuse, etc.) before any type of statement can be made about the overall link between ACEs and ADV. Furthermore, with only one study evaluating the link between ACEs in aggregate and ADV, it is difficult to interpret the significant findings of this study as they compare to the findings from the other studies examining individual types of ACE. More research is needed to determine if the total number of ACEs has a significant impact on ADV, as it does on several other deleterious outcomes [21], above and beyond the impact of each selected ACE.

This review is not without limitations. First, given the nature of review studies, it is possible that some articles were missed in the search. Additionally, given that the ACEs were searched by ACE category rather than by each individual type, both for feasibility purposes, but also to capture all the potential ACEs included within those classifications to date, it is possible that additional articles were missed that looked at a select few variables and did not refer to them as ACEs. Finally, given the sweeping scope of this review, in-depth detail about how each ACE impacts ADV, as well as how the protective and risk factors included affected the overall information, was not included. Future research should consider breaking down these areas individually in more targeted reviews.